OPINION AND ORDER

I. INTRODUCTION

In this consolidated multi-district litigation (“MDL”), plaintiffs seek relief from contamination, or threatened contamina*584tion, of groundwater from various defendants’ use of the gasoline additive methyl tertiary butyl ether (“MTBE”) and/or tertiary butyl alcohol, which is a product formed by the natural degradation of MTBE in water. Plaintiff, the City of New York (“City”),1 filed an action against various defendants — including the only defendants remaining in this case, ExxonMobil Corporation, ExxonMobil Oil Corporation, and Mobil Corporation (collectively, “ExxonMobil”) — in 2003. Because of the size and complexity of this case, five of the dozens of wells at issue were selected for a “bellwether” trial. On October 19, 2009, after an eleven week trial, a jury awarded the City approximately one hundred and four million dollars. ExxonMobil has now renewed its motion for judgment as a matter of law on the City’s claims and moved, in the alternative, for a new trial and/or remittitur.

II. BACKGROUND

The City asserts both state law tort claims and a federal law claim under the Toxic Substances Control Act (“TSCA”) against ExxonMobil. However, because the TSCA claim is not specific to the five focus wells at issue in this case, only the state claims were tried. Although the trial was originally divided into four phases, because I dismissed the City’s punitive damages claim as a matter of law,2 only three phases were tried.

A. Phase I

All five focus wells are located within an uncompleted facility in Jamaica, Queens known as “Station 6” that is not presently, and has never been, used to distribute water to New York City residents. Because the City is not presently using the Station 6 wells, the jury was asked in Phase I to determine whether the City intended to use Station 6 in the future. Specifically, the jury was asked (1) “whether the City has proven by a preponderance of the evidence that it intends in good faith to begin construction of the Station 6 facility within the next 15 years”3 and (2) “whether the City has proven by a preponderance of the evidence that it intends in good faith to use the water from the Station 6 wells within the next 15 to 20 years, either to supply drinking water to its residents or to serve as a backup source of drinking water in case of shortage of supply.”4 The jury answered “yes” to both questions, but determined that the City only intends to use the Station 6 wells as a backup source of drinking water.5

B. Phase II

In Phase II, the jury was asked to make several findings relating to whether there will be any MTBE in the Station 6 wells if and when the City begins to use them as a backup source of water. These interrogatories were modeled around a series of rulings I made prior to trial. First, the City is not injured by the mere presence of MTBE in the groundwater. Instead, “the City is injured by a concentration of MTBE in the groundwater [only] if a reasonable water provider would take action to monitor, test and/or treat groundwater *585containing that level of MTBE.”6 New York state has set a Maximum Contaminant Level (“MCL”) for MTBE in drinking water of ten parts per billion (“ppb”).7 As such, the City is injured as a matter of law when the contamination in the combined outflow of the Station 6 wells exceeds ten ppb. If, however, the concentration level in the combined outflow of the Station 6 wells is at or below ten ppb, whether the City has suffered an injury is a question of fact properly decided by the jury.8

Second, although the sufficiency of the City’s injury is measured by the amount of MTBE in the combined outflow of the Station 6 wells, the location of that injury is the groundwater that will be drawn into the Station 6 wells when they begin operation.9 This area is known as the “capture zone” of the Station 6 wells. Thus, “[if] the City can show that its wells will become injured immediately upon turning them on,” i.e., that MTBE is within the Station 6 capture zone, “it need not go through the curious exercise of turning the wells on to injure itself.” 10

The jury was asked two questions corresponding to these holdings. It was first asked “whether the City has proven, by a fair preponderance of the credible evidence, that MTBE will be in the groundwater of the capture zone of the Station 6 wells when they being operating as a backup source of drinking water.”11 It was then asked “at what peak level MTBE will be found in the combined outflow of the Station 6 wells and when that will occur.” 12 In answering the latter question, the jury was directed to select one of several concentration ranges; less than one ppb; one to three ppb; three to five ppb; five to ten ppb; or more than ten ppb.13

The City introduced the testimony of its hydrogeology expert, David Terry, to aid the jury in answering these questions. Terry explained to the jury that he had created two groundwater models designed to evaluate how MTBE will impact Station 6. He created a flow model, which shows “where the groundwater flows” and “how fast it moves,” 14 to predict the likely size and shape of the Station 6 capture zone. This model relied on the “proposed pumping scenario” — e.g., the location of the pumping wells, the pumping rates of the wells, and the schedule on which the wells would pump — provided by New York City planners for both the Station 6 wells and other surrounding wells.15 As Terry ex*586plained, the size and shape of the Station 6 capture zone depends heavily on the assumed pumping scenario.16

Terry then described a transport model he had created to show “how [MTBE will] move through the groundwater system.”17 The transport model, which “rides on top of the flow model,”18 is used to make “numerical predictions” regarding the amount of MTBE that will enter the Station 6 wells.19 As with the flow model, the transport model relies on specific assumptions — e.g., the locations of MTBE discharges and the levels of those discharges — that can greatly affect the model’s results.20

Terry ran two different transport models based on varying assumptions. He ran an “Analysis 1” model — which used recorded MTBE concentrations from groundwater samples taken in 2004 — to determine the peak concentration of MTBE in the Station 6 wells and the date of that peak concentration.21 On the basis of this analysis, he predicted that the maximum MTBE concentration in the combined outflow of the Station 6 wells would be thirty-five ppb in 2024.22

Terry also ran an “Analysis 2” model using the same transport and flow models but with different assumptions about the amount of MTBE present in the groundwater. As Terry explained, the purpose of Analysis 2 was to see how long MTBE contamination will last at Station 6.23 To do so, Terry input three separate scenarios which assumed that the amount of MTBE gasoline spilled at each of twenty-two known release sites was on average fifty gallons, five hundred gallons or two thousand gallons.24 The peak concentration values predicted by Analysis 2 ranged from non-detect (fifty gallons per release) to six ppb (five hundred gallons per release) to seventeen to twenty-three ppb (two thousand gallons per release).25 Terry concluded that both the two thousand gallon and five hundred gallon scenarios would result in an “impact to Station 6.”26

After deliberating for two days at the close of Phase II, the jury informed the Court that it was unable to come to a unanimous decision on the second question — the level and timing of MTBE’s peak concentration.27 Because the jury was in*587structed that it should only reach the second question if it answered the first question affirmatively, I inquired as to whether the jury had reached a verdict on the first question. The jury informed me that it had reached a unanimous verdict on that question28 and that it had found that MTBE would be present in the Station 6 capture zone when the City began to operate those wells.29 After taking the jury’s partial verdict, I then gave the jury an Allen charge on the second question.30 In summing up that charge, the jury was instructed:

Do not hesitate to reexamine your own views and to change your opinion if you are convinced you were wrong, but do not surrender your honest belief as to the weight and effect of the evidence solely because of the opinion of your fellow jurors or for the mere purpose of returning a verdict.31

Shortly afterward, the jury submitted another note to the Court — querying, inter alia, whether the second question could be restructured or rephrased.32 In response, I instructed the jury that it did not need to select one of the concentration ranges provided by the Court and that it could choose any concentration level that it was able to agree upon unanimously.33 After further deliberation, the jury reached a verdict. It did not settle on any of the suggested ranges — finding instead that the concentration of MTBE in the combined outflow of the Station 6 wells will peak at ten ppb in 2033.34

C. Phase III

In Phase III, the jury answered specific interrogatories relating to: (1) whether the City is, or will be, injured by the MTBE that will be in the combined outflow of the Station 6 wells when the City begins to use them as a backup source of water; (2) whether ExxonMobil was a cause of that injury (if it exists) as either a direct spiller of MTBE gasoline or as a manufacturer, refiner, supplier or seller of MTBE gasoline; (3) whether ExxonMobil was liable under the various state law causes of action asserted by the City; and (4) if ExxonMobil was hable, the amount of compensatory damages that should be awarded to the City.

