MEMORANDUM-DECISION AND ORDER

MORDUE, District Judge.

Plaintiff is a trade association, the members of which include major producers of methyl tertiary butyl ether (“MTBE”), an oxygenated fuel additive (“oxygenate”) used in motor vehicle fuel to improve combustion for the purpose of reducing emissions pollution. In this action under the Supremacy Clause, U.S. CONST., Art. VI, cl. 2, the Clean Air Act (“CAA”), 42 U.S.C. § 7401 et seq., and the Civil Rights Act, 42 U.S.C. § 1983, plaintiff challenges the constitutionality of a New York law (“N.Y. MTBE Law”) prohibiting the use, sale, or importation in New York of gasoline containing MTBE beginning January 1, 2004. Plaintiff seeks judgment declaring that New York MTBE Law is preempted by CAA and regulations promulgated thereunder and thus is unconstitutional under the Supremacy Clause of the United States Constitution. U.S. Const. art. VI, cl. 2.

The Court assumes familiarity with the earlier decisions in this case. See Oxygenated Fuels Ass’n, Inc. v. Pataki, 158 F.Supp.2d 248 (N.D.N.Y.2001) (“OFA v. Pataki ”); as amended on reconsideration by Memorandum-Decision and Order of May 16, 2002; motion for certification to Second Circuit denied by Memorandum-Decision and Order of December 6, 2002; defendants’ motion for dismissal and plaintiffs cross-motion for summary judgment denied by Memorandum-Decision and Order of October 3, 2003.

The Court held a bench trial on October 8, 9, 10, 14, 15 and 23, 2003. For reasons set forth herein, the Court finds that plaintiff has not proven its case and awards judgment in favor of defendants dismissing the case in its entirety.

BACKGROUND

Complaint

In its complaint, filed July 11, 2000, plaintiff claims that N.Y. MTBE Law vio *173 lates the Supremacy Clause because it legislates in a field preempted by Congress, and that it violates plaintiffs civil rights under 42 U.S.C. § 1983. Plaintiff seeks judgment declaring that the law is invalid because, in CAA and the regulations promulgated thereunder, Congress and the Environmental Protection Agency (“EPA”) expressly preempted legislation in the field of fuel and fuel additives for whatever purpose; because Congress impliedly preempted that field; and because N.Y. MTBE Law is “conflict-preempted” by CAA, that is, N.Y. MTBE Law would be an obstacle to the accomplishment and execution of the full purposes and objectives of Congress in CAA. Plaintiff further seeks injunctive relief prohibiting New York from taking any steps to implement or enforce N.Y. MTBE Law and from taking any future action to ban the use of MTBE in gasoline in New York.

Issue at trial

In denying plaintiffs motion for summary judgment on May 18, 2001, the Court held that N.Y. MTBE Law was not expressly or impliedly preempted by federal law. See OFA v. Pataki, 158 F.Supp.2d at 255, 257-58. The Court further held that the issue of conflict preemption could not be determined on the record before the Court and must be tried. See id. at 258. Accordingly, the Court held a bench trial on October 8, 9, 10, 14, 15 and 23, 2003, on the sole issue of conflict preemption.

CAA and the RFG program

The stated purpose of CAA is “to protect and enhance the quality of the Nation’s air resources so as to promote the public health and welfare and the productive capacity of its population.” CAA § 101(b)(1), 42 U.S.C. § 7401(b)(1)- In furtherance of this purpose, Congress enacted the reformulated gasoline (“RFG”) program in 1990, CAA § 211(k), 42 U.S.C. § 7545(k), to address ozone pollution in areas which have failed to attain National Ambient ■ Air Quality Standards (“NAAQS”) for ground-level ozone (“non-attainment areas”). The RFG program requires the sale in nonattainment areas of RFG, i.e., gasoline having certain properties, for the purpose of reducing ozone-causing exhaust and evaporative emissions. See CAA § 211(k)(2), (5).

