Lee-Bolton v. Koppers Inc.

Full title: LISA LEE-BOLTON, et al., Plaintiffs, v. KOPPERS INC., f/k/a KOPPERS…

Court: United States District Court, N.D. Florida, Gainesville Division

Date published: Mar 20, 2017

319 F.R.D. 346 (N.D. Fla. 2017)

Opinion

[Copyrighted Material Omitted] [Copyrighted Material Omitted] [Copyrighted Material Omitted]

          For LISA LEE-BOLTON, ON BEHALF OF THEMSELVES, RICHARD BOLTON, ON BEHALF OF THEMSELVES, NANCY KAREN DAY, ON BEHALF OF THEMSELVES AND ALL, OTHERS SIMILARLY SITUATED, Plaintiffs: BENJAMIN DEAN ADAMS, WILLIAM STUART CALWELL, JR, DAVID HARLEY CARRIGER, LEAD ATTORNEYS, CALWELL PRACTICE PLLC - CHARLESTON WV, CHARLESTON, WV; PAUL STUART ROTHSTEIN, LEAD ATTORNEY, PAUL S ROTHSTEIN PA, GAINESVILLE, FL; AMANDA CHRISTINE BROADWELL, ELLIOTT TUBBS, III, PETER JAMES CAMBS, SR, PARKER WAICHMAN LLP - BONITA, SPRINGS FL, BONITA SPRINGS, FL; WILLIAM JOSEPH DUBANEVICH, LAW FIRM OF WILLIAM J, DUBANEVICH LLC - NEW YORK NY, NEW YORK, NY.

         For CHARLES DARBYSHIRE, AS GUARDIAN OF THE PERSON AND, PROPERTY OF ROY GEIERSBACH, BETTY HEWITT, ON BEHALF OF THEMSELVES AND ALL, OTHERS SIMILARLY SITUATED, JOHN HEWITT, ON BEHALF OF THEMSELVES AND ALL, OTHERS SIMILARLY SITUATED, CECIL BARNES, ON BEHALF OF THEMSELVES AND ALL, OTHERS SIMILARLY SITUATED, Plaintiffs: WILLIAM STUART CALWELL, JR, LEAD ATTORNEYS, CALWELL PRACTICE PLLC - CHARLESTON WV, CHARLESTON, WV; PAUL STUART ROTHSTEIN, LEAD ATTORNEY, PAUL S ROTHSTEIN PA, GAINESVILLE, FL; AMANDA CHRISTINE BROADWELL, PETER JAMES CAMBS, SR, PARKER WAICHMAN LLP - BONITA, SPRINGS FL, BONITA SPRINGS, FL; WILLIAM JOSEPH DUBANEVICH, LAW FIRM OF WILLIAM J, DUBANEVICH LLC - NEW YORK NY, NEW YORK, NY.

         For KOPPERS INC, A PENNSYLVANIA CORPORATION, formerly known as KOPPERS INDUSTRIES INC, BEAZER EAST INC, A DELAWARE CORPORATION, formerly known as KOPPERS COMPANY INC, Defendants: BENJAMIN HARVEY HILL, III, JOSHUA CLARK WEBB, DENNIS PARKER WAGGONER, HILL WARD & HENDERSON PA - TAMPA FL, TAMPA, FL; CAL RICHARD BURNTON, WILLIAM R ANDRICHIK, STEPTOE & JOHNSON LLC - CHICAGO, IL, CHICAGO, IL; LIBRETTA PORTA STENNES, STEPTOE & JOHNSON LLP - WASHINGTON DC, WASHINGTON, DC.

         For BEAZER LIMITED, A FOREIGN CORPORATION, Defendant: COURTNEY KNEECE GRIMM, JOHN A DEVAULT, BEDELL DITTMAR DEVAULT ETC - JACKSONVILLE FL, JACKSONVILLE, FL.

         For KOPPERS INC, A PENNSYLVANIA CORPORATION, Cross Defendant: BENJAMIN HARVEY HILL, III, DENNIS PARKER WAGGONER, HILL WARD & HENDERSON PA - TAMPA FL, TAMPA, FL.

         For RICHARD BOLTON, ON BEHALF OF THEMSELVES, NANCY KAREN DAY, ON BEHALF OF THEMSELVES AND ALL, OTHERS SIMILARLY SITUATED, ROY GEIERSBACH, ON BEHALF OF THEMSELVES, LISA LEE-BOLTON, ON BEHALF OF THEMSELVES AND ALL OTHERS SIMILARLY SITUATED, Counter Defendants: BENJAMIN DEAN ADAMS, DAVID HARLEY CARRIGER, WILLIAM STUART CALWELL, JR, LEAD ATTORNEYS, CALWELL PRACTICE PLLC - CHARLESTON WV, CHARLESTON, WV; PAUL STUART ROTHSTEIN, LEAD ATTORNEY, PAUL S ROTHSTEIN PA, GAINESVILLE, FL; PETER JAMES CAMBS, SR, PARKER WAICHMAN LLP - BONITA SPRINGS FL, BONITA SPRINGS, FL; WILLIAM JOSEPH DUBANEVICH, LAW FIRM OF WILLIAM J DUBANEVICH LLC - NEW YORK NY, NEW YORK, NY.

         ORDER

         M. CASEY RODGERS, CHIEF UNITED STATES DISTRICT JUDGE.

         Plaintiffs¹ own residential property in Gainesville, Florida, located near a former wood-treatment facility operated by the Defendants and their predecessors.² They filed this putative class action lawsuit claiming that their properties are contaminated as a result of chemical activities at the facility and requesting classwide damages for the cost of remediating the contamination and for the diminished value of their properties due to the stigma of contamination.³

1 The named plaintiffs include Lisa Lee-Bolton, Richard Bolton, Nancy Karen Day, Betty Hewitt, John Hewitt, Charles Darbyshire as Guardian of the person and property of Roy Geiersbach, and Cecil Barnes.

2 The American Lumber Company built and began operating the facility in 1916. Koppers Company, Inc. purchased the operation in 1954 and bought the property in 1984. Koppers Inc. became the owner in 1988, and Beazer East, Inc. has been the owner since 2010. Koppers, Inc. and Beazer East, Inc. are the only defendants remaining in the case.

3 There is no allegation that anyone is sick or has suffered health problems from the alleged contamination, and Plaintiffs have abandoned their medical monitoring claim. Nonetheless, they claim to suffer an increased health risk due to the contamination on their property and request remediation.

         Pending before the Court is Plaintiffs' Motion for Class Certification, ECF No. 778, pursuant to Rule 23 of the Federal Rules of Civil Procedure. Both the motion and the response in opposition are supported by expert testimony. Each side has moved to exclude the others' experts pursuant to Rule 702 of the Federal Rules of Evidence and Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), and those motions are also pending.⁴ See ECF Nos. 716, 717, 719, 721, 723, 724, 725, 726, 727, 728, and 729. The Court held an eight-day evidentiary hearing on all motions.⁵ Now, having fully considered the law, the voluminous record, and the arguments of the parties, the Court rules as follows.⁶

4 There are eleven Daubert motions pending. The Court previously granted a motion to strike as untimely the Plaintiffs' rebuttal expert reports of Dr. Michael J. Wade, Dr. William R. Sawyer, and Dr. George C. Clark (attached to ECF No. 705), and thus these reports have not been considered. See ECF No. 797 (striking the exhibits attached to ECF No. 705, consisting of rebuttal reports served by Plaintiffs on August 14, 2015).

5 At the class hearing, Plaintiffs presented live testimony from Dr. George Flowers, Dr. William Sawyer, Dr. Michael Wade, Dr. Jurgen Exner, and Dr. John Kilpatrick., and the deposition testimony of the proposed class representatives--Nancy Day, Betty Hewitt, and John Hewitt. The defense presented live testimony Dr. Allen Uhler; Michael R. Corn; Dr. Shari Libicki; Dr. Scott Phillips; Wayne Gripp; Trever Phillips; and Mitchell Brourman. Also admitted and relied on at least in part by both parties were the depositions of Plant Manager Randall D. Collins and Dr. Patricia Cline.

6 The case was originally filed in 2010 in state court and was removed to federal court, where it has experienced significant delays. By the time the case was reassigned to the undersigned in April 2013, it had already been stayed for several years at the request of the parties. The case was further prolonged by a period of jurisdictional discovery, which was necessary to address motions to dismiss. Then, following an extended period of class discovery and time for adequate briefing, the court held an eight-day hearing on Plaintiffs' Motion for Class Certification and the eleven pending Daubert motions. The undersigned regrets the amount of time it has taken to decide the pending motions, but, as the parties are well aware, the record is voluminous and the issues complex.

         I. Background

         A. The Koppers Site

         From 1916 through 2010, Koppers, Inc. and its predecessors operated a wood treatment facility on a 90-acre parcel of real property located at 200 NW 23rd Avenue in Gainesville, Florida, known as the Koppers Site, which is now owned by Beazer East. The western edge of the Site is adjacent to a residential neighborhood in Gainesville, known as the Stephen Foster neighborhood. At various times over the years, a number of different chemical preservatives were used to pressure treat wood on the Site, including creosote (1916 through 1991), chromate copper arsenate (1936 through 2009), and pentachlorophenal (" PCP" ) (1969 through 1987). PCP is known to produce chlorinated dioxins, specifically, polychlorinated dibenzodioxin and polychlorinated dibenzofuran (" PCDD/F" ), and is classified as a probable human carcinogen. Pla. Ex. 5 (Health Consultation Report July 2014); Def. Ex. 33.043 (EPA 2011 ROD). As owners and operators of the Koppers Site, Defendants have long known of PCP's adverse effects. In 1970, Dow Chemical issued an internal report warning Koppers of PCP's potential health and environmental hazards and of the consequent need to contain dust from its wood treating processes utilizing PCP.⁷ See Pla. Ex. 1. Despite the warning, it is undisputed that Koppers treated wood with PCP from 1969 through 1987, and it is also undisputed that soil and groundwater at and surrounding the Koppers Site were contaminated when PCP seeped into the soil and was deposited in unlined lagoons.⁸ Just how far off-Site the PCDD/F migrated is at the center of this case.

