In Re TMI Litigation Cases Consolidated II , 911 F. Supp. 775 ( 1996 )

911 F.Supp. 775 (1996)

In re TMI LITIGATION CASES CONSOLIDATED II.
This Document Relates to All Plaintiffs.

Civil Action No. 1:CV-88-1452.

United States District Court, M.D. Pennsylvania.

January 5, 1996.

Arnold Levin, Philadelphia, PA, Lee C. Swartz, Hepford, Swartz & Morgan, Harrisburg, PA, Dusan Bratic, Bratic & Portko, Dillsburg, PA, Louis M. Tarasi, Jr., Tarasi & Johnson, P.C., Pittsburgh, PA, Peter J. Neeson, LaBrum & Doak, Philadelphia, PA, Robert S. Mirin, Harrisburg, PA, Shawn A. Bozarth, Harrisburg, PA, Joseph D. Shein, Philadelphia, PA, for plaintiffs.

Alfred H. Wilcox, Pepper, Hamilton and Scheetz, Philadelphia, PA, Fred Speaker, Camp Hill, PA, for defendants.

Charles B. Zwally, Michael D. Reed, Mette, Evans & Woodside, Harrisburg, PA, Augustine V. Cheng, Associate General Counsel, Office of the General Counsel, New York City, for George Colbert.

MEMORANDUM

RAMBO, Chief Judge.

Presently before the court is Defendants' motion in limine to exclude the testimony of Plaintiffs' experts on dose. Specifically, Defendants seek to exclude some or all of the testimony of the following experts: Richard Webb, David A. Lochbaum, Ignaz Vergeiner, Charles E. Armentrout, James E. Gunckel, Victor Neuwirth, Vladimir A. Shevchenko, Steven B. Wing, Douglas Crawford-Brown, and Bruce Molholt. The issues have been briefed, and the court has conducted extensive in limine hearings with respect to these experts.^[1] Accordingly, Defendants' motions in limine are ripe for disposition. In addition, the court previously issued an order, unaccompanied by a memorandum of law, with respect to the following experts: Dr. L.N. Smierennyi, Dr. Ivannovich (L.N.U.), Professor Vasilenko, Ronald Kerman, Zinovy Reyblatt, Professor Eggar, Professor Scharr, Victor Neuwirth, A. Tascev, G. Kozubov, V. Popov, A. Portman, and O. Tarasenko. This memorandum of law will set forth the court's rationale with respect to those rulings.

I. BACKGROUND

The events precipitating this litigation began to unfold at approximately 4:00 a.m. on March 28, 1979, within the Unit-2 reactor at the Three Mile Island nuclear power facility ("TMI"). The facility is located on an island in the Susquehanna River not far from Goldsboro, Pennsylvania, and the events are now known to have led to the nation's most infamous nuclear reactor accident. Despite the diligent research efforts of many persons, including persons affiliated with the instant action, many important questions remain regarding the accident. In the context of this litigation, the parties have endeavored to answer these questions by explaining the course of events that constitute the TMI accident. The in limine hearings conducted by the court in November of 1995 focused on the parties' theories regarding the dose of radiation emitted from TMI and allegedly received by Plaintiffs. The United States Court of Appeals for the Third Circuit has recently ruled that Defendants violated the relevant standard of care by allowing radioactive releases to occur. In re TMI, 67 F.3d 1103, 1117-18 (3d Cir.1995). The issue that remains is that of the magnitude of release for purposes of proving causation and damages. Id. at 1119. ("Defendants violated the standard of care.... Plaintiffs' exposures to radiation remain relevant, but only to prove causation and damages.")

Plaintiffs posit that a "blowout" occurred within the TMI Unit-2 reactor, thereby forcing high quantities of radioactive noble gases into the atmosphere. Plaintiffs have proffered the testimony of a variety of experts in support of this theory. These experts opine on subjects as diverse as nuclear reactor physics, meteorology, chromosomal abnormalities, cancer incidence, and plant biology. Through their experts, Plaintiffs seek to demonstrate that after the blowout occurred within the reactor, winds carried a dense yet narrow plume of radioactive noble gases through the atmosphere. This plume, according to Plaintiffs, made contact with land in areas north and east of TMI, with the most significant contact occurring at higher elevations.^[2] Supporting their theory, Plaintiffs contend, is evidence of tree damage, chromosome abnormalities in humans, increased cancer incidence rates, and evidence of radiation sickness in both humans and animals.

Defendants do not deny that some radioactive noble gases escaped from the reactor building during the accident. In re TMI, 67 F.3d at 1118 ("Defendants conceded that they violated § 20.106, [therefore we find] they violated their duty of care.") However, Defendants contend that Plaintiffs cannot prove either that the release was significant and dangerous or that their alleged injuries were caused by radiation. Defendants challenge Plaintiffs' proffered expert testimony on dose as lacking scientific reliability.

II. LEGAL STANDARDS FOR THE ADMISSION OF EXPERT SCIENTIFIC TESTIMONY

In Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), the Supreme Court outlined the standards and reasoning that a district court must apply in determining whether expert scientific testimony is admissible at trial. The Third Circuit Court of Appeals has interpreted Daubert to characterize the district court's role as that of "gatekeeper." In re Paoli Railroad Yard PCB Litigation, 35 F.3d 717, 732 (3d Cir. 1994) ("Paoli II"). Both Daubert and Paoli II require the district court "to act as ``gatekeeper' and to assure that the scientific methodology upon which the expert opinion is founded is reliable, i.e., that the expert's conclusion is based on good grounds (the methods and principles of science)." Paoli II, 35 F.3d at 732 (discussing Daubert). This court's analysis of the proffered expert scientific testimony will be guided by the Third Circuit's exhaustive discussion and interpretation of Daubert in Paoli II.

Rule 702 of the Federal Rules of Evidence contains two basic requirements. The person proffered to testify as to scientific knowledge must be an expert, and the expert's proffered scientific opinion must be reliable. Paoli, 35 F.3d at 741-42. A proffered expert's qualifications should be evaluated under a rather liberal standard extending to the "substantive as well as the formal qualification[s]" of the expert. Id. at 741. Consequently, an expert who has little or no formal training in a given area may be found qualified based upon his life experience or extensive work history in that area. The reliability prong of the Rule 702 inquiry requires a district court to probe the scientific validity of proffered expert testimony. Id. at 742. When making this reliability inquiry, "a district court should take into account all of the factors listed by either Daubert or Downing^[3] as well as any others that are relevant." Paoli II, 35 F.3d at 742. Specifically, the court must consider the following factors:

(1) whether the method consists of a testable hypothesis; (2) whether the method has been subject to peer review; (3) the known or potential rate of error; (4) the existence and maintenance of standards controlling the technique's operation; (5) whether the method is generally accepted; (6) the relationship of the technique to methods which have been established to be reliable; (7) the qualifications of the expert witness testifying based on the methodology; (8) the non-judicial uses to which the method has been put.

Id. at 742 n. 8.^[4]

In addition to these factors, this court has apprised the parties that it also finds the following factors to be relevant:

Falsifiability. — Scientific explanations must be capable of falsification; that is, the logical form of a hypothesis must make it amenable to empirical testing ... [the factor is important enough that] the Supreme Court listed [it] first among its factors.

Logical Consistency. — A valid hypothesis cannot be self-contradictory.... [I]t is evident that a hypothesis that contradicts itself is logically ill-formed and cannot be tested.

. . . . .

Consistency with Accepted Theories. — Scientific knowledge tends to be cumulative and progressive, and a hypothesis that is not consistent with accepted theories should be regarded with great caution, whether or not the hypothesis ultimately proves true.... Though the Supreme Court rejected the Frye test, it did retain acceptance as one of the factors to be considered in evaluating science.

. . . . .

Precision. — Broad generalizations are far more difficult to corroborate than precise statements and have little explanatory power.... If severe and varied tests are the best indicator of validity, it follows that broad generalization that can account for any possible state of affairs, and thus cannot be empirically tested, are not as good.

In re TMI, No. 88-1452 (M.D.Pa. November 9, 1995) (order explaining the factors the court would consider when ruling on in limine motions) (quoting Bert Black, et al., Science and the Law in the Wake of Daubert: A New Search for Scientific Knowledge, 72 Tex.L.Rev. 715, 783-84 (1994)).

Finally, Rule 702 also requires that there be a "fit" between the testimony offered and a fact in issue. Paoli II, 35 F.3d at 743.

"``Fit' is not always obvious, and scientific validity for one purpose is not necessarily scientific validity for other, unrelated purposes." Thus, even if an expert's proposed testimony constitutes scientific knowledge, his or her testimony will be excluded if it is not scientific knowledge for purposes of the case.

Id. (quoting Daubert, 509 U.S. at ___, 113 S.Ct. at 2796 (citation omitted) (emphasis in original)). Once the party seeking to admit the testimony demonstrates that the proffered testimony is reliable by a preponderance of the evidence, Paoli II, 35 F.3d at 744 n. 11 (citing Daubert), the court must evaluate the proffer pursuant to Rule 403.

Rule 403 requires the district court to consider whether the admission of proffered testimony might overwhelm or confuse the jury. As explained in Paoli II:

[A] district court cannot exclude a scientific technique as too confusing and overwhelming simply based on its conclusion that scientific techniques by their very nature confuse and overwhelm the jury. There must be something about the particular scientific technique such as its posture of mythic infallibility that makes it especially overwhelming.

Id. at 746. Thus, "in order for a district court to exclude scientific evidence, there must be something particularly confusing about the scientific evidence at issue...." Id.

With the foregoing framework in mind, the court will reach the merits of Defendants' motion in limine. Embarking on this task, the court empathizes with the plight of the United States Court of Appeals for the Ninth Circuit as they recently ruled on the remand of Daubert. Before beginning their analysis, the Ninth Circuit reflected as follows:

[T]hough we are largely untrained in science and certainly no match for any of the witnesses whose testimony we are reviewing, it is our responsibility to determine whether those experts' proposed testimony amounts to "scientific knowledge," constitutes "good science," and was "derived by scientific method."

The task before us is more daunting still when the dispute concerns matters at the very cutting edge of scientific research, where fact meets theory and certainty dissolves into probability. As the record in this case illustrates, scientists often have vigorous and sincere disagreements as to what research methodology is proper, what should be accepted as sufficient proof for the existence of a "fact," and whether information derived by a particular method can tell us anything useful about the subject under study.

Our responsibility, then, unless we badly misread the Supreme Court's opinion, is to resolve disputes among respected, well-credentialed scientists about matters squarely within their expertise, in areas where there is no scientific consensus as to what is and what is not "good science," and occasionally to reject such expert testimony because it was not "derived by scientific method."

Daubert v. Merrell Dow Pharmaceuticals, 43 F.3d 1311, 1316 (9th Cir.1995) ("Daubert II"). The scientific issues before this court are similarly daunting. As a nation, we are grateful that we have experienced only one nuclear power reactor accident of the severity of the TMI accident. In the context of this litigation, however, this translates into a lack of concrete data and knowledge regarding the specifics of such an accident. To an extent, then, this will be a journey into uncharted waters.

III. RICHARD WEBB

A. Dr. Webb's Blowout Theory

Dr. Webb, who holds a Ph.D. in Nuclear Reactor Physics, is proffered by Plaintiffs to discuss reactor operations during the accident, and to explain how fission product noble gases were released into the atmosphere. Dr. Webb's report expresses his opinion that a "hydrogen blowout" occurred within the TMI Unit-2 reactor, causing a chain reaction culminating in the atmospheric release of fission product noble gases through the Reactor Coolant System and the Letdown line.^[5] Richard Webb, Analysis of the Three Mile Island Nuclear Accident With Respect to the Release of Noble Gas Fission Product Radioactivity into the Atmosphere, June 19, 1993, at 19 (hereinafter "6/19/93 Webb Report"). The 6/19/93 Webb Report characterizes the blowout as follows:

The lack of forced coolant flow allowed the water in the RCS [(Reactor Coolant System)] to lay and collect in lower parts of the system by gravity. The water level in the 1A cold leg pipe of the RCS is specially [sic] important; for near the base of that pipe is where the Letdown pipe is connected to the RCS. In the period between 10 hours and 13.5 hours into the accident, the 1A cold leg pipe probably emptied of water, due to venting of the reactor coolant through the PORV valve (loss of water from the RCS), and drainage of the water by the letdown out-flow. Because of the large drop in water inventory in the RCS by the PORV venting (during which time the injection of "make-up" water was minimal, and the make-up water was injected into the B-loop cold legs, not the A-loop cold legs), the water level in the reactor vessel probably dropped below the elevation of the inlet and outlet nozzles of the reactor vessel, thereby stopping the water flow (pour) into the 1A cold leg from the vessel, and thus allowing the 1A cold leg pipe to drain completely, due to the continuous letdown out-flow.

Once the water level dropped below the intake of the Letdown pipe, the gas and water vapor in the gas space of the Reactor Coolant System could then enter the Letdown pipe (under high pressure) and blow out from the system.

(Id. at 19 (citation omitted).) Dr. Webb did not testify at the in limine hearings, although his deposition is of record. Significant portions of Dr. Webb's deposition are troubling, however, as they evidence Webb's uncertainty about his own hypothesis, methods and report. (See, e.g., Webb Dep. at 125 ("Q. What was in between the upper and lower crust?" "A. The consolidated region, yes, the consolidated region they call it. If you notice they give .2 percent is metal. So apparently I just neglected it. I don't see it in my notes.") (emphasis added); Webb Dep. at 225 ("And so I haven't really finished my investigation of that second blowout possibility or additional hydrogen generation, therefore more release of hydrogen noble gases earlier in the accident. So I have to make it more complete or accurate about the matter of hydrogen generation.") (emphasis added); Webb Dep. at 256 ("Q. On page two of your report you say that ``A more detailed critique of Woodard's analysis has been written for inclusion as a section of this present report, but it has not been possible to type this critique in time for submission in this present report.' Did you ever type and submit that critique?" "A. Not yet. I've been involved in tremendous research...."); Webb Dep. at 372 ("You got to keep in mind, this result I'm getting of release by the model is not a, is not presented and represented to be a very accurate plus or minus 1 percent sure established prediction.") (emphasis added); Webb Dep. at 529 ("Q. Now I'm really confused. Should that number be 163 or 171 or 213 or something else?" "A. You have a right to be confused...." "Q. What should it be?" "A. Well, I'll, as I said, I wasn't fully, my latest calculation is the number 171 I gave you." "Q. All right." "A. Now that 213, maybe that's based on the 71 gallons per minute and I neglected to correct it. Again there was a very extreme deadline and I worked like 16-hour days and the last few days was like, like three hours' sleep, this kind of thing the last few days." (emphasis added).)

On December 6, 1995, the court became aware of a message left by Dr. Webb on the voice mail of A.H. Wilcox, co-counsel for Defendants. Through this message, Dr. Webb appears to have disavowed the underlying premise of his blowout theory. In relevant part, the message provides as follows:

Chub Wilcox, Richard Webb calling from Germany. As a courtesy to you I want to tell you why I haven't answered your questions at the deposition. I did prepare a letter to send you but it, in fact was ready to mail out. But I wanted to pursue questions I had of my own mathematical analysis which I've done in a very intensive and several months of deep research and I've remodeled and perfected my release models and I don't get the high releases that I got before, so it's like 25 million curies instead of 106 million curies. There's no blowout, gas blowout, that I estimate now that I had done earlier. There's a smaller one earlier in the accident that I'm predicting now, but it's during the time when the weather conditions dispersed the material much more.

So it's entirely, I always wanted to get back to the release model after I did all the dose and health effects research, get back to the release model that I did before and the deposition was an occasion to get back into that scene and pursue some of the questions I still had about which I noted in my original treatise. So I was going to make a report and submit it to the court about three weeks ago but I had to stop all that work after learning that the United States attacked Bosnia back in September ... so I've now been working full time the last three weeks to write up a paper on constitutional law about this war to give my government.

Transcript of Voice Mail Message received by A.H. Wilcox on 12/6/95 from Dr. Richard Webb (emphasis added).

B. Daubert/Paoli Analysis

In defining the phrase "scientific knowledge" as it pertains to Rule 702 of the Federal Rules of Evidence, the Supreme Court noted that "the word ``knowledge' connotes more than subjective belief or unsupported speculation. The term ``applies to any body of known facts or to any body of ideas inferred from such facts or accepted as truths on good grounds.'" Daubert, 509 U.S. at ___, 113 S.Ct. at 2795 (citation omitted). The Court went on to conclude that "the requirement that an expert's testimony pertain to ``scientific knowledge' establishes a standard of evidentiary reliability." Id. Pursuant to Rule 104(a) of the Federal Rules of Evidence, this court must determine "whether the expert is proposing to testify to (1) scientific knowledge that (2) will assist the trier of fact to understand or determine a fact in issue." Id. at ___, 113 S.Ct. at 2796.

In evaluating the Webb proffer, the court encountered difficulty deciphering the substance of the proposed testimony. During June of 1993, Dr. Webb submitted a series of reports and affidavits in support of his evaluation of the TMI accident. These filings are replete with statements by Dr. Webb that they are draft versions, and that more complete reports and explanations will follow at a later date. (See, e.g., 6/19/93 Webb Report at 1 ("[I]t has not been possible to write a full treatise which would contain the complete details of the analysis and proof in the time required for submitting the Affidavit. In place of such a treatise, which can be written later, the following report of the analysis is offered...."); 6/19/93 Webb Report at 24 ("I have written separate chapters ... which contain my critical evaluation of Daniels and Woodard treatises and the Presidents' Commission Report ... but it has not been possible to type and print these chapters for submission of this present report.").) No "final" report was filed prior to the close of fact discovery on dose experts. Moreover, as noted above, during his deposition Dr. Webb appeared particularly uncertain of his methodology and the means by which he reached his conclusions. Dr. Webb's reports, standing alone, are inadequate to serve as a basis for finding that Dr. Webb would testify to scientific knowledge. Likewise, Dr. Webb's deposition testimony does little to solidify or clarify his analysis; in fact, it only adds to the confusion. Finally, Plaintiffs chose not to have Dr. Webb testify during the in limine hearings. As such, the court was unable to probe Dr. Webb himself regarding the uncertainties expressed in his report and deposition testimony.

Based upon the aforementioned discussion, the court was prepared to exclude the testimony of Dr. Webb as unreliable pursuant to Rules 702 and 104(a) of the Federal Rules of Evidence. Whiting v. Boston Edison Co., 891 F.Supp. 12, 18 (D.Mass.1995) (excluding expert's dose calculation based on formula of his own invention which, although superficially impressive, was "so riddled with factual inaccuracies and unproven assumptions that no reasonable jury could give his opinion credence."). Receipt of the transcript of Dr. Webb's voice mail message to Mr. Wilcox only reinforced the court's decision on this matter. As Dr. Webb himself is unable to stand behind his hypothesis or to explain how he reached his conclusions, it is impossible for the court to see how he could assist a jury in understanding the complex facts of the captioned action. Accordingly, the court will grant Defendants' motion in limine as it applies to the proffered testimony of Dr. Webb.

IV. DAVID A. LOCHBAUM^[6]

David Lochbaum is a nuclear engineer with fifteen years experience in the commercial reactor field. In his expert report, Lochbaum reviews Dr. Webb's report and reaches the conclusion that "Webb's methodology provides a valid approach for the determination of the release of fission product noble gases." (Lochbaum Rpt. at 5.) Lochbaum did not testify during the in limine hearings, and does not advance any relevant theory or hypothesis independent of his review of Dr. Webb's report. Because the court has found Dr. Webb's testimony to be unreliable and therefore inadmissible, Lochbaum's testimony no longer "fits" within the case. That Lochbaum can opine as to the methodological soundness of Dr. Webb's report, or point to flaws in Dr. Webb's methodology, is of no consequence as Dr. Webb's report will not be admitted. Based upon the foregoing, the court will grant Defendants' motion in limine to exclude the testimony of David Lochbaum as lacking the requisite Rule 702 fit with a fact in issue.

