In Re Methyl Tertiary Butyl Ether ("MTBE") Products Liability Litigation ( 2008 )

(2008)

In re: METHYL TERTIARY BUTYL ETHER ("MTBE") PRODUCTS LIABILITY LITIGATION.
This document relates to: County of Suffolk et al. v. Amerada Hess Corp. et al., 04 Civ. 5424.

Master File No. 1:00-1898. MDL No. 1358 (SAS). No. M21-88.

United States District Court, S.D. New York.

May 13, 2008.

OPINION AND ORDER

SHIRA A. SCHEINDLIN, District Judge:

I. INTRODUCTION

Methyl tertiary butyl ether ("MTBE") is a gasoline additive that has contaminated groundwater throughout the United States, primarily as a result of leaking underground storage tanks ("USTs"). Because MTBE "dissolves and spreads readily in the groundwater ... resists biodegradation, and is difficult and costly to remove from groundwater," it "has caused widespread and serious contamination of the nation's drinking water supplies."^[1] Among those contaminated water supplies are the wells of plaintiffs Suffolk County Water Authority ("SCWA") and the County of Suffolk, who supply drinking water to over one million people in Suffolk County from a large aquifer system that lies beneath the ground in Long Island, New York.^[2] MTBE has been detected in over one hundred fifty of SCWA's nearly five hundred active wells, and scientists predict that other wells are threatened by contamination.^[3]

Seeking compensatory and punitive damages, as well as other remedies, plaintiffs sued various companies in the oil industry that have manufactured, refined, marketed or distributed MTBE or gasoline containing MTBE.^[4] In their complaint, plaintiffs assert claims for (1) violation of Section 8(e) of the Toxic Substances Control Act ("TSCA"); (2) public nuisance; (3) strict liability for design defect and/or defective product; (4) failure to warn; (5) negligence; (6) private nuisance; (7) trespass; and (8) violation of the New York Navigation Law.^[5] To facilitate the jury trial of the numerous and complicated factual issues raised in the case, I ordered a bellwether trial of claims related to ten wells contaminated with MTBE. The number of wells that will be considered at trial has since grown to eighteen wells.^[6]

The parties agree that all of the focus wells have been contaminated with MTBE. The issue in dispute is one of causation: Where did the MTBE in each well come from, and who bears the responsibility for its presence? Due to the unique and complicated facts of this case, the means of proving causation has been a highly contested issue.^[7] In particular, plaintiffs face two independent obstacles in identifying which defendant's product caused their injuries.

The first obstacle is that many of the spills and leaks of gasoline that may have caused contamination of plaintiffs' well water occurred long ago and beneath the ground. From these spills and leaks, MTBE then migrated toward the wells through water flowing beneath the ground. In many cases it is difficult for plaintiffs to identify the gasoline releases from which the MTBE contamination originated (i.e., the leaking UST at a particular retailer).

The second obstacle is that the gasoline distribution system in the United States requires manufacturers to mix their products together for transportation in a common pipeline system. Because gasoline is commingled, it is impossible to identify with certainty the refiners of the gasoline released from a leaking UST.

Defendants have brought two omnibus motions for summary judgment on plaintiffs' tort claims based on inability to prove causation. The first motion argues that for half of the focus wells, plaintiffs cannot prove that any particular leaking UST at a retail gas station caused the contamination of the well, and thus the companies that own those retail stations cannot be liable. Defendants further argue in that motion that the companies that manufacture, market or distribute MTBE or gasoline containing MTBE should not be liable because plaintiffs cannot prove their role in causing the harm.^[8] The second motion argues that where plaintiffs can identify the source of contamination in a well, only the owners and/or operators of the stations named by plaintiffs' expert as the source of contamination can be liable.^[9]

In addition, defendants filed a separate motion for partial summary judgment as to the Navigation Law claims, again arguing that only those companies whose spills were found by plaintiffs' expert to have caused contamination can be liable for such claims.^[10]

Finally, a small subset of defendants filed individual motions for summary judgment, arguing that for various reasons, their gasoline or MTBE could not have caused plaintiffs' injuries. These companies include: (1) Lyondell Chemical Company ("Lyondell") and Equistar Chemicals LP ("Equistar"), (2) Crown Central LLC ("Crown"), (3) Getty Properties Corp. ("Getty"), (4) Giant Yorktown Inc. ("Giant"), (5) Irving Oil Limited and Irving Oil Corporation ("Irving"), and (6) Total Petrochemicals USA, Inc. ("Total").^[11]

For the reasons below, defendants' motions are granted in part and denied in part.

II. STANDARD FOR SUMMARY JUDGMENT

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 judgment as a matter of law."^[12] An issue of fact is genuine "`if the evidence is such that a reasonable jury could return a verdict for the nonmoving party.'"<sup>[13]</sup> A fact is material when it "`might affect the outcome of the suit under the governing law.'"^[14] "It is the movant's burden to show that no genuine factual dispute exists."^[15]

In turn, to defeat a motion for summary judgment, the non-moving party must raise a genuine issue of material fact. To do so, it must do more than show that there is "`some metaphysical doubt as to the material facts,'"<sup>[16]</sup> and it "may not rely on conclusory allegations or unsubstantiated speculation.'"<sup>[17]</sup> However, "`all that is required [from a non-moving party] is that sufficient evidence supporting the claimed factual dispute be shown to require a jury or judge to resolve the parties' differing versions of the truth at trial.'"^[18]

In determining whether a genuine issue of material fact exists, the court must construe the evidence in the light most favorable to the non-moving party and draw all justifiable inferences in that party's favor.^[19] However, "[i]t is a settled rule that lc]redibility assessments, choices between conflicting versions of the events, and the weighing of evidence are matters for the jury, not for the court on a motion for summary judgment.'"^[20] Summary judgment is therefore inappropriate "`if there is any evidence in the record that could reasonably support a jury's verdict for the non-moving party.'"^[21]

III. CAUSATION

The requirement that the defendant's actions must be the cause of the plaintiff's injury is common in the law and applies to all of plaintiffs' claims other than TSCA.^[22] At issue here is whether plaintiffs' evidence could support a reasonable jury's finding that defendants caused the alleged harm under either traditional or alternative theories of causation, and in what circumstances is it appropriate to allow plaintiffs to prove causation through the use of alternative theories of liability.

A. Traditional and Alternative Methods of Proving Causation

Tort liability usually depends on proof that a defendant's conduct was both the factual cause and the proximate cause of a plaintiff s injury.^[23] For example, in a negligence claim, "[t]o carry the burden of proving a prima facie case, the plaintiff must generally show that the defendant's negligence was a substantial cause of the events which produced the injury."^[24] The substantial factor standard for causation, adopted in New York, recognizes that often many acts can be said to have caused a particular injury, and requires only that defendant's actions be a substantial factor in producing the injury.^[25] A plaintiff need not eliminate every other possible cause, and the fact "[t]hat another possible cause concurs with defendant's negligent act or omission to produce an injury does not relieve defendant from liability."^[26]

Plaintiffs usually bear the burden of proving causation, like every other element of a prima facie case, by a preponderance of the evidence. In other words, plaintiffs must show that it is more likely than not that the defendant's actions caused their injury. In addition, "identification of the exact defendant whose product injured the plaintiff is ... generally required."^[27] "The identity of the manufacturer of a defective product may be established by circumstantial evidence."^[28] Such evidence cannot be "speculative or conjectural," it must be reasonably probable that it was defendant's product that caused the injury.^[29]

It is sometimes impossible to identify the exact defendant who caused the injury even though the plaintiff can establish the other elements of a prima facie case against a number of defendants. In such situations, courts have occasionally allowed plaintiffs to prove causation through alternative means,^[30] such as shifting the burden to the defendants to prove that they did not cause the harm.^[31] Courts have developed various doctrines, deemed "alternative liability theories," which depart from the requirement that plaintiffs prove by a preponderance of the evidence that defendant's conduct was a substantial cause of the events which produced the injury.

B. Applicability of Causation Theories in this Case

In denying defendants' motion to dismiss in this action, I held that New York courts would allow plaintiffs to prove their claims through an alternative liability theory.^[32] Since then, much confusion has arisen regarding the applicability of various theories of alternative liability, particularly market share liability.^[33]

To some extent, the confusion is due to prior rulings in this case that were made without a fully developed evidentiary record. Plaintiffs have now submitted evidence about the gasoline distribution system, including third-party depositions of gasoline pipeline and terminal company representatives, which has shed light on how manufacturers' gasoline is commingled for transport to retail gas stations. On the basis of this evidence, as explained below, I am now able to clarify several issues with respect to alternative liability. First, market share liability need not be applied in this case to prove plaintiffs' claims, and indeed the theory may not even be applicable on these facts, because the product alleged to have caused the harm is undeniably a blended product manufactured by multiple defendants, rather than a product manufactured by a single defendant that cannot be identified.^[34]

Second, plaintiffs may rely on the commingled product theory, which this Court developed to address the particular facts of this case, to prove their claims against gasoline and MTBE manufacturers.^[35] As discussed below, the commingled product theory, while still an alternative means of proving causation, is closer to traditional causation than to market share liability. Under this theory, a reasonable jury could conclude, based on the evidence in the record, that all defendants contributed to the commingled gasoline that caused contamination in plaintiffs' wells. Defendants may still exculpate themselves by showing that their product could not have been part of the commingled gasoline spilled in Suffolk County, but the burden shifts to them to do so.

Third, where a reasonable jury could conclude—under traditional causation principles—that particular gasoline spills caused the contamination in each well, summary judgment is denied for claims against the defendants responsible for those spills.^[36]Fourth, because all entities in the chain of distribution may be liable for product liability claims,^[37] the defendants proven to have spilled the gasoline that caused contamination in a well may be held jointly and severally liable for these claims with the defendants shown under the commingled product theory to have manufactured the gasoline that spilled. In other words, the jury's conclusion that certain defendants spilled gasoline that caused contamination in a well, and are thereby liable as retailers of a defective product, does not preclude the jury from also concluding that certain defendants manufactured the gasoline that was spilled, and are thereby liable as manufacturers of a defective product. Fifth, and finally, even when the jury concludes that plaintiffs have not met their burden to prove that any particular spill of gasoline caused contamination in a well, plaintiffs may still prove their claims against gasoline manufacturers under the commingled product theory.

IV. LIABILITY OF GASOLINE AND MTBE MANUFACTURERS^[38]

The parties do not dispute that gasoline containing MTBE is a fungible product, nor do they dispute that gasoline manufacturers—known as refiners—mix their products together for transportation and distribution, so that the gasoline sold at any given retail station contains the product of multiple refiners. Defendants argue that because plaintiffs cannot identify which refiners were responsible for producing the particular gallons of gasoline that were released into the environment and caused contamination in the wells, their claims against the refiners must fail. In response, plaintiffs argue that all refiner defendants contributed to all gasoline released into the environment in Suffolk County, creating a fact issue for the jury under traditional causation principles. In the alternative, plaintiffs argue that the commingled product theory and/or market share liability should shift the burden of proof to the defendants to exculpate themselves from liability.

Plaintiffs' argument that a reasonable jury could conclude, under traditional causation principles, that it was likely that all defendants' product was part of every gallon of gasoline that caused MTBE contamination, is not supported by the evidence in the record. Rather, it is likely that some of each defendant's gasoline was spilled somewhere in Suffolk County, leading to contamination in some of the wells. Moreover, because of the blended nature of the gasoline, it is impossible to determine whose product was in any particular spill. However, defendants' argument that this impossibility is fatal to plaintiffs' claims is contrary to New York policy and precedent, and to prior rulings in this case. Instead, as explained below, the commingled product theory provides plaintiffs with a means to prove their claims against the manufacturers.

A. Background on the Pipeline System

"The name on the service station sign does not tell the whole story. The fact that you purchase gasoline from a given company does not necessarily mean that the gasoline was actually produced by that particular company's refineries."^[39] In fact, the gasoline is almost never produced exclusively by that company's refineries. Instead, the gasoline sold at each service station in a particular area is usually the same: a blend of a large number of refiners' products, due to the complex gasoline distribution system in the United States.

The system includes various actors:

· MTBE manufacturers, who create MTBE and sell it to be blended into gasoline;

· gasoline refiners, who manufacture gasoline and other petroleum products from crude oil, and who blend MTBE into gasoline;

· marketers, who purchase and sell gasoline, acting as middlemen between refiners and retailers; and

· retailers, who sell gasoline to the public at service stations.

Other participants in the system include those who transport or store gasoline, such as pipeline operators, terminal owners, and jobbers, who transport gasoline from terminals to retailers. Many oil companies are vertically integrated, which means that a single company may own and operate refineries, own terminals, market gasoline, and sell gasoline to the public from retail stations that it owns or on which it places its brand. "The process of manufacturing and distributing petroleum products involves complex arrangements whereby defendants trade, barter, or otherwise exchange product for delivery throughout parts of the country."^[40]

The complex business relationships in the oil industry cannot fully be addressed here.^[41] Instead, this section discusses how gasoline traveled from the manufacturers to the retail stations in Suffolk County from which it leaked into the soil. First, MTBE is manufactured and blended into gasoline at the refinery. Second, the gasoline is transported via pipeline and waterway to New York Harbor. Third, it is stored in primary terminals at the New York Harbor, from which it is redistributed by ship and by pipeline to secondary terminals in Long Island. Fourth, and finally, at Long Island secondary terminals, trucks load the gasoline from common tanks for transport to retail stations in Suffolk County.

1. Gasoline Refining

Crude oil is converted to petroleum products, including gasoline, at refineries. Many of the defendants in this case own and operate petroleum refineries. At the refinery, MTBE—which may have been manufactured by a separate petrochemical company and purchased by the refinery, or may have been manufactured at the refinery itself—is blended into the portion of gasoline that is destined for areas where oxygenated fuel is sold.

2. Pipeline Transport

Most U.S. refineries are located in the Gulf Coast, although regional refineries are scattered throughout the country. A system of underground pipelines transports gasoline from refineries to regional markets. The pipeline serving the East Coast is the Colonial Pipeline.^[42] Stretching from the Gulf of Mexico to New York Harbor, the Colonial Pipeline is the "world's largest volume-refined petroleum products pipeline system."^[43] It includes 5,519 miles of pipeline and serves over eighty customers.^[44]

Refiners load gasoline and other petroleum products into the pipeline from their facilities in Texas, Louisiana, Mississippi and Alabama.^[45] Volumes of gasoline are unloaded at the 267 marketing terminals along the length of the pipeline, most of which are near major population centers.^[46] The Colonial Pipeline terminates at the New York Harbor in Linden, New Jersey, where there are a number of petroleum terminals.^[47]

Most gasoline shipped in the pipeline is fungible. When a refiner designates its gasoline as fungible, it is not kept separate from other refiners' gasoline but is instead mixed with all other refiners' fungible gasoline for shipment. Each refiner loading fungible gasoline into the pipeline must ensure that its product is identical according to pipeline specifications.^[48] When a company takes a volume of gasoline out of the pipeline at a terminal, it is not the same gasoline it put in, but merely an equivalent amount. As Colonial Pipeline explains, when shipping gasoline as fungible, companies unloading their product "will receive equivalent product but may not get back the actual product shipped."^[49]

Some proprietary grades of gasoline are shipped as a segregated product, in which case "shippers receive the same product they injected into the system."^[50] Segregated product represents a small portion of the gasoline shipped in the pipeline, however. According to plaintiffs, "only a few gasoline products ... were not commingled with other gasoline products— Amoco's Silver and Ultimate premium gasolines, and in the early 1990s, Mobil's proprietary premium grades and certain product refined at Exxon's Bayway facility."^[51]

3. Terminals in New York Harbor and Long Island

Terminals are "the primary gateways for petroleum product distribution."^[52] The first distribution hub for gasoline sold in the New York metropolitan area, as well as much of the Northeast, is the set of primary terminals in the New York Harbor. Terminals are owned by individual companies, but like the pipeline companies the terminal owners do not own the gasoline stored at and shipped from the terminal.^[53] The primary terminals receive gasoline and other fuels from the Colonial Pipeline and from waterway shipment, store the gasoline in large tanks, and redirect it toward secondary terminals serving regional markets.