7 Dow's internal report recommended that the on-site PCP handling system be totally enclosed or equipped with ventilation controls to capture the dust, mist and/or fumes from the exhaust system. Pla. Ex. 1.

8 The record reflects that Koppers Company, Inc. discontinued its use of PCP at the Site in 1987 but continued pressure treating wood there through 2009, using other chemicals. This case is only concerned with the use of PCP.

         In 1983, the United States Environmental Protection Agency (" EPA" ) designated the Koppers Site as a Superfund Site and placed it on the EPA's National Priorities List.⁹ The EPA began investigating contamination at the Site in 1984 in collaboration with Beazer East, as well as the United States Department of Health and Human Services Agency for Toxic Substances and Disease Registry (" ATSDR" ), the Florida Department of Environmental Protection (" FDEP" ), and the Florida Department of Health (" FDOH" ). In 1990, after approximately six years of investigation, the EPA issued a Record of Decision (" ROD" ), finding that chemicals used in the wood-treating activities had migrated into the soil and groundwater on the Site, causing contamination. The EPA required soil excavation at the Site, among other remedial measures.¹⁰ The ROD also recommended continued investigation to evaluate potential off-Site contamination. As a result, throughout the intervening years, various testing campaigns of off-Site soil and dust have been conducted by the EPA, Beazer East, and consultants of the Plaintiffs, as discussed in more detail below, and soil remediation has been completed in a portion of the Stephen Foster neighborhood that is adjacent to the western boundary of the Site.

9 The EPA determined that the Cabot/Koppers Site (consisting of two adjacent companies, the Cabot Carbon Corp. and Koppers) met the criteria for Superfund designation in accordance with the provisions of the Comprehensive Environmental Response, Compensation, Liability Act of 1980 (" CERCLA" ), 42 U.S.C. § § 9601-9657, as amended by the Superfund Amendments and Reauthorization Act of 1986 (" SARA" ). See Def. Ex. 33 (EPA Record of Decision Summary of Remedial Alternative Selection Feb. 2011). The Superfund National Priorities' List is a national list of facilities throughout the United States known to have released hazardous substances. Using established criteria, the EPA sets priorities for taking remedial response actions, including investigation and cleanup. See https://www.epa.gov/superfund/superfund-national-priorities-list-npl. Much of the Cabot Carbon/Koppers Site has now been remediated, with some continued groundwater remediation ongoing. The Cabot Carbon portion of the Site, located east of Koppers, was a pine tar and charcoal generation facility that produced liquid solvent products on property zoned for commercial use, and it is now a shopping mall and parking lot for a car dealership. The Koppers portion, zoned industrial, is now vacant.

10 Today, the soil remediation at the Koppers Site is complete and a groundwater extraction system has been installed.

         B. Dioxin

         Some general information about dioxin may be helpful at this point.¹¹ Dioxins and furans are often referred to generically as " dioxin," but the term more accurately describes a large class of related chemical compounds and mixtures of congeners.¹² A dioxin molecule has eight positions (specifically numbered 1 through 8) where a halogen¹³ may attach, and the precise position and type of halogen is important in determining the toxicity of a specific dioxin molecule. Of particular concern in this case are chlorinated dioxins. Because there are eight positions where the chlorine can attach, there are 210 different congeners of chlorinated dioxins (and furans).

11 This chemistry discussion is grounded in undisputed expert testimony and reports presented by both sides. See generally Pla. Ex. 2 (" Understanding Dioxin-Like Compounds in Indoor Dust" ); Pla. Ex. 5 (Health Consultation on Indoor Dust prepared by the FDOH under a cooperative agreement with the ATSDR); Def. Ex. 123 (Expert Report of Michael Corn); Def. Ex. 137 (Expert Report of Dr. Allen D. Uhler). The following chemical abbreviations are used throughout this Order: pentachlorophenal--PCP; polychlorinated dibenzodioxin and polychlorinated dibenzofuran--PCDD/F; polybrominated dioxin/furan or dibenzo-p-dioxin/furan--PBDD/F; dioxin-like polychlorinated biphenyl--PCB; and 2,3,7,8 tetrachlorodibenzo-p-dioxin--TCDD.

12 Dioxins and furans share a similar molecular structure of two benzene rings fused together by either one oxygen atom (dibenzofuran) or two oxygen atoms (dibenzodioxin). For simplicity in this discussion, the Court's use of the term " dioxin" includes furans. A " congener" is " a chemical substance related to another" chemical substance. https://www.merriam-webster.com/dictionary/congener . A " compound" consists of atoms of two or more elements bound together, whereas a " mixture" consists of two or more different elements and/or compounds physically intermingled. https://www.chem.purdue.edu/gchelp/atoms/elements.html .

13 Merriam-Webster Dictionary defines " halogen" as " any of the five elements of fluorine, chlorine, bromine, iodine, and astatine that form part of group VIIA of the periodic table." https://www.merriam-webster.com/dictionary/halogen .

         Only 17 congeners, those with the chlorine atom in the 2, 3, 7, and 8 positions of the molecule, are considered toxic and are evaluated for associated health risks. Pla. Ex. 2, Pla. Ex. 5; Def. Ex. 137. The most toxic is 2,3,7,8 tetrachlorodibenzo-p-dioxin (" TCDD" ) (with the numbers showing the positions of the chlorine halogen within the molecule). The World Health Organization (" WHO" ) and the EPA have identified 2,3,7,8 TCDD as a probable carcinogen, and the toxicity of all other dioxins are measured in relation to it. See Pla. Ex. 170 (EPA summary discussing the health hazards of 2,3,7,8 TCDD). The WHO has studied chlorinated dioxins extensively in order to assess health risks. It has developed a standardized Toxicity Equivalence Factor (" TEF" ) to be assigned to the 17 toxic congeners based on a comparison to TCDD (the most toxic).¹⁴ The total " Toxicity Equivalence" (" TEQ" ) of a mixture is then determined by multiplying the concentration of each congener by its respective TEF and adding them together.¹⁵ The TCDD-TEQ value is then used to assess health risks based on the toxicity of the chlorinated mixture or compound.

14 Pla. Ex. 2 includes Table 1-2, which shows the World Health Organization's consensus TEFs for chlorinated dioxins/furans.

15 According to Dr. Patricia Cline, " [t]he TCDD-TEQ is calculated by multiplying the concentration of each of the 17 congeners by its respective TEF. The TCDD-TEQ of the mixture is calculated by summing the TEQs of each of these 17 congeners." Pla. Ex. 2 (" Understanding Dioxin-Like Compounds in Indoor Dust" ); see also Def. Ex. 137.

         The toxicity of PCDD/F, which refers generally to the group of chlorinated dioxins (as distinguished from a brominated dioxins) can be accurately measured and quantified by a TCDD-TEF equivalency factor. PCDD/F is indisputably associated with the wood-treating activities at the Koppers Site. According to several of the parties' experts, PCDD/F is also widely accepted as a by-product of various other industrial activities as well, including waste incineration, wastewater, and automobile and diesel combustion. See e.g., Pla. Ex. 5; Cline Depo. Additionally, the experts agree that PCDD/F is a known byproduct of residential activities, including indoor or outdoor wood burning in stoves, fireplaces, and grills, and is produced by cigarette smoke and combustion from automobiles and lawnmowers. For this reason, PCDD/F is considered ubiquitous in an urban environment, where it is widely distributed at low concentrations.

         When the halogen in dioxin is a bromine, i.e., polybrominated dioxin/furan or dibenzo-p-dioxin/furan (" PBDD/F" ), as opposed to a chlorine, its toxicity is notably different. PBDD/F has not been as extensively studied as chlorinated compounds (including PCDD/F), and consequently, the EPA continues to study its impacts. Dr. Cline explained that there are no standardized TEFs particular to PBDD/F, which results in uncertainties when interpreting its relative toxicity levels, although it can be assigned a TCDD-TEQ. Pla. Ex. 2. Like PCDD/F, PBDD/F also exists in an urban environment. It is associated with common flame retardants in household furnishings and upholstery, and therefore, PBDD/F is commonly found in varying degrees within a household.

         It is undisputed that PCDD/F is associated with the Koppers Site and PBDD/F is not, and thus, Defendants are not responsible for any contamination from PBDD/F. Therefore, when considering the toxicity of dioxin contamination in soil or dust for purposes of determining liability in this case, it is critical to distinguish between PCDD/F and PBDD/F. Also, because PCDD/F can originate from different sources in an urban environment, and Defendants can be liable only for PCDD/F that originated from the Koppers Site, it is necessary to identify the precise source of any PCDD/F found on Plaintiffs' properties. Through a process known as " chemical fingerprinting," it is scientifically possible to identify the source of PCDD/F in a given sample. This process allows chemists to analyze and compare the specific structures of congeners within PCDD/F to establish its chemical fingerprint and thereby connect it to a particular source.¹⁶

16 In this case, Arcadis performed chemical fingerprinting in its report for Beazer East. In support of the Motion for Class Certification, Plaintiffs presented the testimony of a chemical fingerprinting expert, Dr. Michael Wade, whose testimony is challenged by Defendants' chemical fingerprinting expert, Dr. Allen Uhler, as discussed below.

         C. Dioxin Testing

         In this case, two types of dioxin testing were performed on soil and dust samples. The tests are (1) EPA Method 1613 High Resolution Gas Chromatography/High Resolution Mass. Spectrometry (" GC/MS" ) and (2) EPA Method 4435, entitled " Screening for Dioxin-Like Chemical Activity in Soils and Sediments using the CALUX® Bioassay and TEQ Determination" (" CALUX" ).¹⁷ GC/MS testing separates the individual components of each congener of a particular compound, measures the concentrations of each congener, and calculates the total TCDD-TEQ of the 17 chlorinated congeners in dioxin (PCDD/F). GC/MS can be used to separately detect brominated (PBDD/F) toxicity values as well, but with less accuracy. The FDOH considers GC/MS to be a " long-established method [and] the definitive standard for determining dioxin concentrations in soil." Pla. Ex. 5. In fact, most experts, including those in this case, agree that GC/MS is the " gold standard" for detecting and accurately quantifying chlorinated dioxin toxicity values. Indeed, plaintiff and defense experts alike testified that GC/MS results can be used for accurately quantifying PCDD/F as well as identifying sources of PCDD/F through forensic chemical fingerprinting.