V. IGNAZ VERGEINER

Dr. Vergeiner is a meteorologist affiliated with the University of Innsbruck in Innsbruck, Austria. He holds undergraduate degrees in math and physics, and a Ph.D. in meteorology. For the past twenty years he has taught meteorology at both the undergraduate and graduate levels. Plaintiffs proffer Dr. Vergeiner as an expert in meteorology, specifically boundary level meteorology in alpine regions,^[7] for the purpose of explaining how the alleged plume of radioactive noble gases travelled and dispersed after being released during the TMI accident. Defendants have filed a motion in limine seeking to exclude all of Dr. Vergeiner's proffered testimony.

A. Proffered Testimony

Dr. Vergeiner's testimony can be subdivided into three distinct areas. First, he has been offered to testify on the subject of boundary layer meteorology. If permitted, Dr. Vergeiner will opine regarding the weather conditions at TMI and in surrounding areas at the time of the accident and during the period immediately following the accident.^[8] Based upon his own assumptions and source term data supplied to him by Plaintiffs' counsel,^[9] Dr. Vergeiner produces estimates regarding plume dispersion. A "plume movie" and water model^[10] are presented to assist Dr. Vergeiner in explaining his hypothesis. The "movie" is actually a series of sketches drawn by Dr. Vergeiner aimed at depicting the possible course of plume dispersion during the accident. (Tr. at 617-21, 634-36, 654.) Similarly, the water model is used to visually demonstrate how a plume might have dispersed over the TMI region during the accident. (Vergeiner Dep. at 235-38, 241.) Defendants seek to have the plume dispersion testimony excluded as scientifically unreliable and speculative. Further, Defendants seek exclusion of the plume movie and water model video on the ground that they will confuse the jury. Defendants fear that the jury will construe the examples as what did happen rather than what could have happened.

Second, Dr. Vergeiner is proffered to give his estimates of dose in certain areas surrounding TMI. These dose estimates are based upon his plume dispersion models, and on source terms given to him by Plaintiffs' counsel. Defendants seek to exclude this portion of Dr. Vergeiner's testimony as scientifically unreliable. During his deposition Dr. Vergeiner was unable to explain the genesis of his source terms. (Vergeiner Dep. at 248-49.) Further, at the hearing, he characterized his own models as a hybrid of art and science. (Tr. at 638.)

Finally, throughout his report, Dr. Vergeiner expresses his opinions regarding subjects wholly unrelated to meteorology. These opinions center around a general distrust of the "original" plant data from TMI, and around an apparent suspicion of the nuclear power industry. Defendants object to this testimony as being outside Dr. Vergeiner's area of expertise.

The court will analyze each of the above mentioned areas of proffered testimony under the Daubert/Paoli II framework set forth in Section II of this memorandum.

B. Daubert/Paoli Analysis

1. Is the Methodology Based Upon a Testable Hypothesis?

The logical starting point for an analysis of Dr. Vergeiner's methodology is his own somewhat convoluted discussion of that methodology. When asked during the in limine hearings whether his method consisted of a testable hypothesis, Dr. Vergeiner responded as follows:

I would say this. When I started this work, I had no finished hypothesis. How should I? There was no way I could have. So I started drawing together the data and forming a picture — analyzing the data and forming a picture of the weather situation. In the end, I believe what I do have you may call a hypothesis, or a choice of hypotheses, and I would explain them this way.

One hypothesis is that most of the time that we are interested in, when releases occurred, plumes were narrow and relatively little diluted. So the concentrations were very high, depositions were very localized and you have to face the consequences.

The other hypothesis would be that plumes were overwhelmingly or exclusively as wide as Mr. Woodard has computed them to be, right? They were so diluted already before they hit any hills or mountains that impaction could not occur, that these features had practically no influence on what happened at that — of course, you must also then draw the consequences from that.

And I am convinced out of my work that there is strong reasons for the hypothesis that plumes were most of the time were very narrow and undiluted, and I will explain that. Of course, the best proof of that would be either if we had seen these plumes or if we had been able to have detailed measurements up in the air, documenting these plumes. Right? ... So we don't have that kind of information.

. . . . .

So I'm trying to obtain a picture of how such a hypothesis might be verified. And of course, it is also true that when I think of what I would have needed to really verify any hypothesis, some of these, many of these data are not, do not exist. For example, it would have been great to have many, many measurements of air concentration, because what the model, the Gaussian model, the first thing it gives you is concentration of the release, right? Of the substance. But there is no such measurement. There are some few measurements of iodine, but there is no measurement of the noble gases, as far as I remember, if I remember correctly, to until day eight or nine after the release, eight or nine days after the release, after the start of the accident.

So this type of information is absolutely lacking. And I did look at, just to confirm some assumptions, some scenarios, I call them, I did look at concentration data....

(Tr. at 618-21.) From this explanation, the court surmises that the Vergeiner hypothesis provides that a narrow and highly concentrated plume was emitted from TMI during the accident, and that the plume remained concentrated for a prolonged period of time. (Pls.' Findings of Fact and Conclusions of Law at 10 ("Dr. Vergeiner's hypothesis is that the plume from the TMI accident did not disperse in accordance with a Gaussian plume model, but instead remained in a concentrated form and travelled in a narrow plume manner.") (hereinafter "Pls.' Findings").) This hypothesis is testable insofar as another expert could use the data employed by Dr. Vergeiner to reach an independent conclusion.^[11] However, because Dr. Vergeiner elected to "eyeball" his plume movie drawings they would necessarily be difficult to accurately test or replicate. (See Vergeiner Dep. at 312-13.)

Additionally, Dr. Vergeiner's dose estimates do not lend themselves to empirical testing. The source terms used by Dr. Vergeiner as a basis for his estimates were provided by Plaintiffs' counsel.^[12] Dr. Vergeiner himself could not explain how the figures were derived. Consequently, any expert endeavoring to test the validity of Dr. Vergeiner's dose estimates would be relegated to relying blindly on the source terms supplied by Plaintiffs' counsel. As such, it would be impossible to subject the dose hypothesis to rigorous and unbiased testing.^[13]

Based upon the foregoing, the court finds that Dr. Vergeiner's hypothesis cannot be empirically tested. The bulk of Dr. Vergeiner's testimony consists of broad generalizations and suppositions regarding what might have happened during the accident. While such generalizations could provide a valid starting point for a detailed analysis, Dr. Vergeiner's opinions never move beyond this starting point. Thus, no hypothesis is advanced that is capable of being tested. Accordingly, this factor weighs against admitting Dr. Vergeiner's proffered testimony.

2. Has the Methodology Been Subject to Peer Review?

Insofar as boundary layer meteorology and synoptic analysis^[14] are standard techniques, Dr. Vergeiner's use of these standard techniques in a standard manner has been the subject of significant peer review. To the contrary, Dr. Vergeiner testified that the treatise that he prepared for this litigation has not been peer reviewed. (Tr. at 623.) However, Dr. Vergeiner stated that the principles underlying the conclusions reached in his treatise have been subject to peer review. (Id. at 624-26.) That Dr. Vergeiner has "been unable or unwilling to publish ... [his] work undermines plaintiffs' claim that the findings [Dr. Vergeiner] proffer[s] are ``ground[ed] in the methods and procedures of science' and ``derived by scientific method.'" Daubert II, 43 F.3d at 1318 n. 9 (quoting Daubert, 509 U.S. at ___, ___, 113 S.Ct. at 2795, 2796). The court finds nothing to indicate that Dr. Vergeiner's plume movie or water model methodology has been subject to peer review.^[15] Similarly, it is axiomatic that Dr. Vergeiner's methodology of adopting source terms presented to him by Plaintiffs' attorneys has not been subject to peer review.

This factor will weigh in favor of the admission of Dr. Vergeiner's basic meteorologic and synoptic analysis testimony, and against the admission of Dr. Vergeiner's plume movie, water model and dose estimate testimony.

3. Is There a Known or Potential Rate of Error?

During his testimony, Dr. Vergeiner did not discuss whether there existed any known or potential rate of error with respect to his methodology. Rather, he expounded upon his personal belief that "the general picture and the general magnitude of the concentrations of these plumes would be right," but that the exact course and concentration of the plume could not be calculated by anyone. (Tr. at 622-23.) The court finds the potential rate of error to be high. Much of Dr. Vergeiner's report, as mentioned above, is based upon speculation and estimates. Moreover, although Dr. Vergeiner stated that he believes that it is important to test a hypothesis, he did not thoroughly test his hypothesis. According to Dr. Vergeiner, he tested his hypothesis "to the limited extent that seemed sensible" to him. (Tr. at 663.) When asked to explain the limitations, Dr. Vergeiner noted that he "did not compare ... [his] plume movie with TLDs because ... [he] believe[d] that the TLDs have very bad instruments, very insensitive and very not useful for discriminating plumes." (Id. at 664.); cf. Whiting, 891 F.Supp. at 18 n. 24 (discussing the flawed methodology where expert disregarded primary and secondary data due to his unsubstantiated mistrust of the accuracy of TLDs). In qualifying this answer, Dr. Vergeiner concluded that "[w]e have to get away from the idea that anything that is called data or measurement is beyond criticism." (Id. at 664-65.)

In the long run, Dr. Vergeiner may be proven to be correct. TLDs^[16] may eventually prove to be highly ineffective at measuring plume concentration. However, the court takes judicial notice of the fact that the relevant scientific communities regularly use TLDs and find them to be a valid means of measuring radioactivity. See NUREG-0637, "Report to the Nuclear Regulatory Commission from the Staff Panel on the Commission's Determination of an Extraordinary Nuclear Occurrence ("ENO")" at A1-A4 (January 1980) (general discussion of TLDs with specific attention paid to their use at TMI in conjunction with evaluating the TMI-2 accident). As such, the court finds that within the context of this litigation, Dr. Vergeiner's refusal to consider such information calls into question the reliability and scientific validity of his conclusions. At the very least, Dr. Vergeiner could have compared his results with the TLD readings and then explained to the court why his results were scientifically more sound than the TLDs. Instead, Dr. Vergeiner has disregarded arguably valid information without any rational explanation. The court finds that his failure to consider primary data in combination with his use of speculation, assumptions and "eyeballing" of figures, exposes Dr. Vergeiner's methodology to a potentially high rate of error. Moreover, the court finds that the failure of the water model to take into account the actual topography of the TMI area increases the risk that any demonstration run through the water model will be subject to a high rate of error. (10/18/95 Vergeiner Aff. at 6, ¶ 6 (discussing limitations of the water model).)

This factor will weigh against the admission of Dr. Vergeiner's testimony including his plume movie and water model.

4. Were There Standards Controlling the Technique's Operation?

The court finds this factor to have only tangential relevance to Dr. Vergeiner's proffered testimony. Insofar as Dr. Vergeiner "eyeballed" his plume movie, the court holds that there were few standards controlling the operation of that technique. Further, the court finds that in making his dose estimates, Dr. Vergeiner failed to employ basic standards to control his technique's operation.^[17] It appears that there were available standards to control the operation of the water model, see Water Model Report at 9; however, this will prove irrelevant in light of the other flaws which the court has already touched upon.

Based upon the foregoing, the court will not accord any weight to this factor when determining the admissibility of the proffered testimony.

5. Is the Methodology Generally Accepted?

Certain of Dr. Vergeiner's methodologies are generally accepted while others appear to be directly contrary to generally accepted methods. Dr. Vergeiner's methods with respect to his basic meteorological and synoptic analysis testimony are generally accepted. The parties do not dispute this. The plume dispersion and dose estimate methodologies cannot be categorized as generally accepted based upon a common flaw. At the hearing, Dr. Vergeiner testified that he discarded standard mathematical computer models^[18] often used to estimate concentrations when actual concentration measurements are unavailable, see EPA 1992 at 22,912; NRC 1995 at 36, on the ground that the models did not take into consideration the complicated wind and topographical conditions at TMI. (Tr. at 562-66.) In his report, however, Dr. Vergeiner gives a less scientific explanation for his disregard of generally accepted methods:

So why not straightaway model the flow numerically using the power of modern computers? Let me remind the reader of the enormous complexity of such a task. Transport and dispersion models exist to the hundreds, many of them in the nuclear industry or in the scientific "grey zone" around it.... Even to list the respective acronyms would fill pages.

Quite a few of these models are global in scale, and some apparently have succeeded in simulating the path and contaminating action of the Chernobyl clouds reasonably well, after years of tuning and verification on the many observations available.

. . . . .

Documentation is a problem, as well as the need for special graphics and internal routines, or compatibility of various computer languages. Not all applications have been successes ...

It was my judgement, therefore, not only that it did not seem feasible to obtain access and results within a limited time span and financial frame, but that relatively simpler, well-tested, more robust and accessible models might be just as good, even preferable. This may appear to do injustice to the more than 50 man-years' [sic] expert work condensed in this enormous structure. There is no doubt that each of these models is capable of computing flow structures very suggestive of real nature, but I couldn't convince myself that the enormously increased expense would bear a sound relation to similarly improved results.

Ignaz Vergeiner, Treatise on the TMI-2 Accident of March 28, 1979, Particularly its Meteorological Aspects Including Transport and Dispersion of the Radionuclides Released, at 50 (July 1994) (hereinafter "Vergeiner Treatise I").

Dr. Vergeiner admits to discarding the generally accepted computer models with respect to his dose calculations and plume dispersion analysis. Moreover, he fails to provide the court with anything other than his bare assertion that the self-styled models he does employ are generally accepted. Dr. Vergeiner has not met his Daubert burden in this respect. Daubert II, 43 F.3d at 1319.

Based upon the foregoing, this factor will weigh in favor of excluding Dr. Vergeiner's testimony with respect to the plume movie and dose estimates. This factor will not weigh against the admission of the water model testimony.

6. Is There a Relationship Between the Technique and Methods That are Established to be Reliable?

Portions of Dr. Vergeiner's methodology resemble methods established to be reliable while other portions appear novel. Defendants do not challenge, and the court does not question, that Dr. Vergeiner's broad meteorological statements and synoptic analysis comport with established and reliable meteorological principles. Less can be said with respect to his plume movie, water model and dose estimates. Using computerized models, such as the Gaussian model, as the "established and reliable" method against which Dr. Vergeiner's methods must be judged,^[19] the court finds no relationship between the two. Further, the court finds no recognizable similarities between established methods for dose reconstruction and Dr. Vergeiner's dose estimates. The literature does not support the notion that a scientist would blindly base a dose reconstruction upon an assumed and unverified source term supplied by attorneys. See, e.g., Radiation Dose Reconstruction at 16 ("A full description of the source term includes what was released and in what form and where and when the release occurred. These factors must be described with enough accuracy and detail both to satisfy the scientific requirements for the design and conduct of epidemiologic studies and to address the public's concern."); id. at 18 ("The credibility of a comprehensive source term study depends on confirming that all pertinent documents have been seen and evaluated. Complete records are essential in identifying the source term.")

The court finds that this factor weighs in favor of excluding Dr. Vergeiner's plume movie, water model and dose estimate testimony.

7. Qualifications of the Expert Based Upon the Methodology

Dr. Vergeiner's curriculum vitae evidences his qualifications as a meteorological expert. Moreover, Defendants do not challenge his credentials. The court, however, questions Dr. Vergeiner's ability to opine as an expert with respect to his dose estimates. A review of his curriculum vitae reveals no specialized experience tending to qualify Dr. Vergeiner as an expert in the area of dose reconstruction.^[20] Thus, while the court agrees with Plaintiffs that "Dr. Vergeiner's meteorological credentials unquestionably qualify him for his analysis of weather-related plume dispersion", the court disagrees that this experience also qualifies him to discuss dose estimates. (Pls.' Brief in Opposition at 74.)

Based upon the foregoing the court finds that this factor weighs against the admission of Dr. Vergeiner's dose estimate testimony. It weighs in favor of his meteorological and synoptic analysis testimony, and will not weigh against the admission of the plume movie and water model.

8. Non-Judicial Uses of the Methodology

Insofar as Dr. Vergeiner's methods reflect standard meteorological analysis, they are widely used in non-judicial settings. The methodology followed with respect to the plume movie, water model and dose estimates deviates from standard methodology and appears to have been derived solely in connection with this litigation. This factor will weigh against the admission of Dr. Vergeiner's plume movie, water model and dose estimate testimony.

9. Other Factors

In conjunction with the above Daubert/Paoli II analysis, the court indicated that Dr. Vergeiner's explanations are, by and large, not subject to falsification. Further, while his hypothesis is generally logically consistent, it is not consistent with accepted theories. Most striking is Dr. Vergeiner's disregard of primary data. While Dr. Vergeiner might prove to be ahead of his time in his disdain for TLDs, TLDs are currently accepted devices for measuring atmospheric radiation. As such, Dr. Vergeiner's cursory disregard of such measurements is unacceptable. Finally, the court finds that Dr. Vergeiner's plume movie and water model lack precision. "[B]road generalization[s] that can account for any possible state of affairs ... cannot be empirically tested, [and thus] are not as good." Science 72 Tex.L.Rev. at 784. Dr. Vergeiner himself indicated that his models are nothing more than generalizations and suppositions about what might have happened during the TMI accident. (Tr. at 635-36; Vergeiner Dep. at 278-79.) As such, the models are necessarily imprecise.

C. Rule 702 Fit

The court's analysis of the Daubert/Paoli II factors militates in favor of excluding Dr. Vergeiner's plume movie and water model. While the court could exclude the proffered testimony based solely on those grounds, the court finds the more appropriate ground for exclusion of the plume movie and water model to be an absence of "fit." The plume movie and water model are not accurate depictions of plume dispersion during the TMI accident. Plaintiffs do not dispute this fact. These models, instead, are proffered to demonstrate what may have happened. The models might have "fit" within the case if they bore any meaningful relationship to the actual topography of the TMI area or to the primary data.

The plume movie does not "fit" within the case because it is based upon an undefined source term, fails to account for weather readings on the TMI weather tower, and fails to incorporate all primary data. Thus, it cannot be found to bear a valid relationship to the TMI accident. Similarly, the water model does not take into account the actual topography of the TMI area. Any demonstration performed using the water model, therefore, would not bear a close relationship to the way a substance would be dispersed into the atmosphere around TMI. "Federal judges must ... exclude proffered scientific evidence under Rules 702 and 403 unless they are convinced that it speaks clearly and directly to an issue in dispute in the case, and that it will not mislead the jury." Daubert II, 43 F.3d at 1311 n. 17. Neither the plume movie nor the water model speak clearly to any issue in dispute in this case.^[21] Because neither purports to provide an accurate representation of plume dispersion during the accident, the models cannot "clearly" speak to anything. Furthermore, there is a high potential that such models would mislead a jury insofar as the jury might perceive the models to accurately represent the manner in which plumes did disperse during the accident. The court will exclude the plume movie and water model testimony pursuant to Rules 702 and 403. Robinson v. Missouri Pacific R.R. Co., 16 F.3d 1083, 1089 (10th Cir.1994) ("Concerning future similar issues under Rule 702 we suggest that as ``gatekeeper' the district court carefully and meticulously make an early pretrial evaluation of issues of admissibility, particularly of scientific expert opinions in films or animations illustrative of such opinions.").