Waterway shipments from both domestic and foreign sources represent a significant portion of supply to the New York Harbor terminals.^[54] Although shipments arriving at the terminal may contain gasoline produced by a single refiner, upon arrival that gasoline is placed in the same large storage tanks as the fungible gasoline from the pipeline. Therefore, it becomes mixed with gasoline from other refiners at the terminal itself.

From the New York Harbor terminals, gasoline is sent to secondary terminals throughout the Northeast, including terminals on Long Island where Suffolk County is located, either through the Buckeye Pipeline, an underground pipeline system serving the Northeast, or by barge.^[55] Some gasoline refined in New Jersey is also placed directly into the Buckeye Pipeline for transport to Long Island secondary terminals. Most gasoline placed in the Buckeye Pipeline has already been blended either in the Colonial Pipeline or in storage tanks at the primary terminal. Gasoline placed directly by the refiner into the Buckeye Pipeline, if designated as fungible, is mixed together with the other gasoline, which itself is already mixed.

Secondary terminals on Long Island supply all the gasoline found in Suffolk County—both at retail stations as well as non-retail customers.^[56] As with primary terminals, gasoline arriving at secondary terminals by pipeline and barge is stored in large tanks. The terminals have tanks designated for various grades of fungible gasoline, proprietary gasoline, and other petroleum products.

4. Distribution to Retail Stations

From "racks" at the secondary terminal tanks, tanker trucks load gasoline for distribution to retail stations and other users. The trucks delivering the gasoline may be a company's branded trucks taking it to their branded stations, or they may be independent companies who contract with refiner-marketers to supply branded stations, or who supply independent, non-branded stations.^[57] Unless the truck is delivering a proprietary brand, the gasoline loaded onto each truck comes from the same tanks, which contain gasoline from various refiners mixed together.

There are approximately six hundred retail gasoline stations in Suffolk County, about eighty percent of which are branded.^[58] A branded station is not necessarily owned by the company under whose brand it sells gasoline, and in Suffolk County only about forty-five percent of the branded stations are owned by the branding company.^[59] The stations may be leased or run under franchise agreements, in which case independent owners of branded stations are typically required to purchase gasoline from the refiner whose brand is on the station.^[60] Non-branded stations also purchase gasoline from refiners at the secondary terminals, or from jobbers who purchase it from refiners and re-sell it to station owners.^[61] At gas stations, the trucks load the gasoline into USTs from which it is pumped into vehicles upon sale. These USTs are often the site of gasoline leaks.

B. Proof of Claims Against Manufacturers Under Traditional Causation Principles

Evidence in the record that describes the gasoline distribution system serving the New York regional market includes a report by defendants' expert John O'Brien,^[62] confidential deposition testimony from representatives of the Colonial and Buckeye pipelines,^[63] and deposition testimony from representatives of companies that own, operate or lease six of Long Island's eight secondary terminals, most of which are also confidential.^[64] This evidence illustrates that gasoline sold in Suffolk County is supplied according to the system described above and becomes commingled in the process, making it "the joint product of all who have contributed to it along the way from refineries through pipelines and tankers to primary terminals and out again via pipelines and barges to secondary terminals."^[65]

Testimony from pipeline and terminal employees makes clear that all gasoline delivered to Suffolk County gas stations, with the exception of a few proprietary brands, was necessarily transported through the commingled distribution system described above.^[66] In addition, evidence in the record, including interrogatory responses from refiner defendants^[67] and confidential records of barrels of gasoline placed into the pipeline and taken out of the pipeline by each refiner defendant by year from 1979 to 2003,^[68] demonstrates that during certain years each refiner defendant contributed to the commingled gasoline distributed in Suffolk County.

However, even drawing all inferences in plaintiffs' favor, no reasonable jury could find, by a preponderance of the evidence, that each defendant's gasoline caused the contamination of each well. Many refiners supply the New York area—almost fifty were named as defendants in this case— and overseas companies not named as defendants also supply gasoline to the New York regional market. Because of the large number of refiners contributing product, defendants' expert opines that "no gallon of gasoline stored in, or released from, a UST in Suffolk County would ever have contained molecules from every gasoline manufacturer who made product in a particular year."^[69] There is simply no way to identify, by direct or circumstantial evidence, the refiners of the gasoline that eventually leaked from a UST.

C. Proof of Claims Against Manufacturers Under the Commingled Product Theory

The evidence can, however, support an inference that each defendant's gasoline containing MTBE was, at least during certain years, delivered to Suffolk County gas stations in a commingled state. A reasonable jury could conclude that most defendants' gasoline contributed to contamination in at least some of the wells at some point. To exempt defendants from liability, when plaintiffs have proven the other elements of their claims, simply because plaintiffs are unable to deconstruct the molecules of the commingled gasoline to identify the manufacturers of each gallon of spilled gasoline is unjust. To avoid such a result, New York courts have often "modif[ied] the rules of personal injury liability, in order `to achieve the ends of justice in a more modern context' and . . . to overcome the inordinately difficult problems of proof' caused by contemporary products and marketing techniques."^[70]

In accordance with that policy, I have previously held that under the commingled product theory,

when a plaintiff can prove that certain gaseous or liquid products (e.g., gasoline, liquid propane, alcohol) of many refiners and manufacturers were present in a completely commingled or blended state at the time and place that the harm or risk of harm occurred, and the commingled product caused plaintiff s injury, each refiner or manufacturer is deemed to have caused the harm.^[71]

In addition, "[a] defendant must be able to exculpate itself by proving that its product was not present at the relevant time or in the relevant place, and therefore could not have been part of the commingled or blended product."^[72]

The commingled product theory lies somewhere between market share and concurrent wrongdoing.^[73] It is similar to concurrent wrongdoing—a theory that allows multiple tortfeasors to be held jointly and severally liable when each tortfeasor's independent actions combine to produce the same wrong—because it addresses a situation in which multiple defendants have contributed to an indivisible injury. It is similar to market share in that it shifts the burden to defendants to exculpate themselves from liability.

The theory is different from market share liability, however, in an important way. Market share liability was developed in the context of plaintiffs' inability to identify which manufacturer had produced the defective product—diethylstilbestrol ("DES") pills. Each plaintiff in the DES cases had ingested pills that were manufactured by only one defendant, but no one could determine which of a small number of manufacturers made those exact pills. When holding all manufacturers of the generic pill liable under market share, courts recognized that all but one of them did not cause the plaintiff's injury.

Here, by contrast, because the gasoline that has contaminated plaintiffs' wells was undeniably the commingled product of numerous manufacturers, there is a good chance that many of the defendants held liable, if not the majority, actually did cause plaintiffs' injury. In this sense, the commingled product theory is closer to traditional causation than market share liability.

Based on the evidence described above, a reasonable jury could conclude that gasoline from "many refiners and manufacturers [was] present in a completely commingled or blended state at the time and place that the harm or risk of harm occurred,"^[74] and therefore the manufacturer defendants could be held liable under the commingled product theory. Each defendant, of course, may offer evidence to exculpate itself from liability "by proving that its product was not present at the relevant time or in the relevant place."^[75]

Clarification as to the operation of the commingled product theory is required. First, the time that the risk of harm—the contamination of groundwater—occurred is an issue of fact for the jury. It is admittedly difficult to determine the date of groundwater contamination.^[76] However, plaintiffs have presented sufficient evidence about the spills alleged to have caused contamination in each well and/or the dates MTBE was first detected in each well to allow a jury to estimate a range of time during which the contamination occurred.

Second, the place the harm or risk of harm occurred is the capture zone of each well, where the MTBE now contaminating the well must have first contaminated the groundwater.^[77] A reasonable jury could conclude that each defendant's gasoline was present within the well's capture zone even if the jury concludes that it cannot identify the source of the spill that caused the well's contamination.^[78] It is not necessary for plaintiffs or the jury to identify the particular spill(s) that caused contamination in a well for manufacturers to be liable under the commingled product theory, because identifying the source of a spill does not provide any additional information about the manufacturers of the gasoline contaminating the well.

Third, and finally, I have previously stated that alternative liability theories—particularly market share liability— should not be applied unless plaintiffs have no other remedy. If a jury determines under traditional causation principles that an identified tortfeasor caused the contamination in a well, I held that plaintiffs cannot pursue other tortfeasors with respect to that well under alternative liability theories.^[79] While this reasoning may apply to some claims (e.g., public or private nuisance), I now conclude that plaintiffs may pursue product liability or negligence claims against manufacturers for placing a dangerous product into the stream of commerce, regardless of whether plaintiffs can identify a retailer whose leaking tank spilled gasoline into a well's capture zone.

This is true for several reasons. To begin with, negligence claims against manufacturers are distinct from negligence claims against the retailer who spilled gasoline—the manufacturer's breach would be distributing a dangerous product, or failing to provide warnings as to the use of a product, while the spiller's breach would be failing to prevent leaks on its property. Moreover, for product liability claims New York courts favor liability for entities higher in the chain of distribution, reasoning that "[m]anufacturers are in the best position to know when products are suitably designed and properly made, as well as to diffuse the cost of safety in design and production."^[80] Retailers and distributors, on the other hand, are often "innocent conduits in the sale of the product" and are held liable only because product liability is strict.^[81] Finally, because the commingled product theory is far closer to traditional causation principles than market share liability, the limitation on the use of market share as a method of proof is unnecessary.

Defendants also argue that plaintiffs should not be permitted to prove causation under an alternative theory of liability because the New York Oil Spill Fund ("the Fund") provides a complete remedy for injuries to all of plaintiffs' wells. I am not convinced that the Fund would, in fact, provide such a remedy. Although defendants compare the Fund to a compensation scheme created under the National Childhood Vaccine Injury Act of 1986, which was the basis for the New Jersey Supreme Court's decision that plaintiffs could not apply market share liability to recover for injuries allegedly resulting from a vaccination,^[82] the two schemes are vastly different in purpose, scope, and policy. While the vaccine compensation scheme was intended to provide individuals with a substitute for tort remedies, the Fund is intended to provide resources to pay for the prompt cleanup of oil discharges by the state.^[83] Unlike the vaccine compensation scheme, a reimbursement claim to the Fund does not preclude the claimant from pursuing tort remedies. Furthermore, claims to the Fund have historically been for amounts that are minuscule in relation to the damages plaintiffs seek here, averaging $15,000 to $20,000.^[84] Because it is doubtful that the Fund would provide a complete remedy to plaintiffs, it does not preclude application of the commingled product theory to prove claims against manufacturers.

D. Liability of MTBE Manufacturers: Lyondell and Equistar

Lyondell and Equistar manufactured MTBE and sold it to various gasoline refiners to be blended into gasoline. Unlike other defendants in this action, they never refined, distributed, or marketed gasoline, nor did they own or operate gasoline retail stations. Lyondell and Equistar's summary judgment motion argues that because plaintiffs cannot identify the manufacturer of the MTBE contaminating any of the focus wells, their claims must fail. Although their arguments are similar to those made by gasoline refiners in the omnibus motions, the facts are slightly different and merit separate discussion. Nevertheless, as discussed below, the commingled product theory applies to claims against Lyondell and Equistar,^[85] and operates in the same way as it does for claims against gasoline refiners.

"All MTBE that is blended into gasoline is chemically the same, regardless of the manufacturer."^[86] This Court has recognized that MTBE "lacks a `chemical signature' that would enable identification of the refinery or company that manufactured" it.^[87]

Evidence in the record establishes that Lyondell and Equistar sold MTBE to various refiners, including nearly all refiner defendants in this action, between 1979 and 2003.^[88] Moreover, the evidence discussed above in section B supports an inference that most of these refiners placed gasoline in the Colonial Pipeline for distribution to the New York regional market.

Lyondell and Equistar point out that many other companies manufactured MTBE for blending into gasoline, including refiners.^[89] Once it is blended into gasoline that is commingled in the distribution system, however, MTBE from various producers also becomes commingled. Indeed, MTBE from various producers is probably commingled before it enters the distribution system, because each refiner appears to have blended MTBE from multiple producers into its gasoline.

Plaintiffs cannot prove, either through direct or circumstantial evidence, that any particular molecules of MTBE contaminating their wells were manufactured by Lyondell or Equistar. A reasonable jury could conclude, however, that because refiners blended Lyondell and Equistar's MTBE into gasoline that was placed in the Colonial Pipeline or shipped to the New York Harbor, some of Lyondell or Equistar's MTBE was likely found in groundwater within the capture zones of various focus wells in Suffolk County. For this reason, under the commingled product theory Lyondell and Equistar can be held liable for the contamination in the wells, unless they are able to prove that their MTBE was not in the relevant place at the relevant time. Lyondell and Equistar's motion is therefore denied.

E. Individual Exculpation Motions

In addition to Lyondell, a handful of defendants—Crown, Getty, Giant, Irving and Total—have filed individual summary judgment motions. The parties have referred to these motions as "exculpation motions," presumably referring to the fact that under the commingled product theory a defendant may exculpate itself from liability.

Plaintiffs make two arguments in opposing these motions. Plaintiffs' first argument is that defendants have ignored "this Court's admonition that individual defendants avoid making individual motions if the omnibus motion addresses the same issues."^[90] Plaintiffs are correct that many of defendants' arguments repeat those that are made in the omnibus motion, particularly about their liability as retailers. This is especially true of the motion filed by Getty, which is only being sued as a retailer.^[91] All of the issues raised by Getty are made in the omnibus motion.^[92] The other defendants make the same error when focusing on their liability as retailers of gasoline. I have resolved these issues in Part V of this opinion, which discusses the liability of retailers for spills of gasoline alleged to have contaminated plaintiffs' wells.

Plaintiffs' second argument is that defendants have ignored commingled product theory as it has been developed in this case. For example, Irving argues that "plaintiffs cannot meet their fundamental burden under New York law of proving causation and product identification,"^[93] while Total contends that it "is Plaintiffs' burden to prove causation."^[94] Such arguments ignore the fact that under commingled product theory, the burden lies with the defendant to exculpate itself once plaintiffs have shown that the product was in a completely commingled or blended state at the time and place that the harm or risk of harm occurred. A defendant "must be able to exculpate itself by proving that its product was not present at the relevant time or in the relevant place, and therefore could not be part of the commingled or blended product."^[95] In this case, defendants have ignored their burden and thus failed to present evidence that would permit a reasonable jury to find that their product was not present in either the Colonial pipeline or the New York Harbor.

Certain defendants, however, do argue that they must be dismissed on the product liability claim because their product "was not present at the relevant time . . . and therefore could not have been part of the commingled or blended product."^[96] For example, Giant argues that it "should be granted summary judgment because it had not yet begun operations when the wells at issue in this case were allegedly contaminated with MTBE."^[97] In particular, Giant did not begin operations until May 14, 2002.^[98]

Plaintiffs agree that Giant cannot be held liable for spills or releases of gasoline with MTBE that occurred prior to that date.^[99] Plaintiffs note, however, that "Giant may still be liable for MTBE released and detected in wells after that date."^[100] For example, one of the eighteen focus wells, Crystal Brook Hollow No. 3, began operating in 2004. Because MTBE manufactured or otherwise distributed by Giant could have contaminated this well, plaintiffs may use the commingled product theory to pursue Giant for damages resulting from contamination in that well.

The same reasoning applies to the other defendants as well. On the one hand, if the jury finds that the contamination of a particular well occurred prior to the date the defendant entered the market, then that defendant may not be held liable as a matter of law. On the other hand, if the contamination of a particular well occurred after the defendant's product entered the market, then the jury must resolve whether that defendant is liable for damage to that well. Of course, such a determination will need to be made by the jury on a well-by-well basis given that the date of contamination is a fact-intensive question.

Finally, the individual defendants argue that they should not be liable because they manufactured a de minimus amount of MTBE or gasoline containing MTBE. Total claims that "the amount of gasoline containing MTBE [that Total] placed into pipelines that could reach the New York Harbor market represents an extremely small percentage of the total gasoline transported to that market,"^[101] and Crown argues "MTBE gasoline delivered into New York Harbor from 1979 to 2003 is infinitesimal relative to the total amount of gasoline that [was] delivered to New York Harbor during that time."^[102]

Yet, there is no de minimus exception to liability under the commingled product theory. Defendants concede that their product was present in the Colonial pipeline or the New York Harbor at some point and thus their product commingled with gasoline that was delivered to Suffolk.^[103] If a defendant supplied only a small amount of gasoline to the market, this should be reflected in the amount of damages assessed against that defendant.

