Suffolk County Water Authority v. The Dow Chemical Company and Vibrantz Corporation, Formerly Known as Ferro Corporation

Opinion

UNITED STATES DISTRICT COURT EASTERN DISTRICT OF NEW YORK enema seeeewe eenesannenccanwec cesarean XK SUFFOLK COUNTY WATER AUTHORITY, OPINION & ORDER Plaintiff, — . 17-cv-6980 (NG) (JRC) ~against- THE DOW CHEMICAL COMPANY AND VIBRANTZ CORPORATION, FORMERLY KNOWN AS FERRO CORPORATION, Defendants. Heenan □□□ seenannsenrecrrenewe meee ewensaneeensconaone X GERSHON, United States District Judge: In this water contamination case, defendants Dow Chemical Company and Vibrantz Corporation (collectively, “Defendants”) challenge the proof which plaintiff Suffolk County Water Authority (“Suffolk”) proffers through its expert witnesses that dioxane-stabilized TCA is the source of the contaminant 1,4-dioxane in what it calls the TCA Claim Wells. Suffolk Cnty. Water Auth. v. Dow Chem. Co., 2025 WL 1907206 (E.D.N.Y. July 10, 2025). Defendants do so through the opinions of their own expert witnesses John A. Connor and David T. Adamson. Suffolk does not challenge the bulk of the Connor/Adamson opinions, which are rebuttals to Suffolk’s experts. Rather, Suffolk moves to exclude one subset of those opinions, namely, those opinions which affirmatively assert well-by-well conclusions as to source identification of the 1,4- dioxane in the wells. I. Expert Qualifications John A. Connor is a Senior Principal Engineer employed by GSI Environmental Inc. (“GSI”) with over forty years of experience in environmental engineering, including specialization

in “environmental site investigation, human health and ecological risk assessment, corrective action design, water resource development and remediation, and design and permitting of treatment facilities.” Expert Rebuttal Report of John A. Connor and David T. Adamson (“Connor/Adamson Report”) at 1. He received an M.S. in Civil Engineering from Stanford University and is a registered Professional Engineer, a licensed Professional Geoscientist, and a Diplomate in the American Academy of Environmental Engineering. Jd. Mr. Connor has authored technical publications, investigated sites, and completed numerous projects “involving chemical impacts on soil, groundwater, and surface water resources.” Id. David T. Adamson is a Principal Engineer and Vice President at GSI. Jd. at 2. He received his Ph.D. in Civil and Environmental Engineering from the University of Iowa and has over twenty-two years of research and environmental consulting experience. /d. He has conducted research and authored technical articles on a wide range of topics related to subsurface contamination, including serving as Principal or co-Principal Investigator on research projects sponsored by the U.S. Department of Defense on topics such as 1,4-dioxane fate, transport, and treatment. /d. Dr. Adamson has also consulted on a broad range of topics, including “chemical fate and transport, site investigation, remedy screening, risk assessment, remedial design, drinking water distribution, and litigation matters.” Jd. The expert report at issue was prepared by GSI and signed by both Mr. Connor and Dr. Adamson. Il. The Connor/Adamson Report Mr. Connor and Dr. Adamson opine that TCA is the source of the 1,4-dioxane in only a small fraction of the TCA Claim Wells. Rather, they assert, the predominant source is septic system and wastewater releases. This conclusion appears in Section 3.3 of their Report, which

-2-

consists of pages 70-76 out of 160 pages, plus a mass of appended charts. Mr. Connor and Dr. Adamson opine that they identified ten relevant “lines of evidence” and that they used a “weight- of-evidence” methodology to assess the likely source of 1,4-dioxane in each of the TCA Claim Wells. The lines of evidence include, for example, TCA detections in each well, and land use within the capture zone.! Section 3.3 contains an explanation of the lines of evidence as well as the experts’ ultimate conclusions that the weight of the evidence is consistent with a septic system and wastewater source of 1,4-dioxane in the vast majority of well fields containing TCA Claim Wells. Upon review of the Connor/Adamson Report and the appended charts, I find that there is in fact no well-by-well analysis in the Report or in the charts. To explain this conclusion, a closer examination of what appears in the Report and charts follows. In the Report, Connor and Adamson do not assign numeric weight to the various lines of evidence or follow a decision tree to reach an ultimate conclusion. February 27, 2026 Oral Argument Tr. (“2/27/26 Tr.”) at 38:1-17; see also Adamson Dep. Tr. at 314:16-315:14 (“I do not explicitly describe an assessment on a well-specific basis for each line of evidence. But I don't need to do that for purposes of this evaluation.”). At oral argument, Defendants accepted my suggestion that the ten lines of evidence are more properly considered “factors” and that there was no weighing of one line against another. 2/27/26 Tr. at 35:9-12. In other words, Connor and