D. Conclusion

The foregoing analysis reveals that a significant portion of Dr. Vergeiner's proffered testimony is unreliable and therefore inadmissible pursuant to Rule 702. Specifically, Dr. Vergeiner's dose estimates will be excluded at trial as lacking scientific validity and reliability. Further, the court will exclude Dr. Vergeiner's plume movie and water model as lacking the necessary "fit" with the instant litigation pursuant to Rule 702, and as potentially confusing to the jury pursuant to Rule 403. McKnight v. Johnson Controls, Inc., 36 F.3d 1396, 1402 (8th Cir.1994) (excluding simulation "where some principles of some kind may be demonstrated but in a fashion that looks very much like a recreation of the events that gave rise to the trial"); Robinson, 16 F.3d at 1089 n. 7 (noting that Rule 403 prejudice is a ground that a court should consider for exclusion of computer simulations and animations and the like); Fusco v. General Motors Corp., 11 F.3d 259, 264 (1st Cir.1993) (excluding accident replication videotape where the drama of the videotape was capable of overcoming the logic of the distinctions raised by opposing experts). As Dr. Vergeiner is an expert in meteorology, and can offer insights regarding the weather conditions during and immediately following the accident, the court will not exclude his testimony in this narrow area. The court will, however, require Plaintiffs to give a proffer as to how this narrow area of testimony "fits" absent Dr. Vergeiner's other proposed testimony. (See Defs.' Findings at 54, ¶ 8 (challenging the "fit" of the synoptic analysis presuming the inadmissibility of the plume movie and dose estimates).)

VI. CHARLES E. ARMENTROUT

Professor Armentrout has masters degrees in Physics and Radiological Physics. According to his curriculum vitae, he is currently an Associate Professor at the University of Southern Maine. Professor Armentrout's proffered testimony can be divided into four distinct areas as follows: (1) a discussion of what Armentrout calls "anomalously high" radiation background readings measured in Portland, Maine a few days after the TMI accident (2/20/95 Armentrout Rpt. at 2-4); (2) the results of soil sample analyses performed by Victor Neuwirth on soil samples extracted from TMI and its surrounding areas (id. at 4-23); (3) a discussion of radiation survey meter readings allegedly taken by a local resident during the TMI accident; and, (4) an explanation of the rate-dependant behavior of Geiger-Muller radiation detectors. (Id. at 26.)^[22] The court has already precluded Professor Armentrout from testifying as to the substance of Victor Neuwirth's report (e.g. soil sampling techniques and results). In re TMI, No. 88-1452 (M.D.Pa. November 9, 1995) (preliminary order ruling on portions of Defendants' motion in limine). Professor Neuwirth filed his own report and testified at the in limine hearings. Accordingly, Plaintiffs' proffer with respect to the Armentrout/Neuwirth soil sample analysis will be evaluated pursuant to the court's discussion of the Neuwirth proffer. The court will now evaluate the remaining areas of Professor Armentrout's proffered testimony.

A. Proffered Testimony

1. Section I of the Armentrout Report

On March 31, 1979, following a discussion with a former student, Professor Armentrout and that student assembled radiation detection equipment on the roof of the science building at the University of Southern Maine. (2/20/95 Armentrout Rpt. at 3.) The equipment recorded two "bursts" of what Professor Armentrout identifies as anomalous radiation readings. (Tr. at 1248-49.) Professor Armentrout memorialized his observations in a letter to the President of the University of Southern Maine. (Tr. at 1252-53.) The letter has been made part of the record. (2/20/95 Armentrout Rpt., Ex. A.) In his report, Professor Armentrout claims that government installations around Portland, Maine recorded similar radiation readings during the same time frame as his readings. (2/20/95 Armentrout Rpt. at 4.) Further, he notes that a Science magazine article stated that unusual radiation measurements were taken in Albany, New York shortly after the accident. Professor Armentrout contends that these reports verify the readings that he recorded and indicate that the TMI plume passed over the area. (2/20/95 Armentrout Rpt. at 4 & n. 4.) Despite these alleged "verifications," Professor Armentrout testified that his readings do not tell him with a reasonable degree of scientific certainty what radionuclides were released at TMI or how large the release was. (Tr. at 1290.)

Defendants contend that this portion of Professor Armentrout's proffered testimony should be excluded as it will not be helpful to the trier of fact. (Defs.' Findings at 81, ¶ 15.) Plaintiffs' findings do not directly address this section of the Armentrout report, but rather, focus on the soil sampling techniques and results. The court agrees with Defendants insofar as the testimony does not "fit" with any material fact in issue. Equipment on the roof of the science building at the University of Southern Maine registered radiation "spikes" shortly after the TMI accident began. Professor Armentrout suggests that the readings support the theory that the TMI plume passed over Maine. This contention is made absent reliable scientific corroboration or verification, and thus, cannot tie the radiation readings to any occurrence at TMI. Accordingly, the court finds that the proffered testimony will not assist the jury in determining any fact in issue. Professor Armentrout will be precluded from testifying as to the substance of section I of his report.^[23]

2. Section III of the Armentrout Report

In this section, Professor Armentrout speaks to his attempts to verify alleged radiation readings taken by civilians living near TMI during the accident. After repeated phone contact from a resident of Etters, Pennsylvania,^[24] Professor Armentrout attempted to verify the resident's alleged radiation readings. Professor Armentrout devised a means to calibrate the resident's ion chamber monitor, and came to the Harrisburg area to meet with the gentleman and verify the alleged readings. However, when Professor Armentrout arrived in the area, he could not locate the gentleman who had contacted him. Instead, Professor Armentrout spoke to a neighbor who remembered the that he and the missing gentleman had purchased identical ion chamber monitors^[25] from a local store. Further, this neighbor remembered that he made an indoor reading during the accident that sent his monitor off-scale. (Armentrout Rpt. at 25.) Professor Armentrout contends that this incident verifies the first gentleman's reports of the high readings that he took during the accident.

Defendants challenge this section of the report "because experts in the field of dose reconstruction would not rely on hearsay statements which even Armentrout admitted were unreliable." (Defs.' Findings at 81, ¶ 16.) Again, Plaintiffs do not directly address this section of the Armentrout report, instead electing to concentrate on the soil sample analysis. The court must agree with Defendants. This anecdotal evidence does not meet the most basic standards of scientific validity or reliability. Accordingly, the court will exclude the testimony proffered in section III of the Armentrout report.

3. Conclusion

Based upon the foregoing, the court finds the testimony proffered in section I of the Armentrout report to lack scientific validity, reliability and the requisite "fit" with the instant litigation. The testimony proffered in section III similarly lacks scientific validity and reliability; therefore, the testimony will be excluded. The court will now turn to the Neuwirth proffer and determine whether the testimony will be admitted at trial. Evaluation of the Neuwirth proffer will necessarily involve discussion of that portion of section II of the Armentrout report authored by Professor Armentrout.

VII. VICTOR J. NEUWIRTH

Victor Neuwirth is a laboratory associate and Professor of Chemistry at the University of Southern Maine. Professor Armentrout enlisted the support of Professor Neuwirth to analyze soil samples from the TMI area to determine whether he could extract potentially radioactive materials from the samples. Professor Neuwirth performed the analysis on the soil samples and authored the portion of the Armentrout report dealing with the soil sample analysis. Additionally, Professor Neuwirth testified during the in limine hearings. Defendants challenge the proffered testimony as lacking a Rule 702 "fit" because no connection can be made between the results of the soil sampling and the 1979 TMI accident. (See Defs.' Findings at 87, ¶ 1.)

A. The Proffered Testimony

If permitted, Professor Neuwirth will testify as to the methods he employed and the results he obtained in analyzing certain TMI soil samples. At the direction of Professor Armentrout, Professor Neuwirth concentrated soil samples provided to him by Professor Armentrout and used a sodium iodine detector to make integrated counts of radionuclides.^[26] (Tr. at 1065-66; Armentrout Rpt. at 7-9.) Professor Neuwirth found certain samples to contain radioactive materials, but was unable to identify specific radionuclides. Despite his inability to identify specific radionuclides, Professor Armentrout directed him to calculate the half life of each of the soil samples as a whole (rather than calculating the half lives of specific radionuclides within each soil sample).^[27] Professor Neuwirth took measurements of the same soil samples at a one year interval, with no intermediate readings. Based upon these two data points, he made a generalized half life calculation for each of the samples. Professor Armentrout used Professor Neuwirth's half life calculations to make dose estimates for the TMI Plaintiffs.

B. Daubert/Paoli Analysis

1. Does the Method Consist of a Testable Hypothesis?

Plaintiffs contend that Professor Neuwirth's hypothesis may be tested. Defendants do not directly address this issue. Professor Neuwirth's hypothesis is that radioactive decay observed in the TMI soil samples is attributable to TMI fission products.^[28] Plaintiffs argue that this hypothesis may be falsified insofar as the chemical extraction process would either prove or disprove the notion that the soil samples were radioactive. The court agrees that this portion of the hypothesis is testable, and that it has been tested. However, the effect of the testing was negated as Professor Neuwirth did not use the test results to modify his hypothesis. After completing his soil sample analysis, Professor Neuwirth sent the samples to the Data Chem laboratories to obtain more specific radionuclide readings. The Data Chem results falsify Professor Neuwirth's hypothesis in that they demonstrate that the samples contain little other than naturally occurring background levels of radiation.^[29] (Tr. at 295-303, 1050-52, 1296, 1298, 1300, 1303.) A proper methodology would have incorporated these findings and appropriately modified the underlying hypothesis. Professor Neuwirth neither acknowledged the findings nor modified the initial hypothesis.

Of more concern to the court are Professor Neuwirth's half life calculations. Because he was unable to identify specific radionuclides within any sample, Professor Neuwirth was forced to make a generalized half life estimate for each soil sample as a whole. The estimate was based upon taking two separate readings of each sample with one year separating the two readings. Based upon the decay in the sample over an entire year, half life values were calculated. Professor Armentrout then used the Neuwirth half life values to derive dose estimates for TMI residents.

Based upon the foregoing, this factor will weigh against the admission of the soil sample and half life calculation testimony.

2. Has the Methodology Been Subject to Peer Review?

The parties do not dispute that the chemical extraction process employed by Professor Neuwirth has been subject to peer review. It is doubtful, however, that the half life calculation technique utilized by Professor Neuwirth has been subject to peer review. Defense expert Dade Moeller testified that to determine the half life of a material, measurements must be taken at periodic intervals over time. (Tr. at 293-94.) He further testified that it is important to have more than two data points from which to calculate the half life. (Id.) Plaintiffs provide no evidence that the half life calculation approach utilized by Professor Neuwirth on the TMI soil samples has been peer reviewed. Consequently, the court finds that this factor weighs in favor of the admission of the chemical extraction testimony and against the admission of the half life calculation testimony.

3. Is There a Known or Potential Rate of Error?

Professor Neuwirth testified to the rate of error for his soil extraction calculations. (Tr. at 1033.) Based upon the testimony, the court believes that this is a standard rate of error for what the court has already found to be a standard extraction procedure. This factor will not weigh against the admission of the extraction testimony. To the contrary, the court finds the potential rate of error with respect to the half life calculations and dose estimates to be high. As discussed above, Professor Neuwirth used only two data points when making his half life estimates. Consequently, the calculation does not accurately categorize the decay rate of the sample. See Allen v. United States, 588 F.Supp. 247, 276 (D.Utah 1984); (Tr. at 293-94.). Because the decay rate is not properly categorized, the potential rate of error in estimating the half life value is high. Moreover, any error in the half life calculation carries over into Professor Armentrout's dose estimates as the half life calculations were integral to the dose calculations. Thus, this factor will weigh against the admission of the half life calculation and dose estimate testimony.

4. Were There Standards Controlling the Techniques Operation?

"A well-designed experiment shows how one variable responds to changes in variables under the control of the experimenter." Federal Judicial Center, Reference Manual on Scientific Evidence 347 (1994) (hereinafter "Reference Manual"). The court believes that the Neuwirth soil analysis can be categorized as an observational study. The Reference Manual notes the following regarding observational studies:

In an experiment the investigators select certain units for treatment. In an observational study the investigators have no control over who or what receives the treatment.

. . . . .

In such observational studies, investigators may speak of control groups ... and of controlling for potentially confounding variables ... [h]owever ... the causal inferences that can be drawn from such analyses rest on a less secure foundation than that provided by a controlled randomized experiment.

Research Manual at 351. During the hearings, Professor Neuwirth acknowledged that he and Professor Armentrout had yet to obtain a suitable control group from the TMI area. This control group, when found, would be used to confirm test results and assure that their equipment was functioning properly. (Tr. at 1022-23.) Professor Neuwirth also indicated that he never compared his samples with soil samples taken at or near the time of the TMI accident to determine whether his data was accurate or whether Professor Armentrout's half life assumption was correct. (Tr. at 1045-46.) Professors Neuwirth and Armentrout have started with a less reliable methodology, and by their failure to find a suitable control group, placed the question of reliability into greater doubt. This factor will weigh against the admission of the proffered testimony.

5. Is the Methodology Generally Accepted?

Although Defendants argue that use of gamma spectrometry, rather than use of a sodium iodine detector, is the preferable means of identifying specific radionuclides, Defendants do not suggest that sodium iodine detectors are not generally accepted. (Defs.' Findings at 70, ¶ 8.) Indeed, Professor Neuwirth testified that his soil extraction techniques are generally accepted standard techniques. (Tr. at 1024-25.) The court finds that the technique utilized by Professor Neuwirth is generally accepted, and that the issue of whether it is the preferred technique goes to weight rather than admissibility. This factor will not weigh against the admission of the soil sample testimony.

The half life calculation methodology employed by Professor Armentrout does not enjoy the same level of acceptance as Professor Neuwirth's extraction methods. There is nothing before the court to indicate that it is generally accepted to make a broad half life calculation of a sample in lieu of delineating specific radionuclides within the sample and assigning half life values to each radionuclide. An expert's self-serving assertion that his methodology is standard of generally accepted will not withstand the rigors of a Daubert/Paoli II analysis. Daubert II, 43 F.3d at 1319 ("[P]laintiffs rely entirely on the experts' unadorned assertions that the methodology they employed comports with standard scientific procedures.... We've been presented with only the experts' qualifications, their conclusions and their assurances of reliability. Under Daubert that's not enough.") Accordingly, this factor will weigh against the admission of Professor Armentrout's half life calculation testimony.

6. Is There a Relationship Between the Technique and Methods That are Established to be Reliable?

The court finds this factor most applicable to Professor Armentrout's half life calculations. Dr. Moeller testified as to the proper and established techniques for calculating half lives of radionuclides.^[30] Specifically, Dr. Moeller noted as follows:

In order to determine the half-life of a radioactive material, you must count it at periodic intervals, counting it over, you know, a period of time, and certainly you have a number of points, and then you plot those points on a curve. And if you plot the points on semi-log paper where the activity is on a logarithmic scale and the time is on an arithmetic scale, the points will follow a straight line if, indeed, you have one radionuclide and you're following its half-life and the half-life is short enough so that it has decayed over the time period in which you're counting it. He used his first and his last points and drew a straight line between them, and although he had at least in one case, an intermediate point, he did not bother to plot it. It would not have plotted on the straight line. And on the basis of that, and assuming that he had one radionuclide, he estimated a half-life.

(Tr. at 293.) Professor Armentrout's methodology bears little relation to the established and reliable methodology outlined above by Dr. Moeller. The court finds that this factor weighs against the admission of Professor Armentrout's half life calculations.

Similarly, Professor Neuwirth's soil extraction techniques lack a key component evident in accepted and reliable methodologies — a suitable control sample. Without an appropriate control sample, it is impossible for Professors Armentrout and Neuwirth to verify their results in any meaningful way. This factor weighs against the admission of the soil sample analysis testimony.

7. Qualifications of the Expert Based Upon the Methodology

Defendants do not challenge the qualifications of either Professor Neuwirth or Professor Armentrout. As the court agrees that Professor Neuwirth is qualified to testify as to the soil sample analysis and that Professor Armentrout is qualified to testify as to his half life calculations, no further evaluation of this factor is necessary. This factor will not weigh against the admission of the proffered testimony.

8. Non-Judicial Uses of the Methodology

In a general sense, sodium iodine detectors are utilized and half life calculations are made in a variety of non-judicial settings. Indeed, but for the unique nature of this case, it is questionable whether these types of methodologies would ever have judicial uses. This factor will not weigh against the admission of the proffered testimony.

C. Conclusion

Based upon the foregoing analysis, the court will exclude Professor Armentrout's testimony on half life calculations and Professor Neuwirth's testimony regarding his soil sample analysis. The testimony is not "derived from scientific method", and does not represent "good science." Daubert, 509 U.S. at ___, ___, 113 S.Ct. at 2795, 2797.

VIII. JAMES E. GUNCKEL

Dr. Gunckel is a Ph.D. biologist now retired from a long-time position at Rutgers University, where he specialized in studying the effects of radiation on plants. A significant portion of Dr. Gunckel's work experience involves his collaborative studies with Dr. Sparrow and the Brookhaven National Laboratory regarding the effect of radiation on plants. Dr. Gunckel is not a health physicist, nor does he have expertise in the area of tree biology.

A. Proffered Testimony

Dr. Gunckel's report consists of his evaluation of damaged trees in the TMI area, study of human and animal health in the TMI area, and estimates of dose based upon the tree, human and animal studies. The Gunckel hypothesis is as follows:

Radiation induced growth effects in trees would occur in areas where residents experienced symptoms indicating exposures to radioactivity at the time of the TMI accident and that those tree effects would occur in several species showing relative sensitivities (slight and severe growth effects, and lethality) corresponding to those determined in Brookhaven.

(10/23/95 Gunckel Aff. at 12.) Based upon his interviews with humans who complained of health irregularities during and after the accident, Dr. Gunckel concluded that a dose of 300-360 rem^[31] was "the top dose at TMI that we could recognize by our yardsticks." (Tr. at 1606.) This 360 rem dose is equal to "LD-50/60" — the dose at which "50 percent of the population would die from that dose in 60 days." (Tr. at 1608.) Further, Dr. Gunckel indicated that he chose to study damaged trees located in the same areas as humans with health complaints to establish a method of comparing the erythema doses^[32] experienced by humans with the doses presumably experienced by the trees. (Id. at 1606-07.) Through his study, Dr. Gunckel has concluded that the dose to trees was in the area of 1000 rem. (5/13/93 Rpt. at 10.) Dr. Gunckel is unable to adequately explain how doses of 360 rems to humans and 1000 rems to trees were delivered without significant human casualties. (See Tr. at 1611.)

The court will now evaluate Dr. Gunckel's proffered testimony in light of the Daubert/Paoli II factors. The trees studies will be viewed separately from the human and animal studies. Additional facts related to each of these areas will be set forth where the court determines such explanation is necessary or helpful.

B. Daubert/Paoli II Analysis

1. Does the Method Consist of a Testable Hypothesis?

Dr. Gunckel has advanced a hypothesis capable of falsification. However, no effort has been made to verify either methodology or the conclusions reached.

a. Study of Human and Animal Health

In connection with this study, Dr. Gunckel interviewed approximately fifteen TMI residents regarding health effects they reportedly experienced at the time of the TMI accident. (Tr. at 1600-02.) During direct examination at the hearing, Dr. Gunckel described the nature and execution of his human health study as follows:

We went to these sites, and this was the case of revisiting the sites where people had complained of some sort of health effect. So we were there to do two things: First, to talk with them and to reaffirm whether their alleged effects were real, possible, or not, not trying to say as a doctor would be qualified to say that they had suffered from this, that and the other thing. We were simply asking, sort of trying to get a composite view of what had gone on. So we were not passing on identifying whether what they complained of was true or false. We were simply trying to associate people with possible episodic evidence with a plant indicator. That is, if we could — well, the ideal thing, of course, is if we had one of these indicators on every site and we could look at the indicator and see what the indication of damage was, and then going to our Brookhaven data bank, we could translate that effect into a dose. [Further we] ... found people weren't very well informed, frankly, about what had happened to them. So you really had to go armed with a list of possibilities and see if they could recall or whether they had had anything on their medical records to authenticate what they thought they had experienced.... Specifically, we had a lot of conditions like, oh, let me see. Nausea, vomiting ... petechiae.... low white blood counts.... abnormal tastes and smells at the time.

(Tr. at 1600-01.) Based upon his interviews,^[33] Dr. Gunckel concluded that five of the fifteen study subjects were exposed to erythemic doses.