1. Injury

In Phase III, the jury determined that the City had “proven, by a fair preponderance of the credible evidence, that it is, or will be, injured by the MTBE that will be in the combined outflow of the Station 6 wells.”35 In making this determination, the jury was instructed that “the question is whether the [C]ity has proven by a fair preponderance of the credible evidence that a reasonable water provider in the [CJity’s position would treat the water to reduce the levels or minimize the effects of the MTBE in the combined outflow of the Station 6 wells in order to use the water as a back-up source of drinking water.”36 I also informed the jury that I had previously ruled that the concentration of MTBE need not exceed New York’s MCL to constitute an injury,37 but that it was up to the jury “to determine whether the level of MTBE [it] ha[d] found will be in the Sta*588tion 6 wells in the future will constitute an injury to the [C]ity.”38

2. Causation

The City asserted three separate theories of causation at trial — direct spiller causation, manufacturer/r efiner/supplier/seller causation, and a commingled product theory of causation. As I explain in more detail below,39 the commingled product theory of causation has been developed during the course of this MDL as an alternative to traditional theories of toxic tort causation.40 The jury found that the City proved, by a fair preponderance of the credible evidence that ExxonMobil was a cause of the City’s injuries both as a direct spiller of gasoline containing MTBE and as a manufacturer, refiner, supplier or seller of gasoline containing MTBE.41 Because the jury found that the City met its burden under these traditional theories of causation, it did not determine whether the City met its burden under the commingled product theory of causation.42

a. Direct Spiller Causation

The jury was instructed that Exxon-Mobil was the owner or controller of underground storage tank (“UST”) systems located in specific stations in Queens County, New York and that ExxonMobil was liable as direct spiller if it found:(a) that “[a]t any time that [ExxonMobil] owned or controlled some or all of these underground storage systems, they leaked gasoline containing MTBE”; and (b) “[t]hat these leaks caused or will cause an injury to the [C]ity’s Station 6 wells.”43 The jury was also instructed that “[a]n act or omission is regarded as a cause of an injury if it is a substantial factor in bringing about the injury; that is, if it has such an effect in producing the injury that reasonable people would regard it as a cause of the injury.”44

b. Manufacturer/Refiner/Supplier/Seller Causation

The jury was instructed that ExxonMobil was a cause of the City’s injury in its capacity as a manufacturer, refiner, supplier or seller if its “conduct in [these capacities] was a substantial factor in causing the [C]ity’s injuries.”45 As before, a substantial factor was defined as having “such an effect in producing the injury that reasonable people would regard it as a cause of the injury.”46 In making this decision, the jury was told that it “should consider how much, if any, of the gasoline containing MTBE that was delivered to the locations that are the sources of the MTBE that injured or will injure the Station 6 wells came from gasoline containing MTBE that was manufactured, refined, supplied, or sold by [ExxonMobil].” 47

3. State Law Causes of Action

Relying on its injury and causation findings, the jury found ExxonMobil liable under the City’s public nuisance,48 trespass,49 direct spiller negligence50 and failure to warn (product liability)51 claims. Howev*589er, it found that ExxonMobil was not liable under the City’s defective design (product liability)52 and private nuisance53 claims. With respect to the City’ design defect claim, the jury was instructed that the City must prove three elements by a fair preponderance of the credible evidence: (1) that “gasoline containing MTBE was not reasonably safe for its intended or reasonably foreseeable purpose or in light of the reasonably foreseeable harms caused by its use;”54 (2) that “there was a safer feasible alternative design at the time it was marketed;”55 and (3) that “the defective design was a substantial factor in causing the City’s injury.”56 Although the jury determined that the design was not reasonably safe,57 it found that there was no safer, feasible alternative design to MTBE.58 Accordingly, it did not reach the causation question.59 The jury was not asked to answer specific interrogatories relating to the City’s private nuisance claim.

4. Affirmative Defense: Statute of Limitations

The jury also considered one affirmative defense — whether the City had brought its claims in a timely manner — asserted by ExxonMobil. To succeed on this defense, ExxonMobil was required to show by a fair preponderance of the credible evidence that, prior to October 31, 2000, (1) “there was a sufficient level of MTBE in the capture zone of the Station 6 wells such that if the wells were turned on, the level of MTBE in the combined outflow of the Station 6 wells would have injured the City at that time,”60 and (2) “the [C]ity knew at the time or reasonably should have known that there was a sufficient level of MTBE in the capture zone of the Station 6 wells prior to October 31, 2000 to cause an injury.”61 The jury was again instructed that the “[C]ity’s water was not injured, or otherwise harmed, merely because it has been contaminated by a small amount of MTBE,”62 but that “ExxonMobil need not necessarily prove that the [CJity’s water was contaminated at that time by a concentration of MTBE that exceeded the maximum contaminant levels set by regulatory authorities.”63 The jury determined that ExxonMobil had failed to prove that the City brought its claims in an untimely manner.64

5. Damages

The jury was instructed that if it found ExxonMobil liable under any of the City’s causes of action, it was required to award the City compensatory damages sufficient to compensate the City for losses caused by ExxonMobil.65 The jury made its damages determination in four stages. First, it was asked what sum of money will compensate the City for the actual losses it has sustained, or will sustain in the future, as a result of contamination at Station 6.66 In making this determination, the jury was given the following guidance:

*590Generally, the measure of damages to property is based upon the value of the property at the time of the injury. When, as in this case, the property damaged has no reasonable market value, plaintiff may recover the difference in money between the value to plaintiff of the property before and after the damage. In determining the amount of such loss, you must consider the evidence presented with respect to its utility, its general condition at the time of its damages, whether it may be treated, and, if so, the expense of treating it, the likelihood that it will be treated if the money is available to do so, and its value as so treated as compared to the value before its damage, together with all other evidence presented to establish its value to the City and the extent of the City’s damage.67

The jury found that the City will be fairly and reasonably compensated by an award of $250,500,000. In doing so, the jury evidently adopted the opinion of Marnie Bell, the City’s damages expert — who testified that, assuming contamination at ten ppb, it will cost the City $250,450,000 to remove MTBE from the Station 6 wells.68 This estimate was based on three components: (1) $59,990,000 in initial capital to construct the Station 6 treatment plant; (2) $48,870,000 in equipment replacement costs; and (3) $141,590,000 in operation and maintenance costs.69 The operation and maintenance costs were based on the “worst case scenario” that the Station 6 wells will have to be operated continuously for forty years (which is the approximate amount of time that MTBE contamination at the Station 6 wells is predicted to endure).70

Second, ExxonMobil contended at trial that the groundwater in the Station 6 capture zone is polluted by other contaminants and that these other contaminants will require the City to treat the Station 6 wells regardless of whether there is MTBE contamination.71 Accordingly the jury was instructed:

If you find that ExxonMobil has shown, by a fair preponderance of the credible evidence, that the costs of treating the other contaminants in isolation can be fairly estimated, then you must reduce the City’s damage award for treating MTBE by the cost of treating these other contaminants in isolation.72

The jury found that the City’s award should be reduced by seventy million dollars due to the pre-existing contamination.