The RFG legislation directs EPA -to “promulgate regulations ... establishing requirements for reformulated gasoline to be used in gasoline-fueled vehicles” in non-attainment areas, CAA § 211(k)(l), and sets forth detailed requirements that EPA must include in its regulations. CAA § 211(k)(2), (3). One such requirement is an oxygen content of at least two percent, CAA § 211(k)(2)(B), which is obtained by the addition of oxygenates such as MTBE or ethanol. The statute empowers EPA to determine which fuels may be sold in non-attainment areas and prohibits the sale or dispensing of all other fuels in those areas. CAA § 211(k)(4), (5). EPA may prescribe sampling, testing and record-keeping requirements and impose penalties, CAA § 211(d), (k)(5), and may adjust or waive requirements of the program under certain circumstances. CAA § 211(k)(2)(A), (B), (D). Section 211(c)(1) authorizes EPA to control or prohibit the sale of any fuel or fuel additive if in its judgment any emission product of that fuel or fuel additive causes or contributes to air pollution which may endanger public health or welfare. State regulation of any component or characteristic of a fuel or fuel additive for purposes of emissions control is proscribed except with EPA approval under narrowly defined circumstances. CAA § 211(c)(4)(A)(i), (ii).

N.Y. MTBE Law

It is undisputed that MTBE has a great affinity for water, that it has been detected in groundwater in New York state, and *174 that its taste and odor can be perceived at very low levels. New York enacted N.Y. MTBE Law on May 24, 2000, by amending N.Y. Agriculture and Markets Law (L.2000, c. 35, § 2) to add section 192-g, which provides:

1. For the purposes of this section, “gasoline” shall mean any fuel sold for use in motor vehicles and motor vehicle engines, and commonly or commercially known or sold as gasoline.

2. No person shall import into, or sell, dispense or offer for sale any gasoline which contains methyl tertiary butyl ether.

3. Any person who violates the provisions of this section shall be liable for a civil penalty of not less than five hundred dollars nor more than ten thousand dollars.

New York also amended section 19-0301(3) of N.Y. Environmental Conservation Law (L.2000, c. 35, § 2) to add a new subsection B, which states: “No provision of this subdivision shall be deemed to authorize the use of methyl tertiary butyl ether as an oxygenate in any motor fuel imported into, or sold or offered for sale in this State.” The amendments are effective January 1, 2004 (L.2000, c. 35, § 3). That the purpose of the N.Y. MTBE Law is to protect New York’s groundwater from contamination is not seriously contested by plaintiff.

DISCUSSION

The sole issue before the Court is whether N.Y. MTBE Law is “conflict-preempted” by federal law. Conflict preemption occurs when “compliance with both federal and state regulations is a physical impossibility,” Florida Lime & Avocado Growers, Inc. v. Paul, 373 U.S. 132, 142-43, 83 S.Ct. 1210, 10 L.Ed.2d 248 (1963), or when state law “stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress.” Hines v. Davidowitz, 312 U.S. 52, 67, 61 S.Ct. 399, 85 L.Ed. 581 (1941). Courts must narrowly construe a federal law which is claimed to preempt an exercise of state police power. See Cipollone v. Liggett Group, Inc., 505 U.S. 504, 518, 112 S.Ct. 2608, 120 L.Ed.2d 407 (1992). Indeed, when Congress legislates in a field traditionally within the police powers of the states, such as matters related to public health and safety, there is a presumption that the state law is not invalidated under the Supremacy Clause. See Hillsborough Co. v. Automated Med. Labs., Inc., 471 U.S. 707, 715, 105 S.Ct. 2371, 85 L.Ed.2d 714 (1985).

It is undisputed that N.Y. MTBE Law, which has the purpose of protecting New York’s groundwater from contamination, is a proper exercise of New York’s police power to regulate and control matters related to public health and safety. See id. Accordingly, in considering plaintiffs claim that N.Y. MTBE Law is preempted, the Court must narrowly construe CAA and must bear in mind the presumption that the state law is not preempted.

The stated purpose of CAA is “to protect and enhance the quality of the Nation’s air resources so as to promote the public health and welfare and the productive capacity of its population.” CAA § 101(b)(1). Thus, enhancement of air quality is clearly the overriding goal of CAA. In the Court’s view, however, evidence that a state law would have a relatively minor impact on air quality would be insufficient to support a finding that it stands as an obstacle to the accomplishment and execution of the full purposes and objectives of CAA.

