Pacificorp v. Northwest Pipeline GP (879 F. Supp. 2d 1171, 2012)

OPINION AND ORDER

PAPAK, District Judge:

Plaintiff PacifiCorp brings this action alleging state law claims for negligence, negligence/res ipsa loquitor, trespass, and nuisance against Northwest Pipeline (“Northwest”) and negligence /res ipsa loquitor and breach of contract against Gas Transmission Northwest (“GTN”), arising from shut-downs of the natural gas power generation plant in Hermiston, Oregon purportedly caused by compressor oil in the gas supply that contaminated the generator turbines. Now before the court are motions for summary judgment by PacifiCorp, GTN, and Northwest, (# 253, # 251, # 243), motions for sanctions based on spoliation filed by GTN and Northwest, (# 284, # 279), GTN’s motion to stay proceedings and refer issues to the Federal Energy Regulatory Commission (“FERC”) (#240), and PacifiCorp’s motions to compel against GTN and Northwest (# 196, # 194). For the reasons described below, Northwest’s motion for summary judgment is granted, GTN’s motion for summary judgment is granted in part and denied in part, PacifiCorp’s motion for summary judgment is granted in part and denied in part, Northwest’s spoliation motion is denied as moot, GTN’s spoliation motion is granted in part and denied in part, GTN’s motion to stay is granted, PacifiCorp’s motion to compel evidence from Northwest is denied as moot while PacifiCorp’s motion to compel evidence from GTN is granted in part and denied in part.

BACKGROUND

I. The Parties

Plaintiff PacifiCorp is an Oregon utility company that sells electricity to retail customers. (First Amend. Compl., # 61, ¶ 2.) PacifiCorp and Hermiston Generating Company, L.P. (“HGC”) each own a 50% undivided interest in the Hermiston Station, a natural gas power generation plant in Hermiston, Oregon. Id. at ¶ 5. Under a Long-Term Power Sales Agreement and an Ownership and Operating Agreement, PacifiCorp is entitled to purchase the electric power generated by the Hermiston Station from HGC. Id. The Hermiston plant consists of two General Electric (“GE”) 7FA gas turbines, referred to as Unit 1 and Unit 2, that generate power by combusting natural gas. PacifiCorp is responsible for providing all fuel for the plant. (Skelton Deck, # 248, Ex. 1) (Tenney Dep., at 153:20-154:12.)

In general, gas combusted by the plant originates in Canada, travels through pipelines owned and operated by defendant GTN, flows into a seven-mile-long local pipeline system operated by Cascade Natural Gas Corporation (“Cascade”) at the Hermiston Meter Station, and arrives at the Hermiston plant. GTN provides gas to the Hermiston Meter Station through two parallel pipelines, the “A” and “B” lines. GTN’s predecessor entered into a transportation agreement with HGC (the “FT agreement”) providing for the transport of natural gas from Canada to the Hermiston Meter Station. (Roscher Aff., # 23, Ex. 1.) The FT agreement incorporated the terms and conditions of GTN’s FERC gas tariff. Id., Ex. 1 at ¶ 1.2. The tariff sets certain quality standards for the gas that GTN transports on behalf of the “shipper” of the gas. Id., Ex. 2, at ¶ 3.1. In general, the tariff requires the gas to be “merchantable.” Id. The tariff also requires that the gas: “Shall be commercially free from sand, dust, gums, crude oil, impurities and other objectionable substances which may be injurious to pipelines or which may interfere with its transmission through pipelines or its commercial utilization.” Id. at ¶ 3.1(b)(1). In addition to these general requirements, the tariff sets more specific gas quality standards, such as for hydrocarbon dew-point, temperature, hydrogen sulfide, sulpher, carbon dioxide, water vapor, and oxygen. Id. at ¶ 3.1(b).

Defendant Northwest Pipeline also operates a pipeline transporting natural gas from Canada. Northwest’s pipeline interconnects with GTN’s “A” line at the Stan-field Interconnect, approximately 12 miles upstream from the Hermiston plant. Northwest can both deliver gas to GTN and receive gas from GTN at Stanfield. Northwest typically provides gas to GTN in the late spring through early fall and receives gas from GTN during the rest of the year. Northwest does not have a contract to provide fuel for HGC or PacifiCorp. However, Northwest and GTN have an Interconnect & Operating Agreement governing gas quality at Stanfield. (Richards Deck, # 246, Ex. 4.) That agreement provides that the gas quality tariff provision of the party receiving gas governs the quality of gas delivered at Stan-field — when GTN receives gas from Northwest, GTN’s tariff applies, and when Northwest receives gas from GTN, Northwest’s tariff applies. Id. at 8-9, 23-24.

II. Compressors and Compressor Oil

Because the Northwest pipeline operates at a lower pressure than the GTN pipeline, GTN must compress gas that it receives from Northwest at Stanfield using a wet seal compressor. (Sawyer Deck, #265, Ex. 4, at 1). Northwest also employs reciprocating gas compressors on its pipeline, and some wet seal compressors not far from Stanfield. Id. Both the wet seal compressor and reciprocating compressors require oil for either lubrication or sealing. (Sawyer Deck, #265, Ex. 5) (Zaitz Dep., at 117-118). By design, a small amount of compressor oil enters the gas stream such that compressor oil is known to be a normally occurring substance in natural gas pipelines. Id. (Zaitz Dep., at 101,118,141).

III. Gas Conditioning

Before arriving at the Hermiston plant, gas passes through several sets of gas conditioning equipment to remove liquids from the gas stream. GTN operates a scrubber at the Stanfield metering station within the Stanfield Interconnect to removed liquids. GTN also operates a gas heater and particulate filter at the Hermiston metering station. (Skelton Decl., #248, Ex. 7) (Plaster Dep., at 57:5-19). In addition, the gas passes through Cascade’s particulate filter before reaching the plant. Upon entering the plant, the gas passes through two fuel gas scrubbers, located upstream of each turbine. (Skelton Decl., #248, Ex. 11, at 3) (Black & Veatch Report.)

IV. Oil and Fuel Nozzle Problems Before 2007

Hermiston plant manager Frank Glasgow states that, from the plant’s opening in 1996 until January 2007, “the Plant had operated without any problems attributable to equipment associated with a fuel nozzle, and it had never experienced a forced outage attributable to gas quality.” (Glasgow Deck, # 325, ¶ 7.) However, Hermiston plant employees raised concerns to GTN about the presence of oil in the gas supply as early as 2001, but GTN reported no presence of oil in the white paper filters at the Hermiston meter station. (Skelton Deck, # 248, Ex. 24) (GTN submissions to FERC). Moreover, GE maintenance records starting in 2001 document a number of fuel nozzle problems. In 2001, GE inspections revealed burned fuel nozzle tips. (Skelton Deck, #248, Ex. 25.) Similarly, an inspection in 2003 found “a couple of burned fuel nozzle tips.” (Walker Deck, #305, Ex. 24, at 7.) Then, in 2005, GE found fuel nozzle assemblies to be in “poor condition” due to gas tip diffusion, erosion, end cover insert leakage, and end cover heavy hydrocarbon contamination. (Skelton Opp. Deck, # 310, Ex. 1, at 1.)