Third, the jury was asked whether other companies (defendants that were voluntarily dismissed pursuant to settlement agreements) were also at fault in producing the City’s injury.73 The jury was given a list of these companies and was required to “decide the percentage of the total fault borne by these other companies as compared to ExxonMobil’s fault.”74 In total, the jury found that forty-two percent of the fault for the City’s injury should be apportioned to these other companies.

Fourth, and finally, the jury was asked to determine whether the City was negli*591gent as a spiller of MTBE gasoline and, if so, whether the City’s negligent conduct was a substantial factor in causing its own injury.75 The jury found that the City had no responsibility for causing its own injury.

In sum, the jury determined that the City’s actual damages, taking into account the amount the City will have to pay to remediate pre-existing contamination in the groundwater, are $180,500,000 and that ExxonMobil was responsible for fifty-eight percent of those damages. Pursuant to these findings, ExxonMobil is liable to the City in the amount of $104,690,000.

6. Juror Misconduct

Two issues of juror misconduct arose during the jury’s Phase III deliberations. The first was the revelation that one juror had conducted Internet research during the Phase III deliberations. He was subsequently excused from the jury. The second was an apparent instance of uncivil behavior during the Phase III deliberations — which again resulted in the dismissal of a juror.

a. Internet Research

On October 9, 2009, shortly after the jury was charged in Phase III, I received a note from the jury foreperson that one of the jurors had conducted research about the case on the Internet. That note stated: “Juror No. 8 ... has admitted to researching on the Internet. He proclaimed that the information he found strengthened his opinion. We feel that he will be biased in our deliberations. This information was brought to our attention on Wednesday, October 7th, and we are concerned.”76

Shortly after receiving the note, and conferring with counsel, I conducted a voir dire of Juror No. 8. He informed the Court that his children had performed Internet research relating to this case and shown him the results.77 Specifically, Juror No. 8 admitted to viewing reports relating to the jury’s verdicts in Phase I and Phase II,78 a prior opinion of this Court describing the commingled product theory of causation,79 and some information about one of the City’s lawyers, Victor Sher. When I conducted a second voir dire of Juror No. 8, he opined that the other jurors did not want him on the jury anymore because he was “usually the last hold out on all the questions.”80 However, he made clear that he was not indicating the substance of the jury’s deliberations, and I informed him that he should not do so.81

Although Juror No. 8 initially stated that “everybody” had performed Internet research and that some jurors had driven to Station 6,82 after being pressed by the Court on this claim, he clarified that he did not know of any jurors who had performed Internet research and that he only knew of one juror who had tried to travel to Station 6 (but had not found it).83 Nevertheless, to determine what information the jury had been exposed to, I conducted a voir dire of the remaining jurors.84 Each juror was asked individually if he or she had performed any outside research or learned any information from another juror who had performed outside research.85

*592In the course of this investigation, several of the jurors admitted that they had been exposed to some limited outside information relating to the case. First, Juror No. 4 indicated that he had tried to drive over to Station 6 but that he could not find it.86 Second, Juror No. 5 stated that Juror No. 8 informed him that he had found out “that Exxon was the last oil company to kind of hold out and wasn’t settling out of court”87 and that the City had received about one million dollars from each company.88 Juror No. 5 asserted that he had not performed any outside research himself.89 Third, Juror No. 6 conducted some research about me on Wikipedia and read an article in the New York Times about water contamination caused by the coal industry.90 She also stated that Juror No. 8 had told her that plaintiffs in other cases had received money, but that “he didn’t really say the details of the rulings.”91 Fourth, and finally, Juror No. 11 stated that Juror No. 8 had informed him that the trial was going to have a fourth phase92 and that Juror No. 8 had learned that by conducting Internet research.93

After conducting this investigation, I excused Juror No. 8 from the jury. However, because I did not believe that the rest of the jury had been prejudiced by its exposure to extra-record information, I did not excuse any of the other jurors and denied ExxonMobil’s motion for a mistrial,

b. Uncivil Behavior

On the evening of October 15, 2009, I received a phone call from Juror No. 2— who, clearly shaken by the experience, informed the Court that Juror No. 1 had insulted her and threatened to “cut” her during deliberations.94 In response to that allegation, on October 16, 2009, I questioned every juror individually about the alleged incident and the general level of civility in the jury room. While a few jurors noted that discussions among jurors had become heated,95 no other juror claimed that he or she felt threatened or was unable to continue deliberating.96 When Juror No. 2 was questioned individually, however, she continued to assert that she felt threatened and felt as though she could not freely make her own decisions.97

Because Juror No. 2 continued to assert that she felt threatened, I subsequently excused her from the jury. I did not, however, find it necessary to excuse any of the other jurors. As I stated to counsel:

So, as you hear, by voir diring each juror one by one, I am confident that every juror, other than Juror No. 2, feels he or she has not been threatened, that it is just a natural part of deliberations, that people discuss things and do *593so in a vigorous way. But I am absolutely confident nobody feels threatened other than Juror No. 2, and she says she no longer feels she can reach her own verdict. So it strikes me that she ought to be excused .... 98

Both ExxonMobil and the City agreed that Juror No. 2 should be excused.99

In addition, ExxonMobil moved that Juror No. 1 “be excused for threatening a juror with physical violence,”100 and requested that I question Juror No. 2 again to determine whether she had actually been threatened with physical violence.101 Although I doubted that Juror No. 1 should be dismissed, I acceded to Exxon-Mobil’s request in order to create a record for appeal. When questioned for a second time, Juror No. 2 asserted again that she had been mistreated: “I was called names yesterday. I was called — at the end of the day, they said I was stupid, I can’t form my own opinion because it doesn’t match the rest of them. And I feel — I feel that I am not safe.”102 When prodded to explain further why she felt unsafe, she explained that Juror No. 1 had “threatened to ... cut [her] and ... threatened [her] with a fork.”103

ExxonMobil subsequently moved for a mistrial “based on the further developing ... fact [that] ... an actual instrument was used in the jury room....”104 I denied ExxonMobil’s motion — noting that none of the other jurors had expressed discomfort or indicated that Juror No. 1 had brandished a fork in a threatening manner.105 Given these facts, I deemed it permissible for the jury to continue deliberating — albeit after I instructed them on the importance of civility and asked them to refrain from “shouting,” “cursing” or engaging in any “threatening behavior.”106

III. LEGAL STANDARD

A. Motion for Judgment as a Matter of Law Under Rule 50 and Motion for a New Trial Under Rule 59

A court may render judgment as a matter of law when “a party has been fully heard on an issue during a jury trial and the court finds that a reasonable jury would not have a legally sufficient evidentiary basis to find for the party on that issue.”107 The standard for granting judgment as a matter of law “mirrors” the standard for granting summary judgment.108 Accordingly, in ruling on such a motion, the trial court is required to

consider the evidence in the light most favorable to the party against whom the motion was made and to give that party *594the benefit of all reasonable inferences that the jury might have drawn in his favor from the evidence. The court cannot assess the weight of conflicting evidence, pass on the credibility of the witnesses, or substitute its judgment for that of the jury.109

A jury verdict cannot be set aside lightly. A court may not grant judgment as a matter of law unless (1) there is such a “complete absence of evidence supporting the verdict that the jury’s findings could only have been the result of sheer surmise and conjecture” or (2) there is “such an overwhelming amount of evidence in favor of the movant that reasonable and fair minded [persons] could not arrive at a verdict against [it].” 110