In theory, Dr. Gunckel's methodology is testable. He could have examined the medical records of the study subjects to determine whether their health complaints were consistent with a pre-existing medical condition.^[34] Further, Dr. Gunckel could have made a differential diagnosis of each study subject to rule out potential causes of their symptoms other than radiation.^[35] Most important for this court's purposes, however, is that Dr. Gunckel made no effort to falsify and then modify his hypothesis. This factor will weigh against the admission of the proffered human/animal study testimony.

b. The Tree Study

After identifying areas in which persons described health effects allegedly related to the TMI accident, Dr. Gunckel looked for tree damage in these identified areas. At the hearing, Dr. Gunckel testified that he selected 80 damaged trees located in the fifteen areas where there was anecdotal evidence of human and animal health effects. (Tr. at 1608, 1610.) In describing how he selected the trees for his study, Dr. Gunckel noted the following:

[T]he kind of test I used, a very vague one, is the rule of thumb that most every observer uses on this sort of thing. If the tree, if the crown of the tree showed signs of thinning, in other words, you can see it better on a deciduous tree than you can on a pine tree, showed signs of thinning, in other words, it's losing its leaves, that's the first thing that comes down. And then if beyond that, you found that you had a color change, a color change is very significant in indicating that the tree is not terribly happy.

. . . . .

And I'll certify every one of those trees as having received a thousand roentgen dose, which was adequate to kill the tip of the plant, the leader, the apical meristem of the leader on the tree.

(Tr. at 1608-09.) Notably, Dr. Gunckel did not attempt any type of differential diagnosis with the trees to determine whether their injury could have been caused by something other than radiation. (Gunckel Dep. at 188.) Thus, the court is again presented with a hypothesis that could be tested but that has not been tested. This factor will weigh against the admission of the proffered tree study testimony.

2. Has the Methodology Been Subject to Peer Review?

a. Study of Human and Animal Health

Nothing before the court indicates that the human and animal health study methodology has been subject to peer review.

Peer review and publication do not, of course, guarantee that the conclusions reached are correct; much published scientific research is greeted with intense skepticism and is not borne out by further research. But the test under Daubert is not the correctness of the expert's conclusions but the soundness of his methodology. That the research is accepted for publication in a reputable scientific journal after being subjected to the usual rigors of peer review is a significant indication that it is taken seriously by other scientists, i.e., that it meets at least the minimal criteria of good science.

Daubert II, 43 F.3d at 1318. It is most curious to the court that in the sixteen years that this litigation has been pending, none of Plaintiffs' experts' reports, including this one, have been submitted for publication. This type of report, one purporting to show that persons and animals in the TMI area during the accident received erythemic doses, would surely be of interest to the relevant scientific community and the public at large. Yet, as Judge Kozinski of the Ninth Circuit observed of the Bendectin cases, "[i]t's as if there were a tacit understanding within the scientific community that what's going on here is not science at all, but litigation." Id. (footnote omitted). This factor will weigh against the admission of the human and animal study.

b. The Tree Study

Plaintiffs contend that all methodologies utilized in the tree study have been the subject of extensive peer review. This argument is based on Plaintiffs' characterization of the TMI tree study as being similar to studies performed by Dr. Gunckel and the late Dr. Sparrow at the Brookhaven National Laboratory. The court cannot agree with Plaintiffs' characterization insofar as the TMI tree studies bear no functional relevance to the Brookhaven studies.^[36] At Brookhaven, Drs. Gunckel and Sparrow irradiated trees with known doses of radiation using a gamma field. They then observed the effects exhibited by the trees following the known exposure. Finally, the doctors created a data base of the observations made and based upon the known dose exposures. To the contrary, the instant study involved looking at damaged trees and using the Brookhaven database to infer maximum possible dosage.^[37] Accordingly, the fact that the initial Brookhaven studies were subject to extensive peer review is of little relevance to the study presently before the court.

As with the human and animal study, the court finds it odd that Dr. Gunckel's tree study has never been submitted for publication. This factor will weigh against the admission of the proffered tree study testimony.

3. Is There a Known or Potential Rate of Error?

a. Study of Human and Animal Health

Dr. Gunckel does not assign any specific rate of error to his human and animal health study. The court, however, finds the potential rate of error to be high. According to Dr. Gunckel's in limine testimony, his opinion on dose at TMI is based primarily on the erythema allegedly experienced by five individuals he interviewed.^[38] Dr. Gunckel did not review medical records or perform differential diagnosis. Moreover, as the court will discuss infra, Dr. Gunckel does not have the expertise in human or animal medicine sufficient to draw conclusions regarding medical conditions. Dana Corp. v. American Standard Inc., 866 F.Supp. 1481, 1501 (N.D.Ind.1994) ("An expert's opinion ``must be an "expert" opinion (that is, an opinion informed by the witness's expertise) rather than simply an opinion broached by a purported expert.'" (citing United States v. Benson, 941 F.2d 598, 604 (7th Cir.1991)). Finally, Dr. Gunckel did not employ any type of systematic survey, or case-control comparison of the TMI residents. (See Tr. at 1600.) These flaws in the methodology evidence to the court a potentially high rate of error. This factor will weigh against the admission of the human and animal health study testimony.

b. The Tree Study

During his in limine testimony, Dr. Gunckel proclaimed the rate of error with respect to his tree study to be five percent (5%). (Tr. at 1623.) It is not entirely clear how Dr. Gunckel calculated this rate of error, but his explanation indicates that this is the rate of error assigned to the work conducted by Dr. Gunckel and Dr. Sparrow at Brookhaven. (See id. (witness's explanation).) As discussed by the court above, the instant tree study bears little methodological relationship to the Brookhaven studies. Therefore, the court cannot credit the rate of error assigned by Dr. Gunckel. Because no efforts were made to falsify the Gunckel hypothesis, the court finds that his conclusion can amount to nothing more than speculation and conjecture. As such, the potential for error is high. This factor will weigh against the admission of the tree study testimony.

4. Were There Standards Controlling the Techniques Operation?

a. Study of Human and Animal Health

Important standards that should have been controlling Dr. Gunckel's techniques were notably absent from his study of human and animal health. Specifically, Dr. Gunckel made no differential diagnosis, did not examine all relevant medical records, based significant portions of his opinion on anecdotal evidence and did not attempt to make any type of systematic survey of all TMI residents. This factor will weigh against the proffered testimony on the human and animal health study.

b. The Tree Study

The tree study is somewhat more controlled than the human and animal health study; however, it too appears to be missing important control standards. A question of fact has arisen as to whether Dr. Gunckel made a differential diagnosis with respect to the trees. His initial report and deposition indicate that he did not, while an affidavit filed prior to the in limine hearing and his hearing testimony indicate that he did perform such a diagnosis. The court will presume that Dr. Gunckel's testimony at the hearing was truthful. (See Tr. at 1608-09.) As such, the court finds that there were some standards controlling the operation of the tree study. This factor will neither weigh against the admission of the tree study testimony nor weigh strongly in favor of the admission of that testimony.

5. Is the Methodology Generally Accepted?

a. Study of Human and Animal Health

Nothing before the court indicates that the methodology employed by Dr. Gunckel in conducting the human and animal health study is generally accepted by the relevant medical communities. Generally accepted medical opinions are not based solely upon anecdotal evidence.^[39] The court has been presented with no other evidence aside from Dr. Gunckel's self-serving statements, that his human and animal study methodology is generally accepted. Accordingly, this factor will weigh against the admission of the proffered testimony.

b. The Tree Study

The court cannot dispute that the methodologies employed by Dr. Gunckel and Dr. Sparrow at Brookhaven are generally accepted. Unfortunately, the tree study conducted by Dr. Gunckel in conjunction with the instant litigation bears little relation to the Brookhaven studies. Nothing on the record supports the contention that the specific methodology employed by Dr. Gunckel in the instant tree study is generally accepted. Dr. Gunckel's assurances to this effect are not enough. This factor will weigh against the admission of the proffered testimony.

6. Is There a Relationship Between the Technique and Methods That are Established to be Reliable?

The court has briefly discussed this factor in the context of other factors mentioned above. Dr. Gunckel's methodologies in both the study of human and animal health and the tree study differ in important respects from methodologies established to be reliable. In the health study, Dr. Gunckel relies extensively on anecdotal information and fails to give a differential diagnosis.^[40] At the hearing he admitted to not reviewing any medical files (Tr. at 1645-46). Moreover, what the court finds most unreliable and unusual about Dr. Gunckel's medical testimony is that he is not a medical expert. Reliable methods allow qualified personnel to evaluate the data. In the tree study, Dr. Gunckel looks at damage in trees and extrapolates back to determine dose. This technique bears little relationship to his own pioneering standards of observing effects of known radiation doses through cellular and subcellular analysis. Based upon the foregoing, the court will weigh this factor against the admission of the proffered health and tree study testimony.

7. Qualifications of the Expert Based Upon the Methodology

a. Study of Human and Animal Health

Dr. Gunckel testified at his deposition that he is not a medical doctor, an epidemiologist, an immunologist, or a veterinarian. (Gunckel Dep. at 181-82.) He has not studied at all in the fields of human or veterinary medicine. (Id.) He has never diagnosed or treated people with acute radiation illness. (Id. at 169.) Moreover, his knowledge of animal and human health effects of radiation, by his own admission, was derived solely from literature reviewed in connection with his involvement in this litigation. (Id.) Although well-credentialed in his own field, Dr. Gunckel is not qualified to testify in the fields of medicine, epidemiology, immunology or veterinary medicine. Dr. Gunckel is therefore not qualified to give an expert opinion with respect to his human and animal health study.

b. The Tree Study

It is without question that Dr. Gunckel is a pioneer in the area of studying radiation effects on plants. Defendants only attack on Dr. Gunckel's qualifications in this area is that Dr. Gunckel's work was conducted on seedlings rather than on mature spruce trees. (Gunckel Dep. at 165.) This is the second instance where Dr. Gunckel's deposition testimony is contradictory. As before, the court will construe the facts most favorably to Plaintiffs and presume that he has studied the effects of radiation exposure on mature trees. (See Dep. at 162.) Based upon the foregoing, the court finds Dr. Gunckel qualified to testify as to his tree study. This factor will weigh in favor of the admission of the proffered testimony.

8. Non-Judicial Uses of the Methodology

The court finds no evidence that the novel approach taken by Dr. Gunckel with respect to his methodology for the health study enjoys any use beyond the scope of this litigation. Similarly, the court finds there to be little practical application of the tree study methodology outside the confines of this case. That being said, the court finds this factor to be of little relevance to its overall analysis. Part of the reason that these methods enjoy no use beyond this litigation is that, thankfully, nuclear accidents are an aberration rather than the norm. This factor will not weigh against the admission of any of the proffered testimony.

9. Other Factors

In its November 9, 1995 order, the court noted that the internal logical consistency of any hypothesis proffered would be a factor considered by the court when ruling on Defendants' motion in limine. Defendants contend that the dose that Dr. Gunckel establishes as the threshold for the observed tree damage is logically inconsistent with the doses he estimated for humans, with doses estimated by Plaintiffs' other experts, and to a large extent with the existence of human life. This criticism refers to Dr. Gunckel's claim that the observed tree damage resulted from a dose of approximately 1000 rem. Dr. Gunckel further testified that a dose of 360 rem would be fatal to fifty percent of the human population within sixty days ("LD50/60"). Thus, Dr. Gunckel's hypothesis regarding tree damage is logically inconsistent with the fact that the human mortality expected at a dose of 1000 rem did not occur. Dr. Gunckel attempts to explain the inconsistency by noting that the trees would have been outdoors during the entire accident and thus subject to a continuous dose of radiation. To the contrary, humans had the ability to go indoors and thereby lower the dose of radiation to which they were exposed. No testimony was offered to support the contention that trees could receive a dose equal to 1000 rem in the same area where persons staying indoors would receive a dose significantly less than 360 rem.

The court agrees with Defendants that Dr. Gunckel's hypothesis is logically inconsistent on this point. Standing alone, this factor would not militate in favor of excluding the proffered testimony. Taken together with the court's analysis of the Daubert/Paoli II factors, however, this factor adds to the factors already weighing in favor of the exclusion of the proffered testimony.

C. Conclusion

Based upon the foregoing analysis, the court finds that the Daubert/Paoli II factors weigh decisively in favor of the exclusion of all of Dr. Gunckel's proffered testimony. Application of the factors demonstrates that Dr. Gunckel's methodologies lack scientific validity and reliability pursuant to Rule 702. Moreover, Dr. Gunckel lacks the qualifications to opine as a medical expert with respect to his human and animal health study. Because Dr. Gunckel is unable to verify his conclusion that the damage to the trees occurred at the time of the TMI accident, rather than at some earlier or later date, the proffered evidence also lacks the requisite Rule 702 "fit" with the instant litigation. Specifically, a jury would be unable to infer from Dr. Gunckel's testimony that the tree damage he discusses was in fact caused by the TMI accident. Finally, the portion of Dr. Gunckel's hypothesis related to dose is logically inconsistent with the lack of human casualties in the areas where tree damage was noted. The court will exclude all of Dr. Gunckel's proffered testimony.

IX. VLADIMIR A. SHEVCHENKO

Professor Shevchenko holds a Ph.D in Biological sciences, and has particular expertise in the area of radiation genetics; specifically with respect to the cellular and subcellular effects of radiation on plants. (Shevchenko Dep. at 7-9.) Much of his expertise has been developed through his practical experience studying radiation effects in the Eastern Ural Radiation Belt region, at the site of the Kyshtym atomic weapons plant accident, at the site of the Chernobyl nuclear power plant accident, at the Semipalatinsk Polygon, and at the sites of nuclear experiments in the Alti Region. (7/6/94 Shevchenko Rpt. at 2-3.) Professor Shevchenko has filed two reports with the court. The first report, filed July 6, 1994, discusses the tree study performed by Professor Shevchenko on damaged trees in the TMI area. The second report, filed February 21, 1995, summarizes four studies conducted by other Russian scientists.^[41] On November 9, 1995, this court issued a preliminary order which, inter alia, prohibits any expert from testifying to the substance of a report conducted by a different expert. In re TMI, No. 88-1452 at 2 (M.D.Pa. November 9, 1995) (preliminary order granting limited portion of Defendants' motion in limine). Based upon the 11/9/95 order, Professor Shevchenko was prohibited from testifying as to the substance of the Kozubov, Popov and Tarasenko reports. See infra, § XII. The court did ultimately permit Professor Shevchenko to opine as to the Snigiryova report as the court found the substance of this report to be within Professor Shevchenko's area of expertise. As such, Professor Shevchenko was capable of interpreting the data utilized and results obtained by Dr. Snigiryova, thereby affording Defendants the opportunity for meaningful cross-examination.

The court will next evaluate the proffered Shevchenko testimony pursuant to the Daubert/Paoli II factors. Separate analyses will be conducted for both the 7/6/94 tree study report and the 3/15/95 cytogenetic study of Dr. Snigiryova. Additional facts will be incorporated into the court's analysis where necessary.

A. Daubert/Paoli II Factor Analysis

1. Does the Method Consist of a Testable Hypothesis?

a. 7/6/94 Tree Study Report

Professor Shevchenko visited the TMI area to study trees allegedly damaged during the TMI accident. While in the area, Professor Shevchenko was directed to the location of some of the damaged trees by Dr. Gunckel and Norman Aamodt.^[42] (Shevchenko Dep. at 62-67.) Professor Shevchenko claims that he also selected some of the trees on his own. (Id.) In his report, Professor Shevchenko opines that tree damage observed in the TMI area is consistent with observations he has made at the Russian accident sites that he has studied. According to Plaintiffs, Professor Shevchenko's hypothesis is that "the data reflected a release of radioactivity that was of a sufficient harmful dose to cause injury and radiation effects to trees, and ... [that] quantification of the amount of radioactivity released could be made from the data provided by the evaluation of the trees." (Pls.' Findings at 48.) This hypothesis is testable insofar as a scientist could examine the subject trees and determine whether there are other potential causes for the damage exhibited by the trees. A portion of the methodology, however, is not as readily testable.

Professor Shevchenko's methodology was described by Plaintiffs as follows:

The methodology consisted of visual observation of trees in the TMI area, comparing the types and kinds of damages of the trees in the TMI area to trees observed in Chernobyl and damages in the gamma fields and making a professional judgment as to whether the damages and effects manifested by the trees in the TMI area reflected radiation exposure and dose within a specific range.

(Id. at 48-49.) This portion of the hypothesis could only be tested by a person with the same type of background as Professor Shevchenko. Only someone who has been privy to the extensive on-site observation of nuclear accidents as Professor Shevchenko would be capable of judging whether radiation damage occurred based merely upon viewing a tree. Even with credentials and experience as impressive as Professor Shevchenko, the court questions the ability of any scientist to make a "radiation diagnosis" based upon little more than a visual comparison. Moreover, Plaintiffs have not identified evidence showing that Professor Shevchenko has developed a database of his tree observations, or that other scientists could access such a database if necessary.

The question as to whether the tree study methodology is testable is close. However, Professor Shevchenko did testify that he examined and ruled out all other potential causes of tree damage akin to the damage exhibited by the TMI trees. (Tr. at 1169-70.) This differential diagnosis, combined with Professor Shevchenko's vast experience lead the court to believe that his methodology is testable and that it was tested by Professor Shevchenko. Accordingly, this factor will not weigh against the admission of the tree study testimony.

b. 3/15/95 Snigiryova Cytogenetic Report

Through its order dated November 9, 1995, this court permitted Professor Shevchenko to testify to the substance of the Snigiryova cytogenetic report. The hypothesis with respect to this report was that the TMI accident resulted in exposures and dose to persons in the TMI area reflected by an elevated number of chromosomal abnormalities. Specifically, Dr. Shevchenko believed that if radiation levels were higher than naturally occurring background levels, this would be reflected in a higher than average rate of chromosomal abnormalities. The methodology employed by Professor Shevchenko to test his hypothesis involved analyzing blood samples taken from persons living in the TMI area and looking for chromosome aberrations.

The blood samples for the cytogenetic study were collected by Plaintiffs' attorneys and provided to Professor Shevchenko and Dr. Snigiryova. (Tr. at 1388.) The cytogenetic method of dose reconstruction involves the identification under the microscope and scoring of chromosome aberrations in circulating lymphocytes in the blood. (3/15/95 Snigiryova Rpt. at 1-2.) The particular type of chromosome aberrations that the study focused on are called dicentrics.^[43] (Id. at 6-7.) A highly elevated frequency of dicentrics in a study group is indicative of radiation exposure. (Id. at 23.)

The court finds the hypothesis to be testable in that anyone with access to the blood samples could perform a similar analysis to verify the hypothesis. This factor will weigh in favor of the admission of the proffered cytogenetic testimony.

2. Has the Methodology Been Subject to Peer Review?

a. 7/6/94 Tree Study Report

As with Dr. Gunckel, Plaintiffs contend that the published works of Dr. Gunckel and Dr. Sparrow evidence the peer review of the methodology employed by Professor Shevchenko. However, as the court discussed above, the methodology employed in examining trees in the TMI area bears little resemblance to the methodology peer reviewed in the Gunckel/Sparrow publications. Further, articles written about morphological damage observed in radiation-exposed trees, at best, establish a relation between exposure and effect, not that such observations are useful as a dose reconstruction method. The court cannot find evidence that the methodology of basing an opinion as to possible radiation exposure almost completely on visual inspection has been peer reviewed. Accordingly, this factor will weigh against the admission of the proffered testimony.

b. 3/15/95 Snigiryova Cytogenetic Report

Evidence has been presented indicating that the methodology employed by Dr. Snigiryova in the cytogenetic analysis is a standard methodology. (See Tr. at 343-401 (Testimony of Dr. Michael Bender).) The method has been subject to and withstood extensive peer review. (Tr. at 1171, 1406.) Defendants have raised an issue as to the propriety of the control group utilized by Dr. Snigiryova; however, as will be discussed infra, the court does not find this factor to substantially effect the validity of the methodology. This issue goes to weight rather than admissibility. Consequently, this factor will weigh in favor of the proffered cytogenetic testimony.