V. Claimed Outages

A. January 2007 Outage and Ensuing Investigation

From January 25, 2007 to January 28, 2007, the Hermiston plant experienced the first alleged forced outage of Unit 2. (Glasgow Deck, # 325, ¶¶ 19-20.) The plant initially detected high carbon monoxide emission levels. Id. at ¶25. When the plant could not reduce the levels through tuning, the plant shut down Unit 2 on January 26, 2007. Id. at Ex. 14, at 5 (GE Inspection Report). GE staff removed the Unit 2 burner assemblies, installed a spare set of assemblies that the plant had on hand, and restarted Unit 2 on January 28, 2007. Id. In late February 2007, Glasgow watched while GE conducted flow testing and disassembled the Unit 2 fuel nozzle assembly at GE’s inspection facility. Id. at ¶¶ 25-26. Glasgow observed what he described as “coking” on the inner fuel nozzles. Id. at ¶26. Two photographs were taken of a burner and burner tube, representing the “average condition” of the inner fuel nozzles after the January outage, two photographs were taken of the inside of the fuel nozzle orifices, and a series of photographs were taken while a GE engineer used a horoscope to inspect the annulus ring within the fuel nozzle assembly. Id. at ¶¶ 26, 28, 29, Ex. 1, Ex. 3, Ex. 5. Given the “coking” downstream from the annulus and the dark substance Glasgow observed in the annulus ring, he concluded that stains visible in the annulus were likely dried compressor oik Id. at ¶ 29. GE, however, did not mention compressor oil in its inspection report. Instead, GE noted the horoscope inspection found debris “resembling masking tape particles” in two end covers and concluded simply that “it is likely that contamination restricted the flow of fuel to the combustion chamber, causing abnormal combustion.” Id. at Ex. 14, at 9.

On March 14, 2007, HGC apparently convened a meeting with a GTN engineer, Doug Zaitz, and employees of Cascade and the Hermiston plant to discuss concerns about gas contamination. (Sawyer Decl., # 265, Ex. 5, at 13) (Zaitz Dep.) Representatives of the Hermiston plant told Zaitz that over the past few years they noticed that up to a few quarts of oil per year had collected in the separator filters at the Hermiston plant. (Sawyer Decl., #265, Ex. 20, at 3-5.) This caused Zaitz to write that GTN was concerned that “oil may be passing through [GTN’s] metering facility undetected and unencumbered.” Id. Zaitz communicated with the company that sold the plant’s filters, inquiring about ways to protect the plant’s turbines from contamination, including installing coalescing filter elements. Id. at 4. On April 10, 2007, Frank Glasgow reported to Zaitz that the filter socks at the Cascade pipeline were “saturated” and the scrubber at the plant contained “about 1 [gallon]” of oil. " Id. at 2. About two weeks later, on April 25, 2007, Zaitz wrote to thé filter sales representative that GTN was concerned that “old compressor lube oil ... may be the source of the oil that is appearing [in] the cyclone filter separators” at the Hermiston Plant. Id., Ex. 21, at 1. Zaitz noted that GTN had seal oil systems on its compressors until the early 1990s, but also stated that GTN’s filters at the Hermiston station were dry of liquids. Id. Zaitz later testified, however, that he lacked any independent evidence about the nature of the liquid, as GTN had not collected any of the liquid at that time. (Sawyer Deck, # 265, Ex. 5, at 16) (Zaitz Dep.)

In May 2007, Zaitz continued assisting the Hermiston plant in GTN’s investigation, searching for a lab that might be able to analyze the “mystery fluids” found in the plant filters. (Sawyer Deck, #265, Ex. 4, at 3-8.) Ultimately, Zaitz located Core Laboratories in Houston, Texas, which could perform the appropriate analysis. Id. at 3. Also in May 2007, Zaitz explained to a GTN supervisor that “lost” compressor seal oil was the only source of oil in GTN’s pipeline system, although GTN’s only remaining wet seal compressor was losing little, if any, oik (Sawyer Deck, # 265, Ex. 15, at 2.) Zaitz also raised the possibility that compressor oil was coming from Northwest Pipeline gas transported in GTN’s pipeline. Id. During this time period, Frank Glasgow perceived that GTN was working with the plant to prevent increased amounts of compressor oil from reaching the turbines. (Glasgow Deck, # 325, ¶ 59.)

Also in May 2007, the plant shut down Unit 2 for regular maintenance, photographing what Glasgow describes as “coking” below the inner fuel seal rings and in the fuel nozzle flow orifices. (Glasgow Deck, # 325, ¶ 31, Ex. 7.) GE visually inspected the fuel nozzles and observed “[s]treaks of carbon/oxides,” “grossly deteriorated” gas tip packing seals, and “[h]eavy [h]ydrocarbon [c]ontamination.” (Glasgow Deck, #325, Ex. 8.) GE also collected a sample of about one-half cup of “carbon and packing seal residue.” Id.

B. July 2007 Outage and Further Investigation

On July 30, 2007, the Unit 2 turbine “tripped” due to a high exhaust temperature spread and the plant discovered a “burn through” on one of the burner nozzles. (Glasgow Deck, # 325, Ex. 15, at 5) (GE Inspection Report). The plant replaced the burner nozzle and several other parts, but shut down the unit the next day after observing other problems. Id. GE recommended sending the fuel nozzle assemblies to its shop in Jacksonville, Florida for repair. Id. at 8. GE installed a set of refurbished nozzles and Unit 2 resumed operation on August 2, 2007. (Glasgow Deck, #325, Ex. 15 at 5.) A few weeks later, on August 15, 2007, GTN closed the valve on its A pipeline at the Hermiston metering station, preventing any gas from the Northwest pipeline from being delivered to the Hermiston plant, to help determine the source of the liquids reportedly arriving at the Hermiston plant. (Skelton Deck, # 248, Ex. 41). The next day, Glasgow took samples of the gas at Stanfield, the first of three such samples in August and October 2007. (Skelton Deck, # 248, Exs. 41, 42, 43.) He reported seeing “oil vapor” on the rocks around the sampling area and in the gas stream. Id., Ex. 41. Glasgow sampled the gas a second time on August 23, 2007, this time employing a consultant, McHale & Associates, to take samples of gas and liquid from several locations on the Cascade and GTN pipelines and within the Hermiston plant. Id., Ex. 43. The third sampling occurred on October 5, 2007. Id., Ex. 42.

The samples were analyzed by Texas Oil Tech and found to contain a “lubrication or compressor range oik” (Sawyer Deck, #265, Ex. 25.) Texas Oil Tech did not attempt to quantify the amount of the substance in the August gas samples, but reported that the October sample contained 44 parts per million of residue. Id. at 8. Glasgow, however, admitted that the samples had been collected from the bottom and sides of the pipelines to increase the likelihood that liquids would be captured. (Skelton Deck, #248, Ex. 40, 10-11) (Glasgow Dep.); (Skelton Deck, # 248, Ex. 45 at 2). After Texas Oil Tech analyzed the samples, they were destroyed.