A “court may, on motion, grant a new trial on all or some of the issues ... after a jury trial, for any reason for which a new trial has heretofore been granted in an action at law in federal court.” 111 The legal test for granting a new trial is less stringent than for granting judgment as a matter of law. “Unlike a motion for judgment as a matter of law, a motion for a new trial may , be granted even if there is substantial evidence to support the jury’s verdict.”112 Nevertheless, in practice courts do not grant new trials as freely as the language suggests. “ ‘A motion for a new trial ordinarily should not be granted unless the trial court is convinced that the jury has reached a seriously erroneous result or that the verdict is a miscarriage of justice.’ ”113

Under Federal Rule of Civil Procedure 50(b), “[n]o later than 28 days after the entry of judgment ... the movant may file a renewed judgment as a matter of law and may include an alternative or joint request for a new trial under Rule 59.” “In ruling on the renewed motion, the court may: (1) allow judgment on the verdict, if the jury returned a verdict; (2) order a new trial; or (3) direct entry of judgment as a matter of law.”

B. Motion for a New Trial on Damages or Remittur

“If a district court finds that a verdict is excessive, it may order a new trial, a new trial limited to damages, or, under the practice of remittitur, may condition a denial of a motion for a new trial on the plaintiffs accepting damages in a reduced amount.” 114 A district court must apply New York law to evaluate whether awards in cases decided under New York law are excessive.115 Pursuant to Section 5501(c) of the New York Civil Practice Law and Rules (“CPLR”), a court should reduce an award when it “deviates materially” from “reasonable compensation.”116 *595“[The New York] standard is less deferential to the jury and thus more favorable to the party challenging the award than is the federal ‘shocks the conscience’ [standard].” 117

IV. SUBSTANTIVE LAW

A. Federal Preemption

The doctrine of federal preemption is rooted in the Supremacy Clause of the United States Constitution, which provides that federal law made pursuant to authority granted by the Constitution “shall be the supreme Law of the Land.”118 “[This Court’s] inquiry into the scope of a statute’s pre-emptive effect is guided by the rule that ‘[t]he purpose of Congress is the ultimate touchstone’ in every pre-emption case.’ ” 119 “The Supreme Court has identified three situations that show congressional intent to preempt state law: (1) where Congress expressly states its intent to preempt [express preemption]; (2) where Congress’s scheme of federal regulation is sufficiently comprehensive to give rise to a reasonable inference that it leaves no room for the state to act [field preemption]; and (3) where state law actually conflicts with federal law [conflict preemption].”120

Conflict preemption, which is at issue in this case, occurs “where it is impossible for a private party to comply with both state and federal requirements, or where state law stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress.” 121 “Absent clear congressional intent to the contrary, federal preemption of state law is not favored, especially in areas of law traditionally occupied by the states.”122 Impossibility requires “physical impossibility”123 — e.g., if federal law says do X, and a state law says do not do X. Although the “obstacle preemption” prong is arguably broader than the “physical impossibility” prong, it is by no means easy to satisfy. It requires clear evidence of congressional intent and “a sharp conflict between state law and federal policy.” 124

B. Statute of Limitations

“The statute of limitations is normally an affirmative defense on which the defendant has the burden of proof.”125 The common law property tort claims for which the City seeks damages are governed by the three year limitations period of section 214(4) of the CPLR. This provision, however, is altered by the discovery *596rule of section 214-c(2), which modifies the date on which the three year limitation period begins to run:

Notwithstanding the provisions of section 214, the three year period within which an action to recover damages for ... injury to property caused by the latent effects of exposure to any substance or combination of substances, in any form, upon or within ... property must be commenced shall be computed from the date of discovery of the injury by the plaintiff or from the date when through the exercise of reasonable diligence such injury should have been discovered by the plaintiff, whichever is earlier.

In interpreting New York’s discovery rule in prior rulings, I have explained that the City’s state law claims “accrue when it first knows of both (1) the presence of MTBE at a level sufficient to constitute an injury and (2) the harmful impact of MTBE on drinking water.” 126

C. Causation

As a general rule, the plaintiff in a tort action must prove by a preponderance of the evidence that the defendant’s conduct was the factual and proximate (or legal) cause of its injury. This principle has raised an astonishing number of issues in this MDL, and I have spent a great deal of time and effort determining the applicability of both traditional and alternative theories of causation. Although the jury in this case was only charged on New York’s traditional rules of causation and the commingled product theory of causation, I also describe the concurrent wrongdoing theory of liability and the market share theory of liability as they are important to understanding the arguments asserted in these post-trial motions.

1. Traditional Causation

a. Factual and Proximate Cause

As stated, tort liability usually depends upon proof that a defendant’s conduct was the legal cause of a plaintiffs injury.127 The substantial factor standard for causation, adopted in New York, requires that the plaintiff prove by a preponderance of the evidence that the defendant’s conduct was a substantial factor in producing plaintiffs injury.128 The Second Restatement of Torts outlines three considerations that are important in determining whether defendant’s conduct was a substantial factor in bringing about a harm: (1) “the number of other factors which contribute in producing the harm and the extent of the effect which they have in producing it;” (2) “whether the actor’s conduct has created a force or series of forces which are in continuous and active operation up to the time of the harm, or has created a situation harmless unless acted upon by other forces for which the actor is not responsible;” and (3) the “lapse of time” between the conduct and the injury.129

*597b. Identification of Injurious Product

In addition, and of relevance to this case, New York generally requires “identification of the exact defendant whose product injured the plaintiff.”130 “The identity of the manufacturer of a defective product may be established by circumstantial evidence.”131 However, such evidence cannot be “speculative or conjectural.” 132 “The 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.” 133

2. Alternative Theories of Liability

In limited circumstances, courts permit alternative theories of liability where traditional principles of causation do not establish liability in order to protect the interests of plaintiffs and properly apportion liability among defendants. First, under the concurrent wrongdoing theory of liability, “[w]hen two or more tortfeasors act concurrently or in concert to produce a single injury, they may be held jointly and severally liable.”134 Significantly, in concurrent wrongdoing cases, plaintiffs need not prove that each defendant’s conduct, taken alone, would have injured the plaintiff.135 Instead, “each tortfeasor is responsible for the entire result, even though [its] act alone might not have caused it.” 136

Second, New York has on occasion permitted plaintiffs to proceed on a market share theory of liability. This theory of liability was first adopted by the New York Court of Appeals in Hymowitz v. Eli Lilly & Co. to determine liability and apportion damages in diethylstilbestrol (“DES”) cases.137 DES is a synthetic substance taken by women during pregnancy for the purpose of preventing miscarriages.138 It was banned in 1971 by the Federal Drug Administration when studies established a link between women who took the drug and harmful latent effects in their offspring.139 For various reasons, including the long latency period of DES injuries, women who used DES generally did not know which company had manufactured the DES pills they ingested. To protect the interest of those injured by DES, the Court of Appeals determined that liability may be apportioned according to defendants’ national market share in cases in which identification of the manufacturer of the drug that injured the plaintiff is impossible.140 Thus, the market-*598share theory “provides an exception to the general rule that a plaintiff must prove that the defendant’s conduct was a cause-in-fact of the injury.” 141 A defendant may be held partially liable under the market-share theory without any showing that the defendant caused, or contributed to, the injury.142