The individual exculpation motions are therefore denied.

F. Proof of Claims Against Manufacturers Under Concert of Action Liability

Plaintiffs also argue that refiner defendants who are not implicated as a source of the spilled gasoline contaminating a particular well may still be vicariously liable for another defendant's actions causing contamination of that well under the concert of action theory.^[104] They assert that evidence regarding the operation of the gasoline distribution system, which requires companies to produce gasoline according to the same specifications, and lobbying activities by an industry association that may have concealed known risks associated with MTBE, establish that defendants acted in concert or pursuant to a tacit agreement.

To establish liability under concert of action, plaintiffs must show that: (1) each defendant acted tortiously; (2) defendants had an understanding, express or tacit, to participate in a common plan; and (3) at least one defendant committed a tortious act in furtherance of the plan that constitutes a tort.^[105] The alleged tortious act of each defendant is manufacturing and marketing a defective product, and one defendant's spill of gasoline causing contamination in a well constitutes a tort.

The New York Court of Appeals has held that parallel development and marketing of a product by various companies, in and of itself, does not constitute a common plan.^[106] Therefore, the mere fact that refiners all added MTBE to their gasoline in order to transport it as a commingled product through the pipeline system is not sufficient to prove concert of action.

A concerted effort by refiners to conceal known risks of MTBE from the government and the public, however, could constitute a common plan under this theory.^[107] To support their claim, plaintiffs have submitted various internal memoranda and letters to the EPA from the "MTBE Committee,"^[108] as well as several internal memoranda from Exxon in which an Exxon environmental specialist, Barbara Mickelson, describes the risk of groundwater contamination associated with increased use of MTBE in gasoline.^[109]

But plaintiffs have not identified any company, including any defendant in this action, as a member, funder or supporter of the MTBE Committee.^[110] Although all inferences must be drawn in favor of the non-moving party, the broadest inference to be drawn here is that the MTBE Committee was an industry association in which many oil refiners participated. Many of the remaining refiner defendants in this action are small companies, however. Irving Oil, for example, is a very large refiner but is based in Canada, and cannot be assumed to have been a member of the MTBE Committee. Giant could not have been a member because the company was formed in 2002, many years after the MTBE Committee made the statements at issue.

In addition, there is no evidence that ExxonMobil's knowledge of the risks posed by MTBE, as illustrated by the Mickelson memoranda, was shared with the MTBE Committee or other defendants in this action. Therefore, although the evidence indicates that Exxon knew about the dangers MTBE posed to groundwater and that an industry association represented to the government that there were no dangers involved in the use of MTBE in gasoline, it is not sufficient to support the conclusion that all refiner defendants participated in a common plan to conceal the risks of MTBE from the public.

V. LIABILITY OF RETAILERS FOR SPILLS AND SALE OF GASOLINE CONTAINING MTBE

As explained above, a jury may hold the manufacturers and refiners liable for contamination in a well regardless of whether the spill that caused that contamination can be identified. Nonetheless, it remains important for plaintiffs to prove which spills caused the contamination as the spillers may be liable for negligence in failing to prevent leaks, and as retailers of gasoline containing MTBE, they may be jointly and severally liable with manufacturers for products liability claims. Further, only the spillers may be liable for the discharge of gasoline under the New York Navigation Law.^[111] To prove that certain spills caused contamination in each well, plaintiffs rely largely on the report of their expert hydrogeologist, Charles Sosik. The expert report identifies a "capture zone" for each well that encompasses all potential sources of contamination. The expert report also identifies certain retail stations as the sources of contamination for some, but not all, wells within a "reasonable degree of scientific certainty."

Sosik's expert report also contains information about each spill culled from government and private investigations of the spills. This information includes, inter alia, investigation reports detailing the circumstances of the release, remedial steps taken, and the directional flow of groundwater from a spill site toward the well.

Defendants concede that when plaintiffs' expert has identified a specific source of the contamination, a reasonable jury could find that this retailer was the cause of plaintiffs' injury. Of course, this does not mean that the jury must reach such a conclusion as defendants may present evidence at trial that refutes plaintiffs' argument. Nonetheless, on this summary judgment motion, the parties agree that plaintiffs' expert report creates a fact issue on the issue of causation that the jury must resolve.

With respect to any claim in which the expert does not identify a source of the contamination within a reasonable degree of scientific certainty, defendants argue that no reasonable jury could hold them liable. In contrast, plaintiffs argue that non-expert information is still sufficient to support a jury's finding that a particular defendant caused the contamination of each well.

A reasonable jury could find that a particular retailer was the source of the contamination even if it was not specifically identified in the expert report. However, in some instances plaintiffs' non-expert information is insufficient to support a reasonable jury's verdict. The evidence regarding spills must be examined on a well-by-well basis. As explained below, the evidence is insufficient to support a finding by a reasonable jury that any known gasoline spills caused the contamination of three of the eighteen wells: (1) Horseblock Road Well No. 1, (2) Dare Road Well No. 1, and (3) Strathmore Court Well No. 1.

A. Releases From Underground Storage Tanks

Leaks from USTs are the primary sources of MTBE groundwater contamination.^[112] Before federal regulations of USTs went into effect in 1988, the EPA estimated that between ten and thirty-five percent of USTs were leaking.^[113] "Until the mid-1980s, most USTs were made of bare steel, which is likely to corrode over time and allow UST contents to leak into the environment."^[114] Because of UST leaks and other spills, "[e]ach year approximately 9 million gallons of gasoline (the equivalent of a full supertanker) are released to the environment in the United States from leaks and spills, according to an estimate by the Alliance for Proper Gasoline Handling."^[115]

Federal regulations adopted in 1988 required substantial upgrades to most of the nation's USTs, most importantly the installation and maintenance of leak detection systems.^[116] However, for many reasons UST owners were slow to upgrade their tanks to comply with the regulations.^[117] Even after 2000, when an estimated eighty-nine percent of all USTs had received the mandated upgrades, as many as twenty-nine percent of USTs "were not being operated or maintained properly, increasing the risk of soil and groundwater contamination."^[118]

Without properly functioning leak detection systems, leaks from USTs are rarely noticed immediately because the gasoline contaminates soil beneath the surface and is not visible. Most UST leaks are only discovered well after they began, either when groundwater impact is reported or when excavation occurs for other reasons. Sosik notes that "[i]n the vast majority of cases the spill is discovered when gasoline impacted soil is encountered during the removal or upgrade of an underground storage tank. No one knows when the leak began, how long it had been leaking or the volume of fuel lost."^[119]

A recent study of MTBE contamination in Long Island illustrates the prevalence of unreported leaks. The study, funded by the EPA, was designed to "better define the extent of MTBE contamination stemming from previously unidentified and/or unreported MTBE blended gasoline releases."^[120] Researchers inspected USTs at fifty-two gasoline retail stations in Nassau and Suffolk Counties that had no known prior releases of gasoline, and sampled groundwater at the stations for MTBE contamination.^[121] Thirty-two previously unknown gasoline releases were discovered at the fifty-two sites, meaning that over fifty percent of the sites had released gasoline without reporting it to the state.^[122] In addition, MTBE was found in groundwater at levels over the New York State maximum contaminant level ("MCL") of 10 ppb in thirty-four percent of stations in Suffolk County and fifty-three percent of stations in Nassau County.^[123]

B. Migration of MTBE in Groundwater

MTBE gasoline released from USTs initially leaks into the soil surrounding the tank. Because MTBE is highly soluble in water, it quickly dissolves into groundwater when rain or other water passes through the contaminated soil. Once MTBE dissolves into the groundwater it migrates underground away from the release site:

As rain and other water move through the soil, water-soluble contaminants may be carried along with the moving water. The region of contamination created by such a water flow is called a "plume." Plumes vary in size and shape based on, inter alia, the terrain, the qualities of the local soil, and the speed and volume of water flowing through the soil. To determine the size and direction of a plume is to "delineate" the plume. Eventually, if a plume reaches an underground basin or other aquifer from which water is drawn, the water in that aquifer will become contaminated by the plume. Thus, any wells drilled into that aquifer would produce contaminated water.^[124]

According to Sosik, a key characteristic of plume development observed on Long Island is that "MTBE plumes continue to advance and do not stabilize or retreat."^[125] These plumes are generally narrow and travel far from a spill site due to the characteristics of the soi1.^[126] Sosik states that in cases where there was either a large initial release or a continuous smaller release, the plume will remain attached to the spill site, whereas contamination stemming from "older, smaller volume releases will travel as a detached plume. In almost all cases a detached plume will go undetected until it impacts a private or public supply well or is accidentally discovered in a monitoring well installed for some other purpose."^[127]

C. How Plaintiffs Identified Sources of Contamination

The key evidence supporting plaintiffs' claims is Sosik's expert report, which:

provide[s] opinions relating to the hydrogeology of the Long Island aquifer; the source(s) that have contributed and are contributing to the contamination of wells in the Suffolk County Water Authority drinking water well system; the timing of releases at the source areas and arrival at the wells; the extent of MTBE contamination in a given well; and the anticipated duration of the MTBE contamination in the well.^[128]

Phase I of Sosik's expert report addresses source attribution. Sosik gathered a list of all gasoline spills in Suffolk County that were reported to the New York State Department of Environmental Conservation ("NYSDEC").^[129] He then analyzed each of the known spills that occurred within each well's "capture zone," and based on certain factors determined which of the gasoline releases are the source of MTBE in each of the eighteen wells.

One of the key factors in Sosik's analysis was whether the gasoline spilled at each site had contaminated the groundwater directly beneath the spill or elsewhere on the gasoline station property.^[130] Because only those spills where gasoline constituents entered the groundwater at the spill site could cause contamination in a well some distance away, Sosik "reviewed each file for confirmation of groundwater contamination with gasoline constituents and eliminated those which did not have such confirmation in documents contained in the files."^[131] At many spill sites, however, neither the gas station owners nor the NYSDEC investigators tested the groundwater to determine whether gasoline constituents had leached from gasoline-contaminated soil into the groundwater. Sosik therefore "had to eliminate suspect sites, sites with confirmed gasoline releases to the subsurface" because of the lack of thorough investigation and/or documentation by the responsible party and the NYSDEC at the time the spill was discovered.^[132]

Factors Sosik considered when analyzing spills where testing confirmed the presence of gasoline constituents in groundwater at the site included:

BTEX^[133] and MTBE concentrations within the source area and at the property line, the presence of free phase gasoline, the degree of residual contamination in soil, the nature and date of the release, if known, the direction of groundwater flow, the distance from the site to the wellfield, type, degree, start date and success of remedial actions both at the site and off-site, if implemented, the use or lack of control measures to prevent off-site migration of MTBE from the site and timing and correlation with MTBE detections in the test well and other wells within the wellfield or in the area.^[134]

Based on his analysis of the available information, Sosik concluded that certain spills were likely sources of contamination in nine of the eighteen focus wells, but opined that there was insufficient information for him to determine which gasoline spills were the source of MTBE in the remaining nine wells, although these wells are known to be contaminated with MTBE.^[135]

It is important to understand that Sosik eliminated many "suspect sites ... with confirmed gasoline releases to the subsurface" only because there was no evidence allowing him to confirm that the spills had contaminated the groundwater.^[136] Sosik did not conclude that these spills had not caused contamination in the wells. Indeed, other information exists about these releases with respect to many of the factors Sosik evaluated for spills with confirmed groundwater contamination, such as "the degree of residual contamination in soil, the nature and date of the release, if known, the direction of groundwater flow, the distance from the site to the wellfield, type, degree, start date and success of remedial actions both at the site and offsite, if implemented."^[137] These spills may still be the cause of contamination in some of the wells. Or, an undiscovered release may have caused the contamination, because most gasoline leaks occur beneath the ground and are discovered only when a site is excavated or investigated for some reason.^[138]

Another fact important to the issues in these motions is when the gasoline releases occurred. Dates of releases are difficult to estimate for several reasons. Sosik notes that "releases of MTBE-containing gasoline are rarely a single, isolated release. Rather, releases at source sites typically involve more than one release over a period of time."^[139] Further, most releases are only discovered long after they began, so the date of the spill report to the NYSDEC does not reflect the date of the actual spill. Despite these uncertainties, Sosik estimates that "[i]n the vast majority of the identified source sites, the most likely release date was in the mid to late eighties and early nineties."^[140]

D. Plaintiffs' Evidence Linking Gasoline Spills to Well Contamination

Defendants acknowledge that Sosik's identification of spills at certain gas stations that caused or contributed to contamination in certain wells creates a fact issue with respect to whether the owner or operator of those gas stations caused contamination in the well.^[141] Accordingly, defendants whose gas stations are identified by Sosik as sources of contamination in certain wells do not seek summary judgment on claims arising from those wells. Rather, defendants argue that if plaintiffs' expert does not conclude that a defendant caused contamination in a well, neither could a reasonable jury.

Plaintiffs respond that additional evidence about other known gasoline releases within each well's capture lone—reports from spill investigations, information regarding the circumstances of the release, the directional flow of groundwater from a spill site toward the well—could support a jury's conclusion that these other gasoline releases caused or contributed to MTBE contamination in a well, even though Sosik did not reach that conclusion. The issue for the Court is therefore whether the totality of plaintiffs' evidence, without expert testimony, could support a jury finding that known gasoline releases within a well's capture zone caused contamination in a well.

Plaintiffs' argument rests on the methodology Sosik employed to reach the conclusions in his report. Although Sosik did not consider gasoline releases where the groundwater at the release site was never tested for gasoline contamination, he never concluded that these spills could not have caused or contributed to contamination in the nearby wells. Plaintiffs point out that he simply was unwilling to "confirm" a spill as a source of contamination without conclusive proof that gasoline constituents entered groundwater at the spill site.^[142]

There appears to be no reason why groundwater was tested at some spill sites but not at others.^[143] But while the presence of gasoline constituents in groundwater at a spill site helps prove that the spill contaminated a plaintiffs well water, because it shows that gasoline from that site has reached the area's groundwater, such evidence is not required to prove causation. Taken together, other circumstantial evidence about a gasoline release could support a finding that a particular spill of gasoline within a well's capture zone caused or contributed to MTBE contamination in that well. Such evidence includes the direction of groundwater flow from the spill site toward the well, the estimated volume of gasoline released, whether the release occurred on the ground's surface or beneath the ground, where it would be more likely to affect groundwater as it reached greater depths and/or as rainwater passed through the contaminated soil, the correlation between the estimated date of the release and the date of MTBE detections in the well, and whether and to what extent the spill was cleaned up.

Neither party supports its argument with any cases discussing summary judgment of groundwater contamination claims based on the failure to prove causation. Defendants quote a product liability case, Healey v. Firestone Tire & Rubber Co., stating that "circumstantial evidence ... must establish that it is reasonably probable, not merely possible or evenly balanced, that the defendant was the source of the offending product."^[144] The facts of that case, however, shed little light on this motion.^[145] This Court has found only a few cases on point, from other jurisdictions, but those cases show that courts have found various types of circumstantial evidence linking a defendant's spill to plaintiffs contamination sufficient to create an issue of material fact for a jury.^[146] Circumstantial evidence of causation has included the proximity of the alleged spill site to the plaintiff's well,^[147] the direction of groundwater flow,^[148] and indications of a gasoline leak such as odors or stained soil at the spill site^[149] or unexplained financial losses or inventory discrepancies at the gas station.^[150]

Further, in many cases the issue at summary judgment is whether defendant actually released gasoline,^[151] whereas the spills at issue here are all known releases of gasoline documented by the NYSDEC, at sites from which Sosik has determined that groundwater flows toward a wellfield.^[152] The NYSEC investigation reports for each release generally document the amount of contaminated soil found beneath a UST, the date of its discovery, and remediation efforts, if any, taken to prevent offsite migration of gasoline constituents. Based on the totality of the evidence about a spill, a reasonable jury could conclude that at certain spills not identified by Sosik as contamination sources, MTBE from the gasoline found in the soil beneath a UST dissolved into groundwater and migrated toward the wellfields, and was later detected in well water.