17 " CALUX" stands for Chemical Activated Luciferase Expression. In this Order, CALUX measurements are stated in terms of " CALUX-TEQ" and GC/MS measurements are stated in terms of " TCDD-TEQ."

         The CALUX bioassay is an EPA-approved method of screening for dioxin contamination, estimating dioxin concentrations, and identifying a need for further testing.¹⁸ CALUX responds to all " dioxins and dioxin-like" chemicals and gives a total toxic potency (" CALUX-TEQ" ) without separately quantifying PCDD/F TEQ values or distinguishing between chlorinated or brominated dioxins or other dioxin-like compounds.¹⁹ Dr. Patricia Cline testified that CALUX test results merely report all dioxin and dioxin-like chemical concentrations " in bulk," i.e., without distinguishing between PCDD/F, which is attributable to Koppers, and PBDD/F, which is not, nor can it identify sources of PCDD/F. Cline Dep. at 48-49. This is not in dispute among the experts, and no expert in this case knew of a means of extracting PBDD/F results from a total CALUX measurement.

18 CALUX is described by the EPA as " a bio-analytical screening procedure," and " a relatively rapid screening method capable of estimating a [CALUX-]TEQ concentration for dioxin-like chemicals in a sample." See Pla. Ex. 10. This bioassay measures the response of all dioxin and dioxin-related chemicals (multiple chemicals that may show dioxin-like responses) that activate the Aryl Hydrocarbon Receptor. The bioassay provides one total measure of dioxin-like toxicity but does not distinguish between chlorinated and brominated responses. Pla. Ex. 2. The EPA has expressly validated the CALUX as a screening method for soil but has not validated it for dust. According to Dr. Cline, the bioassay " can provide valuable information on the total dioxin-like response of various environmental mixtures, leading to additional studies that identify specific compounds and/or additional toxicity studies." Pla. Ex. 2.

19 For example, polychlorinated biphenyl (" PCB" ) is an " other dioxin-like" chemical that responds to CALUX. However, a process exists for separating PCB from the CALUX total dioxin measurement. Some of the CALUX-tested samples in this case were pretreated to exclude PCB, although it is not clear whether they all were. Nonetheless, PCB is not at issue, and what is important for purposes of this case is that, unlike PCB, there is no process for separating PBDD/F from PCDD/F in the CALUX testing.

         D. Agency Action

         From 2008 through 2010, Beazer East, in coordination with the EPA, conducted independent off-Site soil testing through an environmental consultant company, Arcadis U.S., Inc. (" Arcadis" ), which collected and analyzed the off-Site soil samples for the purpose of determining how far contamination from the Koppers Site may have migrated into the residential neighborhoods surrounding the Site.²⁰ Arcadis analyzed the samples for TCDD-TEQ values using EPA Method 1613, GC/MS, and determined that within the first 100 feet of the western boundary of the Site, extending into a portion of the Stephen Foster neighborhood, approximately 73% of the residences sampled had TCDD-TEQ toxicity values in soil above Florida's residential target cleanup level of 7 ppt²¹ (GC/MS), but not exceeding EPA limits.²² The toxicity values of PCDD/F decreased dramatically approximately 100 feet away from the Site, and beyond. Arsenic and creosote were also chemicals attributable to Koppers, but Arcadis's testing showed that only chlorinated dioxin (PCDD/F) TCDD-TEQ values exceeded state regulatory levels. The Arcadis report included a " chemical fingerprinting" analysis, which showed that " other sources of PCDD/Fs exist in the vicinity of the Site and are influencing TCDD-TEQ concentrations in several off-Site samples." Def. Ex. 89.010.

20 See Def. Ex. 137, Report of Allen D. Uhler, Ph.D. and Def. Ex. 89, Arcadis Off-Site Data Summary and Fingerprinting Evaluation; Def. Ex. 90, Arcadis Updated Soil Off-Site Data Summary and Fingerprinting Evaluation.

21 The record inconsistently refers to measurements in terms of pg/g, ng/kg, or parts per trillion (ppt). An EPA letter explains that pg/g is the equivalent of ng/kg, Def. Ex. 159, and both represent parts per trillion. Therefore, for the sake of consistency, the Court will refer to all measurements as ppt.

22 Florida's soil cleanup standard of 7 ppt TCDD-TEQ was established in 2005 and is associated with a statistical health risk of 1 x 10-6 (1 in 1,000,000). See Pla. Ex. 2. The EPA's then-existing residential preliminary remediation goal was 1,000 ppt TCDD-TEQ. See Pla. Ex. 2, Def. Ex. 89, at 89.009.

         The EPA engaged the Gainesville community in its investigative and remedial efforts. The agency interviewed community members, kept them informed of findings through meetings and reports, and notified them of the availability of a technical assistance grant (TAG), which could be used to employ a consultant to educate the community about the issues and also to provide technical reports to the EPA. In 2009, Gainesville community members organized a nonprofit organization called Protect Gainesville's Citizens, Inc., which obtained a TAG, and in August 2010, used the TAG funding to hire a technical advisor and principal investigator. The group hired Dr. Patricia Cline, an environmental chemist.²³ Dr. Cline provided several independent technical assistance reports to the Gainesville community and to the EPA, explaining the various testing by the agencies and other consultants. In 2010, as discussed further below, representatives of some of the residents in the Stephen Foster neighborhood analyzed indoor dust samples using CALUX and reported TEQ values (CALUX-TEQs) much higher than had been measured in soil (GC/MS TCDD-TEQs). In response to citizen concerns over the possibility that reports of soil contamination were underestimating the potential impact from dust contamination inside homes near the Site, the FDOH formed an Indoor Dioxin Dust Work Group (" Dust Work Group" ) to study the issue of dust contamination. The Dust Work Group consisted of various county, state, and federal agency representatives, including Dr. Cline.

23 Dr. Cline had previously been a consultant for the City of Gainesville in connection with the earlier investigations that were conducted at the Koppers Superfund Site. Dr. Cline felt that the community did not understand the technical aspects of the EPA's information or the significance of distinctions between the soil and dust testing. She therefore left the City to aid the community group more directly.

         In 2011, the EPA issued another ROD which, among other things, listed a cleanup goal for the off-Site soil that had tested above Florida's soil target cleanup level (7 ppt TCDD-TEQ) and stated off-Site testing was ongoing. Def. Ex. 158. In the absence of any regulatory target cleanup level for indoor dust, the EPA established a Site-specific " health protective" level of 190 ppt TCDD-TEQ for evaluating chlorinated dioxins in indoor dust in Gainesville related to the Koppers Site, which represents a statistical health risk of 1 x 10^-5 (1 in 100,000).²⁴ See Def. Ex. 159 (Oct. 4, 2012, EPA Memo of Kevin Koperec, a toxicologist for the Superfund Support Branch of the EPA).

24 The EPA memo establishing this health protective benchmark noted that to achieve a health risk for dust of 1 x 10-6 as Florida had established for soil, the benchmark would be 19 ppt TCDD-TEQ. The EPA toxicologist, Kevin Koperec, found this unnecessary and instead recommended 190 ppt as the health protective level (a risk of 1 x 10-5 ) for evaluating the EPA's indoor dust sampling related to the Koppers Site because it was in the middle of the target cancer risk range. Def. Ex. 159. To put this into context, by FDOH standards, a risk of 1 x 10-5 is considered " very low." See infra Note 32.

         In 2012, the EPA conducted indoor floor dust testing in a number of Gainesville residences located west, northwest, and southwest of the Site in coordination with the FDOH's Dust Work Group. The concern was that contaminated soil and dust in yards were being tracked into houses as floor dust, and thus samples were collected from rugs, carpets, and hard surface floors. A small portion of the Stephen Foster neighborhood in Gainesville nearest to the Koppers Site was the focus of intense regulatory scrutiny because of the Arcadis Report's conclusion that soil was contaminated near the Site but that toxicity levels of PCDD/F ( i.e., TCDD-TEQ levels) decreased dramatically approximately 100 feet away. The EPA collected and analyzed over 30 indoor floor dust samples--17 from houses near the Koppers Site in a portion of the Stephen Foster neighborhood adjacent to the Koppers western fence line and about a block deep (which is also within, but much smaller than, the Plaintiffs' proposed class boundary for this case) and 13 samples from " unaffected" houses in Gainesville (located approximately 2 miles or more northwest of the Koppers Site and southwest of the Site and also outside the Plaintiffs' proposed class boundary), which were used to determine the relevant background for comparison.²⁵ See Pla. Ex. 2; Def. Ex. 74 (EPA Indoor Dust Study Data Report, EPA Region 4, 2013). The samples were then analyzed using GC/MS, giving toxicity values for PCDD/F and PBDD/F, and a subset of samples were analyzed using the CALUX method.²⁶

25 The " background" is the average toxicity of chlorinated dioxin contamination that exists in an area similar to, but presumably unaffected by, the Koppers Site.

26 The EPA's testing was in response to the CALUX screening already performed in 2010 by consultants of some residents. It is unclear whether the EPA used CALUX concurrently with the GC/MS testing as a standard protocol or simply to compare with the 2010 screening results.

         In 2013, the EPA issued a report evaluating the 2012 floor dust test results. Def. Ex. 74. According to the report, GC/MS results showed the presence of chlorinated dioxin (PCDD/F) in all of the floor dust samples, both near-Site (which, again, means a small portion of the Stephen Foster neighborhood closest to the Koppers Site) and in background samples (approximately 2 miles from the Site). The background GC/MS-tested samples ranged from 2.66 - 77.3 ppt TCDD-TEQ, with a geometric mean of 13.6 ppt TCDD-TEQ, whereas the near-Site GC/MS PCDD/F results ranged anywhere from 6.78 - 90.9 ppt TCDD-TEQ, with a geometric mean of 27.9 ppt TCDD-TEQ. Although the geometric mean of the near Site GC/MS TCDD-TEQ values was somewhat higher overall in the near-Site samples (within approximately one block of the Site) than in background samples, the EPA concluded that the near-Site chlorinated TCDD-TEQ levels were still " generally within the concentration range" of background. Notably, no near-Site or background chlorinated TCDD-TEQ level exceeded the 190 ppt Site-specific health protective level established by the EPA for indoor dust in Gainesville, and all were significantly below that benchmark level.