Third, in order to protect the interests of plaintiffs in this MDL while fairly apportioning liability, I have developed the commingled product theory of liability. “[Ujnder the ‘commingled product theory’ of market share liability, 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 plaintiffs injury, each refiner or manufacturer is deemed to have caused the harm.”143 Each defendant is then given the opportunity 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.”144

This hybrid theory has similarities to both market share liability and concurrent wrongdoing liability. Like market share liability, damages are apportioned according to each defendant’s share of the market at the time of injury, and thus, liability is several, rather than joint and several.145 Unlike market share liability, however, a plaintiff must prove by a preponderance of the evidence that the defendant contributed-in-fact to the injury by showing that each defendant’s product was part of the commingled mass that injured the plaintiff.146 In this respect, commingled product liability is similar to concurrent wrongdoing liability. It requires the plaintiff to prove that each defendant’s gasoline was part of the commingled product, but relieves the plaintiff of the duty to prove that each individual defendant’s contribution to that product, taken by itself, was sufficient to have caused an injury.147 “Rather, to establish liability against a particular defendant with respect to an individual well, [the plaintiff] must show that (a) the defendant’s MTBE was present in a commingled product and (b) ‘the commingled product [rather than defendant’s product alone ] caused plaintiffs injury.’ ”148

V. DISCUSSION

ExxonMobil has made a plethora of arguments in its post-trial motions. These can be summarized as follows: (1) it proved its affirmative defenses at trial; (2) the City failed to prove two essential elements of its state law causes of action-— injury and causation; (3) several of the City’s causes of action should be dismissed as a matter of law; (4) the Court made erroneous evidentiary rulings; (5) a certain juror should have been excused; (6) a mistrial should have been granted because *599a juror impermissibly conducted research on the Internet; and (7) the jury’s damages award is excessive.

A. Affirmative Defense: Federal Preemption

In 1990, Congress enacted amendments to the Clean Air Act (“CAA”) that, inter alia, created the Reformulated Gasoline Program (“RFG Program”).149 The RFG Program required gasoline used in specific geographic areas to have a minimum oxygen content — achieved by the addition of oxygenates such as MTBE and ethanol.150 After the passage of the CAA amendments, the Environmental Protection Agency (“EPA”) certified various blends of gasoline for use in the RFG Program, including gasoline containing MTBE, but did not mandate the use of any one oxygenate.151 Congress repealed the oxygenate requirement in 2005.152

ExxonMobil previously sought summary judgment on the ground that the City’s state tort claims — by imposing a duty not to use MTBE — are preempted by the CAA amendments.153 Although I rejected that motion, ExxonMobil argues that the jury’s determination (in the context of the City’s design defect claim) that there was no safer, feasible alternative to MTBE shows that ExxonMobil proved its affirmative defense of conflict preemption at trial. ExxonMobil urges this theory under both prongs of the conflict preemption test-physical impossibility and obstacle preemption.

1. Physical Impossibility

Because the RFG Program did not expressly require the use of MTBE over other oxygenates, ExxonMobil does not maintain that the amendments to the CAA and the City’s state tort claims are incompatible on their face. Instead, it argues that due to the specific factual circumstances of this case (i.e., that there was no safer, feasible alternative to MTBE), it was impossible for ExxonMobil to comply with federal requirements without using MTBE.154 This argument is problematic for two reasons.

First, the “safer, feasible alternative design” products liability standard is not equivalent to the “physical impossibility” preemption standard. In evaluating the feasibility of an alternative design, the finder of fact weighs the costs and benefits of the marketed product against the costs and benefits of proposed alternatives.155 *600This Court is unaware, and ExxonMobil has not cited, any judicial opinion suggesting that physical impossibility in the federal preemption context entails this sort of utilitarian analysis. “That it may have been more convenient or less expensive for the [ExxonMobil] to use MTBE does not mean it would have been impossible for [it] to use other, less-polluting additives.”156 Accordingly, the jury’s feasibility determination does not establish impossibility.

Second, the burden of proof in the design defect context is inconsistent with the burden of proof in the federal preemption context. To prove its design defect claim, the City had the burden of showing that there were no safer, feasible alternative to MTBE.157 By contrast, because federal preemption is an affirmative defense, the burden of proof is on the defendant, ExxonMobil.158 Thus, the jury’s finding that the City failed to prove there was a safer, feasible alternative cannot substitute for a finding that ExxonMobil proved that MTBE was the only oxygenate it could have used to satisfy the RFG requirements.

In addition, regardless of the jury’s verdict, ExxonMobil did not submit evidence at trial sufficient to demonstrate that it was impossible to use another oxygenate. At most, as is consistent with the jury’s feasibility finding, ExxonMobil showed that it would have been more difficult for it to produce reformulated gasoline containing ethanol (the most likely alternative to MTBE).159 It did not establish that it would have been impossible to do so.

2. Obstacle Preemption

ExxonMobil also contends that permitting the verdict to stand will frus*601trate Congress’s intent “to promote the use of MTBE, and other oxygenates.”160 Defendants raised an essentially identical argument in seeking summary judgment161 — -which I denied. Because I do not find that the evidence submitted at trial alters that analysis, I decline to revisit that decision.

Although it is indisputable that Congress intended to promote the use of oxygenates in creating the RFG Program, there is no “indication that Congress or the EPA ‘struck a particular’ balance between water pollution and the ability of the [RFG Program] to expand”162 or that “Congress or the EPA intended to protect the ‘market share of MTBE even if it prove[d] to be inferior to other oxygenates due to environmental considerations other than motor vehicle emissions.’ ”163 Accordingly, I found that allowing plaintiffs to recover damages for “inordinate environmental effects” caused by the use of MTBE does not conflict with federal policy.164 Congress, in fact, considered including a safe harbor provision immunizing MTBE producers and distributors from state tort liability in the Energy Policy Act of 2005, but ultimately chose not to do so.165 While this decision does not provide direct evidence of Congress’s intent at the time it passed the amendments to the CAA, it bolsters the inference that permitting the City’s state tort claims will not impede Congressional objectives.

The jury’s finding that the City failed to prove that there was a safer, feasible alternative to MTBE does not alter that determination. ExxonMobil submitted evidence at trial showing that using ethanol would have been more costly than using other oxygenates. The City’s state tort claims simply provide a counterbalancing economic incentive for ExxonMobil to decrease or eliminate the use of MTBE because of its severe environmental effects. As discussed, Congress did not intend to prohibit states from influencing the decision-making process of gasoline companies choosing among oxygenates. Thus, the evidence *602submitted at trial does not show that state tort claims will frustrate the objectives of Congress.166

Permitting the City’s state tort claims to proceed may seem unfair — as ExxonMobil is being forced to compensate the City for doing something that it arguably would not have done in the absence of federal regulation. However, the touchstone of the doctrine of federal preemption is not fairness to the parties; it is Congressional intent. Without clear evidence that federal policy and state law are in sharp conflict, or that it would have been physically impossible to comply with federal and state requirements, a finding of preemption is inappropriate.

B. Affirmative Defense: Statute of Limitations

The City’s claims are time barred if, prior to October 31, 2000, (1) there was a sufficient amount of MTBE in the groundwater within Station 6’s capture zone to cause a reasonable water provider to remediate that contamination and (2) the City knew of that level of contamination. These were questions of fact on which ExxonMobil had the burden of proof. Accordingly, for ExxonMobil to be entitled to judgment as a matter of law, it must have been unreasonable for the jury to find that ExxonMobil failed to prove both that the City was injured and that the City was aware of that injury.