3. Is There a Known or Potential Rate of Error?

a. 7/6/94 Tree Study Report

No specific rate of error was assigned to the tree study. Professor Shevchenko, instead, merely testified that he had studied enough trees to be confident that his conclusions were reliable to a reasonable degree of scientific certainty. (Tr. at 758.) Because so much of the methodology depends solely upon Professor Shevchenko's comparative observations, an error in his judgment or recollection could significantly alter the conclusions reached. As such, it is impossible for the court to quantify this factor. Finding Professor Shevchenko's vast experience to lend credence to his confidence in his results, this court will not weigh this factor against the admission of the proffered testimony; however, due to the inherently non-precise nature of this methodology, this factor will neither be weighed in favor of the admission of the proffered testimony.

b. 3/15/95 Snigiryova Cytogenetic Report

Dose reconstruction through cytogenetic analysis is an accepted and accurate method of dose reconstruction. The accuracy of the results, however, decreases as time elapses between exposure and testing. Plaintiffs have failed to introduce examples of other dose reconstruction efforts using the dicentric method conducted after so much time has elapsed since the alleged exposure. The passage of more than one decade from the alleged time of exposure to the execution of the dicentric study increases the potential rate of error of the study. Further, questions about the appropriateness of the control group (e.g. whether the historical Russian control group adequately controls for background radiation of a TMI area study population) tend to increase the potential for an erroneous conclusion. The court, however, finds these factors to go more to the weight to be accorded the testimony than to its admissibility. Accordingly, this factor will not weigh against the admission of the proffered cytogenetic testimony.

4. Were There Standards Controlling the Technique's Operation?

a. 7/6/94 Tree Study Report

Plaintiffs contend that there were proper standards controlling Professor Shevchenko's tree study insofar as he studied a control group of trees in addition to the study trees. Further, Professor Shevchenko testified at the hearing, contrary to his report, that he did perform differential diagnosis to determine other possible causes of the tree damage. (Tr. at 1091-92.) The court finds that Professor Shevchenko did employ two of the most important standards to control the operation of his study. Accordingly, despite the absence of other types of controls, this factor will not weigh against the admission of the proffered tree study testimony.

b. 3/15/95 Snigiryova Cytogenetic Report

"Cytogenetic evaluation requires carefully controlled studies using analytical techniques not available in most laboratories."^[44] The methodology as performed by Drs. Shevchenko and Snigiryova did not employ several standards generally used to control the technique's operation. Most significantly, the FISH method was not employed. This is the only method recognized in the literature as useful for assessing dose when so much time has elapsed since the alleged exposure. Further, the study did not attempt to match the presence in controls of potential confounders, such as age, gender or smoking. The data was not partitioned in accordance with these criteria to confirm that a confounder did not explain the presence of high dicentric counts. The court finds that important standards necessary to control the technique's operation were absent from the cytogenetic study. Thus, this factor will weigh against the admission of the proffered cytogenetic testimony.

5. Is the Methodology Generally Accepted?

a. 7/6/94 Tree Study Report

In support of their contention that Professor Shevchenko's tree study methodology is generally accepted, Plaintiffs point to Professor Shevchenko's "Bibliography" which lists the hundreds of publications he has authored and co-authored. While the list is certainly impressive, it has little relevance to the issue of whether or not the methodology employed in the instant tree study is generally accepted. Most importantly, the court notes that while the majority of the listed publications are on the subject of the effects of radiation in plants, most of these publications deal with effects at the cellular and subcellular levels. Moreover, the court was unable to find any publication dealing with the comparative symptomology diagnosis employed by Professor Shevchenko in the instant study. Accordingly, the court finds that Professor Shevchenko's retrospective dose reconstruction through the observation of damage to tree tops is not a generally accepted method for reconstructing dose. This factor will weigh against the admission of the proffered tree study testimony.

b. 3/15/95 Snigiryova Cytogenetic Report

The parties do not disagree that dose reconstruction through cytogenetic analysis is a generally accepted method. Defendants question whether the instant study would fall within the generally accepted category due to the length of time that elapsed between the accident and the analysis of the study blood samples. The court finds that this issue goes more to weight than admissibility. As such, this factor will not weigh against the admission of the proffered cytogenetic testimony.

6. Is There a Relationship Between the Technique and Methods That are Established to be Reliable?

a. 7/6/94 Tree Study Report

The court is unable to find an established and reliable method for back-calculating dose from observation of damage to trees due to the apparent novelty of the method. Accordingly, this factor is not helpful to the court in evaluating the reliability of the proffered testimony. This factor will weigh neither in favor nor against the admission of the proffered tree study testimony.

b. 3/15/95 Snigiryova Cytogenetic Report

The instant report bears a close relationship to cytogenetic analysis methodologies considered to be established and reliable. The most striking dissimilarities appear to be with respect to the length of time that has elapsed since exposure, and with respect to the selection of the control group. As was discussed above, the court finds that these factors go to weight rather than admissibility. Thus, it is of little consequence that the instant methodology differs from established and reliable methodologies in these ways. This factor will militate in favor of the admission of the proffered cytogenetic testimony. However, the weight accorded the testimony may be diminished due to the aforementioned flaws.

7. Qualifications of the Expert Based Upon the Methodology

Defendants have not challenged the qualifications of Professor Shevchenko with respect to either the tree study or the cytogenetic analysis. Accordingly, this factor will weigh in favor of the admission of the proffered testimony.

8. Non-Judicial Uses of the Methodology

a. 7/6/94 Tree Study Report

For decades scientists such as Drs. Gunckel and Sparrow and Professor Shevchenko have been observing the effect of radiation on plants. In general, these studies have consisted of subjecting a plant to a known quantity of radiation and then observing the plant's reaction over time. Professor Shevchenko has also had the opportunity to observe the effect of radiation on trees based upon his visits to radiation exposed sites within the former Soviet Union. All of these studies and observations have been conducted in non-judicial settings. The observations made by Professor Shevchenko in the TMI area were made for the purposes of this litigation. It is unclear whether the technique employed by Dr. Shevchenko has been used extensively in non-judicial settings. The court, however, presumes that Dr. Shevchenko has employed this type of observational analysis when examining trees within exposed regions of the former Soviet Union. Due to the court's own uncertainty with respect to the importance of this factor to the instant analysis, the court will weigh this factor neither in favor nor against the admission of the proffered testimony.

b. 3/15/95 Snigiryova Cytogenetic Report

The parties appear to agree that dose reconstruction through cytogenetic analysis has extensive use outside the present judicial setting. Accordingly, this factor will weigh in favor of the admission of the proffered cytogenetic testimony.

9. Other Factors

Through its November 9, 1995 order, the court identified to the parties factors beyond the Daubert/Paoli II factors that it deemed to be particularly important to this case. These factors included falsifiability, logical consistency, consistency with accepted theories and precision. In re TMI, No. 88-1452 at 4 (M.D.Pa. November 9, 1995) (order explaining factors the court would consider in ruling on the motion in limine). Defendants have attacked Professor Shevchenko's proffered testimony with respect to a number of these factors.

First, Defendants contend that the Shevchenko tree study report is not logically consistent with Professor Shevchenko's testimony at the hearings. (Defs.' Findings at 139-40, ¶ 19.) The court agrees that some of the hearing testimony of Professor Shevchenko and others tends to contradict the Shevchenko hypothesis. These contradictions support the contention that Professor Shevchenko failed to subject his hypothesis to rigorous empirical testing.^[45] While these examples evidence a degree of logical inconsistency with respect to the tree study, the court finds that the inconsistency is not significant enough to militate in favor of the exclusion of the proffered testimony. This factor will not weigh against the admission of the tree study testimony.

Second, Defendants contend that Professor Shevchenko did not attempt to falsify his cytogenetic findings by making dicentric counts of individuals who were measured by whole body counts, or who lived near TLDs, even though he testified that both whole body counts and TLD dosimetry were used in the development of the cytogenetic dose response curves in his Russian work. (Shevchenko Dep. at 221; Tr. at 1373, 1379.) The court agrees that Defendants have pointed to a flaw in Professor Shevchenko's study design, however, the court cannot agree that this flaw dooms his proffered testimony in this area. The effect of his failure to falsify through comparison with TLDs and persons measured by whole body counts will go to weight rather than admissibility. Accordingly, this factor will not weigh against the admission of the proffered cytogenetic testimony.

Next, Defendants raise several issues which do not neatly fit within any of the categories discussed above. The court will now briefly touch upon those issues. First, Defendants raise an issue regarding the alleged time of death of the damaged tree tops. Professor Shevchenko testified that he used the dendrometric method to date the time of death of the tree tops. Nothing in Professor Shevchenko's report indicates that he relied upon this method to determine the time of death of the tree tops. However, as in previous sections, the court will presume that the witness testified truthfully at the hearing. Professor Shevchenko's statement is, however, contradicted by the report of Dr. Kozubov which indicates that the performance of the dendrometric analysis was based upon the assumption that the tree death occurred in approximately 1979. Accordingly, it is unclear how Professor Shevchenko determined when the tree tops died. It is evident that the time of death is related to determining the cause of the tree top death. If Professor Shevchenko merely presumed that the trees died within a few years of the TMI accident, his finding that radiation was the cause of death becomes less reliable. To the contrary, if Professor Shevchenko determined scientifically that the tree top death occurred within a few years of the TMI accident, it would strengthen his finding that radiation exposure was the cause of death. The court finds that this issue goes to the weight rather than the reliability of Professor Shevchenko's testimony. Accordingly, this factor will not weigh against the admission of the proffered testimony.

Second, with respect to the proffered cytogenetic testimony, Defendants raise an issue related to the propriety of the control group. Defendants contend that the cytogenetic analysis is flawed because Professor Shevchenko and Dr. Snigiryova did not employ a contemporaneous and geographically proximate control group. The effect of this, Defendants argue, is that it produces an erroneous background count for "normal" levels of dicentric chromosomes. Plaintiffs did, however, present testimony which showed that the historical Russian controls used by Professor Shevchenko and Dr. Snigiryova had background dicentric counts virtually identical to those of a control group measured at Hershey Medical Center in Hershey, Pennsylvania. (Tr. at 705.) The court finds that Plaintiffs have come forward with sufficient evidence to demonstrate that their control group, while not contemporaneous and geographically proximate, is adequate.

Finally, Defendants question the admissibility of certain testimony in light of Rule 26 of the Federal Rules of Civil Procedure. Defendants argue as follows:

Testimony about methodological steps that were not identified in an expert report violates the disclosure requirement of Rule 26. Federal Rule of Civil Procedure 26(2)(B), as amended in 1993, requires that testifying experts submit a written report which "shall contain a complete statement of all opinions to be expressed and the basis and reasons therefor." (emphasis added). Contrary to the assertion of plaintiffs' counsel during the hearing, the 1993 amendment did not merely recodify the status quo for expert discovery. The Advisory Committee Notes for the amended rule make clear that "a detailed and complete written report must be submitted." Fed.R.Civ.P. 26(a)(2)(B) advisory committee notes (emphasis added). The note further provides that "[t]he information disclosed under the former rule in answering interrogatories about the ``substance' of expert testimony was frequently so sketchy and vague that it rarely dispensed with the need to depose the expert and often was even of little help in preparing for a deposition of the witness. Revised Rule 37(c)(1) provides an incentive for full disclosure; namely, that a party will not ordinarily be permitted to use on direct examination any expert testimony not so disclosed. Id. (emphasis added).

(Defs.' Findings at 132-33, ¶ 2.) Defendants' are most troubled by Professor Shevchenko's hearing testimony that he did compare his study trees to a control area. No such use of a control group was mentioned in his report. While the court agrees with Defendants that Professor Shevchenko should not be permitted to introduce opinions not discussed in his report, the court finds the admission of this particular testimony to be harmless. Defendants have six months to prepare to cross examine Professor Shevchenko on the issue of the control group that he utilized. Further, they will have the benefit of his report, deposition and hearing testimony to assist them. Accordingly, the court finds the Defendants argument to be inapposite to this particular piece of testimony. This factor will not weigh against the admission of the proffered tree study testimony.

B. Conclusion

Based upon the foregoing, the court finds the proffered tree study and cytogenetic testimony to be admissible. Defendants have made strong arguments in favor of the exclusion of this testimony. The court found the majority of these arguments to be credible and well-placed. Heeding the "the ``liberal thrust' of the federal rules," however, the court finds that it cannot exclude the proffered testimony of Professor Shevchenko based upon Defendants arguments. Paoli II, 35 F.3d at 739 (citing Daubert, 509 U.S. at ___, 113 S.Ct. at 2794). Vital to the courts decision on this matter is Professor Shevchenko's experience in the field. Likely more than any other expert before the court, Professor Shevchenko has had extensive firsthand experience examining the effects of radiation exposure. Since the early 1960s, Professor Shevchenko has been involved almost exclusively in studying the aftermath of nuclear accidents and nuclear testing at Kyshtym, the Eastern Ural Radiation Belt Region, Chernobyl, Semipalatinsk Polygon, and the Altai Region. Thus, what his testimony may lack in rigid conformity to technical standards is amply counterbalanced by his extensive experience.

X. STEVEN B. WING

Steven Wing, a Ph.D epidemiologist,^[46] is currently a Professor of epidemiology at the University of North Carolina at Chapel Hill. (Tr. at 904.) He has filed two expert reports in this litigation. The first report, filed in January of 1994, focuses on Dr. Wing's analysis of mortality rates in the TMI area. Steven Wing, "Mortality Trends in Relation to the Accident at Three Mile Island," January, 1994 (hereinafter "1/94 Wing Rpt.")^[47] The second report, filed on February 25, 1995, is a reanalysis of the Susser/Hatch cancer incidence study. Steven Wing, "A Reanalysis of Cancer Incidence Near the Three Mile Island Nuclear Plant," February 25, 1995 (hereinafter "2/25/95 Wing Rpt.") Dr. Wing contends that the mortality and cancer data support the conclusion that persons in the TMI area were exposed to dangerous levels of radiation during the TMI accident. Defendants have challenged the admissibility of both of Dr. Wing's reports and testimony related thereto.

Before proceeding with its analysis of Dr. Wing's proffered testimony, the court notes that it will be guided in its inquiry by the following observations of the United States Court of Appeals for the Second Circuit:

By its nature, epidemiology is ill-suited to lead a factfinder toward definitive answers, dealing as it does in statistical probabilities and the continual possibility of confounding causal factors. In light of the inherent uncertainty shrouding issues of probabilistic causation, the decision of a district court on whether plaintiff's epidemiological evidence is sufficient to get to the jury should be guided by the well-established standards governing judgment as a matter of law — whether, viewed in the light most favorable to the non-moving party, "the evidence is such that, without weighing the credibility of the witness or otherwise considering the weight of the evidence, there can be but one conclusion as the verdict that reasonable [jurors] could have reached."

Applied to epidemiological studies, the question is not whether there is some dispute about the validity or force of a given study, but rather it would be unreasonable for a rational jury to rely on that study to find causation by a preponderance of the evidence.

Asbestos Litigation, 52 F.3d at 1133 (citations omitted). The court will now proceed with an analysis of Dr. Wing's testimony.

A. Dr. Wing's Mortality Study

1. Proffered Testimony

Dr. Wing embarked upon the mortality trend study at the request of Norman Aamodt, a consultant for Plaintiffs. (1/94 Wing Rpt. at 3.) According to Dr. Wing, Mr. Aamodt came to him with the results of a mortality trend analysis that he had performed.^[48] (Id.) According to the Aamodt analysis, mortality rates for 1980 were higher than historical averages. Mr. Aamodt suggested to Dr. Wing that the increase in mortality rates supported his theory that "high doses of radiation to small populations residing on elevated terrain in the path of noble gas plumes from the TMI accident lead to deaths of some older people who were already in poor health...." (Id. at 3.) After re-analyzing the data relied upon by Mr. Aamodt and reaching similar results, Dr. Wing agreed to perform a more extensive epidemiological study of mortality rates in the area.

Using a more sophisticated analysis than Mr. Aamodt, Dr. Wing was able to calculate age-adjusted death rates, and isolate by cause of death, for the time period in question. (Id. at 5.) Additionally, Dr. Wing included within his analysis a study of mortality trends in all Pennsylvania counties and nationwide during the 1970s and 1980s. Based upon his results, Dr. Wing concluded the following:

The major findings of this study that are suggestive of an effect of the TMI accident on mortality are:

1. an elevation of 1980 mortality in the five counties around TMI amounting to about 490 more deaths than would have been expected from 1979 rates;

2. an increase in cancer mortality among young children in the five counties in 1980-82;

3. an increase of infectious disease death rates in the five counties after the accident; and

4. a geographic clustering of areas with higher-than-expected 1980 mortality in parts of Pennsylvania and the Middle Atlantic and New England States.

These results, considered in the context of other studies and uncertainties about exposure estimates, may reflect mortality impacts of high-level radiation exposures from the 1979 accident at Three Mile Island.

(1/94 Wing Rpt. at 11 (emphasis added).)

2. Daubert/Paoli II Factor Analysis

Defendants chose not to analyze the mortality study through the Daubert/Paoli II factors based upon their belief that the study testimony should be excluded because it "draws no conclusion regarding accident doses based on that study, [and] it is [therefore] not helpful to the trier of fact." (Defs.' Findings at 172, ¶ 1.) The court's discussion of these factors will be drawn from the direct and cross examination of Dr. Wing. Defendants' challenge to the proffered testimony will be addressed following the court's review of the Daubert/Paoli II factors.

a. Does the Methodology Consist of a Testable Hypothesis?

At the in limine hearing, Dr. Wing stated his hypothesis as follows:

That high doses of radiation to small populations residing on elevated terrain in the path of noble gas plumes from the TMI accident led to deaths of some older people who are already in poor health, and that a high post-accident death rate in these small groups appears as a relatively small elevation of mortality in the five-county population as a whole.

(Tr. at 923.) The court finds this hypothesis to be testable. Adjusting his data for age, Dr. Wing was able to test and verify his hypothesis. (Tr. at 923.) This factor will weigh in favor of the admission of the proffered mortality study testimony.

b. Has the Methodology Been Subject to Peer Review?

Dr. Wing testified that the particular study at issue has not been subject to peer review. (Tr. at 924.) He added, however, that he has authored similar studies using some of the same methods and data which have been subject to peer review. (Id.) The court finds that the methodology of the TMI mortality study has been peer reviewed. Dr. Wing used basic and accepted epidemiological techniques to conduct the study. This factor will weigh in favor of the admission of the proffered testimony.

c. Is There a Known or Potential Rate of Error?

According to his hearing testimony, Dr. Wing did not apply statistical significance testing to the mortality study. (Id.) Dr. Wing implies that he has used another method to "assess issues of precision of the results." (Id. at 925.) The court is unable to ascertain the specifics of this method. Nevertheless, with respect to the limited and preliminary conclusion that this study presents (discussed infra), the court finds the rate of error to be moderately high. Accordingly, this factor will weigh only slightly in favor of the admission of the proffered mortality study testimony.

d. Were There Standards Controlling the Technique's Operation?

Because Dr. Wing's underlying methodology was elementary and followed standard procedures, the court finds that he employed basic standards to control the operation of the technique. (See Tr. at 925.) This factor will weigh in favor of the proffered testimony.

e. Is the Methodology Generally Accepted?

As the court has alluded to in its analysis of previous factors, the underlying methodology utilized by Dr. Wing is a standard and generally accepted epidemiological methodology. (Id.) This factor will weigh in favor of the admission of the proffered testimony.

f. Is There a Relationship Between the Technique and Other Methods Established to be Reliable?