         The EPA report noted that the brominated dioxin TCDD-TEQ value exceeded the chlorinated values in the same house in over half the samples. Only one floor dust GC/MS result in the near-Site Stephen Foster neighborhood exceeded the EPA's 190 ppt TCDD-TEQ health risk benchmark, and this result was for PBDD/F, which the EPA attributed to sources other than Koppers.

         Regarding the concurrent CALUX testing that was done, the EPA report noted inconsistencies when comparing the CALUX results to the GC/MS results. Overall, the CALUX-TEQ values were higher than the GC/MS-derived TCDD-TEQ values for chlorinated dioxin alone, but not always. In some samples, the CALUX TEQ values reflected a measurement similar to the sum of the GC/MS samples for chlorinated and brominated dioxins, but again, many times it did not. Also, charts in the report show that the average near-Site CALUX-TEQ values (a geometric mean of 31.78 ppt CALUX-TEQ) was virtually indistinguishable from the average of the background CALUX TEQ values (a geometric mean of 31.59 ppt CALUX-TEQ), reflecting virtually no elevation of the near-Site samples over background. See Def. Ex. 74 (Tables 1 & 2).

         Additionally, the EPA report stated that highest estimated increased risk of cancer due to accidental ingestion of floor dust was 1 x 10^-5 or 1 in 100,000, which is considered " very low." See infra Note 32. In conclusion, the EPA found that indoor dust remediation was not required but that the most contaminated soil, which was located within a small portion of the Stephen Foster neighborhood (about one block deep around the Site), should be remediated to Florida's regulatory soil target level of 7 ppt TCDD-TEQ.²⁷ The EPA explained that by remediating the soil, the risk of potential future exposure to residents from soil-related dust being tracked into homes as floor dust would be reduced.

27 Consistently, the Arcadis report, updated in November 2013, also concluded that soil closest to the Site exceeded Florida's target cleanup level but that toxicity values dropped quickly at a distance of approximately 100 feet from the Koppers Site, and beyond that, soil levels were " not associated with unacceptable risks." Def. Ex. 90.

         After the test results were shared with the Gainesville community (including the results of CALUX testing done by Plaintiffs' consultants, as discussed below), some in the community expressed " outrage" because the reported dust dioxin concentrations and the EPA's 190 ppt health benchmark for indoor dust were so much higher than Florida's soil remediation level of 7 ppt TCDD-TEQ.²⁸ In September 2013, Dr. Cline presented a power point presentation to the Community in which she explained that, although the wood treating processes at Koppers had elevated the PCDD/F levels in soil and dust near the Site, and acknowledged that levels appear to be higher overall in dust than in soil, there was no correlation in the data between the soil and dust results, which she said indicated that different sources were impacting the indoor dust results. She further explained that GC/MS testing had confirmed that the higher dust test results were predominantly due to brominated dioxin (PBDD/F) concentrations, which are attributable to sources inside the homes such as flame retardants, electrical cords, cables and computers, and not the Koppers Site.²⁹ See Def. Ex. 75. Dr. Cline stated that the community's concern over reportedly high dioxin concentrations in indoor dust was due to the use of the CALUX method, which, she explained quantifies a total dioxin-like response and could not, and did not, reliably report the PCDD/F in dust from Koppers. According to Dr. Cline, reporting CALUX results as if they were TCDD-TEQs " and inferring [that] the level of confidence in the [bio]assay is equivalent to results using the definitive method [GC/MS], is misleading to the community." Pla. Ex. 2. Dr. Cline said that the Dust Work Group would not recommend the CALUX bioassay for future dust testing without additional GC/MS support because the CALUX-TEQ tended to overstate toxicity. She testified by deposition simply that, " you can't attribute all of [a CALUX measurement] to chlorinated dioxins." Cline Dep. at 48.

28 The EPA notified residents of the test results by letter. Plaintiff Day, a property owner who participated in the dust sampling, received a letter from the EPA Remedial Project Manager dated November 29, 2012, informing her that the level found in her home was 44.9 ppt TCDD-TEQ and that this did not exceed " any applicable EPA health-based benchmarks" given that the EPA risk-based screening value for dioxin exposure to dust in homes was 190 ppt TCDD-TEQ. Def. Ex. 93 (letter of Scott Miller). Therefore, remediation of indoor dust in her home was not necessary. Day's husband, Kim Popejoy, was president of the community group Protect Gainesville's Citizens. He testified by deposition that the EPA's letters caused concern in the community because of the wide disparity between Florida's low soil target cleanup level of 7 ppt TCDD-TEQ and the EPA's stated health protective level of 190 ppt TCDD-TEQ for dust inside their homes. He testified that citizens were especially concerned because the group's technical advisor, Dr. Cline, had advised them that there was no U.S. standard for evaluating the health effects of dioxin in indoor dust. Dr. Cline explained, however, that soil and dust are very different mediums. Notably, Plaintiffs have not argued that Florida' soil cleanup target level of 7 ppt is the appropriate standard for evaluating indoor dust levels.

29 Defense Exhibit 75 also includes a letter attached to the power point, in which EPA Region 4 Director, Franklin Hill, informed a citizen that the testing had shown the presence of dioxins in all house dust samples, including background, but the EPA had concluded that the levels did not require remedial action, and the EPA and Dr. Cline's group were continuing to study dioxins in indoor dust.

         In 2014, the FDOH, under a cooperative agreement with ATSDR, issued a Health Consultation Report based on the EPA's 2012 floor dust testing. Pla. Ex. 5 (July 24, 2014). Reviewing the EPA's data, the FDOH agreed that the CALUX TEQ tended to overstate toxicity and likewise noted a lack of any consistency in the rate of variance between the GC/MS chlorinated TCDD-TEQ values and the CALUX-TEQs.³⁰ See Pla. Ex. 5, Tables 2 & 3. The FDOH found the variability between CALUX results and GC/MS results too great to provide any reliable estimates of toxicity and consequently decided that the CALUX bioassay method could not be used to assess a health risk to the residents in the Stephen Foster neighborhood near the Site. Like Dr. Cline, the FDOH cautioned that the higher dioxin concentrations in near-Site samples did not mean that the Koppers Site was the only source, or even the major source, of dioxins affecting the indoor dust in near-Site homes. Similar to the EPA, the FDOH also calculated a statistical health risk. The FDOH applied a standard risk formula (Risk=Dose x Slope Factor), using the highest indoor dust GC/MS chlorinated TCDD-TEQ level detected and assuming an indoor dust ingestion dose of 11 milligrams per day for adults,³¹ Pla. Ex. 5, at 12-15, and, like the EPA, found that the increased risk of cancer was 1 x 10^-5 or 1 in 100,000, or " very low." Because the FDOH takes action to protect public health only when the estimated increased cancer risk is in the " moderate" risk category or higher, no remedial action was recommended.³²

30 For instance, at House A in the Stephen Foster neighborhood, the PCDD/F sample measured 11.4 ppt TCDD-TEQ (GC/MS) and the CALUX result was slightly higher at 15.30 CALUX-TEQ, whereas at House E, the PCDD/F sample measured only 8.92 ppt TCDD-TEQ (GC/MS) but the CALUX result was much higher at 42.39 ppt CALUX-TEQ. Also, considering the averages of the test results, the CALUX data again was higher, but the background CALUX geometric mean of 31.59 ppt CALUX-TEQ was nearly identical to the near-Site geometric mean of 31.78 ppt CALUX-TEQ. The soil results were different in that the FDOH found some correlation between the CALUX soil results and the GC/MS soil results. No such correlation was identified between the CALUX and the GC/MS floor dust sample results, however.

31 As authority for selecting a dust ingestion rate of 11 mg for adults in the risk calculation, the FDOH report relies on a June 13, 2011 letter from Professors Leah D. Stuchal and Stephen M. Roberts of the University of Florida's Center for Environment & Human Toxicology. Pla. Ex. 222. The June 2011 letter was responding to an inquiry from the FDEP, requesting the development of a set of assumptions to be used for evaluating the risks associated with off-Site contamination from the Koppers Site. The professors considered a dose of 11 mg appropriate for adults based on six then-recent studies summarized in the Dutch National Institute for Public Health and the Environment (RIVM 2008). The EPA's toxicologist, Kevin Koporec, used this 11 mg in his risk assessment in 2012 because it had been agreed to by the Dust Work Group. Def. Ex. 159, Def. Ex. 105.

32 The FDOH evaluates statistical risk using a set of numerical standards. A " moderate" risk category is defined as a statistical risk of 1 x 10-3 or 1 in 1,000. Categories that do not call for action to protect public health include a " low" increased risk of cancer (1 x 10-4 or 1 in 10,000), a " very low" increased risk (1 x 10-5 or 1 in 100,000), and an " extremely low" increased risk (1 x 10-6 or 1 in 1,000,000).

         Also in 2011 and 2012, the FDOH reviewed the cancer rates in Census Tract 3 of Alachua County, which includes the Stephen Foster neighborhood in Gainesville, and issued reports that showed no increased cancer rates over the prior 30 years for the 18 types of cancer studied, although a few elevations were noted for specific types of cancer. The 2012 study concluded in part:

In summary, there appears to be no increase in overall rates for cancers of the liver, kidney, bladder, and pancreas and for non-Hodgkin lymphoma or leukemia during 1981-2008. These are cancer types most frequently associated with exposure and contamination of communities. Examining rates (primarily in 5 year periods) since 1981 revealed only a recent small increase during 2006-2008 in liver and pancreatic cancers (but only in the all ages category). . . . An elevated rate for esophageal cancer among ages 65 and older during 1981-2008 was noted. . . .

Def. Ex. 84. Also, the 2012 report concluded that the review " generally did not reveal elevations in cancers of concern most likely to be associated with chronic exposure to contamination. . . . [And] many cancers showed significantly lower overall rates in the community." Def. Ex. 84.