' The ten lines of evidence are: 1) Detection of 1,4-Dioxane and/or TCA and TCA Breakdown Products; 2) Concentration Trends Over Time; 3) Breakthrough Curves for 1,4-Dioxane and/or TCA and TCA Breakdown Products; 4) Detection of Septic and Wastewater Indicators; 5) Detection of Other Chlorinated Volatile Organic Compounds (“VOCs”); 6) Land Use Within Capture Zone; 7) Sewerage Within Capture Zone; 8) Presence of TCA Detection Site(s) Within Capture Zone; 9) Other Potential Dischargers of 1,4-Dioxane and/or Chemicals of Concern in the Capture Zone; and 10) Water Travel Time Distribution. Connor/Adamson Report at 72-73.

-3-

Adamson’s methodology amounts to a selection of the ten relevant lines of evidence and then using their “judgment” to determine the ultimate conclusion for each well. Jd. at 41:24-42:10. The Defendants have acknowledged that there is no well-by-well narrative analysis in the Report for each TCA Claim Well.? Jd. at 54:20-22. Dr. Adamson stated at his deposition, and Defendants unequivocally asserted at oral argument, that the entirety of the well-by-well analysis used to reach the experts’ ultimate conclusions appears in the charts and data contained within the appendices to the Report. Adamson Dep. Tr. at 41:6-12; 2/27/26 Tr. at 54:8-55:9. According to Defendants, Table 4 (a 210-page spreadsheet containing data points for each well) in Appendix B to the Report contains the analysis in “full form” for every well. 2/27/26 Tr. at 14:7-12. The thousands of selected data points map to Connor and Adamson’s ten lines of evidence and include operational data such as well pumping rates; physical characteristics of the well such as water travel times; frequency of organic compound detections; land use; and so on.

2 Defendants point to three instances where, they assert, the experts provided well-by-well analysis in narrative form. First, at deposition, Dr. Adamson gave an account of how he reached his conclusion for the Morris Avenue 1 Well through examination of four of the lines of evidence. Adamson Dep. Tr. at 163:10-166:13. Second, Defendants attached to their December 2, 2025 Motion to Adopt a Phased Trial Plan (“Phased Trial Motion”) an expert declaration from Mr. Connor, in which he described his intended testimony and provided as an example a three-page analysis of all ten lines of evidence as applied to one well, the Wicks Road well. Phased Trial Motion, Connor Decl. PP 14-19. He stated there that, after he has explained the general concepts and methodologies behind the lines of evidence to the jury, he “will then apply each of these principles to the individual characteristics of each of the wells” that he is asked to assess at trial; he estimates that his direct presentation will take approximately one to three hours per well. /d. P 13. Third, Defendants, for the first time at oral argument, argued that well-by-well analysis appears in the body of the Report for one well, Wheat Path 3. 2/27/26 Tr. at 18:4-8; see Connor/Adamson Report at 62-63. But this purported analysis, which consists of one exhibit and a single paragraph of explanation, is not, as Defendants contend, an explanation of how the ten lines of evidence led to a specific conclusion as to the source of 1 ,4-dioxane in Wheat Path 3. And, even if this example were treated as constituting a particularized well-by-well analysis, no such analysis is present for the other 252 TCA Claim Wells. -4-