However, plaintiffs' argument that fact issues exist with respect to all potential release sources listed in Sosik's report, including spills that Sosik eliminated as sources for various reasons and registered petroleum bulk storage ("PBS") facilities where no known release occurred, is too broad. Sosik concluded that some spills did not affect groundwater. As for certain PBS tanks, there is no evidence that a leak ever occurred. Therefore, the facts about each spill must be analyzed independently. When there is no evidence that a spill caused contamination in a given well, claims arising out of contamination in that well against the defendant who spilled the gasoline must be dismissed.^[153]

The analysis of each spill below employs the following standard: a reasonable jury could find that a spill within a well's capture zone caused or contributed to contamination in the well if the evidence shows that gasoline has contaminated soil beneath the surface of a spill site, allowing MTBE to reach the groundwater that flows toward the well. In most instances, Sosik identified the spills for which such evidence exists and opined that they caused the contamination. However, such evidence exists for certain spills that Sosik did not identify as causes of contamination.

Because other evidence about the spills exists, the fact that Sosik did not identify them as sources is not dispositive. Defendants' argument ignores the fact that the burden of proof in civil cases is only a preponderance of the evidence. Scientific conclusions often require a higher level of certainty, and Sosik's relatively cautious methodology requiring confirmed groundwater contamination reflects such a scientific "reluctance to quantify [his] judgments as to cause and effect."^[154] The tension between standards of certainty in science and in the law has been often noted, especially in cases where scientific evidence is necessary to prove causation.^[155] The courts have acknowledged that the law imposes liability even where "the cause and effect relationship ... [can] not be established with scientific certainty."^[156] This is because "a court proceeding ... is not simply a search for dispassionate truth."^[157] Unlike science, the law is focused on "resolving social problems . . . . [its] concern is whether tort and injury bear a close enough relationship to make it equitable to impose financial responsibility upon a defendant."^[158]

In addition, for most of the wells described below, the jury will consider whether more than one spill caused or contributed to contamination in the well. If the jury finds that more than one spill contaminated a particular well, the retailer defendants may be held jointly and severally liable under the theory of concurrent wrongdoing.^[159] Where a small number of defendants has been shown to have contributed to plaintiffs injury, "`although the act of each, alone and of itself, might not have caused the entire injury,'" it is not fundamentally unfair to hold them jointly and severally liable."^[160]

To summarize, where plaintiffs present evidence that a defendant—or a gas station operating under a defendant's brand, or a station to which a defendant supplied gasoline—released gasoline within a well's capture zone, and the release was of a volume and depth making it likely that MTBE from the gasoline reached the groundwater flowing toward the well, a genuine issue of material fact exists as to whether that defendant caused MTBE contamination in the well. I now turn to the specific evidence regarding spills within the capture zone of each well.

E. Well by Well Analysis^[161]

This analysis includes both wells where Sosik identified certain spills as sources and wells where he concluded there was insufficient information to identify a source. It is important to highlight two points. First, MTBE has been detected in all of the wells at issue, even if the groundwater at the spill sites was never tested for MTBE. Second, because the majority of the spills occurred underground, they were almost never reported contemporaneously. Therefore, the date that a spill was reported to the NYSDEC is almost always significantly later than the date that the spill actually occurred, with the exception of several observable releases such as an overturned tanker truck or intentional dumping of gasoline into a drain.

1. Wheeler Road Well No. 1

MTBE was first detected in Wheeler Road Well No. 1 in 1989, and detections continued intermittently through 1994.^[162] From. 1995 through the present, MTBE detections in the well have been continuous.^[163]

Five reported spills occurred within the capture zone of Wheeler Road Well No. 1 at two locations, both relatively close to the well.^[164] One location is the site of a gas station operated at different times by Texaco-Shell and Citgo; the other site was an Exxon/Mobil station.^[165]

At the first location, a spill was reported in 1991 when gasoline was discharged above the ground and washed down a drywell.^[166] The NYSDEC files do not document any groundwater contamination as a result of the 1991 spill.^[167] A later spill report was made in 2002 when an investigation at the time of Texaco-Shell's sale of the property to Citgo revealed high levels of BTEX and MTBE in groundwater at the site.^[168] There are no records of attempts to remediate or contain the contamination. Finally, another spill was reported at the same site in 2004 when monitoring based on the 2002 detections indicated a new release of gasoline had occurred.^[169] There is no further discussion of the 2004 spill in Sosik's report.

Two spill reports from the Exxon/Mobil station in 1995 and 2001 appear to be related to the same discovery of contaminated soil, which initially occurred in 1995 during a site investigation.^[170] Groundwater contamination was discovered during continuing investigations at the site in 1996 and 1999, at very high levels. Although some remediation occurred, it was not undertaken until 2001, and Sosik opined that it was not sufficient to prevent the migration of MTBE away from the site.^[171] Over two hundred tons of contaminated soil were eventually removed from the site in 2004.^[172]

In light of the confirmed MTBE contamination in groundwater at both spill sites, Sosik opines that spills at both sites have caused the MTBE contamination in the Wheeler Road well.^[173] A reasonable jury could conclude that one, or both, of these spills caused the contamination of this well.

2. Virginia Avenue Well No. 1

MTBE was first detected in the Virginia Avenue well in 1993, but was not detected again until 1997, after which detections continued intermittently through 2000.^[174] Beginning in 2001, detections increased in frequency and magnitude and have continued through the present.^[175]

Seven spills have been reported to the NYSDEC in the vicinity of Virginia Avenue Well No. 1. Sosik reports that five of those spills, which occurred at a Sunoco station (reported in 1990), a former Rally/Amoco station (reported in 1986 and in 1996), and a Mobil station (reported in 1986), caused or contributed to the well's contamination.^[176] High levels of MTBE and/or BTEX were detected in the groundwater at each of these spill sites, either at the time the spill was reported or upon further investigation. Based on the groundwater contamination at the spill sites and the flow of groundwater from the sites toward the well, Sosik concluded that spills at the Sunoco, Rally/Amoco, and Mobil stations all caused or contributed to MTBE contamination in the Virginia Avenue well.

For various reasons, no reasonable jury could find that the two remaining spills contributed to the contamination. One spill occurred at a high school bus garage downgradient from the well, meaning MTBE would have been transported away from the well.^[177] At the other spill site, a Shell station, an investigation revealed very low levels of MTBE in groundwater at the site, indicating insufficient levels of contamination to create a migrating plume.^[178] Plaintiffs also offer evidence showing that Shell supplied over four million gallons of gasoline between 1998 and 2003 to a gas station referred to as the "Lake Grove S/S" at 2822 Route 25, where there are registered PBS tanks but no reported release.^[179] Because there is no known release from this station, no reasonable jury could conclude that it caused the contamination in the Virginia Avenue well, and therefore Shell's delivery of gasoline to the station is immaterial.

Therefore, a reasonable jury could only find that those spills identified by Sosik — the Sunoco, Rally/Amoco, and Mobil station spills—could have caused the contamination in the Virginia Avenue well.

3. Kayron Drive Well No. 1

Low-level detections of MTBE have occurred regularly in the Kayron Drive well beginning in mid-2000.^[180] Two gas stations about five hundred feet apart are located in the capture zone of the well. Four spills—two reported in 1984, one in 1990, and one in 1998—have occurred at one site, which was a Chevron station, then a U.S. Oil station, and eventually a Coastal station, and one spill was reported in 2002 at a Getty station.^[181]

In 2005, a NYSDEC investigation of the Getty station spill revealed MTBE concentrations in groundwater as high as 130,000 ppb.^[182] The agency investigated MTBE migration by installing monitoring wells outside the station property, and determined that the station is a source of MTBE contamination in the area.^[183] Based on the NYSDEC investigation, Sosik concluded that the Getty station is the cause of contamination in Kayron Drive Well No. 1. Plaintiffs offer evidence showing that Shell sold 329,000 gallons of gasoline to this Getty station in 1980.^[184] While this evidence supports an inference that other companies sold gasoline to this station, the twenty year time gap between the Shell deliveries (established through Shell records) and the discovery of the spill in 2000 is too great to support the inference that Shell delivered the gasoline that spilled from the Getty station and was discovered in the soil in 2002. Therefore Shell cannot be liable for its role as a supplier of gasoline to the Getty station.

At the other site, the two spill reports from 1984 when the station was run by Chevron were related to a tank test failure, but there was no further information in Sosik's report. In 1990, however, USTs were removed at this station (then a U.S. Oil station) along with a significant amount of contaminated soil surrounding the tanks.^[185] Soil testing indicated that the contaminated soil ended about thirty feet above the water table, making groundwater contamination unlikely.^[186] However, the NYSDEC file reported that some contaminated soil was left at the site.^[187] Another spill was reported at this site in 1998, when it was a Coastal station, after gasoline spilled during a tank overfill and entered a storm drain in the street.^[188] No information about an investigation was in the spill file.

A reasonable jury could find that the spills reported in 1990 and 1998 contributed to the contamination in Kayron Drive Well No. 1, along with the spill at the Getty station. The timing of the detection in the well in 2000 corresponds with the 1998 spill at the Coastal station where gasoline washed down a drain. The fact that the 1990 excavation at the same site (then a U.S. Oil station) did not fully remove the contaminated soil beneath the ground's surface makes it likely that MTBE absorbed into groundwater when rain passed through the soil. The site is the same distance from the well as the Getty station, and groundwater flows from the site toward the Kayron Drive well. This evidence would allow a reasonable jury to find that the spills at the U.S. Oil/Coastal station contributed to contamination in the well along with the spill at the Getty station.

4. Lakeview Avenue Well No. 1

Very low levels of MTBE were detected in the Lakeview Avenue well beginning in 2000.^[189] In mid-2001, MTBE levels in the well rose slightly, and detections remained continuous to the present.^[190]

Although multiple spills were reported in the vicinity of Lakeview Avenue Well No. 1, a good number of them were at locations downgradient from the well. As a result, Sosik eliminated those spills as possible sources of contamination even if there was confirmed groundwater contamination at the spill site.^[191] The only site upgradient of the well, a Mobil station located at 5665 Sunrise Highway, reported three spills. The first reported spill, in 1996, was a minor surface spill that was very unlikely to have affected the groundwater.^[192] The other two spills, reported in 1997 and 1998, were discovered during a site assessment in 1997 that revealed soil and groundwater contamination.^[193] An investigation in 1998 revealed MTBE contamination in groundwater at the site at levels as high as 320,000 ppb.^[194] Despite some remediation efforts, which Sosik opined were not sufficient to prevent MTBE migration, Sosik identified these two Mobil station spills as the likely cause of MTBE contamination in the Lakeview Avenue well.^[195] Because this is the only site upgradient of the well, none of the other reported spills within the capture zone could support a finding of causation.

5. Montauk Highway Well No. 1A

Initial MTBE detections at the Montauk Highway well occurred in 1997, and later detections occurred on a more regular basis in 2000, 2005, and 2006, all at relatively low levels.^[196] No less than fifteen spills at ten locations have been reported within the capture zone of Montauk Highway Well No. 1 A. Sosik concluded that a spill at the Shell station located at 4076 Sunrise Highway caused the MTBE contamination in the well. The spill was reported in 2001, when a site assessment revealed MTBE and BTEX in groundwater that appeared to be migrating away from the site and toward the well.^[197] Because the contamination was discovered during a site assessment, the gasoline release itself occurred some time before 2001. Sosik also concluded that releases from four other stations where gasoline constituents were found in the groundwater—a Mobil, Amoco, Sunoco, and a second Shell station— have most likely contributed to the contamination.^[198] The spill reports from these stations all date to the early 1990's, when the Department of Transportation was preparing to widen the road. Although some remediation was initiated at the Shell, Amoco, and Mobil stations, Sosik opined that it was insufficient to prevent the migration of MTBE.^[199]

The file on one spill that may have also contributed, a Gulf station, was not found and therefore there is no evidence to support a finding that this spill contributed to the contamination.^[200] Testing confirmed that spills much closer to the well at a second Gulf station, a Texaco station, and an Aeroliner Marine station all had contaminated groundwater, but the stations are located downgradient from the well.^[201] Therefore, no spills other than those identified by Sosik as contributors—the two Shell stations, the Sunoco station, the Amoco station and the Mobil station— could have caused MTBE contamination in the Montauk Highway well.

6. Samuel Street Well No. 4

MTBE was detected in the Samuel Street well in 1989 and probably contaminated the well even before that time.^[202] Early concentrations were quite high, and peaked at 100 ppb in late 1989.^[203] After that, concentrations dropped to below 5 ppb through mid-2000, when MTBE levels began to increase again. In late 2001 and 2002, MTBE levels peaked again at 59 ppb and then 114 ppb; after that, they stabilized at levels between 13 to 45 ppb from 2002 to the present.^[204]

Eleven spills at seven locations were reported within the capture zone of the Samuel Street well, but Sosik eliminated four locations from further consideration because they were cross-gradient and not close enough to the well. Another spill reported in 1998 at a CJS station was eliminated, as it apparently involved the dumping of a "yellow liquid" into a storm sewer.^[205] Sosik and the NYSDEC have both identified a spill reported in 2001 at a Citgo station as a source of the MTBE contaminating the Samuel Street Well No. 4 from mid-2000 to the present.^[206] The NYSDEC investigated the plume migrating away from the Citgo station and determined that the groundwater contamination was moving off-site toward the Samuel Street well.^[207]

Plaintiffs also offer delivery records from the Island Transportation Corporation and bills of lading from Tosco Pipeline Company, purporting to show that Exxon-Mobil delivered gasoline to this Citgo station in 2002 and 2003, and that Gulf delivered gasoline to the Citgo station in 2001 and 2003.^[208] However, these records cannot support an inference that ExxonMobil delivered gasoline to this station, because neither the Island Transportation nor the Tosco records mention ExxonMobil as a transferor or deliverer of gasoline.^[209] Nor can they support an inference that Gulf delivered its gasoline to this particular station, because while the Tosco records show that Gulf loaded or purchased gasoline, they do not identify a delivery destination.^[210] Because there is insufficient evidence that either Gulf or ExxonMobil delivered gasoline to the Citgo station, they cannot be liable for their role as suppliers of gasoline to this station.^[211]

In addition to the Citgo spill, a spill was reported in 1990 at a Metro station located 689 feet from the well, and another spill was reported in 1991 at a Northville station located 724 feet from the well (the location that was later occupied by the Citgo station discussed above). Both stations are upgradient from the Samuel Street well.^[212] At each site, approximately ten cubic yards of contaminated soil were discovered beneath leaking USTs and removed, but no groundwater testing was performed.^[213] It is unclear why NYSDEC closed the files for these spills.

Sosik did not identify these sites as sources of contamination at Samuel Street because groundwater at the sites was never tested for gasoline constituents. However, evidence shows that a significant amount of gasoline leaked into soil beneath the ground's surface, making it likely that MTBE absorbed into groundwater passing through the soil. Further, MTBE detections in this well date back to 1989, before the Citgo station existed.^[214] The timing of both of the spill reports correlates with a 1989 detection, and these sites are located in close proximity to the well.^[215] Therefore, a reasonable jury could find that in addition to the spill at the CITGO station, the spills at the Metro station and the Northville station contributed to contamination in the Samuel Street well.^[216]

7. Church Street (Holbrook) Well No. 2

An isolated detection of MTBE occurred in 1991 in the Church Street well; intermittent detections occurred in 1998 and 1999; and MTBE has been detected more frequently from 2000 to the present, although at low levels.^[217]

Little dispute exists about the source of the MTBE in the well, because only one service station, a Sunoco, is located within its capture zone.^[218] A significant release of gasoline at this station was reported in December 1989, and MTBE concentrations at the downgradient property line of the station were as high as 17,800 ppb in 1991. Although remediation was performed, Sosik opines that it was only partially successful, and that "[c]haracterization efforts by Sunoco's consultants over the years have conclusively demonstrated that spills which have occurred at this station have impacted the wellfield."^[219] Sosik concludes that the Sunoco spill is the source of the contamination in this well.