In defining the goal of CAA specifically *175 with reference to the RFG program, 1 this Court stated that “the congressional goal in enacting the RFG program was to reduce emissions pollution while ensuring an adequate gasoline supply at a reasonable cost, taking into account other health and environmental concerns so far as reasonable.” OFA v. Pataki 158 F.Supp.2d at 259. The Ninth Circuit recently addressed the question of whether the goal of CAA encompassed considerations of gasoline availability and cost in a similar context in Oxygenated Fuels Ass’n, Inc. v. Davis, 331 F.3d 665 (9th Cir.2003) (“Davis ”), an action by the same association as the plaintiff herein. In support of their contention that California’s ban on MTBE gasoline is conflict-preempted by CAA, OFA argued that California’s law would interfere with CAA’s goal to “ensure a smoothly functioning market and cheap gasoline[.]” The Ninth Circuit rejected this contention and dismissed the case, stating: “We take it as true that Congress wanted to reduce pollution caused by motor vehicles, but at the same time did not want to harm the nation’s economy by causing gasoline prices to rise substantially.” 331 F.3d at 673. The Ninth Circuit further stated that it is “questionable” whether a smoothly functioning market is a goal of the Clean Air Act and that “saying that Congress might not have wanted to cause a substantial increase in gasoline prices is not the same as saying that assuring inexpensive gasoline was a goal of the Act.” Id.

In its most recent Memorandum-Decision and Order, decided October 3, 2003, this Court noted that the above-quoted observations in Davis

do not evince an extremely narrow reading of the goals of the Clean Air Act and thus are not necessarily incompatible with this Court’s conclusion that “the congressional goal in enacting the RFG program was to reduce emissions pollution while ensuring an adequate gasoline supply at a reasonable cost, taking into account other health and environmental concerns so far as reasonable.” [OFA v. Pataki] 158 F.Supp.2d at 259.

If, on the other hand, Davis is read as adopting an extremely narrow reading of the goals of the Clean Air Act, the Court declines to follow it. It cannot reasonably be argued that Congress’ goal was to reduce emissions pollution regardless of the cost — even, for the sake of argument, at the cost of total disruption of the gasoline .market, complete unavailability of gasoline in certain regions of the nation, or astronomical increases in prices. This Court does not view the goals of the Clean Air Act so narrowly as to exclude from all consideration the “larger context of market forces, health and environmental impacts, regional priorities, technological feasibility and other considerations.” OFA v. Pataki [ ], 158 F.Supp.2d at 256.

Thus, in this Court’s view, ensuring an adequate supply of gasoline at reasonable cost is not, viewed in isolation, a goal of CAA; rather, CAA’s goal of enhancing air quality must be viewed in the larger context of market forces, health and environmental impaets, regional priorities, technological feasibility and other considerations.

Accordingly, at trial, plaintiff adduced evidence for the purpose of showing *176 (1) that the N.Y. MTBE Law would cause increased air pollution, and (2) that N.Y. MTBE Law would interfere with the existence of an adequate gasoline supply at a reasonable cost. Defendants primarily directed their efforts at trial to challenging plaintiffs evidence. 2

Plaintiffs case, generally

Essentially, plaintiffs contentions at trial were that as a result of N.Y. MTBE Law, refineries will supply RFG containing ethanol, the only viable oxygenate alternative to MTBE; that the use of ethanol RFG will cause increased emissions of volatile organic compounds (“VOC”) and oxides of nitrogen (“NOx”), which are ozone precursors; that the resulting increased emissions of ozone precursors will be in the range of 20 tons per day; that these increased emissions will impair New York’s ability to comply with CAA; and that the economic impacts of N.Y. MTBE Law bolster the conclusion that the law will interfere with the goals of CAA. In support of these contentions, plaintiff presented three expert witnesses.

Plaintiff built its case on the testimony of David Hirshfeld, an expert in refinery modeling. Hirshfeld testified that, using linear programming modeling, he compared the projected emissions properties of ethanol RFG with those of MBTE RFG and concluded that “the average emissions properties of ethanol blended gasoline RFG that New York would receive after a ban will be slightly worse than the emissions properties of MTBE.”

Plaintiff next introduced the testimony of Thomas Austin, an air pollution control expert, who testified that the use of ethanol instead of MTBE will cause increased emissions of ozone precursors from four sources: evaporative emissions, exhaust emissions, commingling and permeation. He attributes these increases in part to the undisputed fact that the use of ethanol RFG increases the vapor pressure of gasoline.

Plaintiffs final witness on its direct case was Gordon Rausser, an economist. Raus-ser testified that N.Y. MTBE Law will result in gasoline supply shortfalls and price increases in New York.