         In mid-to-late 2014, Beazer East undertook to remediate the near-Site soil at approximately 100 residences within the near-Site portion of the Stephen Foster neighborhood, located just west of the Koppers Site, where the soil TCDD-TEQ values were above Florida's residential soil target cleanup level. According to Mitchell Brourman, a geologist who managed the off-Site soil remediation for Beazer East,³³ Beazer East removed all contaminated soil extending one foot deep where the dioxin concentrations exceeded Florida's benchmark and replaced it with clean soil. All vegetation was removed and landscaping was replaced. Additional soil remediation was conducted on a few commercial sites to the south and a city easement along the western border of the Koppers Site. Based on the EPA and FDOH recommendations, Beazer East did not remediate indoor dust in homes within the Stephen Foster neighborhood but provided each homeowner with a monetary payment based on the square footage of their home to cover the cost of ordinary indoor cleaning, including dusting and vacuuming, after the soil remediation process was completed.³⁴

33 Brourman managed the environmental liabilities of the former Koppers Company and worked with the agencies in the cleanup effort for many years as an employee and contract employee of Three Rivers Management.

34 Plaintiffs' remediation expert, Dr. Jurgen Exner, testified that the dust could be effectively removed from the home using a vacuum with a HEPA filter (High Efficiency Particulate Air filter), wiping areas with a wet cloth, and cleaning out ducts.

         Afterwards, Dr. Cline's Dust Work Group conducted some limited post-remediation GC/MS testing of living space dust in five homes in the Stephen Foster neighborhood. See Pla. Ex. 2. These tests confirmed that after soil replacement and normal cleaning in the homes, the chlorinated TCDD-TEQ values in floor dust in four of the five homes re-tested were comparable to the average background GC/MS levels. In the fifth home, the floor dust concentrations had increased, but Dr. Cline attributed this to the use of a wood-burning stove in the home.

         On November 30, 2014, to close out the Protect Gainesville group's TAG grant, Dr. Cline issued a final technical report entitled " Understanding Dioxin-Like Compounds," in which she summarized and explained the work and findings of the EPA and FDOH Dust Work Group. Pla. Ex. 2. Her review of the dust data reflected that " [t]he lowest and highest dioxin TEQs were in homes that were nearly adjacent" to each other, suggesting that a source internal to the home was impacting the concentrations. She reiterated that wood burning stoves, fireplaces and cigarette smoke all can produce PCDD/F. Additionally, she noted that PBDD/F (which also impacts CALUX measurements but cannot be separately identified through CALUX) is common in household flame retardants found in materials such as upholstery and mattresses.³⁵ Although Dr. Cline said she did not have all of the data necessary for detailed chemical fingerprinting of specific PCDD/F congeners because some of the information was provided to her in summaries, she did observe some " contributions of contaminants associated with the former wood treating sites" in the GC/MS data. She cautioned, however, that contamination should not be attributed to a Superfund site " without sufficient confirmatory analyses," and she said it was misleading to use a bioassay such as CALUX and report its results as if it were equivalent to a definitive method.

35 Similarly, in her deposition in this case, Dr. Cline testified that the CALUX data overstated the TCDD-TEQ values and was not reliable to identify the Koppers Site as the source of indoor dust contamination because, unlike soil, there are many factors inside of a home that can contribute to a CALUX reading.

         E. Plaintiffs' Indoor Dust Testing

         Plaintiffs tested indoor living quarters house dust in 2010 and again in 2014. In 2010, Plaintiffs hired consultants Tencon, Inc. (" Tencon" ) to complete independent off-Site testing at residential properties near the Koppers Site but covering a much larger area than the EPA testing covered. Plaintiffs' expert, Dr. George Flowers, designed the testing campaign to measure the toxicity of indoor living quarters dust, hypothesizing that dust had been transported by an atmospheric pathway into homes, as opposed to the EPA's testing, which focused on whether soil being tracked into the home as floor dust. Tencon collected 105 indoor living area dust samples using a High Efficiency Particulate Air (" HEPA" ) filter vacuum, consistent with EPA protocol, from randomly selected homes within a two-mile radius of the Koppers Site (which includes all of the Stephen Foster neighborhood and is larger than the 1.8 mile proposed class boundary).³⁶ The samples were analyzed under CALUX, while a subset of 10 indoor living space dust samples were separately analyzed using GC/MS.

36 The EPA considered homes that were two miles from the Site to be " background."

         Approximately four years later, in early 2014, another Plaintiffs consultant, Water and Air Research, Inc. (" Water and Air" ), collected more samples of soil, indoor living space dust, and attic dust, again using a HEPA filter vacuum, consistent with Tencon's 2010 sampling. Dust samples were collected from residences in 28 zones along a transect extending from the Koppers Site to the west approximately 19 miles and additional samples were again collected from residences generally within two miles of the Site. The majority of the samples were analyzed using CALUX, but a subset of 12 indoor living dust samples and 9 attic dust samples were analyzed using GC/MS. In total, during the 2010 and 2014 testing campaigns, Plaintiffs' consultants gathered a data set of 139 indoor living space dust samples and 34 attic dust samples.³⁷ Plaintiffs' experts analyzed the results of this CALUX testing and formed opinions about contamination levels, the boundary of the proposed class, and source based, as discussed below.

37 The Plaintiffs' consultants also collected and analyzed soil samples. However, because the Plaintiffs request remediation of indoor dust contamination only, not soil, the Court finds it unnecessary to discuss the soil samples for purposes of class certification.

         II. Hearing on Class Certification and Daubert Motions

         As noted, the Court held a lengthy evidentiary hearing on the Plaintiffs' Motion for Class Certification and the parties' Daubert motions.          A. Plaintiffs' Experts

         1. Dr. William Sawyer & Dr. George Flowers

         The proposed class boundary is a polygonal area surrounding the Koppers Site and extending outward approximately 1.8 miles in all directions and includes all current residential property owners within that area.³⁸ See Pla. Ex. 131. It should be noted that the proposed class area is much larger than the area within the Stephen Foster neighborhood containing the 100 residences that were remediated. To establish the proposed class boundary, Plaintiffs relied on the opinions of Dr. William Sawyer, a toxicologist,³⁹ and Dr. George Flowers, a geochemist and environmental engineer,⁴⁰ whose opinions are based on the Plaintiffs' 2010 and 2014 CALUX testing of indoor living quarters dust.

38 In the Second Amended Compliant, the proposed class area identified a 2-mile radius around the Koppers Site. In her deposition, Dr. Cline criticized Plaintiffs' originally proposed 2 mile class boundary as arbitrary and not defensible because the " house dust concentrations are too variable." Cline Dep. at 55. Plaintiffs modified the boundary of the proposed class in their Motion for Class Certification in light of their experts' testimony. Dr. Sawyer testified that dust is not likely to travel beyond 1.8 miles.

39 Dr. Sawyer is a professional toxicologist and holds a doctorate in toxicology and a Master's Degree in cellular and molecular biology.

40 Dr. Flowers has been an environmental consultant for approximately 25 years and also teaches at Tulane University.

         Dr. Sawyer considered all of the Plaintiffs' 2010 and 2014 CALUX results, which included samples collected within two miles of the Site and those collected along the transect extending 19 miles west of the Koppers Site. In the absence of an indoor dust regulatory target level for PCDD/F, Dr. Sawyer established his own benchmark level based on the Plaintiffs' indoor living dust CALUX samples and the EPA's floor dust CALUX samples, and this benchmark was then used to establish a proposed class boundary, as well as to determine whether there was an increased risk of cancer in the proposed class area. Dr. Sawyer first determined the average background CALUX-TEQ by adding Plaintiffs' indoor living quarters dust test results, collected from 1.8 miles away from the Site and beyond, plus the EPA's CALUX-TEQ for floor dust samples, collected 2 miles away and beyond, and finding the arithmetic mean was 40.33 ppt CALUX-TEQ. He then calculated the 95% upper confidence limit of that average to arrive at a benchmark of 57.08 ppt CALUX-TEQ as the upper limit of background.⁴¹ Dr. Sawyer found that the near Site CALUX-TEQ values on average were significantly elevated above background.⁴² Thus, he concluded that homes with CALUX-TEQ values in living quarters dust above this background level should be remediated. He did not analyze any attic dust samples.

41 Dr. Sawyer described the benchmark as " the cutoff value for urban background," meaning there is 95% certainty that homes with contamination above this level cannot be considered background. Thus, according to Dr. Sawyer, homes with contamination above this level are contaminated above a background level found in an average home and need remediation. The FDOH's Health Consultation Report defines " background level" as the " average or expected amount of a substance or radioactive material in a specific environment, or typical amounts of substances that occur naturally in an environment." Pla. Ex. 5.

42 Specifically, Dr. Sawyer compared the background benchmark of 57.08 ppt CALUX TEQ for background (using both EPA and Plaintiffs' samples) against the average of near-Site samples using only Plaintiffs' CALUX-TEQ values, which was 124.79 ppt CALUX-TEQ (Pla. Ex. 7, Table 6). Dr. Sawyer did not include the EPA's near-Site CALUX results, which, by contrast, averaged 31.78 ppt CALUX-TEQ. Dr. Sawyer explained that he combined the sample sets to determine background because he calculated that the two sets were statistically the same. See Pla. Ex. 7 (Tables 2, 3). Notably, however, he did not combine them when determining the average near-Site value, citing a statistical difference between the two near-Site sample sets, but Dr. Sawyer did not explain why it was otherwise acceptable to combine and compare the toxicity results of two different media.

         Dr. Flowers considered Plaintiffs' CALUX results in relation to Dr. Sawyer's benchmark level of contamination, 57.08 ppt CALUX-TEQ, which he also understood was the level above which a home would need remediation. He did not consider the EPA's test results. He applied a spatial distribution analysis to show the geographical distance between the Koppers Site and the location of all of the CALUX samples. See Pla. Ex. 130 (Flowers Report, Figures 11, 12, 14). Dr. Flowers said this allowed him to draw the proposed class boundary at 1.8 miles for indoor dust contamination based on his " examination and judgment." See Pla. Ex. 130 (Flowers Report, Figure 2). Through this analysis, Dr. Flowers identified a pattern or " ramp up" of chlorinated dioxin levels in indoor living quarters dust in relation to the Koppers Site by considering samples collected along a line beginning 19 miles west in a rural area and leading to the Koppers Site in Gainesville. Along this line, he noted an increase in the CALUX-TEQ values closer to the Site. (No other transects were sampled.) From this " ramp up," Dr. Flowers concluded that the Koppers Site is a geographical point source of dioxin, and in fact, a major source. Although Dr. Flowers acknowledged variability in the data, such that some individual homes closer to the Site had lower CALUX-TEQ values than some farther away, he did not consider or analyze any potential alternate sources of dioxin. Instead, Dr. Flowers said that through the use of statistics and a large number of samples, the variabilities from house to house would average out. Similarly, Plaintiffs' expert Dr. Sawyer testified that he also saw no other plausible sources in the area and no other source to explain the " run up" in the data.⁴³ Dr. Sawyer acknowledged that dust sampling is " highly variable" from home to home, but he too disregarded this concern in this case, saying that the variances would tend to average out with this large of a sample set.