8. College Road Well No. 2

The earliest MTBE detections at the College Road well occurred in 1993 and 1995 at low levels.^[220] Beginning in 1998, detections became continuous and increased in magnitude, peaking in 2000 at about 5 ppb, and have continued to the present.^[221]

Nine spills have been reported A eight locations within Sosik's predicted capture zone for College Road Well No. 2.^[222] Of these locations, an Amoco station had to be eliminated because it was located downgradient from the well, and a Mobil station was also eliminated because it was east of the groundwater divide, meaning groundwater would not flow toward the well.^[223]

Groundwater contamination was confirmed in 2002 at a Shell station when a contractor installed a monitoring well as part of a site assessment and discovered "an extensive free-phase gasoline lens" on the property.^[224] Groundwater contamination was also confirmed at a Sunoco station after a spill originally reported in 1990 was more thoroughly investigated in late 1991.^[225] Sosik identified both spills as sources of the contamination in the well.

In addition to the Sunoco and Shell spills, contaminated soil was encountered at spills reported at Centerreach Car Care in 1990, a Texaco station in 1999, and a Department of Transportation site in 1989.^[226] A soil boring at the Texaco site indicated that the contaminated soil did not reach the water table, which eliminates the Texaco spill, but no similar testing was performed at the other sites. Based on the known releases of gasoline into the soil, the correlation of the dates of the spill reports with early, low-level detections in the well, and the location of these sites upgradient from the well, a reasonable jury could conclude that the spills at the Department of Transportation site and at Centerreach Car Care contributed to the contamination in College Road Well No. 2 along with the spills at the Sunoco and Shell stations.

9. Wicks Road Well No. 1

MTBE has been detected intermittently and at low levels in the Wicks Road well beginning in the Fall of 2000 and continuing to the present.^[227] Ten gasoline spills have been reported at six locations within the capture zone of the well.^[228]

In 1999, a Mystic tanker truck overturned in the vicinity of the well, releasing nine hundred gallons of gasoline and two thousand gallons of diesel fuel into storm drains on the street, from which it discharged into a drywell and a recharge basin.^[229] Later investigations at the site of the drywell revealed the presence of a migrating MTBE plume.^[230] Sosik concludes that the Mystic spill is a source of the MTBE in the Wicks Road well. Plaintiffs also offer evidence showing that Texaco supplied the gasoline that was in the Mystic tanker when it spilled,^[231] and on the basis of this evidence Texaco has withdrawn its motion for summary judgment as to this well.^[232]

At a Northeastern gas station, a spill was reported in 1991 after a large amount of contaminated soil was encountered during a UST removal, and groundwater testing indicated that MTBE was migrating away from the site.^[233] Sosik concludes that the Northeastern spill is also a source of the MTBE in the Wicks Road well. Plaintiffs also offer evidence purporting to show that ExxonMobil delivered gasoline to the Northeastern station during 2001 and 2002,^[234] on the basis of which Exxon-Mobil has withdrawn its claim for summary judgment as to this well.^[235] Plaintiffs claim that delivery tickets also show that Gulf delivered gasoline to the Northeastern station on three occasions in 2001, but there is no reference to Gulf on any of the delivery tickets or bills of lading they submitted in support of this claim.^[236] Therefore, ExxonMobil may be liable for damages caused by the Northeastern spill in its role as a supplier of gasoline, but Gulf may not be held liable for damages caused by this spill as a supplier.

Other spills occurred at a Mobil station and a Shell station, but Sosik provides no information about the nature of these spills other than the fact that no groundwater contamination was confirmed at those sites.^[237] Without any further information about these spills, such as the approximate volume of gasoline released or whether they affected the subsurface, a jury could not find that they caused the well's contamination. Another spill at an Exxon station, reported in 1989, was determined by Sosik to be the "least likely to have affected the Wicks Road wellfield" because although contaminated soil was encountered, there was only one isolated detection of low levels of MTBE in a monitoring well.^[238]

Evidence about one other spill, not identified as a source by Sosik, could support a conclusion that gasoline from the spill contributed to the well's contamination. A spill was reported in 1991 at a New York State Department of Transportation ("NYSDOT") site when contaminated soil was encountered during a UST removal and testing revealed very high levels of gasoline constituents in the soi1.^[239] Unlike the Mobil and Shell spills discussed above, the NYSDOT spill clearly affected the subsurface, and the site is located upgradient from the well. Therefore, in addition to the Mystic and Northeastern spills identified by Sosik, for which Texaco and ExxonMobil may share responsibility as suppliers of the spilled gasoline, a fact issue also exists with respect to the NYSDOT spill's contribution to contamination in the Wicks Road well.

10. Church Street (Bohemia) Well No. 1

MTBE was first detected in the Church Street well in the Summer of 1989 and continued through the end of the year.^[240] Detections began again in mid-1995, and continued intermittently until 1999. From 1999 to the present, low-level detections have been continuous^[241] Seven spills have been reported at three different locations within the capture zone of the Church Street well. A spill reported at Ellanef Manufacturing in 1989 appeared to be related to the discovery of oil rather than gasoline in a storm drain, and was therefore eliminated from consideration.^[242]

Four spills were reported at a gas station—initially a Texaco and later a Shell station—between 1990 and 1993.^[243] The first spill report in 1990 was due to a spill of "petroleum" into the soil beneath the station during cleaning of gasoline dispenser equipment.^[244] Later spill reports were related to a tank test failure. No investigations were performed as a result of these reports. Sosik did not identify spills at this site as sources of contamination because no groundwater testing was performed, but he stated that they were "suspect."^[245] The tank test failures in 1992 and 1993 are evidence that gasoline was leaking out of a UST into the soil, even if the tanks were never excavated and the contaminated soil was not uncovered. Further, the dates of the spill reports in 1992 and 1993 correlate with the 1995 MTBE detections in the well. The circumstances and timing of the spills, together with the location of the site upgradient from the well, could support a jury's conclusion that the spills at the Texaco/Shell station contributed to contamination in the well.

Finally, two spills were reported at an Empire gas station, the first in 2002 and the second in 2003.^[246] During cleanup of the 2002 spill, a surface release of thirty gallons of gasoline, a NYSDEC representative noticed that the UST leak detector alarm system was not working.^[247] After further investigation in 2003, a second spill was reported upon the discovery of contaminated soil around the tanks and MTBE in groundwater.^[248] Neither the station nor the NYSDEC attempted to remediate the spill or prevent MTBE migration.^[249] Although the precise date of the release at this station is unknown, Sosik opines that the groundwater data from 2005 and 2006 are consistent with a spill within the five years prior to testing.^[250] In his estimation, the Empire spill occurred between 2000 and 2003, and therefore could not have caused the 1989 MTBE detections in the well.^[251] However, he states that MTBE has left the site and is expected to reach the well in the future.^[252] Therefore a reasonable jury could conclude that the Empire spill caused more recent contamination, in addition to the Texaco/Shell spills.

11. Broadway Well No. 2

MTBE has been detected intermittently in the Broadway well from early 2000 to the present.^[253] Although ten spills were reported between 1990 and 1998 in the capture zone of the well, Sosik concluded that there was insufficient information to identify any spill as a source of the contamination, since no groundwater testing was performed at any of the spill sites.^[254] Nonetheless, evidence indicates that certain spills are likely to have contaminated the groundwater at the spill site and could therefore have caused contamination in the well.

A former Mobil station located about six hundred feet from the well reported three spills in 1992 and 1994, all of which involved contaminated soil. Soil borings from the investigation of the 1992-reported spill revealed the presence of gasoline constituents at twenty-five to thirty-five feet below the surface.^[255] In 1994, spills were reported when contractors discovered contaminated soil beneath the surface during construction at the site.^[256] Due to the proximity of the station to the well and the presence of contaminated soil at these spill sites, a reasonable jury could conclude that the Mobil station spills contributed to contamination in the well.

Other spills in the area also affected the soil beneath the surface. At a Coastal station, two spill reports were made in 1996 when a gasoline tank was overfilled, and an attendant washed the gasoline down a storm drain with water.^[257] At an Advance station, contaminated soil was discovered in 1991 during a UST removal.^[258] Contaminated soil was also discovered during a UST removal in 1998 at another Mobil station further away from the well, and three hundred cubic yards of contaminated soil were excavated.^[259] Evidence that gasoline contaminated the soil beneath the surface or was mixed with water and washed down a drain at each of these spills makes it likely that gasoline contaminated groundwater, which flows from each site toward the well.

In addition, in their opposition to defendants' motion for summary judgment as to the Navigation Law claims, plaintiffs offer evidence regarding gasoline contamination at a Getty station located at 734 Park Avenue. According to plaintiffs, this contamination was never reported to the NYSDEC, and Sosik does not discuss this spill in his report.^[260] The evidence consists of a report by Tyree Brothers Environmental Services that was purportedly submitted to the Suffolk County Department of Health Services ("SCDHS").^[261] The report describes "baydrain drywell closure activities" performed under the auspices of the SCDHS.^[262] Although the report's language is technical and it is unclear exactly what type of leak, spill or dumping of gasoline occurred, the report does describe "levels of volatile organics and metals" in a drywell that "exceeded the maximum levels set by the NYSDEC," and attempts to excavate contaminated soi1.^[263] This language creates a fact issue as to whether a discharge of petroleum occurred at the Getty station that led to contamination at the site. In addition, because the station is adjacent to one of the Mobil stations described above, it is likely that the groundwater also flows from this station toward the well.

Other spills in the capture zone cannot support a finding of causation because there is no indication that gasoline constituents from these spills would have entered the groundwater. Plaintiffs argue that the jury should also be able to consider as possible causes an incident in which gasoline and automotive fluids spilled onto the ground at Alive Parts in 1994, the discovery of "minimal contamination" in soil during removal of USTs at Norman's Service Station in 1990, and a tank failure that was traced to an "air pocket" at a Tartan station in 1991. Because there is no evidence of significant soil contamination beneath the surface or other means by which these spills would have been likely to have contaminated groundwater, however, the evidence about these spills is insufficient for a jury to conclude that they caused the contamination.^[264] Therefore, even though Sosik concluded that there was insufficient information to conclude that any spills were a source of the contamination, a reasonable jury could find that spills at both Mobil stations, the Coastal station, the Getty station, and an Advance station caused the contamination in the Broadway well.

12. Crystal Brook Hollow Road Well No. 3

The Crystal Brook Hollow well began operating in 2004, and MTBE was immediately detected in its water.^[265] Detections have been continuous from the initial detection to the present, with a peak in August 2004 and subsequent increases in magnitude beginning in 2006.^[266]

A total of fourteen spills have been reported at eight locations within the capture zone of the well. Although Sosik concluded that there was insufficient information to determine if any of the spills were sources of MTBE in the well, evidence about certain spills creates fact issues as to causation. Five spills were reported in 1999 at a 7-11 station, which is the closest site to the well, due to monthly inventory discrepancies.^[267] Although there is no evidence of soil contamination, inventory discrepancies support an inference that an underground leak occurred.^[268] Further, the timing of multiple spill reports in 1999 is consistent with the initial detection of MTBE in the well in 2004, and the station is located in close proximity to the well. Therefore, a reasonable jury could conclude that the spills at the 7-11 station contributed to contamination in the well.

Plaintiffs also have submitted numerous delivery tickets that they claim show that ExxonMobil supplied gasoline to this 7-11 station. The Island Transportation delivery tickets state that the "transferor" of the gasoline is the Mobil Oil Corporation.^[269] The tickets report deliveries made between January 2001 and December 2004.^[270] Although the spills at the 7-11 station occurred prior to the period covered by these tickets, they are circumstantial evidence tending to show that Exxon/Mobil was a supplier of gasoline to the 7-11 station, and may have supplied the station in 1999. Therefore, a jury may also consider whether Exxon/Mobil is liable for contamination in the well, under the product liability and Navigation Law claims, as a supplier to the 7-11 station.

In addition, fact issues exist regarding a spill reported at a Northville gas station in 1991, when contaminated soil was discovered beneath the ground during the removal of a gasoline tank.^[271] The report indicates that although the Suffolk County Department of Health Services was not satisfied with the amount of soil removed in the excavation, the file was nonetheless closed.^[272] The soil contamination, especially if not fully remedied, makes it likely that gasoline constituents were absorbed into groundwater.

Two other spills, one at a Texaco station in 2002 and one at a Getty station in 2004, were reported when trace levels of MTBE were detected in monitoring wells at each station.^[273] Sosik did not believe that these detections were due to an older and more significant release, however, because no other gasoline constituents were found in the monitoring wells.^[274] He opined that the trace levels of MTBE alone do not support the inference that there was enough MTBE released at the site to create a migrating plume.^[275] Plaintiffs' argument that these spills should also be considered by a jury must fail as their own expert has ruled them out as possible sources of the well's contamination.

Evidence concerning the remaining spills similarly fails to support a finding of causation. A spill at a Sunoco station was eliminated because the station was not reasonably proximate to the predicted capture zone; the file on a 1987 spill at Island Transportation was destroyed, leaving no evidence for a jury to consider; and a 1997 spill at Marchese Motors involved only gasoline staining observed on the ground's surface.^[276] Therefore, a reasonable jury could find that the spills reported in 1999 at the 7-11 station caused the contamination in the Crystal Brook Hollow well. The jury could also find that ExxonMobil is liable for those spills as a supplier of gasoline to that station, and that the 1991 reported spill at the Northville gas station also caused or contributed to the well's contamination.

13. Dare Road Well No. 1

An initial MTBE detection at the Dare Road well occurred in 1997, and detections continued intermittently in 2000 through 2003.^[277] From 2004 to the present, the detections have become more frequent and concentrations have increased in magnitude.^[278]

Nine spill reports document known gasoline releases within the capture zone of the Dare Road well between 1986 and 1994. The information in these files, however, at least as discussed in the Sosik report, is very minimal, making an evaluation of causation difficult. One spill, at Jay-six Auto Repair, was confirmed to have contaminated groundwater at low levels, but it cannot be a source of the well contamination because groundwater from the site flows away from the well.^[279] Files for the remaining spills are either missing, destroyed, or lack information. This is troubling because a number of the spills apparently involved contaminated soil discovered when USTs were removed from various sites as part of a road widening project.^[280]

Because no groundwater testing was performed at any of these sites, Sosik's report does not discuss the spills in detail and does not indicate which of the spills involved contaminated soil. Therefore, there is not enough evidence in the record about the spills in the vicinity of the Dare Road well for a jury to determine which spills caused or contributed to contamination in the well.

14. Horseblock Road Well No. 1

MTBE was first detected at the Horseblock Road in 1993, and isolated, low-level detections occurred only several times a year from 1993 to 1999.^[281] In early 2000, however, the concentration spiked and detections have been continuous from 2000 to the present, although the concentration levels have dropped again.^[282]

Only two spills have been reported within the capture zone of Horseblock Road Well No. 1, and the evidence does not indicate that either of the spills were likely to have contaminated groundwater. A small surface spill was reported at a Mobil station in 2000 and was apparently cleaned up that day.^[283] The second spill was reported in 1986, at an abandoned station when storage tanks were removed and opened by a bulldozer.^[284] The only evidence that gasoline had been released was a strong gasoline odor at the site.^[285] There is simply not enough evidence for a reasonable jury to conclude that either of these reported spills caused the contamination in the Horseblock Road well. Based on the MTBE detections in the well, there is no doubt that some gasoline release caused the contamination, but that release remains unknown.