Hirshfeld’s testimony

Plaintiffs refinery modeling expert, David Hirshfeld, testified that he predicted the emissions properties of ethanol RFG using a linear programming model known as Advanced Refinery Modeling System (“ARMS”), which his company developed. Hirshfeld explained that EPA provides refiners with a model known as the complex model and that refiners may make RFG with any combination of eight properties, provided that, pursuant to the complex model, the RFG meets the exhaust and evaporative emissions standards for VOC, NOx and toxics. He agreed that “an absolute bedrock premise in [his] modeling work is that refiners comply with the law” and that when MTBE RFG is banned, the *177 refinery sector will provide ethanol RFG which complies with the complex model. Therefore, according to Hirshfeld, using the complex model and taking economic and other considerations into account, the AEMS model can predict the amounts of the eight properties the refineries are likely to utilize in providing ethanol RFG to New York.

Hirshfeld further testified, based on the ARMS model, that the reductions in emissions properties achieved by the ethanol RFG which refiners were likely to provide would be inferior to the reductions achieved by MBTE RFG. In other words, ethanol RFG would produce emissions reductions that were “less good” than those produced by MBTE RFG, although there would still be reductions. He stated that the difference in emissions when ethanol RFG replaces MTBE RFG in New York would be “small.” He further stated that New York overcomplies with the RFG standards and that with ethanol RFG there would be a decrease in overcompliance, although there would still be over-compliance. 3

Hirshfeld’s emissions projections based on his ARMS model were wholly discredited at trial. The Court bases this conclusion on the testimony of defendants’ refinery modeling expert Martin R. Tallett, on defendants’ cross-examination of Hirshfeld upon plaintiffs direct case, and, most importantly, on defendants’ cross-examination of Hirshfeld upon rebuttal. On cross examination upon rebuttal, Hirshfeld admitted that in running the ARMS model for MTBE RFG he had placed numerical constraints on certain of the eight properties of gasoline, specifically aromatics, ole-fins and benzene; in other words, he had predetermined or “fixed” maximum numbers for these properties. He admitted that under the complex model, aromatics and olefins have significant effects on NOx emissions, and in some cases on VOC as well. He did not claim that he imposed any constraints on aromatics or olefins in running the model for ethanol RFG. Hirshfeld gave no credible explanation for the presence of these constraints, referring only in a conclusory manner to unspecified “customary refining practice” and stating that he imposed the constraints as an aspect of his “professional judgment.”

Defendants’ expert Tallett testified that when he ran the same cases after removing the constraints on aromatics, olefins and benzene, he found that Hirshfeld’s ARMS model predicted in some cases that switching to ethanol RFG would reduce VOC and NOx emissions by as much as two percent. Tallett further testified that fixing these properties was inconsistent with the function and purpose of the complex model, which is designed to allow refiners flexibility to vary gasoline properties. Tallett further testified that the constraints were improper because the purported aim of Hirshfeld’s study was to predict the amount of each gasoline property the refiners would use in exercising that flexibility. Further, by limiting aromatics and olefins only for MTBE RFG and not for ethanol RFG, Hirshfeld placed a ceiling on MTBE’s emissions while allowing the ethanol emissions to rise with no limitation. 4

*178 Hirshfeld did not voluntarily disclose that he had placed constraints on the aromatics, olefins and benzene for MTBE RFG, nor did he ever give a credible explanation for having done so. In his testimony throughout the trial he attempted to convey the impression that he had applied the model in an objective manner and that the results, which were favorable to his client, represented a meaningful comparison of the likely emissions properties of MTBE RFG and ethanol RFG. In fact, the effect of the constraints was to produce results which had the appearance of veracity and reliability but which actually proved nothing. Hirshfeld presented these results with the intention that the Court would rely on them in reaching a decision. For this reason, the Court rejects his testimony in its entirety based on its lack of reliability and credibility.

Although it is not necessary to do so in view of the above finding, the Court observes also that Hirshfeld’s credibility and reliability were further damaged by his admission on cross-examination that he had omitted to disclose model results that were unfavorable to plaintiff; he stated that he ran his model thousands of times but only reported about 40 of these runs. He did not satisfactorily explain why he rejected thousands of results or why he selected the particular ones which he reported.

Moreover, Hirshfeld initially ran his model assuming that refiners would choose to use “individual pool” compliance, one of two methods of demonstrating RFG compliance, but he did not disclose the results of these individual pool cases. Tallett demonstrated that these individual pool results were less favorable to plaintiffs case than were the “split pool” cases Hirshfeld reported in his expert reports. Hirshfeld gave no reasonable explanation for reporting the split pool results but not the individual pool results. Tallett also challenged the reliability of Hirshfeld’s methodology and conclusions on a number of other grounds. Neither Hirshfeld’s rebuttal testimony nor any other part of plaintiffs evidence adequately met those challenges.