Sometime before August 27, 2007, GTN apparently sampled liquids at the Hermiston plant and Stanfield scrubbers, and sent them to Core Laboratories for analysis. Id., Exs. 48, 49. The Core Laboratories testing eventually showed that the distillation results for the two samples were “not similar” and the liquids were “probably not” from the same source. Id., Ex. 49. Core Laboratories also stated: “Both samples cover the correct range of distillation for compressor oils. The Hermiston scrubber oil appears to have a much narrower distillation range than is normally seen and could possible [sic] be a synthetic oik” Id.

On August 27, 2007, Zaitz wrote to various employees of GTN and the Hermiston plant that once the gas and liquid samples had been analyzed and the source of the liquid identified, he would be interested in discussing “long term” solutions to remove the liquid from the gas stream at various locations, depending on its source. (Sawyer Deck, # 265, Ex. 14.)

C. August 2007 Outage

On August 29, 2007, Hermiston again shut down the Unit 2 gas turbine. (Glasgow Deck, # 325, ¶ 22, Ex. 16 at 5.) GE technicians arrived several days later and disassembled the nozzles, observing “moderate to heavy carbon deposits and burnt tips” on most of the old nozzles. Id., Ex. 16 at 5, 9. Glasgow also viewed the disassembled nozzles and saw “coking,” which he documented with several photographs. (Glasgow Deck, # 325, ¶ 30, Ex. 6.) Unit 2 resumed service on September 10, 2077. Id., at ¶ 22, Ex. 16 at 5.

A few days prior, at the request of the Hermiston plant, Cascade “pigged” its pipeline, a process where foam gauges are run inside the pipeline. (Skelton Deck, #248, Ex. 10) (Cooper Dep., at 100:7-101:21). One participant stated that only a “puff of dust” and a piece of grease about the size of a finger came out the end of the pipe, and that no free oil was discovered. Id. But Glasgow, who was also present, states that at least 40 pigs removed from the Cascade pipeline were discolored and covered with oil, as documented in photographs he took. (Glasgow Deck, #257, ¶ 12, Ex. 5.)

D. September 2007 Outage

Unit 2 suffered another outage from September 24, 2007 to October 10, 2007. (Glasgow Deck, # 325, ¶ 23.) GE again inspected the fuel nozzle assemblies, finding them in various pre-stages of failure, and noting a “strong possibility of internal diffusion seal deterioration and coking due to liquid hydrocarbon contamination.” (Glasgow Deck, # 325, Ex. 17, at 42.) GE’s report concluded that Hermiston “has experienced fuel gas contamination from their supply system.” Id., Ex. 17 at 4. However, the author of the report testified only that the root cause of the outages “appeared to be” oil contamination, based on “reports” in September or October 2007 of “oil found in the gas piping itself.” (Sawyer Deck, #265, Ex. 40) (Crafts Dep., at 120-121).

Indeed, throughout 2007, Glasgow and other Hermiston plant employees reported seeing oil inside the plant. (Glasgow Deck, #257, ¶¶4-8, 10); (Journot Deck, # 256, ¶¶ 3-5.) At various times between January and July 2007, plant employee Terry Journot reported seeing about a gallon of oil pour out of the filter where the plant receives gas from Cascade. (Journot Deck, # 256, ¶ 4.) Journot also observed oil in the plant piping and turbine equipment for Unit 2 later in 2007 during a cleaning of the gas lines. Id. at ¶ 5. Also during 2007, Frank Glasgow observed what he deemed to be oil at the gas metering orifice, the “witch’s hat” filter, and the pigtails, all of which are downstream of the Unit 2 scrubber but upstream of the Unit 2 turbine. (Glasgow Deck, # 257, ¶¶ 4-7.)

E. FERC Hotline Call and Investigation

On September 28, 2007, PacifiCorp initiated a FERC hotline complaint concerning contamination of gas received at Hermiston station. (Erb Deck, #44, ¶2.) During that call, a FERC attorney suggested PacifiCorp notify GTN and Northwest Pipeline that they were the suspected source of the contamination. Id. at ¶ 3. In February 2008, PacifiCorp withdrew its hotline complaint. Nevertheless, FERC continued its investigation and, in July 2008, FERC determined that prior gas sampling conducted by GTN and PacifiCorp was unreliable and set out a new protocol for collecting three different samples from GTN’s lines A and B. Those tests occurred in November 2008, December 2008, and January 2009. The testing revealed no significant hydrocarbons or residues in the samples. Based on these tests, FERC informed PacifiCorp that its inquiry would be concluded. In March 2010, FERC confirmed by letter that it would take no further action regarding PaeifiCorp’s allegations that GTN’s gas failed to meet tariff quality standards, based on FERC staffs consideration of facts submitted and obtained during its investigation. (Aff. in Supp., # 24, Ex. A). That letter made clear that the FERC staff’s conclusions were not final or binding on FERC as a whole. Id.

F. Scrubber Examination

In early October 2007, after GE employee Sally Perkins hypothesized that the Unit 2 scrubber was malfunctioning, Glasgow visually examined the Unit 2 scrubber and found it to be working properly, removing small amounts of liquid from the gas supply. (Glasgow Deck, # 325, at ¶¶ 49-54.) Indeed, records from that scrubber show that it consistently removed only a small amount of liquid, totaling about .2 gallons in all of 2007.

G. Mitigation Outages

After the forced outages, the plant instituted what it has described as mitigation measures, some of which required interrupting the gas supply and shutting down power generation. During the aftermath of the August 2007 outage of the Unit 2 generator, from September 8 to September 10, 2007, the plant cleaned its gas line, which also required shutting down the Unit 1 turbine. (Glasgow Deck, #325, ¶ 74.) Another cleaning occurred during the September 2007 outage, from October 5 to October 8, 2007, which again shut down generation on the other turbine. Id. at ¶¶ 75-76. During that second cleaning, the plant installed a temporary coalescing filter in the gas yard at the recommendation of its consultant McHale & Associates. Id. at ¶ 65. The plant subsequently ceased operations twice to check on the functioning of that coalescing filter, on October 16 to October 17, 2007 and on November 10, 2007. Id. at ¶ 69. Then, on March 8, 2008, while Unit 1 was off-line for scheduled maintenance, the plant also interrupted Unit 2 generation to replace its original gas scrubbers with a permanent coalescing filter in the main gas yard. Id. at ¶ 70. The plant also apparently installed a permanent coalescing filter specifically for Unit 2 several months later, an additional precaution recommended by consultant Scientific Process Solutions as early as October 2007. (Skelton Deck, #248, Ex. 50.)