43 Defendants objected to Dr. Sawyer's testimony on source because it was not included in his report. Again, the undersigned indicated that the deposition testimony would be reviewed and has confirmed that he did discuss the issue in his deposition; therefore, the testimony is admitted. However, the undersigned will take into consideration that Dr. Sawyer's formal report included no such analysis.

         In addition to a spatial distribution analysis, Dr. Flowers applied a lognormal distribution analysis to estimate the probability that a house in the proposed class area would have contamination above Dr. Sawyer's 57.08 ppt CALUX-TEQ background benchmark. Dr. Flowers concluded in his report that there is a 66% probability that any given house within the proposed class area will contain living quarters dust contaminated with dioxin in excess of the benchmark.⁴⁴ Although Dr. Flowers made no attempt to explain his lognormal distribution calculation or how he arrived at this probability in his report and did not mention it in his hearing testimony, Plaintiffs rely on his opinion of a 66% probability in their Motion for Class Certification. Dr. Flowers conceded at the hearing (as the 66% probability also suggests) that not every property within the proposed class area will have dust contaminated above the 57.08 ppt CALUX-TEQ benchmark. In his opinion, it is more likely than not that any given home in the proposed class will have dust contamination exceeding that level.⁴⁵           Dr. Sawyer, in addition to establishing the CALUX-derived background benchmark of 57.08 ppt CALUX-TEQ, also considered the health risk from indoor living quarters dust in homes with CALUX-TEQ values exceeding that benchmark.⁴⁶ To determine the risk, Dr. Sawyer applied the 57.08 ppt CALUX-TEQ benchmark to the ATSDR's inhalation exposure dose equation.⁴⁷ Estimating factors such as body weight, exposure, daily dose, Pla. Ex. 7, at 28-29, and assuming an ingestion rate of 30 mg, as prescribed in the EPA Exposure Factors Handbook (2011 Edition), Pla. Ex. 18, Dr. Sawyer calculated an increased cancer risk from indoor dust ingestion in the proposed class of 4.09 x 10^-5. Pla. Ex. 7 (Table 12). Dr. Sawyer sharply criticized the EPA and FDOH for using a lower ingestion rate of 11 milligrams per day in their health risk assessment, see Pla. Ex. 222 (letter from University of Florida Center for Environment & Human Toxicology), rather than the 30 mg rate prescribed in the EPA's own handbook, which he said under-estimated the risk.⁴⁸ Dr. Sawyer characterized the risk level he calculated as " 4.1 times higher than the benchmark established by EPA," but he conceded that this level is still within the FDOH's " very low" risk category and would not require action by the FDOH to protect public health.⁴⁹ ( See also supra Note 32). Also, although he acknowledged the epidemiological cancer rate studies from the area that showed no increased rates of cancer in the proposed class area, he nonetheless insisted that the statistical risk he identified poses an " excessive and persistent health risk" that must be remediated.⁵⁰

44 Dr. Flowers testified differently at the hearing, where he said that there is an 83% chance that any home in the class area will have contamination in excess of the 57.08 ppt benchmark. Dr. Flowers did not explain the discrepancy between this 83% figure and the conclusion in his report that, using a lognormal distribution, there is only a 66% probability that any home within the class area would have contamination in excess of the benchmark. Moreover, Dr. Flowers's testimony that 83% of samples within the proposed class exceeded the benchmark is not accurate. It is clear from other portions of his testimony, his report, and the numbers themselves (115 samples out of 139 total exceeded this benchmark) that 83% is the raw percentage of all collected samples (including those from outside the class area) that exceeded Dr. Sawyer's benchmark. Later in the hearing, Dr. Sawyer added to the confusion by testifying that 88% of all samples (background and samples within the proposed class) exceeded the benchmark for contamination. His testimony is based on his determination that only 108 of the 139 indoor dust samples were usable. In any event, these raw percentages of all samples, without more, in fact reveal little about the proposed class in particular.

45 Dr. Flowers also evaluated the CALUX soil results from Plaintiffs' testing campaigns. He identified that 44% of all of the samples (within the proposed class area and outside it) exceeded a " pristine" rural background and, using a conversion factor to compare the soil results against Florida's 7 ppt TCDD-TEQ benchmark, he found that 10% of all samples (within the proposed class area and outside it) exceeded the regulatory limit. These opinions are challenged on reliability grounds. However, the Court finds no reason to discuss the soil data for purposes of the class analysis.

46 Dr. Sawyer articulated a standard three-step risk analysis, in which he (1) identified the chemical of concern; (2) defined the boundaries of the area of concern where there is contamination above background levels; and (3) used analytical data to measure the risk level within that area. He explained that the EPA and FDOH had done the first step in this case, and Dr. Flowers had completed the second step by establishing the proposed class boundary.

47 It is worth noting again that, although the EPA and FDOH analyzed samples under both GC/MS and CALUX, they did not rely on the CALUX measurements to evaluate risk but used only GC/MS results for that purpose due to the limitations of CALUX.

48 Although Dr. Sawyer disagrees with the EPA and FDOH's use of an ingestion rate lower than that recommended in the EPA Handbook, the EPA clearly based its decision to use a rate of 11 mg on guidance from its toxicologist and University professors who had evaluated recent studies on the issue. Also, the letter of University of Florida professors Stuchal and Roberts establishing the 11 mg rate was drafted in June 2011, Pla. Ex. 222, and the Dust Work Group's July 2011 recommendation to the EPA and FDOH relied on this rate. The EPA Handbook was not published until September 2011. Thus, the record does not support Plaintiffs' suggestion that the agencies were operating with an ulterior motive in using a different rate.

49 To the extent " 4.1 times higher" is meant to suggest that each person in the proposed class is four times as likely to develop cancer as someone outside the proposed class boundary, the statement is misleading. Instead, because the increased risks calculated by both the EPA/FDOH and Dr. Sawyer are considered very low (1 out of 100,000 people versus 4 out of 100,000 people), the overall increase in the risk is actually minuscule.

50 Dr. Sawyer maintained that epidemiological studies are not necessary to require remediation. He referenced another wood treatment site in Cass Lake, Minnesota, where remediation was required based on testing of only 10 homes and no epidemiological studies. There, however, legal liability was not being litigated. Instead, remediation at Cass Lake had been a coordinated effort between the EPA and owner and the average floor dust GC/MS PCDD/F concentration was 61 ppt TCDD-TEQ, Def. Ex. 105.06. Here, by contrast, the EPA's near-Site GC/MS average (within the portion of Stephen Foster closest to the Koppers Site) was only 27.9 ppt TCDD-TEQ and existing epidemiological studies show no increases in cancer rates.

         After Dr. Sawyer issued his report in this case, which was based solely on CALUX data, Defendants challenged his opinions because they were based solely on CALUX. Dr. Sawyer then considered the Plaintiffs' GC/MS test results as well, which were collected from within the proposed class boundary. He testified at the hearing that they " fully supported" the CALUX results.⁵¹ Dr. Sawyer conceded at the hearing that the CALUX method has not been validated for dust and that CALUX-TEQ values in general run higher than GC/MS TCDD-TEQ values. He also admitted that " there's no question that there's inflation of results [in the CALUX-TEQs] due to the brominated dioxin and furans." He said that for this reason, he separately considered the GC/MS data to confirm the CALUX results. In doing so, he found the CALUX results " absolutely reliable," explaining that the average of the GC/MS results within the proposed class boundary (58.1 ppt TCDD-TEQ), which includes no brominated compounds, exceeds the calculated CALUX boundary (57.08 ppt CALUX-TEQ), although " not by a lot." Then, because he had calculated the risk in Table 12 of his report using the boundary of background (57.08 ppt CALUX-TEQ), which he considered the " bottom level of risk," he said there would be no difference if he used the GC/MS average of 58.1 ppt TCDD-TEQ because the numbers are so similar. That is, the GC/MS average would produce the same level of risk (4.09 x 10^-5) as the CALUX-TEQ benchmark, which he said confirmed the reliability of his risk calculation in Table 12. At the hearing, Dr. Sawyer withdrew his reliance on all risk assessment calculations except for Table 12 of his report.⁵²

51 There was a great deal of discussion at the hearing about whether Dr. Sawyer had properly disclosed prior to the hearing the fact that he had considered the GC/MS testing after he issued his report. There also was an issue regarding whether he had disclosed his opinion that the EPA used the wrong ingestion rate prior to his rebuttal report of August 14, 2015, which has been stricken. The undersigned decided to review Dr. Sawyer's response affidavits and deposition testimony in order to determine whether these opinions were disclosed prior to the rebuttal report. The undersigned finds that these opinions were expressed by Dr. Sawyer during his deposition and in his affidavits filed in response to the Defendants' Daubert motions; not solely in his rebuttal report. Thus, Dr. Sawyer's testimony in regards to GC/MS testing and the EPA's ingestion rate are admitted. The record reflects, however, that Dr. Sawyer's response affidavits were withdrawn and not admitted at the hearing; thus, although his testimony is considered in full, the undersigned has not considered the withdrawn affidavits.

52 Specifically, Dr. Sawyer withdrew his reliance on the other CALUX-based risk calculations shown in his report at Tables 13, 14, and 15.