The Court further finds that Hirshfeld’s own testimony establishes that Hirshfeld resisted providing important aspects of his complex model to defendants’ refinery modeling experts, Gary Whitten and Tal-lett. Hirshfeld failed to provide the ARMS action menu, the calibration cases he used, and a VMP.zip file with export capability. This resistence continued even after a protective order was in place. The Court finds that Hirshfeld’s explanations for this conduct are neither reasonable nor credible.

In sum, the Court rejects Hirshfeld’s evidence in its entirety based on its lack of reliability and credibility. One final observation regarding Hirshfeld’s testimony: even if the Court were to accept Hirsh-feld’s evidence, his ultimate conclusion was that the average emissions properties of ethanol RFG in New York “will be slightly worse than the emissions properties of MTBE” (emphasis added). In the Court’s view, this is not evidence of a conflict of sufficient magnitude to support conflict preemption of New York’s law.

Austin’s testimony

Thomas Austin, plaintiffs air pollution control expert, testified that the use of ethanol instead of MTBE would cause increased VOC and NOx emissions from four *179 sources, due in part to the undisputed fact that the use of ethanol, the only viable oxygenate alternative to MTBE, increases the vapor pressure of gasoline. First, the increased vapor pressure would increase emissions through evaporation. Second, ethanol RFG use would cause increased exhaust emissions. Third, adding ethanol to gasoline would accelerate the rate of permeation through plastic fuel tanks and rubber hose materials, causing increased emissions. Finally, commingling, which would result from the mixing of ethanol RFG and MTBE RFG, would cause the vapor pressure of the gasoline to rise, resulting in additional emissions.

Austin’s testimony that N.Y. MTBE Law would result in increased exhaust and evaporative emissions was based on Hirsh-feld’s discredited predictions of the emissions properties of ethanol RFG and as such has no probative value. The Court notes also that, based on Hirshfeld’s projections, Austin testified that with ethanol RFG, New York would continue to over-comply with the complex model requirements, although to a lesser degree than with MTBE RFG. Thus, even if the Court were to accept Hirshfeld’s projections and Austin’s testimony based thereon, Austin’s predictions of exhaust and evaporative emissions would not aid plaintiff in proving its case.

Austin’s initial commingling projections were shown to be based on incorrect figures, and he did not present corrected projections. His testimony regarding permeation, although not based on Hirshfeld’s testimony, depended heavily on a five-year-old survey of 324 California residents; for this and other reasons it has negligible probative value. In his rebuttal testimony, 5 Austin introduced revised and significantly higher predictions of permeation emissions. He essentially abandoned his former reliance on Hirshfeld’s calculations by downplaying his exhaust and evapo-rative emissions projections, which were based on Hirshfeld’s calculations, and emphasizing his newly revised permeation projections, which were not based on Hirshfeld’s calculations.

Aside from the credibility issues inherent in the timing of these last-minute upward revisions, Austin’s testimony falls short of supporting plaintiffs claims in a fundamental respect: although he projected an increase in emissions of VOC and NOx in an amount of 20 tons per day, he did not predict the amount of ozone which would result from this increase. Austin admitted that calculating an amount of ozone resulting from amounts of ozone precursors is “not as simple as just a one-to-one linear relationship between precursors and results in ozone” and that translating a change in the amount of ozone precursors into a change in the amount of ozone “is a fairly complex process that involves the use of modeling and having to look at more than just the specific change” in the amounts of precursors. Austin described several methods for performing this “translation” of amounts of ozone precursors into an amount of ozone, including (1) photochemical modeling, which Austin described as the “best” and “most precise” way to calculate the quantity of ozone; (2) *180 a “proportional” analysis, involving looking at the ratio of emissions before and after the change; and (3) the Empirical Kinetic Modeling Approach, which he described as “a simplified way of incorporating the photochemistry that’s known to take place without having to go to the complication of running an actual air shed model.” Neither Austin nor his staff performed any of these methods; Austin stated that it was beyond the scope of what he was trying to accomplish in his reports. Thus, plaintiff did not present projections of any specific amount of increase in ozone which will result from N.Y. MTBE Law. Nor did it provide a context in which the Court could evaluate the significance of this increase. The “empirical data” from California and the Chicago-Milwaukee area does not assist plaintiff in filling this gap; the record contains no basis to find that the limited conclusory evidence pertaining to those areas is relevant to the New York area.