15. Oak Street Well No. 1

MTBE was first detected in the Oak Street well in January of 2000, and detections at low concentrations have been continuous through the present.^[286] Three reported spills have occurred within the capture zone of the Oak Street well. A spill at a Getty station was reported in the Spring of 1990 when a relatively small amount of contaminated soil was discovered during the removal of a UST; about one cubic yard of soil was removed.^[287] Contaminated soil was also discovered at a Shell station during removal of a UST in 1991 and about fifteen to twenty cubic yards of contaminated soil was removed, although some contaminated soil remained after the excavation.^[288] A sample of soil from seventy-five feet below the surface tested negative for gasoline constituents, although MTBE was not included in the analysis and a high threshold for other constituents was employed.^[289]

Finally, an enormous gasoline release from the Northville Terminal, which is located two thousand feet from the well, was discovered in 1987. A one-eighth of an inch hole in a pipe that transported gasoline from the terminal's storage tanks to the loading racks released one million gallons of gasoline over a period of years.^[290] In response to the massive spill, extensive remediation was conducted and the spill site has been carefully studied. Although studies have not indicated that the Northville spill affected the Oak Street wellfield, Sosik states that it is "highly suspect" due to its proximity to the wellfield and the absence of other identifiable sources.^[291]

Each of these spills creates an issue of fact as to causation for the jury to resolve. The Getty and Shell spills both affected soil beneath the surface, and contaminated soil was left behind at the Shell site. The Northville spill occurred in close proximity to the Oak Street wellfield and a jury could conclude that despite remediation, some MTBE from the spill remained behind and is now contaminating the Oak Street well. Therefore, a jury may consider whether each of the known spills is a source of the contamination in this well.

16. Wheat Path Well No. 3

MTBE was first detected in the Wheat Path well in early 2000, and detections have been continuous from late 2000 to the present with a peak in September 2004.^[292] Three spills have been reported at sites within the capture zone of the Wheat Path well. In 2003, a Mobil station reported minimal leaks from a gasoline dispenser.^[293] This type of spill is not likely to have caused any groundwater contamination. The two other spills, both of which occurred at a Northville station in 1990 and 1991, are the same incidents described above for the Crystal Brook Hollow Road well. The 1990 spill was reported when contaminated soil was encountered during the removal of an underground diesel tank.^[294] As discussed above, the diesel fuel that leaked from this tank would not have contained MTBE. The 1991 spill was reported upon the discovery of contaminated soil during the removal of an underground gasoline tank.^[295] The Department of Health Services was not satisfied with the amount of soil removed from the site after the 1991 spill report.^[296] As stated above, the contaminated soil beneath the surface at the Northville station, especially if it was not fully removed, could support an inference that MTBE dissolved into the groundwater at the spill site. Therefore, although Sosik did not opine that any of the known spills were a source of the contamination in the Wheat Path well, a jury could conclude based on the evidence that the 1991 Northville spill caused the contamination of that well.

17. Strathmore Court Well No. 1

Isolated detections of MTBE in the Strathmore Court well occurred in early 1997 and 1998.^[297] From 1999 to the present, detections were more frequent, with a peak in mid-2002.^[298]

Four reported spills occurred within the predicted capture zone of the Strathmore Court well. However, the evidence as to each spill is insufficient to support a reasonable jury's verdict on causation. One of the spills, reported in 2003 at a Mobil station, involved only minimal releases from a dispenser and some staining on the ground.^[299] This type of spill, of a minimal amount of gasoline on the surface of the ground, is not likely to have contaminated groundwater. There is little information in Sosik's report about the other three spills, which were reported at a Hess station in 1986, 1987, and 1999, other than the mention of intentional dumping into a storm drain in 1986.^[300] The file on the 1986 spill was destroyed, however, and the evidence concerning the Hess spills is not detailed enough to support a finding of causation. Nor is there evidence of a leak at the one PBS tank in the area. Therefore, although MTBE has been detected in the Strathmore Court well, the source of the contamination is unknown.

18. Morris Avenue Well No. 2

From 1993 to 1998, low-level MTBE detections occurred in the Morris Avenue well about once a year.^[301] Beginning in 1998, detections became more frequent with peaks in mid-1999 and 2000; detections have been continuous from 2001 to the present.^[302]

Six spills were reported within the capture zone of the Morris Avenue well.^[303] At only one spill, from a Getty station in 2002, was the presence of MTBE in groundwater at the station confirmed.^[304] However, the spill at this Getty station is considered the source of contamination in the Kayron Drive well, and Sosik opines it is likely that well would have captured the MTBE flowing from the spill before it reached the Morris Avenue well.^[305] In addition, he states that the timing of the detections in the Morris Avenue well are not consistent with transport of MTBE from the Getty spill.^[306]

A spill reported in 2002 at an Amoco station involved the removal of four tons of contaminated soil discovered during an upgrade of a gasoline dispenser system.^[307] While little contaminated soil remained after the excavation, no testing was performed to determine if gasoline constituents had entered the groundwater.^[308] Because of the lack of groundwater testing, Sosik did not consider this spill as a source. However, the huge volume of impacted soil beneath the surface — as well as the fact that the dates of the leak are unknown — makes it likely that MTBE absorbed into the groundwater at the spill site prior to the removal of the contaminated soil. Therefore, a fact issue exists with respect to whether the Amoco spill contributed to the contamination in the well.^[309]

In addition, the jury may consider whether gasoline from a tank overfill, which entered a storm drain at a Coastal station in 1998, and whether a spill at a former U.S. Oil/Chevron station, at the same location as the Coastal station, which involved a large amount of contaminated soil, contributed to the contamination.'^[310] Therefore, a reasonable jury could find that the Coastal station spill, the U.S. Oil/Chevron spill, and the Amoco spill caused the contamination in the Morris Avenue well.

VI. CONCLUSION

For the foregoing reasons, defendants' motions are denied in part and granted in part. The Clerk of the Court is directed to close these motions (docket ## 1554, 1619, 1624, 1629, 1640, 1656).

SO ORDERED.

NOTES

[1] Methyl Tertiary-Butyl Ether (MTBE): Advance Notice of Intent to Initiate Rulemaking Under the Toxic Substances Control Act to Eliminate or Limit the Use of MTBE as a Fuel Additive in Gasoline ("Advance Notice of Intent"), 65 Fed. Reg. 16,094, 16,096-97 (Mar. 24, 2000). Oil companies began adding MTBE to gasoline in order to increase its oxygen content in 1979, and increasingly used it after 1992 when federal regulations mandated the use of oxygenated gasoline (known as reformulated gasoline or "RFG") in certain high-smog areas of the country. MTBE was intended to help reduce vehicle emissions that caused air pollution.

[2] The Long Island aquifer system is designated by the Environmental Protection Agency ("EPA") as a "sole source aquifer" under the Safe Drinking Water Act, 42 U.S.C. § 300h-3(e), which means that it is the sole or primary source of drinking water for the area.

[3] See Sixth Amended Complaint, County of Suffolk, et al. v. Amerada Hess Corp. et al., 04 Civ. 5424 ("Compl.") ¶ 70.

[4] Many of these defendants have settled with plaintiffs, thereby resolving all claims. A small number of defendants have not joined in that settlement agreement to date. These are: ExxonMobil, Lyondell Chemical Company and Equistar Chemicals LLP, Crown Central LLC, Getty Properties Corp., Giant Yorktown, Irving Oil Limited and Irving Oil Corporation, Gulf Oil Limited Partnership (GOLP), and TOTAL Petrochemicals USA.

[5] See id. ¶ 4.

[6] See In re Methyl Tertiary Butyl Ether ("MTBE") Prods. Liab. Litig., No. 1:00-1898, 2007 WL 1791258, at * 2 (S.D.N.Y. June 15, 2007) (finding a bellwether trial was "warranted by the sheer size of this action, the need for expeditious resolution, judicial economy, and the convenience of the Court, the jury, and the parties.").

[7] The difficulty concerns "`causation' in the sense of product/defendant identification (i.e., that a given defendant caused MTBE to be produced and distributed in a manner that allowed it to reach plaintiffs' wells)" rather than whether MTBE, and not some other substance, caused the harm plaintiffs allege. In re MTBE Prods. Liab. Litig., 517 F. Supp. 2d 662, 670 n. 39 (S.D.N.Y.2007) (holding that plaintiffs may not recover punitive damages if relying on a market share theory of liability). See also In re MTBE Prods. Liab. Litig., 379 F. Supp. 2d 348 (S.D.N.Y.2005) (denying defendants' motions to dismiss in numerous actions, and finding that New York courts would apply a theory of collective liability); In re MTBE Prods. Liab. Litig., 447 F. Supp. 2d 289 (S.D.N.Y.2006) (holding that plaintiffs may proceed on an alternative theory of liability if they cannot identify a make-whole defendant for a particular well).

[8] See Defendants' Motion for Partial Summary Judgment as to Nine Focus Wells for Which Plaintiffs Cannot Establish Causation. To avoid the use of this title, which is both cumbersome and presupposes an outcome in defendants' favor, I will refer to this motion as "Def. I."

[9] See Defendants' Motion for Partial Summary Judgment as to Those Nine "Traditional" Causation Focus Wells for Which They Are Not Identified as the Responsible Parties. Again, for the purpose of simplicity I will refer to this motion as "Def. II."

[10] Plaintiffs' Navigation Law claims are asserted only against those defendants who are dischargers of gasoline into the environment, and the factual and legal arguments made in the Navigation Law motion are identical to those made in the second omnibus motion on the issue of causation (Def. II). They are addressed in Part V, infra, discussing liability of retailers alleged to have spilled gasoline.

[11] I will refer to Defendant Crown Central LLC's Memorandum of Law in Support of Its Motion for Summary Judgment Motion as "Crown Mem." and its Reply Brief as "Crown Reply." Plaintiffs' brief in opposition will be referred to as "PL Opp. to Crown Mem." Citations to the other briefs filed by individual defendants will have a similar form (e.g., Giant Mem., PL Opp. to Giant Mem., Giant Reply, etc.).

[12] Fed.R.Civ.P. 56(c).

[13] Higazy v. Templeton, 505 F.3d 161, 169 (2d Cir.2007) (quoting Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S. Ct. 2505, 91 L. Ed. 2d 202 (1986)).

[14] McCarthy v. Dun & Bradstreet Corp., 482 F.3d 184, 202 (2d Cir.2007) (citing Jeffreys v. City of New York, 426 F.3d 549, 553 (2d Cir.2005)).

[15] Vermont Teddy Bear Co. v. 1-800 Beargram Co., 373 F.3d 241, 244 (2d Cir.2004) (citing Adickes v. S.H. Kress & Co., 398 U.S. 144, 157, 90 S. Ct. 1598, 26 L. Ed. 2d 142 (1970)).

[16] Higazy, 505 F.3d at 169 (quoting Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 586, 106 S. Ct. 1348, 89 L. Ed. 2d 538 (1986)).

[17] Jeffreys, 426 F.3d at 554 (quoting Fujitsu Ltd. v. Federal Express Corp., 247 F.3d 423, 428 (2d Cir.2001)).

[18] McClellan v. Smith, 439 F.3d 137, 144 (2d Cir.2006) (quoting First Nat'l Bank of Ariz. v. Cities Serv. Co., 391 U.S. 253, 288-89, 88 S. Ct. 1575, 20 L. Ed. 2d 569 (1968)).

[19] See Allstate Ins. Co. v. Hamilton Beach/Proctor Silex, Inc., 473 F.3d 450, 456 (2d Cir.2007) (citing Stern v. Trustees of Columbia Univ., 131 F.3d 305, 312 (2d Cir. 1997)).

[20] McClellan, 439 F.3d at 144 (quoting Fischl v. Armitage, 128 F.3d 50, 55 (2d Cir.1997)). Accord Anderson, 477 U.S. at 249, 106 S. Ct. 2505.

[21] American Home Assurance Co. v. Hapag Lloyd Container Linie, GmbH, 446 F.3d 313, 315 (2d Cir.2006) (quoting Marvel Characters, Inc. v. Simon, 310 F.3d 280, 286 (2d Cir. 2002)).

[22] "The notion that causal connection between agency and harm must be established is ... often implied even when the word ["cause"] is not used." Antony Honoré, "Causation and the Law," Stanford Encyclopedia of Philosophy (2005), http://plato. stanford.edu/entries/causation-law/. The parties do not distinguish among the different claims when discussing causation in their briefs.

[23] See Restatement (Second) of Torts § 433 (1977). For an academic discussion, see Richard W. Wright, Causation in Tort Law, 73 CAL. L.REV. 1735 (1985) (discussing causation as the basis for imposing legal responsibility for another's harm). Causation in strict products liability claims "is determined by the prevailing rules and principles governing causation in tort." Restatement (Third) of Torts: Prod. Liab. § 15, cmt. a (1998).

[24] Mortensen v. Memorial Hospital, 105 A.D.2d 151, 483 N.Y.S.2d 264, 269 (1st Dep't 1984). Accord Derdiarian v. Felix Contracting Corp., 51 N.Y.2d 308, 316, 434 N.Y.S.2d 166, 414 N.E.2d 666 (1980).

[25] See Mortensen, 483 N.Y.S.2d at 270 (citing Dunham v. Village of Canisteo, 303 N.Y. 498, 504, 104 N.E.2d 872 (1952)); see also Capicchioni v. Morrissey, 205 A.D.2d 959, 613 N.Y.S.2d 499 (3d Dep't 1994) (holding that jury instruction implying that defendant must be the sole cause of a car accident was reversible error). In strict products liability, it is the product defect or the failure to warn, rather than the defendant's conduct, that must be shown to be a substantial cause of the plaintiff's injuries. See Voss v. Black & Decker Mfg. Co., 59 N.Y.2d 102, 106, 463 N.Y.S.2d 398, 450 N.E.2d 204 (1983).

[26] Mortensen, 483 N.Y.S.2d at 270

[27] Hymowitz v. Eli Lilly & Co., 73 N.Y.2d 487, 504, 541 N.Y.S.2d 941, 539 N.E.2d 1069 (1989) (citing Morrissey v. Conservative Gas Corp., 285 A.D. 825, 136 N.Y.S.2d 844 (2d Dept 1955); Prosser and Keeton, Torts § 103, at 713 [5th ed.]).

[28] Healey v. Firestone Tire & Rubber Co., 87 N.Y.2d 596, 601, 640 N.Y.S.2d 860, 663 N.E.2d 901 (1996).

[29] Id. at 602, 640 N.Y.S.2d 860, 663 N.E.2d 901.

[30] See, e.g., In re MTBE, 517 F.Supp.2d at 668(clarifying that market share liability is an evidentiary method rather than an independent claim).

[31] Alternative liability, as it was first explained in the seminal case of Summers v. Tice, 33 Cal. 2d 80, 199 P.2d 1 (1948), provides that the burden of proof is shifted to the defendant to prove that it did not cause the injury, where each defendant acted tortiously, but there is uncertainty as to which of the defendants caused the injury.

[32] See In re MTBE, 379 F.Supp.2d at 425.

[33] The doctrine of market share liability was developed to provide a remedy in cases where the plaintiff alleges harm from a product that is fungible, and therefore she is unable to identify the manufacturer of the particular product she used. Under the doctrine of market share liability, the burden of proof shifts to the defendants if plaintiff establishes a prima facie case on every element of the claim except for identification of the actual tortfeasors. In that case, each defendant is severally liable for the portion of the judgment representing its share of the national market at the time of injury, unless it exculpates itself by proving it could not have made the product that caused plaintiff's harm. See Sindell v. Abbott Labs., 26 Cal. 3d 588, 163 Cal. Rptr. 132, 607 P.2d 924 (1980). Accord Hymowitz, 73 N.Y.2d 487, 541 N.Y.S.2d 941, 539 N.E.2d 1069.

[34] In the fact scenario addressed by Sindell and Hymowitz, by contrast, each pill alleged to have caused plaintiffs' injuries was manufactured by only one defendant. Indeed, in a letter dated October 2, 2007, Plaintiffs themselves represented to the Court that they did not intend to pursue their claims under market share liability.

[35] Plaintiffs' claims against manufacturers of MTBE and gasoline containing MTBE allege that the manufacturers created and distributed a defectively designed product, failed to warn of the product's known dangers, negligently placed a product they knew to be dangerous into the stream of commerce, and created a public nuisance. See Compl. ¶¶ 214-250 (Second, Third, Fourth and Fifth Causes of Action).

[36] These claims include product liability for design defect; negligence in the storage and handling of gasoline containing MTBE; trespass; private and public nuisance; and violation of the New York Navigation Law, which holds "dischargers" of gasoline strictly liable for damages resulting from that discharge. Plaintiffs do not argue that alternative liability theories should apply to claims against the parties responsible for spilling or leaking gasoline. See id. ¶¶ 226-254, 260-270 (Third, Fourth, Fifth, Sixth, Eighth and Ninth Causes of Action).