H. 2008 Exponent Report

PacifiCorp subsequently retained Exponent Failure Analysis (“Exponent”) to investigate the root causes of the outages in January, July, August, and September 2007, as well as a Unit 1 shutdown in December 2007 for which damages are not sought in this suit. (Skelton Deck, # 248, Ex. 14.) Exponent considered plant information, reports by PacifiCorp’s consultants, and turbine operational data. Id. Also, at Exponent’s request, the plant provided Exponent with a subset of damaged fuel nozzles from the July 2007 outage for laboratory inspection. Id. In its August 15, 2008 report, Exponent opined that the fuel nozzles installed at the plant were “susceptible to thermal failure” due to their lack of fairings and their propensity to trap liquid hydrocarbons, which, in the presence of minor temperature increases, could form coke and cause an internal failure of the nozzles. Id. at viii. Exponent also found that the tuning of Unit 2 differed from Unit 1, making it “less robust” in preventing intermittent flame flashback, an effect that can raise the temperature of the fuel nozzles sufficiently to transform trapped hydrocarbons into coke. Id. at ix-x. In sum, Exponent concluded that two concurrent conditions likely caused the Unit 2 failures: (1) intermittent flashback due to lack of fairings and differences in tuning; and (2) the presence of trapped liquid hydrocarbons in the fuel nozzles. Id. at x.

LEGAL STANDARDS

Summary judgment is appropriate “if the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law.” Fed.R.Civ.P. 56(c). Summary judgment is not proper if material factual issues exist for trial. See, e.g., Celotex Corp. v. Catrett, 477 U.S. 317, 322, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986); Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986); Warren v. City of Carlsbad, 58 F.3d 439, 441 (9th Cir.1995), cert. denied, 516 U.S. 1171, 116 S.Ct. 1261, 134 L.Ed.2d 209 (1996). In evaluating a motion for summary judgment, the district courts of the United States must draw all reasonable inferences in favor of the nonmoving party, and may neither make credibility determinations nor perform any weighing of the evidence. See, e.g., Lytle v. Household Mfg., Inc., 494 U.S. 545, 554-55, 110 S.Ct. 1331, 108 L.Ed.2d 504 (1990); Reeves v. Sanderson Plumbing Products, Inc., 530 U.S. 133, 150, 120 S.Ct. 2097, 147 L.Ed.2d 105 (2000). On cross-motions for summary judgment, the court must consider each motion separately to determine whether either party has met its burden with the facts construed in the light most favorable to the other. See Fed.R.Civ.P. 56; see also, e.g., Fair Hous. Council v. Riverside Two, 249 F.3d 1132, 1136 (9th Cir.2001).

DISCUSSION

1. Overview

As explained further below, I ultimately conclude that all claims against Northwest (negligence, neghgence/res ipsa loquitor, nuisance, and trespass) must be dismissed, and that the negligence/res ipsa loquitor claim against GTN must also be dismissed, leaving only PacifiCorp’s breach of contract claim against GTN as potentially viable at trial. Moreover, as sanctions for PacifiCorp’s spoliation of fuel nozzle assemblies and other turbine parts after litigation was reasonably foreseeable, PacifiCorp may not introduce expert testimony from Dr. Kemal about the cause of the August and September 2007 outages. PacifiCorp is also subject to an adverse inference jury instruction concerning the numerous fuel nozzles that were not preserved from the July 2007 outage. Given that expert testimony of causation is required for this type of complex mechanical failure, the scope of PacifiCorp’s potential damages for forced outages is narrowed to the January and July 2007 outages. Further, PacifiCorp fails to produce expert testimony to establish that oil contamination caused damage to the combustion liner replaced after the January 2007 outage. PacifiCorp, however, need not introduce expert causation evidence for the mitigation outages and damages based on its mitigation efforts are therefore still available. PacifiCorp’s request for prejudgment interest is not properly pled. Additionally, the doctrine of primary jurisdiction requires this court to stay proceedings pending a referral to FERC, which must construe the tariff language that forms the basis of GTN’s disputed contractual duty. In the meantime, GTN may conduct independent examination and testing of the fuel nozzles in Dr. Kemal’s possession, and PacifiCorp’s is entitled to receive some of the discovery sought in its motions to compel, namely, all communications considered by GTN’s non-reporting expert witnesses.

II. Northwest and GTN Motions for Summary Judgment — Availability of Negligence Claims

A good deal of the parties’ briefing involves the question of whether PacifiCorp may even assert negligence claims against Northwest and GTN. These claims attempt to enforce common law obligations that PacifiCorp asserts are either identical to or “consistent with” the gas quality provisions set out in GTN and Northwest’s FERC tariffs. In general, because these specific FERC tariff gas quality provisions supplant any common law duties concerning gas quality, the negligence-based claims against Northwest and GTN both fail.

A. Northwest

Fazzolari v. Portland School Dist. No. 1J, 303 Or. 1, 734 P.2d 1326 (1987) and its progeny hold that the availability of a negligence claim turns on a two-step analysis. First, court consider whether there is evidence of a special relationship between the plaintiff and the defendant due to “a status, a relationship, or a particular standard of conduct that creates, defines or limits the defendant’s duty.” Buchler v. State By & Through Or. Corr. Div., 316 Or. 499, 504, 853 P.2d 798 (1993) (en banc) (quoting Fazzolari, 303 Or. at 19, 734 P.2d 1326). Such a special relationship defining the defendant’s standard of care can also be found in “a statute or an ordinance.” Little v. Wimmer, 303 Or. 580, 584, 739 P.2d 564 (1987) (en banc). If a special relationship exists, the court analyzes defendant’s duty under that standard. If there is no special relationship, the court proceeds to the second step, analyzing whether defendant’s conduct “unreasonably created a foreseeable risk to a protected interest of the kind of harm that befell the plaintiff.” Buchler, 316 Or. at 504, 509, 853 P.2d 798 (quoting Fazzolari, 303 Or. at 17, 734 P.2d 1326.)

Here, PacifiCorp seems to identify two separate sources of Northwest’s tort liability. At times, PacifiCorp argues that Northwest owed a duty to PacifiCorp not to deliver gas containing compressor oil based on Northwest’s own tariff requirement to deliver gas “free of water and hydrocarbons in liquid form at the temperature and pressure at which the gas is delivered” and on Northwest’s contractual obligation with GTN to adhere to the “merchantable” and “commercially free” gas quality provisions in GTN’s tariff when transporting gas into GTN’s pipeline. (Br. in Opp., #314, at 22, 23.) Elsewhere, PacifiCorp contends that Northwest’s delivery of gas containing any compressor oil at the Stanfield Exchange created a foreseeable risk of damage to downstream users such as PacifiCorp. Id. at 24-25. Indeed, at oral argument, PacifiCorp tacitly admitted that relevant tariffs delineate the defendants’ standard of care concerning gas quality, confirming that both Northwest and GTN owed a standard of care “consistent with” their respective gas quality tariffs.