Due to the above and other shortcomings, Austin’s testimony does not establish that N.Y. MTBE Law will increase ozone pollution to an extent which will interfere with the achievement of the goals of CAA.

Carl Johnson’s testimony; DEC waiver request

Plaintiff urges the Court to view Austin’s prediction of increased emissions in combination with the testimony of defendants’ witness Carl Johnson, DEC Deputy Commissioner, that New York has “very little cushion in terms of meeting attainment” with NAAQS and that the additional emissions from the MTBE ban would “interfere” with New York’s ability to attain the ozone standard. The Court finds, however, that read in context, Johnson’s testimony does not support plaintiffs case. Essentially, Johnson explained that DEC “would have to” address any increase in emissions from the MTBE ban or any other source and that it would do so by exploring every available avenue of emissions reduction, including non-road vehicles, generators, portable gas containers, paint and consumer products. Thus, in effect, Johnson testified that any increased emissions from a source such as the MTBE ban would “interfere with” or “impede” DEC’s efforts only to the extent that it would compel DEC to find compensating sources of reductions, but that nevertheless, DEC “would have to” find these compensating sources, and the reductions would ultimately be achieved.

Likewise, to the extent that plaintiff would have the Court rely on DEC’s statements in its January 6, 2003 request to EPA for a waiver of the RFG oxygenate requirement, the Court accepts the testimony and explanations given by Johnson and declines to draw contrary conclusions from the waiver request. Neither DEC’s waiver request nor any part of Johnson’s testimony adds support to plaintiffs case. Gordon Rausser’s testimony

With respect to economic impacts of N.Y. MTBE Law, the Court has stated that ensuring an adequate supply of gasoline at reasonable cost is not, viewed in isolation, a goal of CAA, but that it does not view the goals of CAA so narrowly as to exclude from all consideration the “larger context of market forces, health and environmental impacts, regional priorities, technological feasibility and other considerations.” OFA v. Pataki, 158 F.Supp.2d at 256. The Court further observed in its October 3, 2003 Memorandum-Decision and Order: “It cannot reasonably be argued that Congress’ goal was to reduce emissions pollution regardless of the cost — even, for the sake of argument, at the cost of total disruption of the gasoline market, complete unavailability of gasoline in certain regions of the nation, or astronomical increases in prices.” A short-term or relatively small impact on prices and/or *181 supply, however, would not support a finding that N.Y. MTBE Law interferes with the achievement of the goals of CAA.

Plaintiffs economics expert, Gordon Rausser, testified that N.Y. MTBE Law would transform New York into a boutique fuel market and cause New York to experience at least a 10% supply shortfall “in the short run” (that is, up to about six months). He further stated that the law would cause greater price volatility, with price spikes between 25% and 100% in the short run and between 6% and 16% thereafter. He based his predictions of supply shortfalls and price increases on the assumption that suppliers will reduce their supply of RFG to New York by 5 to 10 percent if MTBE is banned; plaintiff does not, however, present an adequate eviden-tiary basis for this assumption. Much of his testimony is speculative and has insufficient evidentiary support. Plaintiff has not made its case regarding the economic impacts of N.Y. MTBE Law; accordingly, it is not necessary to resolve factual disputes between Rausser’s testimony and that of defendants’ expert, Robert Reynolds.

In any event, even accepting Rausser’s predictions, the Court finds that the predicted effects are not of sufficient magnitude to support a finding of conflict preemption. At most, Rausser has projected a relatively small overall impact, the most severe elements of which would last no more than six months. Plaintiff has not shown that the economic impacts of N.Y. MTBE Law will interfere with the achievement of the goals of any aspect of CAA.

FINDINGS OF FACT

For the reasons set forth above, the Court finds that plaintiffs refinery modeling expert, David Hirshfeld, is not a reliable or credible witness and that therefore his testimony and his conclusion that the reductions in emissions achieved by ethanol RFG would be inferior to the reductions achieved by MBTE RFG are entirely discredited. Hirshfeld’s testimony,- therefore, does not aid plaintiff in proving its case. Further, even if the Court were to accept Hirshfeld’s testimony, his ultimate conclusion was only that the average emissions properties of ethanol RFG in New York “will be slightly worse than the emissions properties of MTBE” (emphasis added); the Court finds that this does not constitute evidence that N.Y. MTBE Law would cause an increase in emissions of significant magnitude to support plaintiffs case.