In the last fifteen pages of Table 4, Connor and Adamson report their ultimate conclusions as to the source of 1,4-dioxane in each well in two columns titled “Septic and Wastewater Source” and “Potential Industrial Source.” Each well is marked as “Y,” “N,” or “NA” in these two columns, with no additional explanation. But nowhere in the charts is there any analysis or description of a methodology that led the experts from the data points to their conclusion for each well.? At trial, Defendants intend to present the testimony of either Connor or Adamson in a narrative fashion; that is, the experts will explain their reasoning in evaluating the lines of evidence and coming to a particular conclusion individually for each well, rather than relying solely on their identification of the charts. /d. at 26:4-10. They argue that their experts should be permitted to testify in this manner because the proposed narrative explanations will correspond exactly to the “analysis” present in the appendices to the Report. /d. at 55:4-9. But, I repeat, there is no such analysis, and the proffered testimony would be the first presentation of the experts’ reasons for reaching their conclusions. I. Discussion For purposes of this motion, I will assume that any challenges to the validity of the lines of evidence chosen by Defendants’ experts or to the judgment they exercised in choosing the data points within those lines of evidence can be dealt with on cross-examination. That is because a more fundamental problem underpinning these experts’ opinions mandates the granting of Suffolk’s motion. That problem is that there is simply no way the jury could understand from the Connor/Adamson Report and its appended charts by what means, methodology, analysis or

3 Defendants state that “the same well-by-well analysis that is in Table 4” is presented in “a more user friendly way” in the Well Field Summaries in Appendix F. Jd, at 18:21-24, 19:5-10. But no additional information that could be characterized as an analysis is found in the Well Field Summaries. -5-

exercise of judgment the ultimate conclusions of the experts as to source identification were reached. Defendants acknowledge that Section 3.3 contains only (1) a discussion of the ten lines of evidence and (2) Connor and Adamson’s ultimate conclusions as to the likely source of 1,4- dioxane in each of the TCA Claim Wells. They argue that how Connor and Adamson applied the ten lines of evidence to reach their conclusions appears in the appendices. But they acknowledge that the cells in Table 4 marked with “Y,” “N,” or “NA” “may record the conclusion” but are “not the analysis.” 2/27/26 Tr. at 39:8-16. Indeed, review of these appendices reveals a complete absence of analysis as to how the experts reached those final conclusions. Put another way, it is impossible to glean what the experts’ thought processes were in considering the cumulative effect of the thousands of data points that make up the purported “full form” well-by-well analysis in Table 4 in reaching their conclusions. I reject Defendants’ argument that the detailed narratives through which the experts intend to present their testimony, which they estimate could take up to three hours per well, can be discerned merely by looking at the charts. On the contrary, it is clear that Defendants intend to use the trial testimony to supply the missing explanations as to how each line of evidence factored into the ultimate conclusion as to each well. Indeed, if the charts themselves were a sufficient basis for understanding the experts’ reasoning in reaching their ultimate source identification conclusions, as Defendants contend, then—after the experts’ general testimony applicable to all wells—Defendants would need only to present the charts to the jury. Yet when Defendants were asked at oral argument whether there remained a need for the experts to testify in narrative form as to each well, as they had represented to be necessary in their Phased Trial Motion, they said

-6-

there was. 2/27/26 Tr. at 24:8-26:10. It seems that, although they insist the charts are sufficient, Defendants recognize the insufficiency of the charts alone. Most importantly, for this stage of the proceedings, there is no way for the court to evaluate whether the methods the experts used or the judgment they applied was reliable or not. As an example of why this is important, consider the undisputed fact that, with respect to some wells, the experts’ lines of evidence point to inconsistent or conflicting results as to the source of the 1,4 dioxane. Nowhere do the experts provide any analysis of how they resolved—or can resolve— such inconsistencies or conflicts. Rather, it is clear that their application of judgment is separate for each well, yet no effort has been made in the Report to set forth how those individual judgments were made. Defendants argue that case authority supports their position that they have provided enough evidence to find that the experts’ opinions are reliable and admissible, but this is not correct. The basic principles for the admission of expert evidence under Rule 702 of the Federal Rules of Evidence and Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993), are set forth in Suffolk Cnty., 2025 WL 1907206, at *2, and will not be repeated here. As emphasized in Amorgianos v. Nat'l R.R. Passenger Corp., reliability must be shown at every step. 303 F.3d 256, 267 (2d Cir. 2002). It is true that multi-factor analyses can pass muster under Daubert. See In re Zoloft (Sertraline Hydrochloride) Prods. Liab. Litig., 858 F.3d 787, 795 (3d Cir. 2017) (noting that while flexible methodologies such as weight of the evidence are “generally reliable, each application is distinct and should be analyzed for reliability”); see also Gen. Elec. Co. v. Joiner, 522 U.S. 136, 153 (1997) (“It is not intrinsically ‘unscientific’ for experienced professionals to arrive at a conclusion by weighing all available scientific evidence—this is not the sort of ‘junk science’ with