Dr. Wing contends that "these [mortality] studies have been in the past used to address many questions about health effects of agents on populations." (Tr. at 926.) Insofar as the study may be viewed as a method of addressing questions about possible radiation health effects, rather than as a means of generating conclusions about dose, the court finds that it bears a close resemblance to methods established to be reliable. This factor will weigh in favor of the admission of the proffered mortality study testimony.

g. The Qualifications of the Expert Based Upon the Methodology

Defendants do not challenge Dr. Wings qualifications. Moreover, the court finds him amply qualified to opine regarding his mortality study. This factor will weigh in favor of the admission of the proffered testimony.

h. Non-Judicial Uses

It is without question that mortality studies are used on a regular basis in non-judicial settings within the field of epidemiology. This factor will weigh in favor of the proffered testimony.

i. Conclusion

The Daubert/Paoli II factors weigh heavily in favor of the admission of the proffered mortality study testimony.

3. Rule 702 Fit and Rule 403 Confusion

When evaluating whether the proffered testimony "fits" within the case, the court must determine whether the testimony will assist the fact finder in understanding a fact in issue. Paoli II, 35 F.3d at 743. Dr. Wing is proffered as an expert on "dose." Thus, his testimony should assist the trier of fact in determining or understanding something about the dose of radiation received by residents living in the TMI area during the accident. Dr. Wing's own characterization of his mortality study demonstrates the absence of "fit" with respect to the question of dose:

But I emphasize that there are no radiation doses that we have for this analysis, or relative dose estimates. So it's not an analysis of the association between dose and mortality. It's an investigation of an observation that was originally made in Vital Statistics data to understand who experienced this in terms of the age groups, what causes of death in the local area showed this, and which did not.

(Tr. at 920-21 (emphasis added).) Through his report, Dr. Wing can testify about certain changes observed in area mortality rates during the time period that he studied. His conclusions suggest that there may be some correlation between some observed changes in the mortality rates and the TMI accident. Dr. Wing cannot at this stage in his research make any direct correlation between the TMI accident and certain increased mortality trends. As such, the court finds that his testimony would not help the trier of fact understand any fact in issue. Moreover, with respect to Rule 403, the court finds that admission of this testimony would unnecessarily confuse the jury. The study, at this point, is suggestive yet incomplete. Admission of testimony regarding the study results may lend an air of conclusiveness to the study that is not rightfully deserved.

These factors weigh heavily against the admission of the proffered mortality study testimony.

4. Conclusion

The court's analysis of the Daubert/Paoli II factors weigh decisively in favor of admitting the proffered testimony, while the court's Rule 702 fit and Rule 403 confusion inquiries weigh strongly against the admission of the testimony. The court finds, and the Daubert/Paoli II analysis supports the finding, that as far as the mortality study has been taken it is methodologically sound. The study lacks any definitive conclusion tying dose to mortality rates. Consequently, the court views the study as a starting point, or preliminary analysis, meant to be supplemented by further study and testing. Were the study complete, it could prove helpful to the trier of fact. Yet, as the study presently exists, it is incapable of presenting any conclusion relevant to the instant litigation. Accordingly, the court will exclude the proffered mortality study testimony pursuant to Rule 702 as lacking the requisite "fit" with the instant litigation.

B. Dr. Wing's Cancer Incidence Study

1. The Proffered Testimony

Dr. Wing's cancer incidence study is a reanalysis of the data collected in connection with the study of cancer incidence in the area around TMI undertaken by Maureen Hatch and M. Susser of Columbia University. The Susser/Hatch study sought to test the hypothesis that the risk of cancer may have been raised by exposure to radiation from the accident. After dividing the TMI area into geographical study tracts, Drs. Susser and Hatch analyzed pre- and post-accident cancer incidence relative to dose within those study tracts. The Susser/Hatch study concluded that radiation exposure had not caused an increase in post-accident cancer incidence.

The Wing reanalysis of the Hatch data is based on the relationship between cancer incidence and accident exposures characterized by a single dose response line derived from the cancer incidence in all of the study tracts.^[49] The dose response line has a rising slope when increased dose corresponds to an increased incidence of cancer, a zero slope indicating no change in cancer incidence with increased dose, or a negative slope indicating a decline in incidence corresponding to an increase in dose. (2/5/95 Wing Rpt. at 6; Walker Rpt. at 3; Tr. at 857-58, 937-38.) Statistical analysis is used to derive the slope of the dose response line which best fits the cancer incidence data across all study tracts. Dr. Wing determined the difference between the dose response lines in the pre-accident period (1976-1979) and the post-accident period (1981-1985). (Tr. at 937.) The Wing study shows a small increase in cancer incidence for all types of cancer, and more significant increases in the incidence of lung cancer and leukemia. (2/5/95 Wing Rpt. at 11.) Based upon these findings, Dr. Wing concludes that the increases are "consistent with allegations that the magnitude of radiation doses from the TMI accident were much higher than was assumed in the past." (Id.)

2. Daubert/Paoli II Factor Analysis

a. Does the Methodology Consist of a Testable Hypothesis?

Dr. Wing's hypothesis for his cancer incidence study is that accident emissions from TMI were sufficiently large to have caused measurable increases in cancers in the six years following the accident. Defendants agree with Plaintiffs that, in theory, this hypothesis is testable. The court finds that the cancer incidence study methodology consists of a testable hypothesis. Accordingly, this factor will weigh in favor of admitting the proffered testimony.

b. Has the Methodology Been Subject to Peer Review?

The cancer incidence study has not been subject to peer review. Dr. Wing has, however, provided testimony that data reanalyses such as the one he performed on the Susser/Hatch cancer incidence data are common within the field of epidemiology. Aside from Dr. Wing's "unadorned assertions" that the methodologies have been subject to peer review, Plaintiffs have presented no evidence of this fact. Daubert II, 43 F.3d at 1319. Consequently, the court is left to question whether "[p]ersonal opinion, not science, is testifying here." Id. While the court found Dr. Wing to be credible, and has no basis to question his word, the court cannot ignore that Plaintiffs have failed to carry their evidentiary burden. Furthermore, the court finds that the existing legal precedent in this area clearly defines the burden that Plaintiffs must shoulder. This factor will weigh against the admission of the proffered testimony.

c. Is there a Known or Potential Rate of Error?

Plaintiffs contend that "the standard of error contained in the reanalysis is consistent with similar reanalyses that appear in the epidemiological literature." (Pls.' Findings at 35.) This statement is made without further explanation as to the precise rate of error that appears in "the epidemiological literature." Further, the court is given no citation to any of this literature. Tables 1-4, appended to the cancer incidence study identify the standard error in the calculations, yet no information is provided to assist the court in understanding these numbers.^[50] Defendants, to the contrary, argue that the results of Dr. Wing's study are "ascribable to chance as well as to any real association with accident emissions." (Wing Dep. at 270-73.) Specifically, Defendants contend that when confidence intervals are calculated^[51] for Dr. Wing's findings of an increased incidence of all cancers in the TMI population, they include values of a negative association between accident doses and the incidence of all cancers in the post-accident population. (Tr. at 971; Walker Rpt. at 5.) The same result is true with regard to the calculated increase in post-accident leukemia, (Tr. at 971; Wing Dep. at 270-73; Walker Rpt. at 5.); lung cancer was the exception, producing large error bands. (Wing Dep. at 270-73.) Defendants' experts Hoel and Walker testified that lung cancer has never been identified with radiation exposure as an isolated effect, and that one would not expect lung cancer to occur within the time frame of the Wing analysis. (Tr. at 793, 795, 800, 878-79.)

On a related topic, Defendants question the accuracy of the report in light of the alleged failure of Dr. Wing to account for an appropriate latency period with respect to his lung cancer analysis. Evidence was presented at trial that the average latency period for lung cancer is 15 years. (Tr. at 783-84.) Dr. Wing's study results presume a radically decreased latency period. The court agrees that the latency question decreases the credibility of Dr. Wing's findings. Standing alone, this inconsistency is not enough, however, to draw into question the accuracy of all of the study findings.

The court is unwilling to accept Dr. Wing's bald assertion that the potential rate of error is standard, especially in light of evidence produced by Defendants tending to show a potentially high rate of error. Further, the court is cognizant that part of the difficulty in resolving this factor is due to the court's own uncertainty regarding Dr. Wing's standard error calculations. Thus, the court will, for the moment, defer ruling on this factor. If upon completing the remainder of its analysis, the court finds that the outcome of this factor is decisive as to the question of admissibility of the study testimony, the court will order Plaintiffs to provide a more direct and substantiated discussion of the error issue.

d. Were There Standards Controlling the Technique's Operation?

A significant portion of Dr. Wing's report is devoted to discussing how he attempted to rule out factors that could serve to bias the study results. Further, a separate affidavit is provided explaining the care with which Dr. Wing and his associates reviewed the R-base data. At the hearings Dr. Wing testified that "[w]e were very careful with our work in documenting all our data processing. And, in fact, I believe it's because of our care and the thoroughness with which we approached the data that we received, that we discovered the errors and the mistakes in the previous [Susser/Hatch] report."

Despite this, Defendants point to several standards that Dr. Wing failed to utilize. Defendants contend that Dr. Wing failed to consider the biological plausibility of the results, failed to test the risk estimates derived from his statistical analysis against actual trends in cancer rates for the types of cancers known to be radiogenic, and to consider whether there were changes in proportions of such cancers in the population. Additionally, Defendants argue that Dr. Wing failed to consider whether actual mortality patterns in the population confirm the conclusion that there exist increased incidences of cancer in the population attributable to radiation exposure, and that he failed to evaluate the statistical evidence "in the context of the sensitivity of the findings to changes in assumptions and data, coherence of the evidence with other knowledge, the magnitude of associations, their consistency across groups, and temporal relationships of exposure and effect." (Tr. at 877.)

The court finds that although he did not employ all possible standards, there were some standards controlling the operation of Dr. Wing's techniques. This factor will not weigh against the admission of the proffered testimony.

e. Is the Methodology Generally Accepted?

Plaintiffs argue that Dr. Wing has performed a generic epidemiologic re-analysis, and that such re-analyses are common place and generally accepted. Defendants claim that while the statistical technique employed by Wing is generally accepted in the scientific community, his methodological approach regarding the data subject to that statistical analysis and the conclusions to be drawn from that analysis are not. Specifically, Defendants take issue with the methodology because it produces conclusions at odds with what is generally known and accepted about cancer latency periods. In theory, such a finding could prove that the methodology leading to the conclusion is necessarily flawed, or it could prove that previous assumptions regarding latency periods were incorrect. The court finds that this factor goes to the weight of the testimony rather than its admissibility. Only the lung cancer outcomes contradicted standard latency period assumptions. Thus, the leukemia and "all cancers" results could still prove helpful and relevant to a jury. This factor will not weigh against the admission of the proffered testimony.

f. What is the Relationship of the Technique to Methods Established to be Reliable?

In many ways Wing's cancer incidence study closely resembles a standard and reliable epidemiological reanalysis. Yet, in one important way it does not. Dr. Wing's reanalysis produces no conclusive findings. When testifying at the in limine hearings Dr. Wing indicated that he had chosen carefully the wording of the conclusion to his report. (Tr. at 956.) Specifically, Dr. Wing's findings are nothing more than suggestive of an increase in cancer rates consistent with radiation exposure. Dr. Wing could not go so far as stating that his reanalysis suggested a dose response relationship. Thus, where an established and reliable methodology would be carried to the point of producing some form of conclusive findings, Dr. Wing's methodology does not. Accordingly, this factor weighs against the admission of the proffered testimony.

g. What are the Qualifications of the Expert With Respect to the Methodology?

Defendants have not challenged the qualifications of Dr. Wing. As such, and because the court finds him to be well credentialed, this factor will weigh in favor of the admission of the proffered testimony.

h. Non-Judicial Uses

Epidemiological re-analyses of data are performed on a daily basis in a variety of non-judicial settings. Although the court has found that Dr. Wing's re-analysis varied from standard re-analyses in important ways, those variations are not material to this factor. When viewed as the starting point for a more complete epidemiological re-analysis, it is clear that the Wing methodology has non-judicial uses. This factor will weigh in favor of the proffered testimony.

i. Conclusion

The Daubert/Paoli II analysis militates slightly in favor of admitting the proffered testimony. However, the court has deferred ruling on one of the more important factors, that of the known or potential rate of error. Accordingly, should the testimony survive Defendants' Rule 702 "fit" challenge, the court will direct Plaintiffs to submit a more focused and explanatory discussion of the known or potential rate of error for the Wing cancer incidence re-analysis.

3. Rule 702 "Fit"

Defendants challenge the Wing study on the ground that it does not "fit" within the litigation. Specifically, Defendants argue that the study has no bearing on Plaintiffs' central thesis that the accident resulted in a narrowly dispersed plume that caused highly localized segments of the TMI population to be exposed to large amounts of radiation. This argument is based upon the fact that Dr. Wing did not utilize Plaintiffs' dose pattern estimates when conducting his study. Rather, Dr. Wing calculated dose in a manner more closely aligned with the original Susser/Hatch study. The court agrees that in some ways the Wing study does not "fit." The court's rationale is, however, based upon reasons other than those mentioned by Defendants. That certain of Dr. Wing's findings are suggestive of increased cancer incidence consistent with radiation exposure is relevant and important. However, the strength of this finding is somewhat diminished by Dr. Wing's failure to explain the discrepancy between expected latency periods and his lung cancer findings.

Nevertheless, the court finds that a portion of the Wing cancer incidence reanalysis is relevant to a fact in issue, namely with respect to the dose of radiation received by Plaintiffs. The findings regarding leukemia and "all cancers" are suggestive of an increase. Such a finding could assist a jury in evaluating a fact in issue — the precise dose of radiation to which Plaintiffs were exposed. As such, the court will exclude only the study testimony related to the lung cancer findings as lacking the requisite "fit" with the captioned action.

4. Conclusion

Based upon the foregoing analysis, the court will exclude the portions of the Wing report dealing with lung cancer as lacking the requisite Rule 702 "fit." The court will defer ruling on the remainder of the proffered testimony until it is in receipt of Plaintiffs' supplemental discussion of the rate of error issue. Defendants will be entitled to respond briefly to Plaintiffs' supplemental filing should they so desire.

XI. DOUGLAS CRAWFORD-BROWN

A. Proffered Testimony

Dr. Crawford-Brown, presently a full professor at the University of North Carolina, holds a Ph.D. in nuclear science and health physics. (Tr. at 1676-77.) Plaintiffs have proffered Dr. Crawford-Brown as an expert in exposure assessment. Dr. Crawford-Brown testified that exposure assessment is one of the distinct stages of a risk assessment. (Tr. at 1679.) The following definition of risk assessment was provided by Dr. Crawford-Brown:

The formal definition [of risk assessment] that's required in the red book, or the guide for the federal government, Conducting of Risk Assessment, is that it's a process by which you first determine whether a substance or a risk agent is able to produce effects, and then you determine the amount of exposure or the amount of the substance that's in the environment, and the degree to which people are connected to it, and then you determine the relationship between exposure and the probability and severity of effects. And then you summarize that as an estimate of the probability and severity of effects and uncertainty distribution in a distribution of variability in the population.

(Tr. at 1678.) Plaintiffs have defined Dr. Crawford-Brown's role in the litigation as that of reviewer of the totality of the evidence and offerer of a final opinion as to whether Plaintiffs' experts, as a whole, considered the types of evidence necessary to establish what dose or doses of radioactivity were delivered to the residents of the TMI area as a result of the TMI accident. Based upon the "separate considerations of evidence" described in his Affidavit, and "conditional on establishing that the effects noted were in most likelihood caused by radiation exposures," it is Dr. Crawford-Brown's opinion that the asserted effects "are consistent with the claim that individuals located in the vicinity of the points of biological dosimetry" considered by him "received dose equivalents in excess of 100 rem." (2/27/95 Crawford-Brown Aff. at 18.)

Of particular concern to the court throughout this in limine process was the issue of how Dr. Crawford-Brown's final opinion might be altered should the court find any or all of the evidence underlying his opinion to be inadmissible. During the pre-hearing conference, the court indicated to Plaintiffs that it "would like to find out what ... [Dr. Crawford-Brown's] perception is on some of the findings he has relied on, how much confidence he has in them." (Tr. at 57.) Moreover, the court stressed that it "would like to know how his opinion in any way might be altered in light of what he may think about this material." (Tr. at 58.) As the prior sections of this memorandum indicate, a significant portion of Plaintiffs' proffered testimony will be excluded. Consequently, the court will evaluate the proffered testimony of Dr. Crawford-Brown in light of the decisions it has already made regarding Plaintiffs' other experts on dose.

B. Daubert/Paoli II Analysis

1. Does the Methodology Consist of a Testable Hypothesis?

When questioned on this topic during the hearing, Dr. Crawford-Brown testified that his methodology did not involve a testable hypothesis. (Tr. at 1714.) Plaintiffs opine that due to the nature of Dr. Crawford-Brown's testimony, this issue is irrelevant. The court is less eager to offhandedly discard what the Supreme Court has identified as a "key" factor in the Daubert analysis. Daubert, 509 U.S. at ___, 113 S.Ct. at 2796. Instead, the court finds the absence of a testable hypothesis to underscore the fundamental flaws in the Crawford-Brown methodology. As has been a recurring problem with all of Plaintiffs' experts, Dr. Crawford-Brown's testimony is rooted in an unproven assumption. Specifically, Dr. Crawford-Brown has presumed that all of the effects alleged by Plaintiffs (e.g. chromosome dicentrics, tree damage, human and animal health effects) were caused by radiation. It appears, at least in theory, that Plaintiffs assumed that each of their experts would succeed in validating the overall thesis that Plaintiffs' alleged injuries were caused by radiation emitted from TMI during the accident. Thus, Plaintiffs fashioned their expert testimony as a precarious house of cards with each expert relying on conclusions which other experts were responsible for generating and verifying. Because of this, the failure of even one of Plaintiffs' experts to produce a verifiable conclusion necessarily harms the conclusions of other of Plaintiffs' experts. Dr. Crawford-Brown provides the quintessential example of this perilous approach.

Dr. Crawford-Brown's testimony could have been based upon a testable hypothesis. Rather than simply presume that radiation was the cause of the effects noted by Plaintiffs, Dr. Crawford-Brown could have performed a full exposure assessment analysis, including an evaluation of the relative strength or weakness of each of the strands of evidence (e.g. biological dosimetry data) available to him.^[52] Instead, he chose to rely blindly upon the conclusions generated by Plaintiffs other experts. Thus, contrary to Plaintiffs' argument, the court finds that Dr. Crawford-Brown's testimony does not by it's nature prevent the formation of a testable hypothesis. Rather, Dr. Crawford-Brown chose to design his methodology so as not to consist of a testable hypothesis. Consequently, this factor will weigh against the admission of the proffered testimony.

2. Has the Methodology Been Subject to Peer Review?

The methodology utilized by Dr. Crawford-Brown loosely resembles a traditional exposure assessment methodology. Insofar as this resemblance exists, the methodology has been subject to peer review. Dr. Crawford-Brown's methodology differs from the traditional model, however, in that he does not evaluate the relative strength of each of the strands of evidence upon which he relies. This twist on the traditional exposure assessment methodology has not been subject to peer review. Accordingly, this factor will weigh against the admission of the proffered testimony.

3. Is There a Known or Potential Rate of Error?

The rate of error of an exposure assessment is intimately connected to the rate of error of each of the pieces of evidence upon which the assessment relies. Consequently, the potential rate of error in Dr. Crawford-Brown's assessment is tied to the respective rates of error of Plaintiffs' other experts. The court, above, has found those rates of error to be uniformly high, with the exception of the Shevchenko testimony. Because Dr. Crawford-Brown's conclusion presumes the accuracy of the other experts' conclusions, and because the court has determined those experts' methodologies to have potentially high rates of error, the court finds the potential rate of error of the Crawford-Brown methodology to be high. This factor will weigh against the admission of the proffered testimony.