For the reasons set forth above, the Court further finds that' the testimony of Thomas Austin, plaintiffs air pollution control expert, does not support plaintiffs claim that N.Y. MTBE Law would result in increased exhaust and evaporative emissions. This testimony was based on Hirshfeld’s discredited predictions of the emissions properties of ethanol RFG and because Austin'testified that with ethanol RFG, New York would continue to over-comply with the complex model requirements, although overcompliance would be reduced in an unspecified amount. With respect to permeation, the Court finds that Austin’s testimony has negligible probative value because it depended heavily on a five-year-old survey of 324 California residents and because its credibility was significantly undermined by the timing of his upward revisions on rebuttal. Moreover, Austin’s testimony does not support plaintiffs claims because, although he projected an increase in emissions of VOC and NOx in an amount of 20 tons per day, he did not predict the amount of ozone which would result from this increase, thus providing no evidence of the significance of this increase. Nor did he provide a context in which the Court could evaluate the signifi- *182 canee of this increase. As stated above, the Court further finds that the “empirical data” from California and the Chicago-Milwaukee area does not assist plaintiff. Thus, the Court finds that Austin’s testimony does not constitute evidence that N.Y. MTBE Law will increase ozone pollution in any significant amount.

For the reasons discussed above, the Court further finds that neither Carl Johnson’s testimony nor DEC’S January 6, 2003 request to EPA for a waiver of the RFG oxygenate requirement adds support to plaintiffs case.

With respect to economic impacts of NY. MTBE Law, the Court finds that the testimony of plaintiffs economist, Gordon Rausser, was speculative and has insufficient evidentiary support and that therefore plaintiff has not made its case regarding the economic impacts of N.Y. MTBE Law. Moreover, even accepting Rausser’s predictions, the Court finds that the predicted effects are not of sufficient magnitude to support a finding that the economic impacts of N.Y. MTBE Law will interfere with the achievement of the goals of CAA.

CONCLUSIONS OF LAW

Plaintiff has based its challenge to N.Y. MTBE Law on the contention that the law will stand as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress in various aspects of CAA. First, the Court concludes that'plaintiff has not shown that N.Y. MTBE Law will interfere with the goal of the CAA’s RFG program to reduce emissions of VOC, NOx and toxics through the reformulation of conventional gasoline. This goal will be advanced by the use of any approved RFG. Moreover, both Hirshfeld and Austin testified that with ethanol RFG, New York would still overcomply with the RFG emissions standards, albeit to a lesser extent than with MTBE RFG. Thus, when N.Y. MTBE Law goes into effect, New York will continue to advance the goal of the RFG program by using ethanol RFG. 6

Nor has plaintiff shown that N.Y. MTBE Law will interfere with accomplishment of the goals of CAA generally. As noted, Hirshfeld’s evidence was entirely discredited; in any event, his ultimate conclusion that N.Y. MTBE Law would cause a “slight” deterioration in RFG emissions properties does not support plaintiffs case. As discussed above, Austin’s evidence that N.Y. MTBE Law will cause decreased reductions in ozone precursors does not demonstrate that N.Y. MTBE Law will affect ozone to an extent that would interfere with the goals of CAA. Rausser’s evidence was speculative and lacking in sufficient evidentiary support. In any event, Raus-ser did not project economic impacts from N.Y. MTBE Law of sufficient magnitude to support the conclusion that the law would impede accomplishment of the goals of CAA.

Plaintiff further urges that N.Y. MTBE Law will frustrate achievement of the goals of CAA because it will cause New York’s ozone levels to exceed the ozone NAAQS in 2007. The Court is not persuaded that NAAQS constitutes an expression of the goals of CAA such that a state’s failure to achieve NAAQS would necessarily impair achievement of the goals of CAA. Moreover, it is doubtful whether any prediction that a state will fail to achieve NAAQS in 2007 could support a finding of conflict preemption in 2003. In any event, plaintiff has not demonstrated *183 that New York will fail to achieve the ozone NAAQS in 2007. As noted, although Austin predicted a 20-ton per day increase in emissions of ozone precursors, he did not predict any specified amount of ozone increase resulting from the MTBE ban, nor did he provide a context in which the Court could evaluate the significance of this increase. Nor does Johnson’s testimony nor any other part of the record establish that N.Y. MTBE Law will prevent New York from meeting the ozone NAAQS in 2007.