         2. Dr. Michael Wade

         Plaintiffs rely on the opinion of Dr. Michael Wade, a chemical fingerprinting expert, to show causation, that is, that the Koppers Site is the source of the alleged classwide indoor dust contamination. Dr. Wade is an organic geochemist and principal of the forensic geochemistry consulting company, Wade Research. He testified that he was retained to answer a simple question of whether there is " a relationship in the available dioxin data" between on-Site dioxin and off-Site dioxin. Dr. Wade considered the GC/MS data and found a relationship using Principal Component Analysis (" PCA" ). He concluded " that off site transport of PCP-derived PCDD/F had occurred from the Koppers facility to property located to the west" and has been found in soil and dust collected by Plaintiffs. Pla. Ex. 28. Dr. Wade's brief report provided little explanation and no details of his analysis.⁵³

53 To characterize Dr. Wade's report as brief is an understatement. He testified that he drafted the four-paragraph report on an iPhone during a flight while seated " in the middle seat, sandwiched between two people much bigger than I." Although he testified that this did not impact his report, the report includes only summary conclusions with no calculations or statistical data showing his analysis. Dr. Wade, however, did disclose his raw data files to Defendants and said the data could be easily replicated with that information, but he provided no explanation of his analysis.

         Dr. Wade attempted to clarify his report at the hearing. He explained that PCA is a widely accepted and proven methodology for fingerprinting dioxin contamination that uses a statistical multi-linear regression approach in which the principal components of each sample are plotted and examined for interrelationships. Dr. Wade testified that his PCA reflected a central behavior in the GC/MS data supporting a relationship between the samples and a dominant source.⁵⁴ He conceded that departures from the main statistical data trend were present and said that this suggests other factors may have been influencing the distribution of PCDD/F as well. He explained, however, that he was not asked by Plaintiffs to quantify the extent of off-Site contamination or to identify any alternate sources of dioxin contamination that might have influenced the data. He determined only that, according to the data (including soil, floor dust, indoor living quarters dust, and attic dust), PCDD/F had been transported off-Site to properties to the west.

54 Dr. Wade used only the GC/MS results in his analysis, and Plaintiffs conceded that CALUX results were not adequate for chemical fingerprinting analysis.

         3. Dr. Jurgen Exner & Dr. John Kilpatrick

         Plaintiffs also presented expert testimony from Dr. Jurgen Exner, to show the feasibility of classwide remediation, and Dr. John Kilpatrick, to show the feasibility of establishing classwide stigma damages. Both experts assumed the presence of classwide contamination in forming their opinions.

         Dr. Exner is an industrial chemist who specializes in waste management and environmental restoration. He described a labor-intensive remediation process by which all homes would be remediated using an efficient vacuum cleaner equipped with a HEPA filter capable of trapping dust particles on all indoor surfaces and wiping solid surfaces with wet cloths.⁵⁵ He said the process may need to be repeated for highly contaminated residences. Dr. Exner opined that all of the affected homes would be cleaned with this common method at a set cost per foot relative to the size of the house. Dr. Exner estimated a class-wide fixed cost of remediation at $945 per residence but stated that the cost could vary between $2.90 and $3.20 per square foot, depending on the size of the residence. His price was based on his own assumptions regarding contractor costs because specific contractor quotes would not be sought until remediation was certain. Dr. Exner acknowledged that the amount of time and effort, and consequently cost, could vary from home to home depending on several factors. Notably, because the extent of the contamination is not the same in every home, Dr. Exner said labor costs would increase depending on the number of walls and attic space in each home and the need to re-clean any portion of an individual home in order to meet whatever target cleanup standard is set. Dr. Exner conceded he had no idea how many homes had attics or the number of walls in each home within the class.⁵⁶

55 Dr. Exner acknowledged that ordinary vacuuming will reduce the amount of dust contamination in a home.

56 Dr. Exner explained that unfinished attics increase the square footage of the home and would need to be added into the cost analysis.

         Dr. John Kilpatrick, a certified real estate appraiser for over 30 years, testified on the issue of classwide stigma damages. He opined that it is feasible to perform a class-wide mass appraisal approach to determine the diminished value of Plaintiffs' properties based on stigma from the contamination. Dr. Kilpatrick intends to use a software program known as Greenfield AVM (an automated valuation model) for predicting property values, which is a hedonic regression⁵⁷ mass appraisal program he created. Dr. Kilpatrick said the model is capable of calculating an unimpaired value for the impacted homes based on a mathematical formula incorporating several variables, such as the tax assessed value of the property, square footage, lot size, etc., and that he will adapt the model for use in this case. Dr. Kilpatrick has compiled data for the mass appraisal he intends to perform from RealtyTrac, a real estate database containing information on recorded deeds, sales prices, property characteristics, and tax assessor data from Alachua County, Florida, where Gainesville is located. He also gathered geographic information systems data (" GIS" ) through the Florida Geographic Data Library for use in the proposed mass appraisal. The AVM uses this large body of data to compare anywhere from 100 to 2,000 comparable properties. Dr. Kilpatrick touted his AVM as much more reliable than individual appraisals, which he said consider only a limited number of comparable properties. Dr. Kilpatrick explained that after calculating the unimpaired value using the AVM, he will add an " impairment factor" and the model will then calculate an impaired value due to the stigma of contamination remaining after remediation.

57 Dr. Kilpatrick described hedonic regression as a means of comparing pricing data by taking into consideration several characteristics of properties. More specifically, hedonic regression has been described as " a type of multiple regression analysis that attempts to determine mathematically the value or price of the various components or characteristics that add up to the total value or utility." Hartle v. FirstEnergy Gen. Corp., Nos. 08-1019, 08-1025, 08-1030, 2014 WL 1317702, at *8 (W.D. Pa. Mar. 31, 2014).

         Additionally, Dr. Kilpatrick plans to perform survey research including market data from comparable case studies, market data contained in a survey of empirical literature, and also pricing and transactional data within the proposed class area for purposes of validating the diminished value. Damages for each class member whose home needs remediation would be the difference between the impaired and unimpaired value of the home. Dr. Kilpatrick said he has not yet run the model because he needs to know the precise date of valuation and the boundaries of the certified class in order to input the appropriate information, but he testified that the necessary data, some of which has been collected and validated, is readily available.⁵⁸

58 Dr. Kilpatrick had the proposed class boundary in his report, but he felt it was premature to calculate diminution in value until the Court certified a class and established a date for valuation.

         B. Class Representatives

         The proposed class representatives testified by deposition. They each own residential property near the Koppers Site. The CALUX-analyzed dust samples from two of the proposed class representatives' homes exceed Dr. Sawyer's 57.08 ppt CALUX-based benchmark for the class boundary, and all seven named Plaintiffs maintain that their property values and future health have been impacted by dioxin contamination in dust in their homes.

         1. Nancy Day

         Plaintiff Nancy Karen Day owns a rental property located at 3119 4th Street, adjacent to the Koppers Site in the Stephen Foster neighborhood. She testified that her husband Kim Popejoy's real estate company deeded the property to her in 2012 with knowledge that it may have been impacted by Koppers' activities (her husband's company had purchased it in 2006). According to Day, since acquiring the property in 2012, she has never mopped or swept the floors in the house nor has she or anyone else ever dusted, although she stated that her husband removed debris and wiped down surfaces in February 2012, when the prior tenants moved out and the property was deeded to her. In November 2012, the EPA informed Day that GC/MS testing of her property's floor dust showed a GC/MS PCDD/F level of 44.9 ppt TCDD-TEQ, which was below the EPA's risk-based screening level of 190 ppt TCDD-TEQ (Day Dep. at 64-65), but Plaintiffs' CALUX testing of the home in 2014 (before the soil remediation) showed a dioxin dust contamination level of 196.52 ppt CALUX-TEQ (Flowers Report, at 32), which is above Dr. Sawyer's CALUX benchmark. The soil on this property was remediated in mid to late 2014, and Beazer East paid Day $397.80 for normal cleaning. Day took the money but did not use it to remediate the indoor dust because she felt that the amount was insufficient, although she admitted she did not research how much it would cost to properly clean the home. Day did not know whether the rental home had lost value due to contamination from the Koppers Site.

         Day and her husband, Kim Popejoy, own and live in a residence located at 300 NE 13th Avenue, which is not in the Stephen Foster neighborhood but is within the proposed class boundary, approximately .7 miles south of the Koppers Site. Plaintiffs' indoor living dust testing in this home measured 27.23 ppt CALUX-TEQ, which is below Dr. Sawyer's CALUX-based benchmark, but Day felt that her home had lost value because of the Koppers Site contamination, although she does not know by how much. Day and her husband purchased the home in 1996. Her husband and son both smoke cigarettes, but not in the house. Her husband testified that their home has a fireplace and that for the last 15 years there has been a homeless camp on a nearby vacant lot where campfires are regularly burned.

         2. Betty and John Hewitt

         Plaintiffs Betty and John Hewitt reside at 3134 NW 8th Street, four blocks west of the Koppers Site, west of a main thoroughfare and within the Stephen Foster neighborhood.⁵⁹ The Hewitts purchased their home in 1973. Mrs. Hewitt vacuums the home and dusts regularly. Both Hewitts have been smokers, although Mr. Hewitt quit smoking in 1982, and Mrs. Hewitt testified that she no longer smokes inside the house. Mr. Hewitt suffers from esophogeal and colon cancer but testified that he has never been told that either was related to exposure to chemicals used at the Koppers Site. Mr. Hewitt testified that he thought the home was worth $179,000 in 2007 and said it appraised at only $122,000 in 2014. Plaintiffs' CALUX testing of the living space dust in the home measured 90.35 ppt CALUX-TEQ, which exceeds Dr. Sawyer's CALUX-based benchmark.

59 The Hewitts' home is just outside the portion of the Stephen Foster neighborhood that was remediated. Thus, the soil on their property was not remediated and they received no money for indoor dust remediation, although soil remediation occurred at homes across the street from them.

         C. Defendants' Experts

         Defendants challenged the scientific reliability of Plaintiffs' experts' opinions through their own experts, who all testified at the hearing. Dr. Allen Uhler, an environmental chemist; Michael Corn, a consulting environmental engineer; and Dr. Shari Libicki, a chemical engineer, each criticized the Plaintiffs' reliance on the CALUX method in this case for reasons similar to those expressed by Dr. Cline. They all agreed that there is wide variability in the CALUX-TEQ values from home to home and that CALUX tends to overestimate toxicity because it provides only a total dioxin and dioxin-like measurement, not a true TCDD-TEQ value. Corn was particularly critical of Dr. Sawyer's method of combining Plaintiffs' CALUX indoor dust results with the EPA's CALUX floor dust results to determine a composite background for indoor dust, explaining that this is an apples to oranges comparison. And, defense expert Dr. Scott Phillips, M.D., a medical toxicologist, sharply criticized Dr. Sawyer's health risk assessment for its reliance on CALUX, which he said does not distinguish between chlorinated and brominated dioxins and misleadingly elevates toxicity. See Def. Ex. 109.