At trial, ExxonMobil relied on two pieces of evidence to show that the City’s claims are time barred. Neither of these is sufficient to establish that ExxonMobil is entitled to judgment as a matter of law. First, William Yulinksy, the Director of Environmental Health and Safety for the Bureau of Wastewater Treatment, testified that the City anticipated a need to build a treatment plant at Station 6 as early as September 1999.167 However, as I explained in deciding ExxonMobil’s pre-trial motions, “[t]he design and construction costs of the Station 6 treatment plan is not the City’s injury. The injury is the contamination of the City’s groundwater.... [T]he design and construction costs are damages the City seeks to recover in remediating the injury.”168 “Thus, in determining the timeliness of the City’s recurring injury claim, the question is not, as [ExxonMobil] suggests, when the City first knew of its future damages, but rather when the City was first injured by MTBE contamination in its water and first knew that it was so injured.”169 The fact that the City anticipated & future need to build a remediation facility is not conclusive evidence that the City’s groundwater was already contaminated at a level sufficient to injure the City or that the City knew of that level of contamination.

Second, City employees circulated internal memos in 1995 and 1997 suggesting that MTBE contamination was a growing concern as a potential groundwater contaminant.170 This evidence is equally in*603conclusive. A generalized concern that MTBE contamination may pose problems for the City’s water supply does not show that MTBE contamination in the groundwater of Station 6 had reached levels that would cause the City, as a reasonable water provider, to remediate that contamination.

To succeed on its statute of limitations, ExxonMobil was required to prove by a preponderance of the evidence that the City was injured and that it knew of that injury. Given the evidence introduced at trial, it was not unreasonable for the jury to determine that ExxonMobil failed to meet that burden.

C. Injury

New York law requires the City to show that it suffered an actual injury to recover damages under most of its asserted causes of action.171 ExxonMobil raises numerous issues relating to whether the City’s injury is legally cognizable and whether the jury’s finding that the City suffered an injury is supported by the evidence introduced at trial. I have already considered and rejected many of these arguments in deciding ExxonMobil’s pre-trial motions.172 Because they are not affected by the evidence submitted at trial, I do not revisit those determinations. However, I do consider three other issues raised by ExxonMobil. First, ExxonMobil argues that the jury’s finding that the MTBE concentration in the combined outflow of the Station 6 wells will peak at ten ppb lacked any evidentiary basis.173 Second, ExxonMobil argues that the evidence submitted at trial is insufficient to support the jury’s finding that the City will be injured by the predicted level of contamination.174 Third, and finally, ExxonMobil argues that the City failed to show that it has a “good-faith intent to begin construction of Station 6 within the next [fifteen] years.”175

*6041. Jury’s Determination that MTBE Concentration in the Combined Outflow of Station 6 Wells Will Be Ten PPB

The jury, in finding that the concentration of MTBE in the combined outflow of the Station 6 wells will peak at ten ppb, did not adopt the conclusions of Terry — the only expert witness to quantify the amount of MTBE that will be in the Station 6 wells. While Terry opined that the concentration of MTBE would peak at thirty-five ppb in 2024, the jury found that the concentration of MTBE would peak at ten ppb in 2033. According to ExxonMobil, this disparity demonstrates that the jury “ignored the only evidence quantifying the concentration and timing of MTBE impacts to the Station 6 wells, and substituted sheer speculation in its place.”176 Because the jury was entitled to rely on Terry’s testimony without adopting his ultimate conclusions, I disagree.

As Terry explained to the jury, “[a] groundwater model is a tool” used to estimate the amount of contaminant that will travel to a particular water source on the basis of certain assumptions that are entered into that model.177 He informed the jury that he was basing his prediction in Analysis 1 on specific assumptions and that he “did not have perfect information” about MTBE contamination in the geographic area surrounding Station 6.178 In addition, Terry’s- Analysis 2 model showed a range of other possible peak concentrations on the basis of different assumptions about the amount of MTBE gasoline that has been spilled within the Station 6 capture zone. For example, if each recorded gasoline spill released five hundred gallons of gasoline on average, the level of MTBE concentration would peak at six ppb. While it is true that the jury did not select any one of Terry’s specific predictions, it did select from within the range of possible outcomes suggested by Terry.

The parties disagree as to whether the jury could rely upon Terry’s model without adopting his specific conclusions as to the likely peak MTBE concentration. In general, a jury is not required to choose between adopting or rejecting an expert’s testimony wholesale; it is free to accept or reject the expert’s opinions in whole or in part and to draw its own conclusions from it.179 Because juries are rarely asked to provide numerical predictions in the liability phase of a trial, no court that either the parties or I have found has dealt with the precise issue raised in this case. However, triers of fact are routinely asked to make numerical assessments on the basis of expert testimony in the damages phase. In that context, juries have been given significant latitude to alter the amount of damages proposed by experts while still relying on the models developed by those experts.180 While a *605jury’s task in the liability and damages phases do have significant differences, the principle that expert evidence “is to be treated in the same manner as other evidence in the case” applies whether the jury is determining damages or predicting the amount of future contamination.181 The role of an expert is to assist the jury by “providing the groundwork ... to enable the jury to make its own informed decision.” 182 It is not to usurp or invade the jury’s decision-making function.

The jury was given the difficult task of predicting how much MTBE will be in the combined outflow of the Station 6 wells many years into the future and it was not expected to return an answer reflecting scientific precision. In fact, the jury was permitted to choose among several possible ranges of MTBE contamination rather than choosing an exact concentration level. Moreover, as discussed, Terry made clear that even his conclusion as to the likely peak concentration at Station 6 was an estimate based upon imperfect knowledge about real world facts. The jury scrutinized Terry’s testimony, and, most likely finding that several of his express assumptions were inaccurate, selected a lower peak concentration than that suggested by Terry. This does not amount to impermissible conjecture or surmise, but a reasonable approach to a difficult question.

2. Injury at Ten PPB

I have already held in prior decisions that whether contamination at or below the MCL injures the City is a question of fact for the jury. At trial, the City introduced several pieces of evidence to support the inference that a reasonable water provider in the City’s position would treat the water in the Station 6 wells. For example, the City introduced testimony that twenty-five percent of the population can detect MTBE at three to four parts per billion,183 that ten percent of the population can detect MTBE at one to two parts per billion,184 that MTBE may be carcinogenic at low levels of exposure in drinking water,185 and that if one or more of the Station 6 wells is taken offline for any reason the concentration of MTBE in the combined outflow of the remaining wells could spike to fifteen ppb or higher (a violation of the MCL).186 There is nothing unreasonable about the jury determining, on the basis of this evidence, that a reasonable water provider in the City’s position would treat the water in the Station 6 wells.

3. Good Faith Intent

ExxonMobil also argues that the jury did not have sufficient evidence to find that the City has a good faith intent to begin construction of the Station 6 facility within the next fifteen years.1871 disagree. *606Kathryn Garcia, an Assistant Commissioner for Strategic Projects at the New York City Department for Environmental Protection, testified that the Station 6 wells would be used as a backup source of water if other pieces of infrastructure, such as aqueducts currently in use, suffered an outage.188 Graham Fogg, a hydrogeology professor at the University of California, Davis, testified that plumes of MTBE will remain in the geographic vicinity of Station 6 for decades into the future.189 This testimony, which the jury apparently credited, provides a sufficient basis for the jury to infer that the City intends to use the Station 6 wells as a backup water source and that it intends to build a treatment facility to ensure that it is able to use water from those wells if and when needed.