In the first step of the two-step Fazzolari methodology, I conclude that Northwest and GTN’s gas quality tariff provisions set forth precisely the type of “particular standard of conduct that creates, defines or limits the defendant’s duty” obviating resort to a foreseeability analysis. The parties do no dispute that FERC tariffs are effectively federal regulations. (Br. in Opp., # 314, at 23, n. 15); see also California ex rel. Lockyer v. Dynegy, Inc., 375 F.3d 831, 839 (9th Cir.2004) (filed FERC tariffs are the equivalent of a federal regulation). Thus, the FERC tariffs can form the basis for a particular standard of conduct limiting Northwest’s duty. Northwest’s tariff limits its duty concerning gas quality by explicitly narrowing the scope of its obligations to gas delivered to a “Shipper” at a “Delivery Point.” Richards Decl., # 246, Ex. 3, ¶¶ 3.1, 3.2. While Northwest’s tariff gives rise to a duty not to deliver contaminated gas to GTN (a Shipper) at the Stanfield exchange (a Delivery Point), it does not also obligate Northwest to a gas quality standard for gas received by PacifiCorp (not a Shipper) at the Hermiston station (not a Delivery Point). The same analysis also obtains under Northwest’s contractual obligation to satisfy the gas quality provisions of GTN’s tariff when Northwest supplies gas to GTN at the Stanfield exchange. Thus, assuming Northwest delivered contaminated gas to GTN which eventually reached the Hermiston plant, such conduct might give rise to a cause of action by GTN against Northwest, but it would not violate any tariff-created standard of care Northwest owed to PacifiCorp. In sum, I reject PacifiCorp’s attempt to manufacture a negligence claim against Northwest by expanding the scope of FERC tariffs to create a generalized duty of care to all foreseeable downstream users, in contradiction to the tariffs’ express terms.

Although not necessary to my conclusion, I also note that PacifiCorp’s approach also appears to violate a FERC policy statement and subsequent FERC authority holding that upstream gas transporters are only responsible for ensuring gas quality at their tariff delivery points to their direct customers. See, e.g., ANR Pipeline Co., 116 FERC ¶ 61,002, ¶ 61,014 (Jan. 3, 2006); Dominion Cove Point, LNG, LP Dominion Transmission, Inc. 126 FERC ¶ 61,036, ¶ 61,211 (Jan. 15, 2009). For much the same reasons, PacifiCorp’s negligence claim against Northwest also likely violates the filed rate doctrine, which prohibits challenging the tariff rate outside of FERC proceedings, attempted here through PacifiCorp’s effort to impose an obligation to ensure gas quality as to non-Shippers, a requirement not present in Northwest’s FERC tariff. See Brown v. MCI WorldCom Network Services, Inc., 277 F.3d 1166, 1170, 1172 (9th Cir.2002). Given my conclusion above, I need not resolve Northwest’s other grounds for summary judgment, including that PacifiCorp’s negligence claim is conflict and field preempted by federal law, that the doctrine of primary jurisdiction requires dismissal, that PacifiCorp cannot prove Northwest caused its alleged damages, that PacifiCorp’s res ipsa loquitur claim fails as a matter of law, and that PacifiCorp’s prejudgement interest claim also fails.

B. GTN

GTN also moves for summary judgment on the negligence claim against it, arguing that its gas transport agreement with PacifiCorp incorporating the gas quality provisions of its FERC tariff supplanted the common law duty of care regarding gas quality. PacifiCorp, by contrast, insists that a negligence claim still lies because the gas transport agreement with GTN did not alter or eliminate GTN’s common law duty.

The Oregon Supreme Court’s most recent pronouncement on the overlap between contract and negligence claims is Abraham v. T. Henry Const, Inc., 350 Or. 29, 249 P.3d 534 (2011). There, the Court reiterated that Fazzolari still remains the touchstone for determining the existence of negligence liability, even in the presence of a contract between the parties. Id. at 37, 249 P.3d 534. The Court sketched what amounts to a variation on the Fazzolari two-step inquiry that courts should apply when determining whether a negligence claim exists:

we first consider whether plaintiffs alleged that defendants unreasonably created a foreseeable risk of harm to a protected interest, resulting in injury to plaintiffs. If so, we must determine whether the contract between the parties altered or eliminated defendants’ common law duty to avoid harming plaintiffs. If it did not, then the contract does not bar plaintiffs from bringing a negligence action against defendants.

350 Or. at 37, 249 P.3d 534. Consequently, the crucial question here is whether PacifiCorp and GTN’s contract “altered or eliminated” GTN’s common law duty regarding gas quality with respect to PacifiCorp. Id. at 37, 249 P.3d 534. This is the effectively the same inquiry posed by Fazzolari, investigating whether the contract “creates, defines, or limits” the existing common law duty. Id. at 38, 249 P.3d 534.

As discussed above, the transport agreement incorporated the gas quality provisions of GTN’s tariff. I agree with GTN that, in doing so, the contract replaced any preexisting common law duty regarding gas quality with the standards spelled out in the tariff. Thus, the tariff can be said to have “altered,” “created,” “defined,” or “limited” GTN’s common law duty — all different words used by Fazzolari and Abraham to refer to the same concept of replacement.

In briefing, PacifiCorp states conclusorily that “there is nothing in GTN’s contract with PacifiCorp that purports to alter or eliminate GTN’s common law duty under Oregon law to avoid harming PacifiCorp.” (Br. in Opp., # 314, at 17.) In oral argument, PacifiCorp advances several more specific, but ultimately unpersuasive theories. First, PacifiCorp seems to contend that where the tariff provisions are incorporated into the contract, conduct breaching those provisions gives rise to a negligence claim so long as the foreseeability prong of Fazzolari is also satisfied. According to this argument, for example, if the tariff and thus the contract required gas to contain no more than 10 parts per million of compressor oil, and delivery of more than 10 parts per million would also “unreasonably create[] a foreseeable risk to a protected interest of the kind of harm that befell the plaintiff,” Fazzolari, 303 Or. at 17, 734 P.2d 1326, then delivery of gas with 15 parts per million compressor oil would give rise to both a contract and negligence claim, despite the fact that the contract specifically defined gas quality standards for compressor oil. It is clear that this formulation conflicts with the holdings of Fazzolari and Abraham, preserving a contract claim even when the parties agree to a specific standard of care defining their respective duties.

Second, PacifiCorp also suggests that because compressor oil is know to be extremely harmful, unlike other substances naturally occurring in gas for which the tariff provides more specific quality standards, the tariff merely reiterates the common law duty to remove it from gas before delivery. Indeed, the Oregon Supreme Court is clear that merely reciting or incorporating a common law duty into a contract does not somehow replace the common law standard of care. See Abraham, 350 Or. at 42-43, 249 P.3d 534 (“By merely reciting the obligation to build plaintiffs’ house in a reasonably skilled manner and in accordance with the building code — and, by implication, in such a way as to avoid foreseeable harm to plaintiff-defendants did nothing to supplant the common law standard of care.”). But that is not what happened here. PacifiCorp admits that the tariffs requirement that gas be “commercially free” of objectionable substances like compressor oil that would interfere with “commercial utilization” is a specific standard just like the other tariff standards for naturally occurring compounds. There is no principled basis for distinguishing between the tariffs quantitative standards for certain substances and the qualitative standard for compressor oil, which will ultimately be given a more precise construction either by FERC or this court. Both replace the common law standard of care.