4. Were There Standards Controlling the Technique's Operation?

Inasmuch as the experts' upon which Dr. Crawford-Brown relies had standards controlling the operation of their techniques, Dr. Crawford-Brown had standards controlling his. That being said, the court has already highlighted the important standard absent from the Crawford-Brown technique — the weighing of the relative strengths and weaknesses of the evidence he considered.

5. Is the Methodology Generally Accepted?

Exposure assessment is a generally accepted and helpful methodology. It is not a generally accepted practice to discard a key step in the standard methodology. Accordingly, because Dr. Crawford-Brown did not follow the standard methodology, the court cannot find the methodology he did employ to be generally accepted. This factor will weigh against the admission of the proffered testimony.

6. What is the Relationship of the Technique to Methods That are Established to be Reliable?

The comments that the court has made above speak to this factor. Dr. Crawford-Brown's exposure assessment bears some similarity to established and reliable methods. However, he also fails to employ some techniques that would be expected in a traditional exposure assessment. This factor will weigh against the admission of the proffered testimony.

7. The Qualifications of the Expert Based Upon the Methodology

Defendants have not challenged Dr. Crawford-Brown's qualifications. As the court finds him well qualified, this factor will weigh in favor of the admission of the proffered testimony.

8. Non-Judicial Uses

There is no question that exposure assessment is applied with frequency in a variety of non-judicial settings. The version of exposure assessment applied by Dr. Crawford-Brown in this litigation, however, appears specifically tailored to the legal setting. Dr. Crawford-Brown was presented with a set of conclusions and supporting evidence which he was directed to credit as being true. Based upon these assumptions, he was then asked to state whether the assumptions were consistent with the thesis that the TMI population was exposed to unsafe levels of radiation during the TMI accident. With the caveat that he assumed it could be proven that radiation was the cause of the observed effects, Dr. Crawford-Brown conducted a limited exposure assessment which revealed the evidence to be consistent with Plaintiffs' thesis. The court cannot imagine how such a methodology would have any practical use in a non-judicial setting where the objective would seem to be seeking answers to difficult questions, rather than seeking to verify a thesis via circular reasoning.^[53] This factor will weigh against the admission of the proffered testimony.

9. Conclusion

The Daubert/Paoli II analysis weighs decisively in favor of excluding the proffered testimony. As the United States District Court for the Northern District of Indiana noted in Dana Corp.:

[Plaintiff's expert] based his opinions on his understanding of what various depositions reported, and he may do that under Rule 703. But an expert's opinion that is without foundation is worth nothing as evidence, and is so irrelevant for purposes of Rule 402 and not reliable for purposes of Rule 702 as construed by Daubert. In the instances in which the record provides no such foundation, ... [the expert's] opinion adds nothing to the question of [whether plaintiffs were exposed to radiation as a result of the TMI accident].

The proffered testimony of Dr. Crawford-Brown will be excluded as unreliable pursuant to Rule 702 and irrelevant for purposes of Rule 402.

XII. BRUCE MOLHOLT

A. Proffered Testimony

Bruce Molholt has filed several reports with the court. His 4/8/93 report setting forth his opinion that depressed lymphocyte counts could be used to back-calculate a radiation exposure dose was the subject of Defendants' first motion in limine on dose issues. In fact, it was through the briefing of that motion in limine that the court determined that Daubert hearings were necessary. A second report, dated March 13, 1995, purported to perform a risk assessment on the TMI test Plaintiffs to determine whether the cancers they developed following the TMI accident were more likely than not due to radiation exposure as a result of the accident.

(3/13/95 Molholt Rpt. 3.) Finally, in connection with the court's request for supplemental briefing on Defendants' original motion in limine, Plaintiffs submitted a final Molholt report, dated May 1, 1995. It was in this final report that Molholt articulated his "corollary hypotheses" as follows:

Furthermore, as corollary hypotheses, if radionuclides were released during the TMI accident sufficient to expose nearby persons to ... [greater than or equal to] 100 rems, then there should have been additional multiple phenomena observable in the surrounding human population, flora and fauna both immediately and long-term. The expected sequelae from exposure to ... [greater than or equal to] 100 rem include the following additional ten phenomena which expand the data set from which to reconstruct radiation dosimetry:

* Erythema, eye irritation, nausea, vomiting and hair loss,

* Metallic and iodine tastes and smells,

* Radioiodines in excess in local fauna,

* Increased neonatal hypothyroidism from radioiodine emissions,

* Increased incidence of adult thyroid diseases,

* Increased infectious diseases in exposed persons,

* Increased infectious diseases in exposed animals,

* Increased incidence of autoimmune diseases,

* Increased incidence of cancer, and

* Radiation damage to trees.

Supporting relevant scientific data for these ten additional corollary hypotheses are summarized below. Taken together with fulfillment of the two predictions of the first hypothesis, fulfillment of the predictions afforded by these additional ten hypotheses comprise a very convincing biodosimetry data set that radiation exposures exceeded doses of 100 rem during the TMI accident.

(5/1/95 Molholt Rpt. at 4-5.)

At the in limine hearing, Plaintiffs withdrew the proffered immunosuppression testimony of Dr. Molholt. (Tr. at 681.) Plaintiffs made this withdrawal with the caveat that they would not withdraw Dr. Molholt's opinion as to his "corollary hypotheses" to the extent they "addressed the relevance of other indicators of different levels of exposure." (Id. at 684.) Defendants now challenge the "corollary hypotheses" on the ground that they are simply naked assertions of the alleged conclusions of analyses performed by others, and as such, lack any indicia of scientific reliability. Plaintiffs first submit that the discussion of the corollary hypotheses testimony should be continued to the February, 1996, in limine hearings on medical causation. Further, they argue that such postponement is logical insofar as the corollary hypotheses are intertwined with the toxicological aspect of Dr. Molholt's causation report. The court is not entirely convinced of Plaintiffs' argument, but it is willing to give them the benefit of the doubt in reserving ruling on this issue until the conclusion of the February in limine hearings. In reserving judgment on this issue, the court expects that Dr. Molholt will testify at the hearings. Should Plaintiffs fail to produce Dr. Molholt, the court reserves the right to exclude his testimony on that basis alone.

That being said, the court has serious questions regarding the admissibility of the corollary hypotheses now that Plaintiffs have withdrawn the immunosuppression hypothesis. Dr. Molholt states in his report that the corollary hypotheses are meant to be considered in conjunction with the immunosuppression hypothesis. Accordingly, Dr. Molholt will need to offer convincing testimony, based upon his reports that are already of record, that the corollary hypotheses can stand alone. Moreover, Plaintiffs are hereby placed on notice that with respect to Dr. Molholt, the court finds the issues raised in Defendants findings of fact and conclusions of law to be extremely persuasive. To convince this court of the reliability of his testimony, Dr. Molholt, and any other expert that testifies in support of his reports, will have to directly and succinctly rebut the challenges made and flaws exposed in Defendants' findings.

Based upon the foregoing, the court will defer ruling on the admissibility of the Molholt corollary hypothesis pending his participation in the upcoming in limine hearings on medical causation.

XIII. RATIONALE BEHIND THE PRELIMINARY ORDER DATED 11/9/95

On November 9, 1995, immediately following a lengthy prehearing conference, this court issued a preliminary order granting limited portions of Defendants' motion in limine. Due to the time pressures attendant to issuing that order, the court deferred issuing a memorandum of law until a later date. The court is now prepared to explain the rationale supporting its November 9 order.

A. Untimely Filing of Expert Reports on Dose

The testimony of experts named in ¶ 1 of the 11/9/95 order were excluded based upon the untimely filing of their expert reports. Through an order dated November 13, 1994, this court set the deadline for the filing of Plaintiffs' expert reports as March 1, 1995. As the parties are aware, case management has been a recurring problem in this litigation. Exasperated by the parties consistent failure to abide by any case management schedule, the court, with the agreement of the parties, imposed a rigid case management schedule aimed at seeing this case to trial in June of 1996.^[54] The court has repeatedly stressed its unwillingness to depart from this schedule absent the most compelling circumstances. See, e.g., Order dated 11/13/94 ("All deadlines stated in the schedule are binding deadlines. No changes in deadlines, nor extensions of time will be grated absent extreme and compelling circumstances."); Order dated 4/28/95 ("[T]he deadlines set by the November 2, 1994 joint case management schedule are binding.").

Rule 26 of the Federal Rules of Civil Procedure provides the following with respect to the filing of expert reports:

The report shall contain a complete statement of all opinions to be expressed and the basis and reasons therefor; the data or other information considered by the witness in forming the opinions ... the qualifications of the witness, including a list of all publications authored by the witness within the preceding ten years; the compensation to be paid for the study and testimony; and a listing of any other cases in which the witness has testified as an expert at trial or by deposition within the preceding four years.

(C) These disclosures shall be made at the times and in the sequence directed by the court.

Fed.R.Civ.P. 26(a)(2)(B) and (C). This court directed the disclosure of Plaintiffs' expert reports on dose to occur on March 1, 1995. On the eve of the in limine hearing, some eight months later and without an explanation, Plaintiffs attempted to introduce new reports for a total of ten experts. Moreover, the bulk of these reports were to be delivered by previously unidentified experts. In Daubert, the Supreme Court drew an important line in the sand in reference to the evolving nature of scientific testimony:

Scientific conclusions are subject to perpetual revision. Law, on the other hand, must resolve disputes finally and quickly.... We recognize that in practice, a gatekeeping role for the judge, no matter how flexible, inevitably on occasion will prevent the jury from learning of authentic insights and innovations.

Daubert, 509 U.S. at ___, 113 S.Ct. at 2798; see Margaret A. Berger, Procedural Paradigms for Applying the Daubert Test, 78 Minn.L.Rev. 1345, 1372-73 (1994) ("[C]ourts should not allow the ... [proponent of the expert testimony] to rely on affidavits from expert witnesses whose identity and reports were disclosed, but who now wish to raise new issues about which the ... [opposing party] never had notice.")

This court, heeding the Supreme Court's observation, must draw its own line in the sand. Sempier v. Johnson & Higgins, 45 F.3d 724, 734 (3d Cir.1995) ("Under Federal Rules of Civil Procedure and our jurisprudence, district courts have broad discretion to manage discovery."), cert. denied ___ U.S. ___, 115 S.Ct. 2611, 132 L.Ed.2d 854 (1995). For litigation purposes, the facts of this case must at some point become complete. Whether the facts, for other non-litigation purposes, continue to evolve is necessarily of little matter to the court. A decision which has the finality of a legal decision must be based upon a discrete and fixed grouping of facts. This court ordered expert reports on dose to be filed by March 1, 1995. Thus, for purposes of this litigation, the discrete body of factual evidence with respect to the issue of dose became fixed on March 1.^[55] Accordingly, attempts to build on this body of facts after the filing deadline passed were misplaced. Consequently, the court granted Defendants' motion in limine with respect to all expert reports filed after the deadline for filing such reports.

B. Experts Testifying Regarding the Reports of Others

In ¶ 2 of its November 9 order, the court ruled that no expert could testify as to the substance of the report of a non-testifying expert.^[56] Plaintiffs had envisioned that Professor Armentrout would testify as to Professor Neuwirth's soil sample analysis and that Professor Shevchenko would testify as to a barrage of studies conducted by six other Russian scientists. The court found several problems with Plaintiffs' proposed approach. First, neither Professor Armentrout nor Professor Shevchenko were qualified to testify as experts in the fields the subject of the secondary reports. In fact, Professor Armentrout had enlisted the assistance of Professor Neuwirth because he was unable to perform the soil extraction technique on his own. This lack of expertise fed directly into a second problem. Because the testifying experts were not in fact experts, they could not speak intelligently as to the methodology and scientific underpinnings of the secondary reports. This, in turn, frustrated Defendants' ability to effectively cross-examine as to the secondary reports.

Based upon the foregoing, a reasonable compromise was made. All of the secondary reports had been filed in conjunction with either the Armentrout or Shevchenko reports. As such, there was no element of surprise or lack of notice, because Defendants were in possession of these reports. The court ruled that it would permit testimony on the secondary reports, but only if that testimony was provided by the author of the report. Thus, insofar as Plaintiffs were able to produce these "secondary" experts for the in limine hearings, the "secondary" reports were admissible. Execution of the court's order was frustrated by the federal budgeting impasse ongoing during a portion of the in limine hearings. The partial government shutdown attendant to the impasse included the closing of all passport offices. As a consequence, Dr. Tarasenko was unable to obtain a visa to attend the hearings. The court and the parties agreed that Dr. Tarasenko would be made available as soon as she could obtain a visa.

XIV. CONCLUSION

Based upon the preceding analyses, the court will grant in part Defendants' motion in limine and exclude the proffered testimony of Richard Webb, David A. Lochbaum, Charles E. Armentrout, Victor Neuwirth, James E. Gunckel, Douglas Crawford-Brown, Steven Wing with respect to his mortality study, and a portion of the testimony of Ignaz Vergeiner. The court will deny the motion in part and admit the testimony of Vladimir I. Shevchenko. Finally, the court will defer its ruling as to: the proffered testimony of Dr. Tarasenko; the new Lochbaum report; the admissibility of the Wing cancer incidence study pending further briefing on the rate of error issue; and, will defer its ruling as to the Molholt corollary hypotheses proffer until the conclusion of the February in limine hearings, provided that Plaintiffs produce Dr. Molholt to testify at those hearings. The court will issue its ruling with respect to the Kozubov proffer on January 8, 1995.

With respect to other pre-trial matters, the court is aware of outstanding issues relating to Plaintiffs' motion to compel certain discovery, and Defendants recent motion to compel production of Plaintiffs' expert reports on punitive damages. Mindful of the need to have consideration of these matters expedited due to the relative proximity to trial, the court expects to rule on these motions promptly.

Finally, the court feels compelled to note a few observations in closing. Throughout the long and arduous history of this litigation, the court has endeavored above all else to be fair, and to provide Plaintiffs with an opportunity to have their day in court. The court is mindful of the blow that has been dealt to Plaintiffs' case through today's ruling. That being said, the court's supreme responsibility is to apply the law in a just and impartial manner. Supreme Court and Third Circuit precedent on the issue of expert scientific testimony is clear. Moreover, this court has made every effort to apprise the parties not only of factors that it deemed especially relevant with respect to its analysis, but also, to put the parties on notice before the hearings as to areas of proffered testimony that were of particular concern to the court. (Tr. at 2-60 (pre-hearing conference). It was Plaintiffs who carried the burden within this proceeding to demonstrate by a preponderance of the evidence, in light of existing law and the detailed "road map" provided by this court, that the proffered testimony of their experts was scientifically valid and reliable, and based upon good grounds. Unfortunately, for the most part, Plaintiffs were unable to meet this burden and thus must fall victim to rules of evidence "designed not for the exhaustive search for cosmic understanding but for the particularized resolution of legal disputes." Daubert, ___ U.S. at ___, 113 S.Ct. at 2798.

An appropriate order will be issued.

NOTES

[1] The hearings were conducted on November 13-17, 20-22, and 27-30, 1995. Prior to and during a portion of the hearings, the Congress and the President were unable to agree upon a federal budget. As such, portions of the federal government, including all passport offices were required to close. Due to this partial government "shutdown," Dr. Tarasenko was unable to obtain a visa to attend the hearings. The court and the parties have agreed to include Dr. Tarasenko in the next round of in limine proceedings presently scheduled to occur in February of 1996. Dr. Tarasenko's appearance at the February hearing is contingent upon her ability to obtain a visa and upon the ability of Congress and the President to agree upon a budget. If it becomes necessary, other arrangements will be made with respect to Dr. Tarasenko.

[2] Further, Plaintiffs contend that the plume was so narrow as to move between thermoluminescent dosimeters ("TLDs") set up as part of the TMI Radiation Environmental Monitoring Program ("REMP"). Based upon this theory, Plaintiffs' experts have unanimously questioned the accuracy of the TLDs and disregarded their readings.

[3] United States v. Downing, 753 F.2d 1224 (3d Cir.1985). The Supreme Court adopted much of Downing's rationale in its opinion in Daubert.

[4] The court reads Paoli II as requiring it to evaluate all factors with respect to each expert's proffered testimony. Where the court finds a particular factor to be inapposite to its analysis, the court will explain the reasoning behind that decision.

[5] The functioning of the coolant system in TMI's pressurized water reactor has been described as follows (internal numbers in brackets correspond to diagram attached as Exhibit A):

The coolant system of a pressurized water reactor such as the one at Three Mile Island Unit 2 ... is a circuit, or closed loop of distilled water with a small amount of boric acid. The water is kept during normal reactor operation at an average temperature of 575 degrees Fahrenheit ... and pressures high enough (around 2200 pounds per square inch) to keep the water from boiling to steam.... This loop is called the primary reactor coolant system.

The primary coolant water circulates and recirculates through this loop. First it goes from the reactor core [1] — which might be thought of as a furnace where the coolant water is heated by the heat of the fission chain reaction in uranium fuel rods sheathed in cladding made of a zirconium alloy called Zircaloy — then through two "hot leg" [2] pipes which are 3 feet in diameter, to a steam generator [3]. Here in the generator the primary coolant system water transfers some of its heat to water in another, separate loop: the "secondary" or "feedwater loop." Water in the secondary loop is heated to steam without ever coming in direct contact with the primary coolant water. Having lost some of its heat, the primary coolant water then moves back via four pipes known as the "cold legs" [4] to the reactor core [1]. There it is heated once again and repeats the cycle.

The primary coolant is normally kept moving at high speed in the loop by enormous "reactor coolant pumps" [5] ... (there are four of them).

. . . . .

One more detail: there are two coolant loops in the TMI-2 reactor ..., called the "A" and "B" loops. Each loop is pretty much like the other. Each has its own steam generator feeding steam to the turbine that makes electricity. Each loop contains two out of the four reactor coolant pumps.

. . . . .

At a point between the reactor and the steam generator, there is a pipe leading to the bottom of a large vessel or tank called a "pressurizer" [6]. During normal operation, coolant water does not circulate back and forth through this pipe. The pressurizer is kept a little more than half full of coolant water. Above the water is a bubble, or cushion, of steam. This steam cushion at the top of the pressurizer can be made larger or smaller by slight heating or cooling of the pressurizer water just beneath it. The whole apparatus is, in fact, a means of keeping the pressure in the reactor coolant system relatively constant. When the bubble temperature and pressure are increased, it tends to push water from the pressurizer out into the primary reactor coolant loop, thus increasing the pressure in the loop. When the temperature and pressure in the pressurizer are lowered, steam condenses and the bubble shrinks. Water tends to come from the reactor coolant loop up into the pressurizer, and the overall system pressure is decreased.

In a pipe leading out of the top of the pressurizer, at the top of the space normally occupied by the steam bubble, is the "pilot operated relief valve" (PORV) [7], a very important piece of equipment to anyone who would understand the TMI-2 accident. This relief valve is designed to open automatically when the system pressure begins crowding the upper limits of safety. In theory, if the pressure in the coolant system rises very abruptly, the valve will open, some of the steam will rush out (shrinking the bubble), more water will move up into the pressurizer from the primary coolant loop, and system pressure will go down. When the system pressure is back to normal, the PORV is supposed to close automatically, having returned the system to a safer pressure. If the PORV fails to close, there is a block valve [9] between it and the pressurizer that can be closed [8]. In addition to the PROV there are two other safety valves.

Mitchell Rogovin and George T. Frampton, Jr., Nuclear Regulatory Commission Inquiry Group, "Three Mile Island, A Report to the Commissioners and to the Public," NUREG/CR-1250, January, 1980 at 10-11 ("Rogovin Report").

[6] Pursuant to this court's order dated December 14, 1995, David Lochbaum will be permitted to submit a second expert report on the issue of the alleged blowout. The following analysis pertains only to the report currently filed with the court.