D. Causation

The jury found ExxonMobil liable as both a manufacturer, refiner, supplier and seller of gasoline containing MTBE and as a direct spiller of gasoline containing MTBE. ExxonMobil argues that the evidence introduced at trial does not support that determination and that the jury was confused as to the traditional principles of causation due to the Court’s description of the commingled product theory of causation in the jury charge.

1. Insufficiency of the Evidence

At trial, the City relied on Terry’s capture zone model to establish causation. As I explain below, combined with the other evidence submitted at trial, this capture zone model provided a sufficient basis for the jury to determine that ExxonMobil’s conduct both as a manufacturer, refiner, supplier and seller of MTBE gasoline and as a direct spiller of MTBE gasoline was a substantial factor in causing the City’s injury.

a. Manufacturer, Refiner, Spiller and Supplier

At trial, the City produced three pieces of evidence to show that ExxonMobil’s conduct as a manufacturer, refiner, supplier and seller of gasoline containing MTBE was a substantial factor in injuring the City: (1) expert testimony that, due to the commingled nature of the gasoline distribution network, ExxonMobil gasoline would be present in nearly every UST within the Station 6 capture zone;190 (2) expert testimony that ExxonMobil supplied approximately twenty-five percent of the gasoline sold in Queens;191 and (3) *607testimony that, over time, all USTs leak.192 ExxonMobil’s objection to the sufficiency of this evidence focuses on the City’s use of market share data.

First, ExxonMobil argues that the City should not have been permitted to use market share data because it disclaimed reliance on the market share theory of liability.193 This argument conflates application of the market share theory of liability with using market share data as circumstantial evidence of causation. Under the market share theory, liability is apportioned according to a defendant’s national market share even though it is assumed that the vast majority of defendants in any one case did not injure the plaintiff. The City, on the other hand, sought to use market share data to show that “Exxon-Mobil ... represented such a substantial fraction of the total gasoline sold in the Queens area, that ... it was not only present, but present in such quantity at leaking stations” that it was more likely than not a cause of the City’s injury.194 While New York has severely limited the circumstances in which the market share theory of liability may be applied,195 it has never suggested that market share data cannot be used to prove traditional causation, and I find nothing unreasonable about the jury relying on that data to find ExxonMobil liable.

Second, ExxonMobil argues that even if the use of this market share data is permissible, a twenty-five percent market share cannot make it more than twenty-five percent likely that ExxonMobil was a cause of the City’s injury, and thus, reliance on the market share data does not satisfy the preponderance of the evidence standard.196 This argument is incorrect. Given the commingled state of gasoline entering Queens, ExxonMobil’s twenty-five percent market share can support the inference that ExxonMobil’s conduct was more likely than not to have caused twenty-five percent of the contamination in the Station 6 wells. The jury was entitled to find that, by causing this significant portion of contamination in the Station 6 wells, ExxonMobil’s MTBE-containing gasoline was a substantial factor in producing the City’s injury.

b. Direct Spiller

To succeed on its direct spiller claims, the City was required to show that MTBE gasoline released from ExxonMobil-controlled stations was a cause of the City’s injury at Station 6. However, “[t]hat showing need not be made with absolute certitude nor exclude every other possible cause of injury.”197 Instead, the City was only required to prove “facts and circum*608stances ... from which causation may reasonably be inferred.”198

The City attempted to meet that burden at trial by introducing evidence of multiple spills at ExxonMobil-controlled stations within Terry’s predicted capture zone and evidence that MTBE has escaped remediation efforts at many of those sites.199 ExxonMobil argues that Terry’s capture zone model does not provide an adequate basis for proving direct spiller causation because it was based on the assumption that the Station 6 wells will pump water continuously from 2016200 until 2040.201 According to ExxonMobil, that assumption conflicts with the jury’s finding that “Station 6 will be used only as a backup source” 202 and is unsupported by the evidence.203

Regardless of this discrepancy, however, the jury could still reasonably have found that MTBE from ExxonMobil-controlled stations would impact Station 6. At trial, a series of demonstrative maps depicting the evolution of Terry’s predicted capture zone over time were shown to the jury.204 These maps, which were based on exhibits introduced at trial,205 indicate that Exxon-Mobil-controlled stations are situated in nearly every direction from Station 6. Moreover, several of these stations are close enough to Station 6 to be well within the borders of Terry’s predicted capture zone at every stage of its evolution (i.e., even in the earliest stages of the assumed pumping schedule). The City, for example, introduced evidence that significant amounts of MTBE gasoline were spilled206 at an ExxonMobil-controlled station, 113-21 Merrick Boulevard, located within a half-mile of Station Six.207 In conjunction with Terry’s (allegedly imperfect) capture zone model, this sort of evidence- — ie., the existence of several ExxonMobil-controlled stations in close proximity to, and positioned radially in multiple directions from, Station 6 — provides a sufficient basis for the jury to reasonably infer that Exxon-Mobil’s conduct as a direct spiller was a cause of the City’s injury.

2. Jury Confusion

ExxonMobil seeks a new trial on the ground that “[e]ven though the jury did not render a verdict on the commingled product theory ... the erroneous commingled product theory so permeated the trial as to make it impossible for the jury to comprehend traditional causation.” 208 Specifically, ExxonMobil suggests that the Court’s instruction on the commingled product theory “impermissibly relieved the City of its burden to identify the defendant who caused its harm.”209 *609This argument is without merit.

First, as discussed, the commingled product theory of liability, unlike the market share theory of liability, does not allow for the imposition of liability against a defendant in the absence of a finding that the particular defendant’s product contributed to harming the plaintiff. The market share theory assumes that the vast majority of defendants did not actually contribute to the harm incurred by a particular plaintiff (i.e., all defendants supplying DES used to treat women during pregnancy are held liable even if only one defendant manufactured the DES pills that were ingested by the plaintiff). The commingled product theory, in contrast, requires a finding that the product of each and every defendant held liable contributed to plaintiff’s injury (ie., a defendant cannot be held liable unless its gasoline was present in the commingled product that contaminated plaintiffs groundwater).210

Second, the jury was clearly instructed that the commingled product theory is distinct from the traditional theory of causation and informed as to the nature of that difference. For example, in describing the commingled product theory, I instructed the jury:

What sets [the commingled product theory] apart from manufacturer, refiner, supplier or seller causation is that to prove [causation under the commingled product theory], the [C]ity need not show that [ExxonMobil’s] contribution taken alone would have injured the [C]ity.211

There is no indication that the jury did not understand this instruction.212

Third, the jury was entitled to consider testimony that the gasoline supplied to Queens was commingled in determining whether ExxonMobil’s gasoline injured the City under traditional principles of causation. The fact that ExxonMobil gasoline was spread throughout the gasoline distribution network in Queens increases the probability that ExxonMobil gasoline was in the USTs that leaked gasoline into the Station 6 capture zone.213 Evidence of commingling was therefore relevant even if the jury did not ultimately reach the commingled product theory of causation.