[37] See Codling v. Paglia, 32 N.Y.2d 330, 338-40, 345 N.Y.S.2d 461, 298 N.E.2d 622 (1973) (setting forth principle that a manufacturer may be liable for injuries to innocent bystanders caused by its product). Accord Bellevue South Assoc. v. HRH Const. Corp., 78 N.Y.2d 282, 290, 574 N.Y.S.2d 165, 579 N.E.2d 195 (1991) (reasoning that "manufacturers could better sustain the losses which could be spread among all their customers"); Perillo v. Pleasant View Assoc., 292 A.D.2d 773, 739 N.Y.S.2d 504 (4th Dep't 2002).

[38] Lyondell Chemical Company and Equistar Chemicals, both manufacturers of MTBE but not of gasoline, have brought their own summary judgment motion that is addressed in Part IV.D, infra. The other sections in Part IV will focus only on the liability of gasoline refiners.

[39] Energy Information Administration ("EIA"), Primer on Gasoline Sources and Markets, http://www.eia.doe.gov/neic/experts. contactexperts.htm.

[40] In re MTBE, 379 F.Supp.2d at 365.

[41] See 8/2/07 Expert Report of John B. O'Brien ("O'Brien Report"), Ex. F to Plaintiffs' Exhibits in Support of Supplemental Briefing in Opposition to Def. I ("Pl. Supp. Opp. to Def. I"), for a full discussion of petroleum product distribution exchanges and other business relationships.

[42] See ERS Group, Report on Petroleum Products Markets in the Northeast, Prepared for the Attorneys General of Maine, Massachusetts, New Hampshire, New York, and Vermont ("ERS Report on Petroleum Products") at 20 (Sept. 2007). Although neither party cites this report, there is no reason to believe the information it contains is not accurate. The report is admissible as the report of a public agency (the Attorneys General of several states) setting forth matters observed pursuant to a duty. See Fed.R.Evid. 803(8)(B).

[43] Colonial Pipeline Company, About Us, http://www.colpipe.com/ab_main.asp. Again, neither party cites the Colonial Pipeline website, but there is no reason to believe the very basic information cited here is not accurate.

[44] ERS Report on Petroleum Products at 42.

[45] See Colonial Pipeline Company, About Us, http://www.colpipe.com/ab_main.asp.

[46] See id.

[47] See ERS Report on Petroleum Products at 44. A number of smaller pipelines, which are part of the Colonial Pipeline system, run between the various terminals in the New York Harbor and connect to the other pipelines that serve the Northeast region, the Buckeye and Sunoco piplines. See id.

[48] Different grades of gasoline, and different petroleum products, are shipped separately in the pipeline in "batches." "On the Colonial pipeline system, for example, batches range in size from 75,000 barrels to 3.2 million barrels." Id. The batches are generally not physically separated, and some mixing of products often occurs at each end of a batch. See id.

[49] Colonial Pipeline, Frequently Asked Questions, http://www.colpipe.com/ab_faq.asp.

[50] Id.

[51] Plaintiffs' Opposition to Defendants' Partial Motion for Summary Judgment as to Nine Focus Wells for Which Plaintiff Cannot Establish Causation ("Pl. Opp. To Def. 1") at 17 n. 9.

[52] ERS Report on Petroleum Products at 52.

[53] Thirty-one different companies own terminals in the New York City area, including many of the defendants named in this action as well as companies not named as defendants. See id. at 63, Table 3.8.

[54] Domestic refineries currently supply approximately eighty-five to ninety percent of U.S. gasoline consumption. See id. at 26; see also Lawrence Kumins, Gasoline Supply: The Role of Imports, CRS Report for Congress, Sept. 14, 2004, at I. Just over fifty percent of gasoline delivered to the New York harbor by waterway comes from foreign sources. See ERS Report on Petroleum Products at 39, Figure 2.5. A large portion of imports are gasoline components, which are then blended into gasoline by U.S. refiners. See id. at 4, 7. Canada is by far the most significant importer of gasoline to the U.S. and Canadian gasoline supplies a good portion of the Northeast gasoline market. See Kumins, Gasoline Supply at 7. Irving Oil, a defendant in this case, is the largest Canadian refiner. It exports "approximately 175,000 barrels of petroleum products per day to the Northeast, including 100,000 barrels of reformulated gasoline." ERS Report on Petroleum Products at 49.

[55] See O'Brien Report at ¶¶ 126-127. Not all primary terminals in the New York Harbor serve the Long Island secondary terminals. See id.

[56] See id.

[57] See ERS Report on Petroleum Products at 74-77.

[58] See O'Brien Report at ¶ 128.

[59] See id.

[60] See ERS Report on Petroleum Products at 74

[61] See id.

[62] See O'Brien Report, at ¶¶ 125-128, Ex. F to Pl. Supp. Opp. to Def. I.

[63] See 1/18/08 Deposition of James Edward Brown ("Brown Dep."), Ex. I in Support of Pl. Supp. Opp. to Def. I; 1/16/08 Deposition of James V. Scandola ("Scandola Dep."), Ex. J in Support of Pl. Supp. Opp. to Def. I. This testimony is confidential pursuant to MDL 1358, Case Management Order No. 36 (Feb. 15, 2008), which applied the Court's 2004 Revised Confidentiality Order to documents, information and other evidence relating to certain terminals.

[64] See 11/15/07 Deposition of Daniel Gianfalla, Ex. K in Support of Pl. Supp. Opp. to Def. I; 12/20/07 Deposition of David McWilliams, Ex. L in Support of Pl. Supp. Opp. to Def. I; 10/19/07 Deposition of Mario A. D'Antonio ("D' Antonio Dep."), Ex. M in Support of Pl. Supp. Opp. to Def. I; 12/19/07 Deposition of H. Brant Brown, Ex. N in Support of Pl. Supp. Opp. to Def. I; 12/13/07 Deposition of Anthony Cassandro ("Cassandro Dep."), Ex. O in Support of Pl. Supp. Opp. to Def. I; and 1/04/07 Deposition of John Didier ("Didier Dep."), Ex. P in Support of Pl. Supp. Opp. to Def. I.

[65] Pl. Supp. Opp. to Def. I at 22.

[66] See Brown Dep. at 64-67; Scandola Dep. at 57-58, 87; D'Antonio Dep. at 62-63; Cassandro Dep. at 23-25, 41; Didier Dep. at 41. It should be noted that not all terminals in the New York Harbor send gasoline to the Long Island terminals. See Supplemental Expert Declaration of John B. O'Brien ("O'Brien Supp. Decl."), Ex. B to Defendants' Rule 56.1 Statement in Support of Def. I, ¶ 7.

[67] See Declarations of Refiner Defendants in Compliance with Case Management Order No. 4, Ex. C to Declaration of Carla M. Burke in Support of Plaintiffs' Opposition to Def. I.

[68] See Confidential Barrels In/Barrels Out Charts submitted in support of Plaintiffs' Opposition to Def. I for in camera review.

[69] O'Brien Supp. Decl. ¶ 7. This statement is slightly misleading, however, because most spills involved the release of many gallons of gasoline over a period of time.

[70] Hymowitz, 73 N.Y.2d at 507, 541 N.Y.S.2d 941, 539 N.E.2d 1069 (quoting People v. Hobson, 39 N.Y.2d 479, 489, 384 N.Y.S.2d 419, 348 N.E.2d 894 (1976) and Bichler v. Eli Lilly & Co., 55 N.Y.2d 571, 579-80, 450 N.Y.S.2d 776, 436 N.E.2d 182 (1982)).

[71] In re MTBE, 447 F.Supp.2d at 301.

[72] Id. In fashioning the commingled product theory, I also stated that liability is several only and is apportioned by proof of a defendant's share of the market at the time of the injury. See id. The parties have not addressed the issue of apportionment in their briefs, or who bears the burden of proof on this issue, and it is therefore premature to discuss it here.

[73] See In re MTBE, 379 F.Supp.2d at 377.

[74] In re MTBE, 447 F.Supp.2d at 301.

[75] Id. See individual defendants' exculpation motions addressed in Part IV.E infra.

[76] See Phase I Expert Report of Charles H. Sosik, June 22, 2007 ("Sosik Report") at 2.

[77] The parties dispute the boundaries of each well's capture zone. The jury will determine the scope of capture zones based on the evidence presented at trial.

[78] The fact that certain wells may have no identifiable source of contamination has long been understood in this litigation. Nevertheless, it is beyond cavil that some release of gasoline must have caused the MTBE contamination in these wells. As Sosik notes in his report, "[w]hile the source of the release cannot be identified with sufficient certainty, the MTBE in the well is not naturally occurring." Id.

[79] See In re MTBE, 447 F.Supp.2d at 304-05 ("If there is an identifiable defendant (or defendants) and plaintiffs can obtain a make-whole remedy from those parties, then there is no need to turn to an alternative theory of liability to pursue other possible tortfeasors.").

[80] Godoy v. Abamaster of Miami, Inc., 302 A.D.2d 57, 754 N.Y.S.2d 301, 307 (2d Dep't 2003).

[81] Id. at 305.

[82] See Shackil v. Lederle Labs., 116 N.J. 155, 561 A.2d 511 (1989).

[83] See H. Carl McCall, New York State Comptroller, New York State's Oil Spill Fund, March 2001 ("McCall Report"), at I. While neither party cites this report, there is no reason to believe its information is inaccurate, and it is admissible as the report of a public agency documenting its activities under Federal Rule of Evidence 803(8)(A).

[84] See McCall Report at 6. The report also notes that the Fund has operated "in the red" for the past ten years and, because MTBE contamination in groundwater caused oil spill cleanup costs to rise dramatically, "the Fund faces the prospect of not having sufficient funding to complete MTBE cleanups or pay victims of MTBE spills." Id. at 10.

[85] Because Lyondell and Equistar have not spilled gasoline, the claims against them are based only on the manufacture and negligent distribution of a dangerous product. Plaintiffs have stipulated to the dismissal of claims under the New York Navigation Law against Lyondell and Equistar.

[86] Declaration of Plaintiffs' Expert Michael Grabowski, Ex. H to Declaration of Chad A. West ("West Decl.") in Support of Plaintiffs' Opposition to Defendants Lyondell Chemical Company's and Equistar Chemicals LP's Motion for Summary Judgment ("Lyondell Opp."), ¶ 2.

[87] In re MTBE, 379 F.Supp.2d at 365.

[88] See Various Defendants' Responses and Objections to Plaintiffs' First Set of Interrogatories Regarding Defendant Identification, Ex. B to Declaration of Jerry D. Bernstein in Support of Lyondell Motion for Summary Judgment ("Bernstein Decl."). The responses show sales to BP Products from 1993-1997 and 2000-2003; ConocoPhillips from 1997-2003; ExxonMobil from 1986-1996 and 1998-2002; TMR Company (years not specified); El Paso Corporation and Coastal Eagle Point Oil Company (years not specified); CITGO from 1987-2003; Amerada Hess in 1998 and 2001; Shell (years not specified); Koch Industries in 1996; Sunoco from 1995-2003; and Valero in 1989, 1993-1994, 1998, 2000, and 2003. See id. See also Declaration of Defendant Lyondell Chemical Company Pursuant to Case Management Order # 4, Ex. I to West Decl. (listing refiners in Pennsylvania and New Jersey to which Lyondell sold MTBE).

[89] See Local Rule 56.1 Statement in Support of Lyondell Mem. ¶ 3. Defendants state that refiners purchased MTBE from approximately two hundred seventy-five suppliers, but plaintiffs note that only twenty-seven to thirty-two major U.S. MTBE producers existed in the mid-1990's, and only twenty-three major U.S. producers existed in 2003. See id. ¶ 2; "MTBE.MTBE," Chemical Marketing Reporter, 247 n. 14 (Apr. 3, 1995), Ex. J to West Decl.; Kirschner, Mark, "MTBE (Chemical Profile)," Chemical Market Reporter, 263.1 (Jan. 6, 2003): 27(1), Ex. K to West Decl.

[90] Pl. Opp. to Getty Mem. at 1. See also Pl. Opp. to Crown Mem. at 2-4; PL Opp. to Giant Mem. at 2-3; PL Opp. to Irving Mem. at 1-2; PL Opp. to Total Mem. at 2-3.

[91] On January 25, 2008, plaintiffs agreed to dismiss claims against Getty to the extent they are based on commingled product theory because they are not being sued as a refiner of gasoline.

[92] Indeed, having reviewed the briefs, I agree that "Getty should not have submitted this motion at all," since Getty's arguments repeat those made in the omnibus motions. Id. at 1.n.1. I therefore grant Plaintiffs' request to strike Getty's individual motion for summary judgment.

[93] Irving Mem. at 1.

[94] Total Mem. at 1. See also Crown Mem. at 7 (arguing that plaintiffs "must have sufficient proof that an individual defendant's gasoline was `a substantial factor in bring about the injury.'") (quoting New York Pattern Jury Instructions—Civil § 2:70).

[95] In re MTBE, 447 F.Supp.2d at 301.

[96] Id.

[97] Giant Mem. at 1.

[98] See id.

[99] See Pl. Opp. to Giant Mem. at 1.

[100] Id.

[101] Total Mem. at 7.

[102] Crown Mem. at 2.

[103] See, e.g., Irving Mem. at 3("Between 1996 and 2003, Irving Oil imported a total of 130 shipments of gasoline containing MTBE to terminals in New York Harbor."); Crown Mem. at 3 ("Crown transported most of its gasoline via the Colonial Pipeline.")

[104] Defendants contend that this argument, which was made in plaintiffs' supplemental briefing in opposition to summary judgment but not in their original opposition, is untimely and should not be entertained by the Court. I need not decide the issue of timeliness, however, because for the reasons discussed below plaintiffs' theory lacks an evidentiary foundation.

[105] See Rastelli v. Goodyear Tire & Rubber Co., 79 N.Y.2d 289, 295, 582 N.Y.S.2d 373, 591 N.E.2d 222 (1992).

[106] See id.

[107] See City of New York v. Lead Indus. Ass'n, 190 A.D.2d 173, 597 N.Y.S.2d 698, 700-01 (1st Dep't 1993) (declining to dismiss a claim based on concert of action liability where plaintiffs alleged that lead paint manufacturers and their trade association "coordinated their efforts to conceal the hazard, to mislead the public and the government as to that hazard, and to market and promote the use of the product despite their knowledge of the hazard").

[108] See Various Letters and Memoranda from the MTBE Committee, Ex. G to Pl. Supp. Opp. to Def. I. Defendants contend that documents contained in this exhibit are inadmissible hearsay. As discussed below, plaintiffs do not offer any evidence supporting admissibility.

[109] See Memoranda from B.J. Mickelson (dated August 23, 1984, April 19, 1985, and February 22, 1985), Ex. H to Pl. Supp. Opp. to Def. L Defendants contend that this exhibit is inadmissible hearsay. However, the statement might be admissible as a party admission under Federal Rule of Evidence 801(d)(2)(D).

[110] Plaintiffs could have ascertained the membership of the MTBE Committee during discovery. They have not done so, however, and their brief glosses over this problem, stating only that "[b]oth gasoline refiners and manufacturers of neat MTBE act through trade associations to mislead and conceal the dangers from EPA and the public." Pl. Supp. Opp. to Def. I at 19.

[111] There is no basis to hold manufacturers or refiners liable under the New York Navigation Law, N.Y. Nay. Law. § 181(1), which creates strict liability for "dischargers" of petroleum products, unless the manufacturer or refiner delivered product directly to a gas station or other facility that released gasoline and "was in a position to prevent the discharge or effect a cleanup." State of New York v. Montayne, 199 A.D.2d 674, 604 N.Y.S.2d 978, 978 (3d Dep't 1993). Accordingly, plaintiffs have dismissed all claims under the Navigation Law against the following defendants, none of which discharged gasoline from a gas station they owned, operated or directly supplied: Total, Irving Oil, Crown Central, Lyondell and Equistar, and Giant Yorktown. The only remaining defendants against whom plaintiffs assert Navigation Law claims are ExxonMobil, Gulf and Getty, each of which own, operate, or brand gas stations in Suffolk County where gasoline spills or leaks have occurred.