29 <

which Daubert was concerned.”) (Stevens, J., concurring in part and dissenting in part). But it is imperative that experts who apply multi-criteria methodologies, such as weight of the evidence, use and explain “a scientific method of weighting” to ensure that their particular application “is truly a methodology, rather than a mere conclusion-oriented selection process.” In re Mirena lus Levonorgestrel-Related Prods. Liab. Litig. (No. II), 341 F. Supp. 3d 213, 247 (S.D.N.Y. 2018) (quoting /n re Zoloft, 858 F.3d at 796), aff'd, 982 F.3d 113 (2d Cir. 2020); see also Daniels-Feasel v. Forest Pharms., Inc., 2021 WL 4037820, at *7 (S.D.N.Y. Sept. 3, 2021), affd, 2023 WL 4837521 (2d Cir. July 28, 2023). In short, the analyses used must be reliable, and, as just shown, no analyses have been provided here. Defendants are correct that courts have denied the exclusion of experts applying weight of the evidence methodologies in groundwater contamination cases. E.g., Fisher v. Ciba Specialty Chems. Corp., 2007 WL 2302470 (S.D. Ala. Aug. 8, 2007), at *8. But Defendants fail to grapple with the central issue at the heart of Suffolk’s Daubert challenge. The question is not simply whether Connor and Adamson may consider various lines of evidence together in reaching conclusion. Rather, as explained above, the problem is that their Report does not present any analysis of how the various lines of evidence led them to their ultimate conclusions. In these circumstances, the court is unable to determine the reasoned bases for Connor and Adamson’s analysis and whether it is truly reliable “at every step.” Amorgianos, 303 F.3d at 267. In sum, the reliability of a multifactor methodology depends on an expert’s specific application of it to the facts at hand. The cases Defendants rely upon support this conclusion. In Fisher, for example, the court found the expert’s opinion as to the source of chemical contamination at a single site to be reliable only after exploring his reasoned analysis step-by-step. 2007 WL 2302470, at “6-7; see also Dolomite Prods. Co. v. Amerada Hess Corp., 2004 WL

-8-

1125154, at *3 (W.D.N.Y. May 19, 2004) (accepting the expert’s methodology as to a single site after rejecting specific challenges made in that case as not requiring Daubert exclusion without addressing the issue presented here). But here, the Connor/Adamson Report does not provide a reasoned basis for this court to evaluate their conclusions as to each of the sites. Rather, here, as in Open Text S.A. v. Box, Inc., the experts “made no attempt to quantify the effect of each factor.” 2015 WL 349197, at *6 (N.D. Cal. Jan. 23, 2015). As that court held, “{w]hat mandates exclusion here, rather than simply subjecting the opinions to cross-examination, is not that the qualitative factors [the expert] discusses are improper considerations,” or that “the reams of evidence she claims she considered are beyond the pale. Exclusion is required because the link, if any between those inputs and” the expert’s conclusions “is written in invisible ink.” Jd. Finally, Defendants’ suggestion that they can have their experts describe in narrative form how they reached their source identification conclusions well-by-well for the first time at the trial, without having provided their analyses pre-trial, violates fundamental principles of pre-trial expert discovery. Rule 26(a)(2)(B)(i) of the Federal Rules of Civil Procedure provides that a party must provide “a complete statement of all opinions the witness will express and the basis and reasons for them.” It is also fundamental that a party will not ordinarily be permitted to use on direct examination any testimony not so disclosed. Fed. R. Civ. P. 26(a)(2) advisory committee's note to 1993 amendment; Fed. R. Civ. P. 37(c)(1).

-9-

IV. Conclusion For the foregoing reasons, Suffolk’s motion to exclude Connor and Adamson’s well-by- well assessment of dioxane sources is GRANTED.

SO ORDERED.

NINA GERSHON □ United States District Judge March 23, 2026 Brooklyn, New York

a Ole