E. Juror Misconduct: Extra-Record Information

During the jury’s Phase III deliberations, I learned that a member of the jury had performed limited Internet research relating to this case. ExxonMobil argues that I erred in not declaring a mistrial on the basis of this juror’s misconduct.214 Search engines have indeed created significant new dangers for the judicial system. It is all too easy for a juror to find out more than he or she should by typing a few carefully chosen words into a search engine.215 Nevertheless, in this in-*610stance, the jury was not too polluted by the receipt of extra-judicial information such as to prevent it from rendering a fair verdict based on the evidence introduced at trial.

It is axiomatic that juries are required to decide cases on the evidence *611introduced at trial. However, a new trial is not required solely because the jury was exposed to extrinsic information.216 “The issue, as Judge [Henry] Friendly observed, is ‘not the mere fact of [jury] infiltration ... but the nature of what has been infiltrated and the probability of prejudice.’ ”217 While a court may question a jury about what they learned in making this determination, after deliberations have begun, it is inappropriate for a “court [to] inquire into the degree upon which the extra-record information [is being] used in deliberations and the impression which jurors actually ha[ve] about it.”218 “Rather, courts must apply an objective test focusing on two factors: (1) the nature of the information or contact at issue, and (2) its probable effect on a hypothetical average jury.” 219

Because the jury was not exposed to evidence that would prejudice the average juror, it was not error to deny ExxonMobil’s motion for a mistrial. Juror No. 8, who clearly obtained improper information during Phase III deliberations was immediately excused after the Court learned of his misconduct. The jury brought this fact to my attention soon after they became aware of it and the remaining jurors appeared candid and forthcoming in answering my questions about the information they had learned. Although Juror No. 8 initially painted a picture of a jury that was engaging in rampant outside research, his assertion that ‘everybody was doing it’ was nothing more than the defensive tactic of a juror looking for cover after he had been caught with his hand in the proverbial cookie jar. When he and the other jurors were questioned further, it became apparent that the limited information Juror No. 8 had communicated to them was decidedly vague.

Juror No. 11 learned that there was going to be a fourth phase of the trial — which he described as “some other penalty part.” 220 Although the punitive damages phase is often separated from the liability phase because plaintiffs are permitted to introduce evidence in the punitive damages phase that is irrelevant in the liability phase,221 mere knowledge of a possible penalty phase is insufficient to create prejudice. Indeed, when the jury was charged on Phase III of this trial, it was given detailed instructions on how to calculate compensatory damages. If such instructions (which are commonly given) are not deemed to prejudice the jury’s deliberations, then neither should a juror’s vague knowledge that there may be a further penalty phase.

Juror No. 5 learned that ExxonMobil was the only remaining defendant in this case and that many of the other defendants had settled for approximately one million dollars each. The jury, however, was already well aware from the evidence introduced at trial, as well as from this Court’s instruction that it must apportion *612liability among all responsible oil companies operating in the area surrounding Station 6, that other oil companies had caused contamination at Station 6. As such, a juror could easily make the connection that these other defendants had settled without being informed that this was the case by another juror who claimed to have obtained information from the Internet. Moreover, Juror No. 5’s knowledge of these supposed settlements was so limited and vague that it would be highly unlikely to have prejudiced the average juror’s deliberative process.222

In addition, two other jurors performed some limited outside research on their own. Juror No. 4 tried to drive over to Station 6, but failed to find it. Juror No. 6 looked me up on Wikipedia and read an article about water contamination caused by coal companies. As with the information shared by Juror No. 8, the jury’s exposure to this information does not warrant a mistrial. Although it was inappropriate for Juror No. 4 to attempt to find Station 6, his failure to actually find that site prohibits any inference of prejudice. As for Juror No. 6 — the information which she uncovered had only a tangential relationship to this case. Jurors do not enter the courtroom as blank slates and the addition of this limited information to Juror No. 6’s mix of knowledge cannot reasonably be deemed prejudicial.

F. Dismissal of Alleged Holdout Juror

ExxonMobil argues that a new trial is warranted because the “Court erred in dismissing Juror [No.] 2, rather than declaring a mistrial, when it was revealed that she was threatened for being a holdout.” 223 ExxonMobil did not object to the removal of Juror No. 2 at any point until it filed its post-trial motions. In fact, at the time of removal, ExxonMobil agreed that the Court should excuse Juror No. 2.224 Because it would be patently unfair to allow ExxonMobil to object to a decision by this Court (the dismissal of Juror No. 2) that it previously endorsed only after it received an adverse verdict, ExxonMobil has waived its right to make this objection.225 If ExxonMobil believed that Juror No. 2 was being forced off of the jury because she was a holdout, it should have objected to her removal when the evidence allegedly supporting that inference arose.

G. Remaining Liability Issues

ExxonMobil has also raised a series of issues relating to: (1) the jury’s findings of liability on the City’s failure to warn,226 negligence,227 public nuisance,228 and tres*613pass claims 229; and (2) several evidentiary rulings.230 Having reviewed the record thoroughly, and considered all of Exxon-Mobil’s submissions, I find that these arguments — many of which have previously been considered by this Court — are without merit.

H. Damages

ExxonMobil argues that the jury’s damages award warrants a new trial on damages, or, in the alternative, remittitur.231 ExxonMobil points to three purported errors in Bell’s testimony232 — which provided the basis for the jury’s award.233 None of these alleged errors are sufficient to warrant a new trial on damages or remittitur.234

First, ExxonMobil argues that Bell’s capital cost estimate of $59,990,000 was unreasonable given that she had estimated that cost at $9.9 million in 2004.235 However, Bell was cross-examined on this point at trial and testified that she had altered the estimate for a number of reasons — including inflation, the escalation of prices of certain raw materials and a change in design.236 The jury was entitled to accept that explanation.

Second, “the jury implausibly credited [] Bell’s dubious testimony that certain equipment will require replacement every 20 years, regardless of whether that equipment is used or for how long.” 237 Bell opined at trial that the equipment used at Station 6 will likely deteriorate even if it lays fallow.238 As anyone who has left a piece of machinery, such as an automobile, idle for a long period of time is aware, there is nothing unreasonable about the inference that such equipment can become corroded and require replacement parts. Accordingly, I cannot find that the jury acted in error by crediting Bell’s testimony that equipment at Station 6 will have to be replaced twice over the forty year period it will be in use.

*614Third, ExxonMobil argues that “the jury inappropriately awarded $141 million for [operating and maintenance] costs based on nothing more than the ipse dixit statements of [ ] Bell and her speculative assumption that Station 6 will operate continuously for 40 years.” 239 The City introduced evidence at trial that the Station 6 wells will also be used to fulfill the public’s water needs in emergency drought situations or if presently unforeseen infrastructure repairs become necessary.240 Because there is no sure way to estimate how likely these emergencies are to occur, and thus, how often the City will need to use the Station 6 wells, it was not unreasonable for Bell, in assessing damages, to assume that the wells will operate continuously. Indeed, it is appropriate that the wrongdoer (ExxonMobil) rather than the innocent party (the City) should bear the risk that these emergencies will occur frequently. As Bell explained, she assumed the wells would run continuously for forty years because, for planning purposes, the City needs to presume the “worst-case scenario.” 241 The jury did not act unreasonably in accepting this explanation and adopting Bell’s operating and maintenance costs.

Accordingly, because I conclude that the jury’s damages award did not contain specific, quantifiable errors, I reject Exxon-Mobil’s request for a new trial on damages, or in the alternative, remittitur.

VI. CONCLUSION

For the aforementioned reasons, Exxon-Mobil’s post-trial motion is denied. The Clerk of Court is directed to close this motion (Docket No. 610).