Third, PacifiCorp appears to argue that the common law duty of care is not supplanted unless the parties’ contract contains an explicit limitation of remedies for negligence, such as an exculpatory clause. This is surely one way for the parties to signal the unavailability of a negligence claim, but it not the only way. As Abraham recognizes, parties can limit tort remedies either by “contractually limiting or specifying available remedies,” or by “defining their obligations in such a way that the common law standard of care has been supplanted.... ” 350 Or. at 40, 249 P.3d 534. The tariff, and by extension the transport agreement, utilizes the latter method, defining GTN’s obligations concerning objectionable substances in a manner supplanting the common law standard of care. Accordingly, the transport agreement bars PacifiCorp’s negligence claim against GTN. I decline to address GTN’s other arguments against the negligence claim as a whole.

III. GTN’s Spoliation Motion

Both defendants move for sanctions based on PacifiCorp’s spoliation of evidence, although, since I grant Northwest’s motion for summary judgment, I focus solely on GTN’s motion. GTN seeks dismissal of this action as a sanction for PacifiCorp’s willful spoliation of: (1) turbine components that were allegedly damaged, including fuel nozzles, transition pieces, and combustion liners; (2) gas and liquid samples; (3) scrubbers installed in front of turbine Units 1 and 2 that were replaced with coalescing filters in March 2008; and (4) samples collected from fuel nozzles, filter cartridges, a “pig,” and “reference oil”. Alternatively, GTN seeks to have any reference to the allegedly spoliated materials excluded — including any other expert testimony relying on that evidence — and an adverse inference instruction that the destroyed scrubbers were malfunctioning or inadequate.

“A federal trial court has the inherent discretionary power to make appropriate evidentiary rulings in response to the destruction or spoliation of relevant evidence.” Glover v. BIC Corp., 6 F.3d 1318, 1328 (9th Cir.1993). Thus, a district court has the discretion to impose sanctions based on its power “to make discovery and evidentiary rulings conducive to the conduct of a fair and orderly trial.” Unigard Sec. Ins. Co. v. Lakewood Eng’g & Mfg. Corp., 982 F.2d 363, 368 (9th Cir. 1992) (internal citation omitted). Sanctions for spoliation include dismissal of claims, exclusion of evidence, and adverse jury instructions permitting a jury to draw an inference that the destroyed evidence would have been adverse to the party responsible for its destruction. Unigard, 982 F.2d at 368-370. Before a court imposes the “harsh sanction” of dismissal, “the conduct to be sanctioned must be due to willfulness, fault, or bad faith.” Anheuser-Busch, Inc. v. Natural Beverage Distributors, 69 F.3d 337, 348 (9th Cir.1995) (internal citation omitted). Even when a court imposes a lesser evidentiary sanction than outright dismissal, it must find that the party willfully destroyed the evidence. Unigard, 982 F.2d at 368 & n. 2; Glover, 6 F.3d at 1329; see also Akiona v. United States, 938 F.2d 158, 161 (9th Cir.1991).

A party’s destruction of evidence is considered “willful” if the party “has some notice that the [evidence was] potentially relevant to the litigation before [it was] destroyed.” Leon v. IDX Sys. Corp., 464 F.3d 951, 959 (9th Cir.2006) (emphasis in original) (internal citation omitted). Because the relevance of destroyed evidence cannot clearly be ascertained, a party “can hardly assert any presumption of irrelevance” as to the destroyed evidence. Id. (quoting Alexander v. Nat’l Farmers Org., 687 F.2d 1173, 1205 (8th Cir.1982)).

Circuit courts describe the duty to preserve evidence as attaching when a party should know that evidence may be relevant to litigation that is “anticipated,” or “reasonably foreseeable.” Silvestri v. Gen. Motors Corp., 271 F.3d 583, 590, 591 (4th Cir.2001) (citing Kronisch v. United States, 150 F.3d 112, 126 (2d Cir.1998), West v. Goodyear Tire & Rubber Co., 167 F.3d 776, 779 (2d Cir.1999)). As the Federal Circuit recently explained, “[w]hen litigation is ‘reasonably foreseeable’ is a flexible fact-specific standard that allows a district court to exercise the discretion necessary to confront the myriad factual situations inherent in the spoliation inquiry.” Micron Tech., Inc. v. Rambus Inc., 645 F.3d 1311, 1320 (Fed.Cir.2011) (citing Fujitsu Ltd. v. Fed. Express Corp., 247 F.3d 423, 436 (2d Cir.2001)). Although “this standard does not trigger the duty to preserve documents from the mere existence of a potential claim or the distant possibility of litigation, ... it is not so inflexible as to require that litigation be imminent, or probable without significant contingencies.... ” Id. (internal citations and quotations omitted). Some district courts in the. Ninth Circuit express the standard differently, holding that litigation must be “probable,” before the duty to preserve evidence applies. Realnetworks, Inc. v. DVD Copy Control Ass’n, 264 F.R.D. 517, 524 (N.D.Cal.2009); Hynix Semiconductor Inc. v. Rambus, Inc., 591 F.Supp.2d 1038, 1061 (N.D.Cal.2006); In re Napster, Inc. Copyright Litig., 462 F.Supp.2d 1060, 1068 (N.D.Cal.2006).

A. When Duty to Preserve Attached

The first step in the spoliation analysis is to determine the point at which PacifiCorp became obligated to preserve all evidence that might be relevant to reasonably foreseeable future litigation. GTN argues that PacifiCorp’s duty to preserve evidence arose in July 2007, the time when PacifiCorp’s 30(b)(6) designee Shane Holst testified PacifiCorp first became aware that damage to turbine components parts was due to gas contamination. (Suppl. Walker Deck, # 341, Ex. 6, at 303-304). At the very least, GTN contends that the duty arose on August 2, 2007, the date this court previously determined PacifiCorp anticipated this litigation for purposes of work product protection analysis. (Opinion and Order, # 210, at 5.)

PacifiCorp counters that the duty to preserve relevant evidence arose much later. PacifiCorp explains that this court erroneously reported PacifiCorp hired outside counsel on August 2, 2007. In fact, PacifiCorp involved only its in-house counsel on that date, and did not hire outside counsel until October 9, 2007. Thus, PacifiCorp contends that the duty to preserve potentially relevant evidence arose only on October 9, 2007. PacifiCorp insists that between August 2, 2007, when PacifiCorp’s in-house counsel became aware of possible fuel contamination, and October 9, 2007, GTN worked cooperatively with the Hermiston plant to assure no further fuel contamination, which indicated that future litigation was not yet probable.