[7] During his testimony, Dr. Vergeiner indicated that the TMI region consisted of alpine terrain. According to Dr. Vergeiner, for meteorological purposes, a mountain can be any elevation of terrain of 50 meters or more. (Tr. at 639-40, 565-66, 585.)

[8] Weather conditions are critical to any plume dispersion analysis:

Weather has a dominant influence on how and where a fallout cloud is dispersed. Important variables include the altitude to which the cloud rises and the wind speed and direction at the higher altitudes through which the cloud may rise. [For example,] [t]he degree of wind shear ... can materially affect the dispersal of the cloud, yielding a wider, shorter spread of particles than would be observed if no wind shear was present.

Allen v. United States, 588 F.Supp. 247, 302 (D.Utah 1984) (citing S. Glasstone & P. Dolan, The Effects of Nuclear Weapons ¶¶ 9.57-9.74 (3d ed. 1977)).

[9] When asked about the source of his estimated release rates, Dr. Vergeiner indicated that they were supplied to him by Plaintiffs' counsel. (Tr. at 661-62.) Additionally, Dr. Vergeiner stated that although the gross magnitude of his releases were the same as Dr. Webb's, his release times were different from Dr. Webb's. (Id.) Dr. Vergeiner was unable to explain why the time of his releases differed from those of Dr. Webb. (Id.) That an expert would be supplied source term numbers by an attorney is not in itself troubling. Yet, when that expert whole-heartily accepts the source terms without questioning their origin, attempting to verify the terms or assuring himself that they are based upon "good science," the court must question the expert's methodology. See Tulou v. Raytheon Service Co., 659 A.2d 796, 799 (Del.Super.Ct.1995) ("As with any model, the data input is crucial....").

[10] The water model consists of a topographical map which forms the base of a water tank. The tank is then filled with water. A dye is injected into the water and a current is run through the water to simulate air flow. In theory, the tank will demonstrate how a material will disperse in the atmosphere depending on the terrain and air patterns. (See generally Tr. at 31-32 (description of the water model by Plaintiffs' counsel at the pre-hearing conference).) There are two significant design features of the proffered water model which serve to decrease its reliability and fit with the instant litigation. First, the topographical map does not accurately reflect the topography of TMI and its environs. Second, when filled to its highest level, the water in the tank just covers the highest "mountains" in the topography. Accordingly, the tank does not account for the possibility that air flow may travel over, rather than around, these "mountains." (Tr. at 33; Vergeiner Dep. at 237; Ignaz Vergeiner, Treatise on the Meteorological Aspects of the TMI-2 Accident (Part II), at 53-57 (February 1995) (discussion of the water model); IHW, Shallow Water Lab Flow Simulations Applicable to TMI-2 Accident, at 7-9 (February 1995).)

[11] Dr. Vergeiner has been roundly criticized by Defendants for his complete disregard of computer modeling, specifically the Gaussian Plume model, in connection with his plume dispersion analysis. Defendants have presented evidence that mathematical modeling, and the Gaussian model in particular, are accepted methods used within the relevant scientific field. (Tr. at 149.); see, e.g., The Rogovin Report (employing computer modeling to evaluate plume dispersion); The Kemeny Commission Report (same). Moreover, computer modeling is accepted and widely used by the Environmental Protection Agency and the Nuclear Regulatory Commission. See Cincinnati Gas & Elec. Co. v. Costle, 632 F.2d 14, 18-19 (6th Cir.1980) (discussing EPA's use of computer models to monitor air quality). Dr. Vergeiner, to the contrary, argues that a substantial body of evidence has developed within the relevant scientific community which supports the notion that the Gaussian model is inaccurate in a number of circumstances. (Tr. at 610-13.)

[12] Plaintiffs have failed to come forward with any evidence tending to support the scientific validity of the terms which they supplied to Dr. Vergeiner. Presumably, the terms may have been based upon or related to Dr. Webb's source term estimates; however, the court has already found those estimates to be entirely unreliable.

[13] In 1995, the National Research Council published a study entitled "Radiation Dose Reconstruction for Epidemiologic Uses." The purpose of the study was stated as follows: "to identif[y] criteria to be considered when undertaking radiation dose reconstructions studies, to examine the pitfalls encountered in previous studies, and to recommend areas of needed research." Committee on an Assessment of CDC Radiation Studies, Board on Radiation Effects Research, Commission on Life Sciences, National Research Counsel, "Radiation Dose Reconstruction for Epidemiologic Uses" at vii (National Academy Press 1995) (hereinafter "Radiation Dose Reconstruction"). Summarizing their findings with respect to full scale dose reconstruction studies, the Committee noted that "[d]ose reconstruction studies must rely on solid science, state-of-the-art methods, and careful peer review if they are to be viewed as credible. Ultimately, a dose reconstruction study will be judged by the scientific community primarily on the basis of the technical quality of the study and its contribution to science." Id. at 14.

[14] During the hearings, Dr. Vergeiner gave the following explanation when asked by the court to explain "synoptic analysis":

Synoptic meteorology is essentially the — comes out of the realization that when you plot weather observations taken at the same time, you know, which could be nighttime in Europe and late afternoon in this country, then you can draw maps. And you start to interpret them and you find out that, you will find certain prominent features, like weather fronts, and they would move in a coordinated way, right? You would get the rains first in, God knows where, in Chicago, and later on you would get them further east. And this is what's called synoptic meteorology.

And to do that, of course, you don't want to look only at surface observations, but you do want to look at upper air observations, because the air moves faster at upper levels. And in order to interpret your features, you do want both surface and upper air observations.

And for upper air, there is a network of balloons soundings, these big balloons with an instrument package, and they are launched every 12 hours at stations like Pittsburgh, New York — Albany, New York, Washington, D.C., stations at about that distance. And from those radio soundings, you get temperature and winds and pressure at upper levels. They are the backbone of synoptic meteorology.

(Tr. at 592-93.)

[15] The important consideration here is not whether any models similar to those presented by Dr. Vergeiner have been peer reviewed, but whether the methodology he has employed in developing these models has been subject to peer review.

[16] A "TLD" is a thermoluminescent dosimeter. These devices are used to measure airborne radiation.

[17] Recognizing its merely average scientific acumen, and based upon its review of documents before it, the court believes that Dr. Vergeiner should have: (1) made some attempt to verify that the source terms provided to him by Plaintiffs' counsel were derived by a scientist with sufficient expertise using reliable and established methods; (2) run some of the more generally accepted computer models and tested his hypothesis against those models; (3) provided a detailed and scientific comparison between his model and the more accepted computer models and given specific scientifically based reasons why his model should be accepted as reliable.

[18] "Computer dispersion modeling is the mathematical prediction of mixing different concentrations of gases in the atmosphere and the resulting effects in concentrations downwind under varying meteorological conditions." Tulou v. Raytheon Service Co., 659 A.2d 796, 799 (Del.Super.Ct.1995).

[19] The court has chosen to use computerized dispersion and mathematical models as the "established" technique because of their widespread use and acceptance. See supra, note 11.

[20] The court also notes that it finds Dr. Vergeiner's testimony regarding his qualifications in this area to demonstrate his lack of expertise. (Tr. at 561.)

[21] Additionally, Dr. Vergeiner has stated that he did not use the water model in performing his dose reconstruction. (Vergeiner Dep. at 30, 204-05, 323-27.)

[22] Because neither Plaintiffs nor Defendants have addressed this testimony, the court will not rule at this time.

[23] Because the proffered testimony is so obviously lacking in scientific basis, the court will not take the time to trudge through the Daubert/Paoli II analysis. The Daubert/Paoli II factors are geared toward analyzing scientific testimony. As such, they presume a certain level expertise with respect to the organization and structure of a study. Studies wholly lacking in scientific basis are not organized or relayed to the court in a format similar to traditional scientific studies. Thus, they do not lend themselves to application of the Daubert/Paoli II analysis.

[24] No identifying names are used in the following discussion because the court has not been provided with the names of either the Etters resident or his neighbor.

[25] "[I]on chamber monitors ... [have been] used at the government's Nevada test site and elsewhere for measuring the levels of ion radiation fluxes, commonly referred to as background radiation." Dept. of Energy v. White, 653 F.2d 479, 481 (Cust. & Pat.App.1981), cert. denied, 454 U.S. 1144, 102 S.Ct. 1005, 71 L.Ed.2d 296 (1982).

[26] A "radionuclide" is an unstable nuclide that can undergo radioactive decay. (1/20/93 Aff. of Dr. John Fraizer at ¶ 9.)

[27] It is the court's understanding that a variety of radionuclides, with a corresponding variety of half lives, can be found in any soil sample. Thus, it is more accurate to isolate each radionuclide and calculate a half life for that radionuclide, than it is to measure the overall radioactivity of the entire sample and derive a generalized or average half life for the entire sample. See Allen v. United States, 588 F.Supp. 247, 276 (D.Utah 1984) ("Measurement in curies does not, however, define the type or energy of the radiation being emitted. That information is specific to each radionuclide and is important to the evaluation of the risk created by exposure to curie, millicurie, or microcurie in amounts of material."), rev'd on other grounds, 816 F.2d 1417 (10th Cir.1987), cert. denied, 484 U.S. 1004, 108 S.Ct. 694, 98 L.Ed.2d 647 (1988).

[28] "Fission products" are "the nuclear fragments resulting from the splitting of uranium or plutonium nuclei." Allen, 588 F.Supp. at 288.

[29] "Natural background radiation" is radiation arising from radiation sources that occur naturally in the environment. See 10 C.F.R. § 20.1003 (1995). "Natural radioactivity is all around us, in soil, in building materials, and in trace amounts in meat, vegetables, fruits, grains and drinking water." Allen, 588 F.Supp. at 327.

[30] The court credits Dr. Moeller's testimony based upon its consistency with accepted scientific principles. See Allen, 588 F.Supp. at 276.

[31] The unit of dose equivalent, "rem," is equal to 0.01 joule per kilogram in man. The dose equivalent in rems is equivalent to the absorbed dose in rads multiplied by the quality factor. 10 C.F.R. § 20.1004 (1995). A "rad" is the unit of absorbed dose equal to 0.01 joule per kilogram in any medium, 0.01 gray. Id.

[32] "Erythema" is defined as "redness of the skin produced by congestion of the capillaries, which may result from a variety of causes, the etiology or a specific type of lesion often being indicated by a modifying term." Dorland's Illustrated Medical Dictionary at 577 (27th ed. 1988). Here, it is posited that the erythema resulted from radiation exposure.

[33] Dr. Gunckel described the interview process as follows:

We went to these sites, and this was the case of revisiting the sites where people had complained of some sort of health effect. So we were there to do two things: First, to talk with them and to reaffirm whether their alleged effects were real, possible, or not, not trying to say as a doctor would be qualified to say that they had suffered from this, that and the other thing. We were simply asking, sort of trying to get a composite view of what had gone on. So we were not passing on identifying whether what they complained of was true or false. We were simply trying to associate people with possible episodic evidence with a plant indicator.

(Tr. at 1600.)

[34] In his deposition, Dr. Gunckel admitted that some of the information which he relied upon was not contained within the subject's medical records. See, e.g., Gunckel Dep. at 173 ("[The doctor] said it couldn't be erythema because, because Met Ed and NRC said that no, no radiation got out. So he did not put it in her medical record that she had received an erythema dose...." "Q. So you're reporting what she reported her doctor said?" "A. That's right.").

[35] At this point Dr. Gunckel's qualifications would come into play. As will be discussed infra, Dr. Gunckel is not qualified to review medical records, form an opinion, and opine as a medical expert. (See Gunckel Dep. at 16, 169.) Therefore, even if this would have been the correct direction for his research to take, it is likely that Dr. Gunckel would have needed to employ someone with greater medical expertise to conduct this portion of his study.

[36] A key component of the studies performed by Drs. Gunckel and Sparrow at Brookhaven were cellular and subcellular studies on the trees. They used these cellular and subcellular evaluations to determine the extent of damage done to the trees by radiation. These studies did not include viewing morphological damage to trees. (See Tr. at 1595, 1618.)

[37] Defendants have raised an issue as to whether the Brookhaven studies ever involved mature trees. (Defs.' Findings at 94, ¶ 21.)

[38] Speaking to the issue of medical reports prepared for litigation, the Third Circuit has stated:

We do not doubt the propriety of a medical report prepared just for litigation purposes, but a physician who evaluates a patient in preparation for litigation should seek more than a patient's self-report of symptoms or illness and hence should either examine the patient or review the patient's medical records simply in order to determine that a patient is ill and what illness the patient has contracted.

. . . . .

Hence, we think that generally, a doctor only needs one reliable source of information showing that the plaintiff is ill and either a physical examination or medical records will suffice — but the doctor does need at least one of these sources.

Paoli II, 35 F.3d at 762 (emphasis added).

[39] Wade-Greaux v. Whitehall Laboratories, Inc., 874 F.Supp. 1441, 1483 ((D.V.I.1994) ("Anecdotal human data ... have inherent biases that make them unreliable.... [S]uch anecdotal human data do not represent the type of data reasonably relied upon by experts in the field....").

[40] In Paoli II, the Third Circuit noted that "while we think the standard techniques of differential diagnosis are reliable and will allow a doctor who employs them to testify to a novel conclusion, we also think that part of differential diagnosis is using these techniques to rule out alternative causes. Paoli II, 35 F.3d at 759 n. 27.

[41] These reports are as follows:

1. "Report of conducted dendrometric investigated samples of pine and spruce wood collected in the region of accident on NPS Three Mile Island," written by A. Tascaev and G. Kozubov, dated February 17, 1995 (hereinafter "Kozubov Rpt.");

2. "Experimental determination of gammaradiation accrued dose and beta radiation assessment, in porcelain insulators (samples 1-9) after TMI accident," written by V. Popov and A. Portman, dated January 6, 1995 (hereinafter "Popov Rpt.");

3. "Results of immunological analysis of blood of people living around TMI," by O. Tarasenko, dated February 15, 1995 (hereinafter "Tarasenko Rpt.");

4. "Cytogenetic analysis of the people living in the neighborhood of TMI nuclear Power Plant [sic]," by G. Snigiryova, dated March 15, 1995 (hereinafter "Snigiryova Rpt.").

[42] Norman Aamodt is a consultant for Plaintiffs.

[43] A dicentric is an interim aberration of a chromosome. It describes a chromosome that has two centrimeres rather than one.

[44] Brent Carson, "Increased Risk of Disease from Hazardous Waste: A Proposal for Judicial Relief," 60 Wa.L.Rev. 635, 643 (1985); see id. at 643 n. 40 (identifying the Center for Disease Control's follow-up assessment of Love Canal as a carefully controlled study).

[45] As Defendants note, Professor Shevchenko does not explain why the doses found in other studies he commissioned are lower than the threshold doses required to cause the morphological damage he observed in the TMI trees. Moreover, after claiming that the sublethal exposure experienced by TMI trees would take several years to cause death at the tree tops, he attributed damage to tree tops discovered only months after the accident to radiation damage.

[46] "Epidemiology is the study of disease patterns in human populations. ``Attempts to define a relationship between a disease and a factor suspected of causing it.'" In re Joint Eastern and Southern Dist. Asbestos Litigation, 52 F.3d 1124, 1128 (2d Cir.1995) (quoting Brock v. Merrell Dow Pharmaceuticals, Inc., 874 F.2d 307, 311 (5th Cir.), modified on reh'g, 884 F.2d 166 (5th Cir. 1989), cert. denied, 494 U.S. 1046, 110 S.Ct. 1511, 108 L.Ed.2d 646 (1990)).

[47] The court believes that this study can be characterized as an "ecological study." "Studies that collect data about the group as a whole are called ecological studies. Such studies are useful for identifying associations but generally are regarded by epidemiologists as ``weak.'" Research Manual at 132-33 (footnote omitted).

[48] To the best of the court's knowledge, Mr. Aamodt is not an epidemiologist.

[49] In contrast, dose assessments in the Susser/Hatch study were made on the basis of the Beyea dose estimates. These dose estimates were based upon original plant data, meteorologic data from the TMI meteorological tower, and a Gaussian plume model. (See Susser/Hatch Study at 400-01.)

[50] A recent article in the Harvard Law Review explained the importance of the "rate of error" factor in the context of the Daubert analysis. "Confronting the New Challenges of Scientific Evidence," 108 Harv.L.Rev. 1532 (May, 1995) (hereinafter "New Challenges"). In the article, the following is noted:

In considering scientific reliability, judges will often need to address the statistical significance of scientific studies. Although Daubert itself raised the issue whether scientific evidence must meet a threshold level of statistical significance for admissibility, the Court did not speak to statistical significance, perhaps because it concluded that legal reliability embraces only scientific validity. Yet because significance testing, too, is central to a large body of scientific research, statistical inference will frequently factor into Daubert's legal reliability prong.... Judicial inquiry into statistical significance is, therefore, necessary in determinations of scientific, and hence evidentiary, reliability.

Id. at 1535-36 (footnotes omitted). The article went on, observing that despite the importance of the evaluation of statistical significance, many judges were ill-equipped to perform such an analysis; and further, that many experts were often unprepared to discuss the statistical dimensions of their own studies. Id. at 1540.

[51] The measures of standard error appearing in Dr. Wing's report can be used to calculate the size of the "confidence interval" surrounding the risk estimates derived from the data. (Tr. at 970.) A confidence interval sets forth the upper and lower bounds of the likely true result. A confidence interval of 95% encompasses that range of numbers within which the true result will fall 95% of the time. Wade-Greaux, 874 F.Supp. at 1452; Federal Judicial Center, Reference Manual on Scientific Evidence 154-55 (West 1994).

[52] During his deposition, Dr. Crawford-Brown admitted that in making an exposure assessment, ordinarily one would assess or weigh the strength of each of the kinds of evidence available:

Science can't distinguish between the estimates. What it can say is something about the relative strength of each of the estimates — which ones you believe and which ones you disbelieve, to what degree do you believe them or disbelieve them.

And then, in my work, anyway, that is then called a cumulative confidence distribution. You ask, now, considering all the lines of reasoning, how confident are you that it's less than one rem, ten rems, a hundred rems, a thousand rems, and so forth. And that assignment of confidence at any given dose [estimate] depends on how much you're weighing the various lines of evidence.

(Crawford-Brown Dep. at 56-57.) He did not make this type of assessment with respect to the TMI data.

[53] In fact, Dr. Crawford-Brown expressed his reservation that he was "kind of worried about having this circularity" in using incidence data to estimate doses, based on the assumption that doses were the cause of any increased incidence. (Crawford-Brown Dep. at 110-11.)

[54] The Civil Justice Reform Act of 1990, 28 U.S.C. § 473(a), requires federal trial court judges "to implement techniques and strategies designed to dispose of cases in an efficient and inexpensive manner." Capek v. Mendelson, 143 F.R.D. 97, 99 (E.D.Pa.1992).

[55] The court has made two exceptions to this deadline based upon extremely compelling circumstances. First, the court allowed Plaintiffs an additional week to view original plant data from the TMI vault. Plaintiffs were permitted to supplement pre-existing expert reports with information drawn from these original plant documents. Second, following Dr. Webb's recantation of the bulk of his "blowout" theory, the court ruled that David Lochbaum could file a new expert report speaking to the "blowout" issue. Despite its strong reservations, the court permitted this additional filing based upon the importance of this theory to Plaintiffs' case and due to the rather bizarre circumstances of Dr. Webb's sudden recantation.

[56] At this juncture, the court's ruling applies only to Professors Shevchenko and Armentrout. The court wishes to make clear, however, that the reach of this ruling extends to all expert scientific testimony whether the testimony is from Plaintiffs' or Defendants' experts. Additionally, the court notes that it has made an exception to its rule with respect to the report of Dr. Snigiryova. The court allowed Professor Shevchenko to testify to this report because it was in his area of expertise. Consequently, Defendants were offered a meaningful opportunity for cross-examination. See DeLuca v. Merrell Dow Pharmaceuticals, Inc., 911 F.2d 941, 953 (3d Cir.1990).