Likewise, the Court also rejects any contention that N.Y. MTBE Law is conflict-preempted on the ground that it will prevent the state from complying with the State Implementation Plan (“SIP”). Johnson characterized SIP as “the collection[] of actions that the state of New York is going to take in order to meet attainment in 2007.” There is no evidence that New York’s SIP requires the use of MTBE RFG or prohibits the use of ethanol RFG. As with NAAQS, the Court is not persuaded that SIP constitutes an expression of the goals of CAA, nor does the evidence show that N.Y. MTBE Law will prevent New York from achieving SIP.

In enacting N.Y. MTBE Law to protect its citizens from groundwater contamination, the State of New York exercised a power traditionally reserved to the States. See Hillsborough, 471 U.S. at 715, 105 S.Ct. 2371. In urging the Court to strike down this enactment on the ground that it conflicts with federal law, plaintiff must overcome the presumption in favor of the constitutionality of the state law. See id. at 716, 105 S.Ct. 2371. This it has not done. For the reasons set forth above, the Court concludes that plaintiff has not demonstrated that N.Y. MTBE Law stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress in enacting CAA. The Court reaches this conclusion regardless of whether CAA’s goal of enhancing air quality is viewed as limited to air quality alone, or whether it is viewed in the larger context of market forces, health and environmental impacts, regional priorities, technological feasibility and other considerations. See OFA v. Pataki, 158 F.Supp.2d at 256.

It is therefore

ORDERED that judgment shall be entered dismissing the complaint in its entirety on the merits.

IT IS SO ORDERED.

Notes

. In OFA v. Pataki, this Court rejected plaintiff's argument, based on Geier v. American Honda Motor Co., 529 U.S. 861, 120 S.Ct. 1913, 146 L.Ed.2d 914 (2000), that oxygenate choice is a goal of CAA. See 158 F.Supp.2d at 260 n. 6. Subsequent to that decision, the Ninth Circuit rejected a similar argument in OFA v. Davis, 331 F.3d 665, 672 (9th Cir.2003). Plaintiff raises the argument again in its post-trial Memorandum of Law. The Court sees no reason to reconsider its previous conclusion.

. The Court rejects plaintiff's contention that the Court should draw an adverse inference against defendants from defendants' decision not to call one of their experts, Gary Whitten, as a witness at trial. See generally United States v. Caccia, 122 F.3d 136, 138 (2d Cir.1997). Plaintiff has the burden of coming forward with proof supporting its claims. Defendants have no affirmative obligation to prove anything and may prevail simply by attacking essential elements of plaintiff's case — in large part, this what defendants did in this trial. It is not clear what inference the Court could properly draw under the circumstances. At most, the Court can infer that Whitten would not have succeeded in refuting any part of plaintiff's proof and that therefore calling him would have accomplished the same thing as not calling him; certainly, the Court could not reasonably infer that Whitten would have supplied a missing element in plaintiff's case.

. There was evidence at trial that the decreased emissions reductions in New York might be offset by improved reductions in non-attainment areas in the northeast outside of New York State, primarily in northern New Jersey, which is part of the New York-Northern New Jersey-Long Island severe nonattainment area and is upwind of New York. Having concluded that plaintiff has not made its case, the Court does not deem it necessary to consider this or other evidence which tends to contradict plaintiff's proof.

. The Court rejects Hirshfeld's effort to discredit Tallett's results. Hirshfeld merely stat *178 ed that he was told by his partner Kolb that Kolb was unable to replicate Tallett's results. As a result of Hirshfeld’s apparent lack of knowledge about how Kolb reached this con-elusion, Hirshfeld's testimony in this regard lacked probative value and deprived defendants of meaningful cross-examination on the issue.

. Defendants objected that this was not proper rebuttal testimony and moved to preclude or strike this evidence. The Court reserved decision on the motion and permitted plaintiff to introduce Austin's rebuttal testimony. In view of the importance of the issues and the need for a complete record, the time pressures under which the parties and the Court were operating, the latitude which the Court afforded both sides in preparing and presenting their cases, and the fact that at the close of Austin's rebuttal testimony defendants did not seek an opportunity to introduce further evidence to refute it, the Court denies the defendants’ motion.

. Both Hirshfeld and Johnson testified that if MTBE gasoline is banned in New York, the refinety sector will provide to New York ethanol gasoline which complies with the RFG program.