It is clear that at least as early as September 19, 2007, both HGC and PacifiCorp reasonably foresaw litigation against gas suppliers like GTN based on gas contamination. HGC’s claim for reimbursement from PacifiCorp on that date announced that HGC’s preliminary analysis showed the root cause of forced outages occurring on January 25, 2007, July 30, 2007, and August 29, 2007 was “the presence of lubrication/compressor seal oil as a contaminant in the gas fuel supply to the Plant.” (Sasaki Deck, # 324, Ex. 18, at 2.) Further, HGC notified PacifiCorp that “PacifiCorp has the sole right to seek indemnity for all such expenses from the Fuel Gas suppliers....” Id. at 3.

But PacifiCorp apparently predicted litigation even earlier, as of August 2, 2007. In resisting a motion to compel based on the work product protection, PacifiCorp argued that emails and other documents identified as work product beginning on August 2, 2007 were prepared “in anticipation of litigation” and related “to the collection of information necessary to assess, develop, and establish PacifiCorp’s claims and/or to identify and analyze anticipated defenses.” (P.’s Opp. to Mot. to Compel, # 115, at 6.) Indeed, PacifiCorp stated:

PacifiCorp is not claiming work product protection for documents related to this action earlier than August 2, 2007, which is when legal counsel became involved in investigating the outages at the Hermiston Plant.... GTN’s contention that PacifiCorp could not have anticipated litigation prior to September 19, 2007, is simply mistaken.

Id. at 4. In light of these earlier representations, PacifiCorp is judicially estopped from taking the position, advanced presently, that it did not anticipate litigation against GTN in early August 2007 because GTN was cooperating with the plant at that point to find the cause of the gas contamination. See Helfand v. Gerson, 105 F.3d 530, 534 (9th Cir.1997) (judicial estoppel precludes a party from gaining an advantage by taking one position and then seeking a second advantage by taking an incompatible position).

The fact that the court incorrectly identified August 2, 2007 as the date PacifiCorp retained outside counsel is of no moment. PacifiCorp failed to correct, and even repeated, this court’s misstatement of that fact in later briefing in support of its motion for reconsideration. Moreover, I reject PacifiCorp’s hairsplitting attempt to distinguish between the standard for work product protection requiring only “possible” litigation and that for the duty to preserve relevant evidence, which PacifiCorp characterizes as requiring “probable” litigation. Courts analyzing this issue in depth endorse a flexible and fact-specific inquiry into when parties become obligated to preserve evidence, eschewing a hyper-technical reliance on terms like “probable.” Cf. Micron Tech., 645 F.3d at 1320. Finally, even under PacifiCorp’s proposed “probable” litigation standard, PacifiCorp’s assertion that it was assessing, developing, and establishing claims and anticipated defenses in August 2007 surely suffices to establish that litigation was probable at that time. Consequently, PacifiCorp was obligated to preserve any potentially relevant evidence relating to gas quality litigation as of August 2, 2007.

Evidence that PacifiCorp destroyed before that date cannot form the basis for spoliation sanctions. In particular, although it is not clear from the record exactly when the turbine components from the January 2007 outage were destroyed, it is likely that they were refurbished or discarded sometime soon after GE’s inspection report relating to the January 2007 outage, issued on April 22, 2007. By contrast, components from the July 2007 outage were undoubtedly destroyed after August 2, 2007, because GE issued its report recommending the plant send the fuel nozzles damaged in the July outage to Florida for repair on October 31, 2007. (Glasgow Deck, #325, Ex. 15 at 1, 8.)

B. Destroyed Evidence

For each piece of destroyed evidence, I analyze whether the destruction was willful, the degree of prejudice to GTN, and the efficacy of less drastic sanctions than dismissal. Given that all the evidence for which GTN seeks sanctions was destroyed after August 2, 2007, when PacifiCorp anticipated litigation, the willfulness inquiry focuses on the potential relevance of the destroyed evidence. See Leon, 464 F.3d at 959 (destruction willful if the destroying party had notice that the evidence was potentially relevant to the litigation). Further, in assessing prejudice to GTN, I look to whether PacifiCorp’s actions have impaired GTN’s ability to go to trial or achieve a “rightful decision of the case.” Id. Destruction that precludes a party from inspecting physical evidence can create prejudice. Unigard, 982 F.2d at 369. Similarly, forcing a party to rely on evidence selected by an opposing party’s expert creates prejudice, because such evidence generally supports that party’s case. BTO Logging, Inc. v. Deere & Co., 174 F.R.D. 690, 694 (D.Or.1997). However, the existence of secondary evidence can reduce prejudice caused by the inadvertent loss of primary evidence. Med. Lab. Mgmt. Consultants v. Am. Broad. Cos., Inc., 306 F.3d 806, 825 (9th Cir.2002) (digital images of misplaced slides and records retained by lab and doctors examining those slides constitute secondary evidence).

1. Turbine Fuel Nozzles

The destruction of turbine parts is a central dispute in this litigation. I primarily address the alleged spoliation of the turbine nozzles, which raise particularly troubling legal and factual issues. Overall, PacifiCorp preserved only 11 of the 350 fuel nozzles allegedly damaged due to contamination. Instead of preserving the other damaged fuel nozzles, PacifiCorp refurbished them, purportedly to avoid the higher cost of buying new replacements and to prevent plant shutdowns and resulting loss of revenue that would occur if another failure struck while the plant operated without a back-up set of nozzles. Although PacifiCorp never informed GTN of its plans to refurbish the nozzles, PacifiCorp argues that GTN “had to know that the nozzles would be refurbished” because GTN was informed of the outages, yet never objected to the refurbishment. (P.’s Mot. in Opp., # 326, at 26-27.)

Of the 11 nozzles preserved, only five are relevant to the Unit 2 outages for which damages are claimed in this litigation. Those five nozzles came from the July 30, 2007 Unit 2 outage and were provided to Dr. Kemal for examination at his request. As each turbine has 14 combustor assemblies, and each assembly contains five fuel nozzles, these five nozzles constitute only a fraction of the nozzles from one of four outages which Dr. Kemal addressed in his expert report. Dr. Kemal cross-sectioned several of the five Unit 2 nozzles to conduct his analysis, leaving others unaltered. PaeifiCorp’s counsel represented that Dr. Kemal created no lab data or notes during his testing of the nozzles, although Kemal’s recent declaration indicates he did create evidence logs and lab analysis reports, which so far have not been produced to defendants. (Kemal Deck, # 319, ¶ 17.) Dr. Kemal’s expert report, deposition, and declaration offer somewhat different observations about the condition of the cross-sectioned nozzles and the alleged presence of “coke” within them. But Dr. Kemal consistently opines that, based on the comparison of the Unit 2 nozzles he physically examined with photographs of nozzles from three other outages taken by plant employees, all nozzles exhibited similar patterns of coking and inside-out burning. He therefore concluded that the four forced outages of Unit 2 in 2007 were caused by the same mechanism — the presence of liquid hydrocarbons in the gas supply.

Defendants were later provided an opportunity to examine the fuel nozzles in the possession of Dr. Kemal. At that examination, communications apparently broke down when GTN requested permission to bring the nozzles to a different laboratory for additional testing, with PacifiCorp blaming GTN for unnecessarily delaying further examination and GTN blaming PacifiCorp for refusing to allow additional destructive testing. Currently, several unaltered nozzles from the July 2007 outage remain in Kemal’s laboratory.