[112] See Advance Notice of Intent at 16, 100-01; In re MTBE Prods. Liab. Litig., 241 F.R.D. 185, 190 (S.D.N.Y.2007).

[113] Underground Storage Tanks: Technical Requirements, 53 Fed. Reg. 37,082, 37,086 (Sept. 23, 1988). The EPA estimate was based in part on a study of Suffolk County's UST program which indicated that twenty-six percent of the county's six thousand storage tanks were leaking. See id.; E. Blaine Rawson, Are We Properly Controlling our LUSTs? A Review of the Problems with Underground Storage Tank Regulation, 40 Idaho L.Rev. 111, 117 (2003).

[114] USEPA, "Overview of the Federal UST Program," http://www.epa.gov/OUST/ overview.htm.

[115] Advance Notice of Intent at 16,095.

[116] 42 U.S.C. § 6991b(c) directed the EPA to establish minimum requirements for leak detection systems, which are codified at 40 C.F.R. § 280.41.

[117] See Underground Storage Tanks: Technical Requirements, 53 Fed. Reg. at 37,083 (recognizing that the universe of newly regulated systems was "immense" and "most of the facilities to be regulated are owned and operated by very small businesses, essentially `Mom and Pop' enterprises not accustomed to dealing with complex regulatory requirements"). See also 2/1/07 Expert Report of Marcel Moreau, Ex. F. to Declaration of Robin Greenwald in Support of Plaintiffs' Memorandum of Law in Opposition to Def II. ("Greenwald Decl.") at 43 (noting that while major oil companies began to upgrade UST systems in the 1980s, many small station owners did not upgrade until the late 1990s).

[118] U.S. General Accounting Office, Pub. No. GA0-01-464, Environmental Protection: Improved Inspections and Enforcement Would Better Ensure the Safety of Underground Storage Tanks (May 4, 2001) at 2.

[119] 10/31/07 Rebuttal Report of Charles B. Sosik ("Sosik Rebuttal"), Ex. B to Greenwald Deck, at 7.

[120] USEPA MTBE Pilot Project—Objective 2: Investigate Potential Sources of MTBE Contamination on Long Island That Could Impact Water Supplies or Environmentally Sensitive Areas, January 2008, at ii. Again, neither party cites this study, but there is no reason to believe it is inaccurate. It is admissible as a report of a public agency. See Fed.R.Evid. 803(8)(A).

[121] See USEPA MTBE Pilot Project.

[122] See id. at 5.

[123] See id. MTBE was found at levels below the MCL at a much higher percentage of the stations.

[124] In re MTBE Prods. Liab. Litig., Nos. 04 Civ. 4968, 00 Civ. 1898, 2007 WL 700819, at *3 n. 30 (S.D.N.Y. Mar. 7, 2007).

[125] Sosik Rebuttal at 2.

[126] See id. Investigations in Long Island have identified MTBE plumes between six thousand and nine thousand feet long, and have also documented that the plumes will travel at a rate of 0.012 to 0.15 feet downward for each foot of distance traveled horizontally from the spill source. See Sosik Report at 6.

[127] Sosik Rebuttal at 2.

[128] Sosik Report at 1-2.

[129] Under Article 12 of the New York Navigation Law and the Petroleum Bulk Storage Regulations, 6 NYCRR § 613.8, all petroleum spills in New York must be reported to the NYSDEC within two hours of discovery, except spills where the quantity is known to be less than five gallons, the spill was contained and under control of the spiller, the spill has not and will not reach the State's water or any land, and the spill is cleaned up within two hours of discovery.

[130] See Sosik Report at 10.

[131] Id.

[132] Sosik Rebuttal at 1.

[133] Sosik explains that he looked for BTEX as well as MTBE because often site investigations did not test for MTBE before 1994. Another reason is that sometimes spills were reported long after they actually occurred, in which case all the MTBE may have already migrated off the site. " See Sosik Report at 10.

[134] Id.

[135] See id.

[136] Sosik Rebuttal at 1.

[137] Sosik Report at 10.

[138] See id. at 2 ("In almost all cases a detached plume [of MTBE] will go undetected until it impacts a private or public supply well or is accidentally discovered in a monitoring well installed for some other purpose than tracking the plume in question.").

[139] Id.

[140] Sosik Rebuttal at 7.

[141] Sosik traced MTBE in a well back to the gasoline spill where the MTBE was released into the environment, not all the way back to its manufacturer. When he identifies a "source" of MTBE in a well, he refers to the entity that released the gasoline. Therefore, other entities in the chain of supply that may be liable for contamination under product liability law are not named in Sosik's report.

[142] Pl. Supp. Opp. to Def. I. at 4.

[143] Indeed, plaintiffs argue that often it was the responsibility of the defendants, who owned or branded the stations where spills occurred, to undertake groundwater testing and ensure that the spill was adequately contained, but that in many cases defendants failed to take such steps. They also argue that allowing defendants to escape liability on the basis that no groundwater testing was performed at their properties after a spill would be fundamentally unfair, as it would reward defendants' irresponsible behavior and provide an incentive to other companies not to fully investigate spills.

[144] 87 N.Y.2d 596, 601-02, 640 N.Y.S.2d 860, 663 N.E.2d 901 (1996).

[145] Healey concerned a tire rim that had explosively separated from a tire, seriously injuring the plaintiff. The trucking company where plaintiff worked had lost the particular rims believed to be the only ones that could have caused the accident, making it impossible for plaintiff to identify the manufacturer of the defective product. See id. at 602-04, 640 N.Y.S.2d 860, 663 N.E.2d 901.

[146] See, e.g., Masten v. Texas Co., 194 N.C. 540, 140 S.E. 89 (1927) (denying summary judgment where there was evidence that plaintiff's well was downhill from defendant's gasoline storage tank, groundwater flowed from the tank toward the well, an excavator had noticed gasoline stains in soil near the tank, and tank was the only one within a half-mile radius of the well).

[147] See South Cent. Bell Telephone Co. v. Gaines Petroleum Co., 499 So. 2d 521, 523 (La.App. 2 Cir.1986) (defendant's gas station was the nearest possible source of gasoline leak that damaged plaintiff's underground telephone wires).

[148] See Wilson v. McLeod Oil Co., 327 N.C. 491, 521-22, 398 S.E.2d 586 (1990) (holding that evidence of groundwater flow from one suspected spill site toward plaintiffs' wells was sufficient to create issue of fact regarding causation, but testimony that groundwater in a lower aquifer could possibly flow in the other direction was not sufficient to support a finding that a different suspected spill site also contributed to contamination).

[149] See id. at 505-06, 398 S.E.2d 586; South Cent. Bell, 499 So.2d at 523.

[150] See South Cent. Bell, 499 So.2d at 523.

[151] See Gerst v. Marshall, 549 N.W.2d 810 (Iowa 1996) (upholding grant of summary judgment for defendants when plaintiffs were unable to identify how or when a release of gasoline occurred). But see Malone v. Ware Oil Co., 179 Ill.App.3d 730, 128 Ill. Dec. 558, 534 N.E.2d 1003, 1005 (1989) (concluding "evidence was sufficient to support a jury determination that gasoline from defendant's service station had migrated underground to plaintiffs property and caused the contamination" despite plaintiffs expert's inability to identify any particular release of gasoline from the station, where monitoring wells downgradient of station revealed contamination).

[152] Sosik also lists registered petroleum bulk storage ("PBS") tanks where no documented gasoline release has occurred; these are discussed below. In addition, several releases discussed in Sosik's report turned out to be downgradient from the well; as discussed below, these spills could not support a jury finding of causation.

[153] As discussed earlier, MTBE manufacturers and gasoline refiners may still be liable for contamination in the well under the commingled product theory. See supra Part IV. It is also important to point out that in some cases, because many oil companies are vertically integrated, a company may not be liable (as a retailer) for spilling gasoline, but it may be liable as a refiner of gasoline for product liability and negligence claims.

[154] Matott v. Ward, 48 N.Y.2d 455, 460, 423 N.Y.S.2d 645, 399 N.E.2d 532 (1979) (discussing formulation of medical expert's opinion regarding causation, and holding that the issue of causation was properly presented to the jury although the expert testified that he could not say with certainty that car accident was sole cause of plaintiff s condition).

[155] See, e.g., William Glaberson, "The Courts v. Scientific Certainty," N.Y. Times, June 27, 1999, § 4 (Magazine), at 5 ("Science, which never stops searching for answers, has a high threshold for reaching conclusions: 95 percent certainty, some scientists say, is necessary to decide that one thing probably caused another. But the law must stop its search at the conclusion of each case. So juries in civil cases are told that a mere preponderance of the evidence-51 percent—is enough certainty to render a verdict.").

[156] Matott, 48 N.Y.2d at 462, 423 N.Y.S.2d 645, 399 N.E.2d 532.

[157] Justice Stephen Breyer, "Introduction," in Federal Judicial Center Reference Manual on Scientific Evidence (2d ed. 2000). Justice Breyer nevertheless stressed the importance of scientific accuracy to the just disposition of many types of cases.

[158] Matott, 48 N.Y.2d at 460, 423 N.Y.S.2d 645, 399 N.E.2d 532.

[159] Contamination in each well will generally be an indivisible injury, unless the evidence indicates that one spill caused an initial period of contamination and another spill caused a distinct, subsequent period of contamination.

[160] Id. (quoting Ravo v. Rogatnick, 70 N.Y.2d 305, 311, 520 N.Y.S.2d 533, 514 N.E.2d 1104 (1987)).

[161] The facts in the well-by-well analysis are all drawn from Sosik's source attribution report. Defendants dispute many of Sosik's conclusions.

[162] See Sosik Report at 11.

[163] See id.

[164] See id.

[165] See id.

[166] See id. at 12.

[167] See id.

[168] See id.

[169] See id.

[170] See id.

[171] See id.

[172] See id.

[173] See id. at 13.

[174] See id. at 14.

[175] See id.

[176] See id. at 15.

[177] See id. at 14.

[178] See id. at 15.

[179] See Shell Delivery Spreadsheet, Ex. I to Greenwald Decl.

[180] See Sosik Report at 40.

[181] See id.

[182] See id.

[183] See id. at 41.

[184] See Shell Delivery Spreadsheet.

[185] See Sosik Report at 41.

[186] See id.

[187] See id.

[188] See id.

[189] See id. at 50.

[190] See id.

[191] See id. at 51.

[192] See id.

[193] See id.

[194] See id.

[195] See id. at 52.

[196] See id. at 53.

[197] See id. at 55.

[198] See id. at 56.

[199] See id. at 55.

[200] See id. at 55.

[201] See id. at 54.

[202] See id. at 17.

[203] See id.

[204] See id.

[205] See id.

[206] See id. at 18.

[207] See id.

[208] See Island Transportation Delivery Records and Tosco Pipeline Bills of Lading, Ex. J to Greenwald Decl.

[209] See id.

[210] See id.

[211] Of course, these companies may be liable for product liability and negligence claims based on their role as refiners.

[212] See Sosik Report at 17.

[213] See id. at 18.

[214] See id. ("the spill release would have predated Citgo's occupancy of the site").

[215] See id. at 18-19.

[216] Claims arising out of the 1989-2000 MTBE detections may be barred by the statute of limitations, because the very high levels of MTBE detected in this well prior to 2000 most likely triggered the statute of limitations. See In re MTBE, 2007 WL 1601491, at * 12 n. 100. This is the only well at which MTBE was detected at levels exceeding the MCL before 2000, and defendants' summary judgment motion on the basis of the statute of limitations as to this well was denied in a separate opinion. Because the point at which plaintiffs knew or should have known they were injured is a question of fact for the jury, the Metro and Northville spills that may have caused the 1989-2000 detections cannot be eliminated from the jury's consideration.

[217] See Sosik Report at 43.

[218] See id. at 44.

[219] Id. at 45.

[220] See id. at 19.

[221] See id.

[222] See id. at 20.

[223] See id.

[224] Id.

[225] See id. The history of the Sunoco spill investigation is illustrative of the often arbitrary basis of NYSDEC decisions to investigate some spills thoroughly. A spill was reported in 1990 when contaminated soil was found beneath old USTs that were being removed. NYSDEC closed the spill file quickly, without performing any testing, but reopened it a year later when MTBE was reported in a private residential well about fifty feet from the station's property line. Investigations of the MTBE plume by Sunoco contractors revealed MTBE in groundwater at levels of up to 39,000 ppb in 1998, seven years later, and showed a plume migrating toward the wellfield. Without the impact to the private well, however, it is possible that no further investigation of the spill would have occurred.

[226] See id.

[227] See id. at 35.

[228] See id.

[229] See id.

[230] See id.

[231] See Motiva Enterprises Bills of Lading, Ex. K to Greenwald Decl.

[232] See Defendants' Reply in Support of Def. II ("Def Reply II") at 9. The individual defendants do not admit the truth of plaintiffs' claims but withdraw their motion as to this well to simplify the motion. See id.

[233] See Sosik Report at 35.

[234] See Island Transportation Corp. Delivery Records and Bills of

[235] See Def. Reply II at 9.

[236] See id.

[237] See Sosik Report at 35.

[238] Id. at 36.

[239] See id. at 35.

[240] See id. at 45.

[241] See id.

[242] See id. at 46.

[243] See id.

[244] See id.

[245] Id.

[246] See id.

[247] See id.

[248] See id. Again, the NYSDEC investigation illustrates the arbitrary nature of circumstances leading to the detection of UST leaks. Had the surface spill never occurred, it is unlikely that the UST leak would have been discovered.

[249] See id.

[250] See id. at 47.

[251] See id.

[252] See id.

[253] See id. at 32.

[254] See id. at 32, 34.

[255] See id. at 33.

[256] See id.

[257] See id.

[258] See id.

[259] See id.

[260] Sosik relied on NYSDEC spill reports to locate all known spills within each well's capture zone.

[261] See Summary of Underground Injection Control Baydrain Closure Activities at Getty Service Station, 734 Park Avenue Huntington, New York, December 1992 ("Summary"), Ex. C to Declaration of Steven J. German in Support of Plaintiffs' Opposition to Defendants' Motion for Summary Judgment on Plaintiffs' Navigation Law Claim. The report would likely be admissible as a public record describing factual findings resulting from an investigation pursuant to the agency's authority, under Federal Rule of Evidence 803(8)(C).

[262] Summary. The report mentions both the Nassau County Department of Health and the SCDHS as overseeing the project.

[263] Id.

[264] Id.

[265] See id. at 25.

[266] See id

[267] See id.

[268] See South Cent. Bell, 499 So.2d at 523.

[269] See Island Transportation delivery documents, Ex. C to PL Supp. Opp. to Def. I.

[270] See id.

[271] See Sosik Report at 25-26. Another spill was reported at the Northville gas station in 1990 when contaminated soil was discovered during the removal of a diesel fuel tank. Because the diesel fuel that leaked from the tank would not have contained MTBE, the 1990 spill at this station could not have caused the contamination in the Crystal Brook Hollow well. See id. at 25.

[272] See id.

[273] See id. at 26.

[274] See id.

[275] See id.

[276] See id.

[277] See id. at 23.

[278] See id.

[279] See id. at 24.

[280] See id.

[281] See id. at 22.

[282] See id.

[283] See id.

[284] See id.

[285] See id.

[286] See id. at 42.

[287] See id. According to Sosik, there is no other information about this spill in the NYDEC file. See id.

[288] See id.

[289] See id.

[290] See id. at 43.

[291] Id.

[292] See id. at 27.

[293] See id.

[294] See id.

[295] See id.

[296] See id.

[297] See id. at 28.

[298] See id.

[299] See id. at 29. This is the same 2003 Mobil spill discussed above for the Wheat Path well.

[300] See id.

[301] See id. at 38.

[302] See id.

[303] See id.

[304] See id. at 39.

[305] See id.

[306] See id.

[307] See id.

[308] See id.

[309] A jury must also consider the dates the spill may have occurred in conjunction with the dates of detection. It may find that the Amoco spill, reported in 2002, could only have been responsible for later contributions to the contamination, which was first detected in 1993.

[310] See id.