Here, PacifiCorp does not dispute the relevance of the nozzles to the litigation. It is self-evident that the destroyed fuel nozzles are relevant, both since PacifiCorp seeks damages for their refurbishment and for lost revenue allegedly caused by their failure. Thus, I conclude that PacifiCorp’s destruction of nozzles from the July, August, and September 2007 outages was willful, since litigation was reasonably foreseeable when PacifiCorp sent those nozzles to be refurbished.

PacifiCorp, however, argues that no sanctions should be imposed for two other reasons. First, PacifiCorp contends that refurbishing the nozzles was a reasonable measure necessary to mitigate additional damages. PacifiCorp cites several cases supporting its position that refurbishment of the turbine nozzles was proper and necessary to carry out its countervailing obligation to mitigate damages, because preserving the nozzles would have jeopardized continued operation of the plant. See Flint Hills Res. LP v. Lovegreen Turbine Servs., Inc., Civil No. 04-4699 (JRT/FLN), 2006 WL 2472819, at *5 (D.Minn. Aug. 25, 2006) (plaintiffs immediate inspection and repair of gas compressor that was jammed with a cloth rag was an effort to mitigate significant damages from plant shutdown rather than an intentional destruction of evidence, because plaintiff documented the repair with photographs and notes and, importantly, avoided destroying or discarding any damaged property by saving all rag fragments); McLaughlin v. Denharco, Inc., 129 F.Supp.2d 32, 36 (D.Me.2001) (plaintiffs repair of tree delimbing machine did not constitute conscious wrongdoing because he had to maintain the machine to mitigate damages and because defendants were not “responsive”); Baliotis v. McNeil, 870 F.Supp. 1285, 1292 (M.D.Pa.1994) (finding no bad faith by insurer who failed to maintain the scene of a fire until all potential defendants were notified and could conduct inspections, but permitting an adverse inference instruction that lost evidence was unfavorable).

While these cases illustrate circumstances where failure to preserve evidence in its original state is necessary for purposes of mitigation, they do not assist PacifiCorp. First, unlike in McLaughlin and Baliotis, here PacifiCorp never notified GTN that it planned to refurbish the nozzles or offered to let GTN inspect the nozzles before they were irrevocably altered. Indeed, PacifiCorp, as the party in control of evidence relevant to anticipated litigation, had the obligation to notify GTN of that evidence and provide access before destroying it. See Erlandson v. Ford Motor Co., No. 08-CV-1137-BR, 2009 WL 3672898, at *5 (D.Or. Oct. 30, 2009) (“The party in control of the evidence has ‘an obligation to give the opposing party notice of access to the evidence ... if the party anticipates litigation involving that evidence.’ ”) (quoting Silvestri, 271 F.3d at 591).

Further, mitigation was not urgently required here, as in other cases where courts refused to impose sanctions for destruction of evidence on mitigation grounds. For example, in Flint Hills, damages of $1 million per day in lost revenue would necessarily have accrued unless the plaintiff repaired its damaged compressor immediately. Flint Hills, 2006 WL 2472819, at *5. By contrast, here, the plant resumed operation with spare nozzle assemblies and apparently sent the damaged nozzles for refurbishment months after the allegedly forced outages. PacifiCorp therefore had ample time to allow defendants to inspect the damaged nozzles without delaying PacifiCorp’s chosen mitigation strategy.

Second, PacifiCorp insists that GTN suffered no prejudice, as other adequate reliable evidence exists concerning the nature of the fuel nozzle failures. However, the prejudice from PacifiCorp’s conduct is overwhelming, at least as regarding fuel nozzles from the August and September 2007 outages. PacifiCorp’s refurbishment prevents GTN’s experts from examining any damaged turbine nozzles from those outages to investigate the plausibility of alternate causes of damages besides oil contamination — causes such as design defects, plant operation error, or other mechanical error that defense experts raise. See Unigard, 982 F.2d at 369.

Moreover, PacifiCorp applies an incorrect standard by arguing that GTN suffers no prejudice because the remaining evidence of nozzle failure is “adequate and reliable.” The loss of all nozzles from the August and September 2007 outages places GTN in the position of relying on the photographs taken by plant employees and the analyses of the plant’s turbine repair contractor (GE) from those outages, which likely support PacifiCorp’s failure theory while failing to document evidence supportive of other failure theories. See BTO Logging, 174 F.R.D. at 694. These photographs and GE analyses do not constitute secondary evidence that can be easily substituted for the original. It is obvious that where the crucial issues of causation revolve around identifying the presence or indicia of a chemical compound — compressor oil — in a damaged machine, photographs of the machine are hardly an adequate substitute for the thing itself. The GE reports contain photographs that suffer the same flaw as the plant photographs, and the reports’ observations about the condition of the nozzles are even less reliable as secondary evidence of the refurbished nozzles. Finally, the one preserved fuel nozzle assembly from the July 2007 outage cannot replace nozzles from other outages because the two outages are distinct events with potentially different causes. GTN will be prejudiced if it is forced to accept Dr. Kemal’s assertion that the failure mechanism he identified for the July 2007 outage also occurred in August and September 2007 without the opportunity to examine the equipment damaged in those outages.

Because the despoiled fuel nozzles are at the heart of this case and absolutely no physical evidence remains of the parts damaged in the August and September 2007, the most appropriate sanction is exclusion of all expert testimony concerning those damaged parts and whether compressor oil caused their damage. Napster, 462 F.Supp.2d at 1066 (as a sanction for spoliation, a court may exclude “witness testimony proffered by the party responsible for destroying the evidence and based on the destroyed evidence.”). This includes Dr. Kemal’s testimony concerning causation of the August and September outages, plus any other expert’s testimony relying on those conclusions. The Ninth Circuit has affirmed imposition of a somewhat similar sanction in Unigard, where the district court excluded testimony of an expert about the cause of damages where plaintiff destroyed the instrumentality allegedly causing its damages, reasoning that plaintiffs introduction of the expert testimony would unfairly prejudice defendant and preclude a fair trial. Unigard, 982 F.2d at 368. This ruling does not constitute an outright dismissal of PacifiCorp’s breach of contract claim, but rather a limitation on the evidence PacifiCorp may introduce in support of that claim, commensurate with PacifiCorp’s spoliation. As I discuss more thoroughly below, this ruling precludes PacifiCorp from establishing that GTN’s alleged gas contamination was the factual cause of the August and September 2007 outages and, consequently, prevents PacifiCorp from recovering any damages associated with parts, labor, services, or lost power generation flowing exclusively from those outages.

A slightly different analysis applies to the July 2007 nozzles. 65 of those nozzles were refurbished, but importantly, five were given to Dr. Kemal, a few of which were cross-sectioned and others of which remained in their original condition. PacifiCorp’s spoliation of almost every turbine component from the July 2007 outage also might warrant exclusion of Kemal’s testimony concerning the cause of the July 2007 outage, for some of t