         1. Dr. Allen Uhler

         Dr. Uhler testified regarding the source, nature, and extent of off-Site dioxin contamination in Gainesville. Dr. Uhler considered the data through a chemical fingerprinting process known as Diagnostic Ratio Analysis with radar plots. He excluded off-Site dust samples that did not have positive measurements for PCP-derived PCDD/F (which is the only type that can be attributed to Koppers), and then compared the on-Site (Koppers) ratios with the off-Site ratios. He found that the chemical fingerprints (diagnostic ratios) of off-Site dust samples varied, indicating that they are derived from different sources, not solely the Koppers Site. However, he did identify a consistent diagnostic ratio between GC/MS soil samples on-Site (Koppers) and off-Site along the Site's western and southern boundaries, but only within a limited area (extending at most 300 meters from the Site, and beyond that, the soil concentrations were consistent with urban background samples). Again, Dr. Uhler did not observe the same consistency in comparing the patterns of the indoor dust results.

         Dr. Uhler also saw " no recognizable trend" or " ramp up" of increasing chlorinated TCDD-TEQ values in the GC/MS testing based on proximity to the Koppers Site. He also noted that only two samples in the proposed class exceeded the EPA's health protective limit of 190 ppt TCDD-TEQ and that homes near these sample locations or closer to the Site were lower, again suggesting that the Koppers Site is not the source, or not the only source, and that localized sources were impacting the values. (For instance, one home with a 193 ppt TCDD-TEQ was near another measuring 9.9 ppt TCDD-TEQ.) See Def. Ex. 139; Def. Ex. 140. Dr. Uhler said that the majority of the GC/MS TCDD-TEQ values were " variable but typically below--at or below 100 ppt" TCDD-TEQ within the first 500 meters and even very close to the Koppers Site. He concluded from this that " there doesn't seem to be a strong relationship with dust--living quarter dust TEQ concentration with distance" to the Site. Dr. Uhler also observed variability in the attic dust GC/MS toxicity values, which he said showed no correlation to the Site, given that the two attic dust samples collected closest to the Site contained only low levels of PCDD/F.

         Dr. Uhler criticized Dr. Wade's chemical fingerprinting analysis and application of PCA methodology in this case as incomplete and unreliable. He explained that PCA is " an exploration tool that's numerical," and through PCA, variances between toxicity values in samples are plotted in a mathematical way in a two or three dimensional space, and the more variance between them, the larger the numbers become, and where there is little variance, the numbers coalesce together. Dr. Uhler described the PCA analysis as having two forms of output--the factor score plot, which shows the distribution of data in space, and the factor loading chart, which is a table that describes which variables in the congeners are causing either the similarities or the differences.⁶⁰ In Dr. Uhler's opinion, the distribution of the 17 congeners can only be identified when the factor score plot is read together with the factor loading chart so that a determination can be made as to which variables in the congeners are causing the similarities or differences. According to Dr. Uhler, Dr. Wade's analysis is incomplete because he failed to explain whether or how he had evaluated the factor loading chart, which is what gives insight into the congeners driving the differences or similarities.

60 There is some confusion in the record during Dr. Uhler's testimony because the second step has been referred to as both " the factor score plot" and " factor loading plot," which is somewhat confusing. When Dr. Uhler's entire testimony is read, however, the terminology becomes clear--the second step of the PCA analysis is the " factor loading chart."

         2. Michael Corn

         Michael Corn prepared a report discussing the primary pathways for potential PCP and PCDD/F migration off-Site. He testified that dioxin from PCP would have dripped onto the ground with diesel material and it could have been picked up by wind and transported, with larger particles settling closer to the Site. Corn testified that if the Koppers Site was the sole source of PCDD/F, and PCDD/F had been transmitted through the air, he would expect to see similar values in off-Site soil, house dust and attic dust with a linear regression trend relative to the Koppers Site, but he did not observe that type of relationship or pattern in the data, even considering the CALUX data. Instead, in his opinion, dust air transport pathways were limited by a substantial tree line along the western edge of the facility, the fact that crushed limestone was used to cover the yard and roads at the Site, and the fact that the stormwater runoff flowed to the northeast, not to the west.

         3. Dr. Shari Libicki

         Dr. Shari Libicki explained that she visited the Koppers Site and had studied the atmospheric patterns and meteorological data for the area. She gave a number of opinions in her report regarding air transport of chemicals, including that numerous alternative airborne sources of dioxin emissions exist in the Gainesville area other than the Koppers Site, that a vegetative barrier on the western and southern boundary of Koppers and a stand of trees in the southwestern corner reduced airborne transport of dioxin from the Site, and also that homes within a two-mile boundary of the Site are not similarly situated to one another with respect to potential air transport of dioxin from the Site due to the presence of lush and varied vegetation off-Site. See Def. Ex. 110. Dr. Libicki explained that scientific literature dating back to the early 1900s confirms that vegetation removes particulates from the air and that the vegetation at the Koppers Site is very lush.⁶¹ Dr. Libicki also identified several alternative sources of dioxins within the proposed class area and noted that only rarely did Koppers burn PCP contaminated wood and sludge at the Site, although she admitted at the hearing that this was unclear.⁶²

61 Although photographs available from later years, 2007 through 2011, showed changing vegetation at the Site and along the western border, there was one stand of trees that continuously existed as a buffer zone. Wayne Grip, an aerial photographer, also discussed the changes in the physical environment at and near the Koppers Site over the years, based on his interpretation of historical photography and mapping.

62 This testimony is based on Dr. Libicki's review of the deposition of Randall Collins, Plant Manager, who testified that wood waste was burned at the Site until the 1960s when the boiler was converted to natural gas, which was prior to the time PCP was used at the Site. He also stated that waste wood was burned " for a few years" in the early 1980s, Collins Dep. at 96-103, and agreed wood waste would have been treated with PCP at that time. However, in an errata to his deposition, Collins clarified that it was more accurate to say that there were " a few test burns" during those years, or that periodic test burns were conducted. ECF No. 832-4 & Def. Ex. 200. Dr. Libicki also relied on correspondence between Koppers and regulatory agencies regarding a boiler test program and the denial of a permit for burning waste in the early to mid-1980s for her opinion that little or no burning of PCP-contaminated wood took place. See Def. Exs. 201, 202, 203.

                    4. Dr. Scott Phillips

         Dr. Scott Phillips, M.D., testified regarding health risks within the proposed class area and criticized Dr. Sawyer's reliance on a statistical risk assessment for class certification purposes. Dr. Phillips opined that Dr. Sawyer's statistical risk assessment is unhelpful and does not show the existence of a " persistent and excessive health risk" present in each home in the proposed class. According to Dr. Phillips, to determine whether a person has an actual increased risk of cancer, a clinical risk assessment is required, which includes consideration of individual factors such as exposure, dose, and susceptibility.

         Dr. Phillips considered existing epidemiological studies and public health assessments to determine whether there was a defined health impact from the Site. Specifically, he considered the FDOH's 2011 study of the rates of 18 different types of cancer in the Stephen Foster neighborhood in Alachua County, census tract 3, dating back to 1981, Def. Ex. 82, and a 2012 study reviewing cancer rates in Alachua County from 2000 to 2008, Def. Ex. 84, and testified that no increased pattern of cancer occurrences are reflected in these reports. Dr. Phillips said that, although a slight increase in liver cancer was noted between 2006 and 2008, this was not statistically significant overall. He also considered the FDOH public health assessments, which set a " very low" increased risk of cancer from indoor dust contamination in residences near the Koppers Site. Def. Exs. 86, 87. In light of the cancer studies confirming that there has been no elevated risk of cancer in Gainesville's Stephen Foster neighborhood over the past 30 years, Dr. Phillips opined that dust in the homes did not create a health risk needing remediation, despite the " very low" statistical increase identified by Dr. Sawyer and the FDOH. In Dr. Phillips's opinion, classwide remediation could not lower the risk of cancer where the actual cancer studies show there has been no increase in cancer.

         5. Trevor Phillips

         Defendants also presented expert testimony from Trevor Phillips, a real estate appraiser, criticizing Dr. Kilpatrick's AVM mass appraisal method for evaluating stigma damages. In Phillips's opinion, Dr. Kilpatrick's AVM cannot reliably develop a market value for individual properties, and it fails to satisfy several standards set forth in the Uniform Standards of Professional Appraisal Practice (" USPAP" ). Phillips testified that the mass appraisal model was unreliable because it uses potentially out-dated tax assessor data. He explained the individualized nature of an appraisal, noting that two properties may look comparable on paper but be very different in actual sales values. In his opinion, property value diminution in the proposed class area is not supported by the actual market data.

         III. Discussion

         Plaintiffs rely on expert testimony to meet their burden under Federal Rule of Civil Procedure 23 for class certification. Both sides challenge each other's experts as unreliable. Therefore, before reaching the class issues, the Court must first resolve the pending Daubert motions.

         A. Expert Challenges

          The admissibility of expert testimony is governed by Rule 702 of the Federal Rules of Evidence and Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). The party offering the expert testimony bears the burden to lay a proper foundation for its admission and must prove its admissibility by a preponderance of the evidence. See Allison v. McGhan Med. Corp., 184 F.3d 1300, 1306 (11th Cir. 1999). " Vigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof [remain] the traditional and appropriate means of attacking shaky but admissible evidence." Daubert, 509 U.S. at 596. Determining the admissibility of expert testimony is within the district court's discretion, and the district court is given " considerable leeway" in making the decision. Rink v. Cheminova, Inc., 400 F.3d 1286, 1291 (11th Cir. 2005) (quoting Kumho Tire Co. v. Carmichael, 526 U.S. 137, 152, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999)).

          District courts in the Eleventh Circuit must perform a " full Daubert " analysis " when an expert's testimony is critical to the class certification." Local 703, I.B. of T. Grocery & Food Employees Welfare Fund v. Regions Fin'l Corp., 762 F.3d 1248, 1258 n.7 (11th Cir. 2014) (quoting with approval Am. Honda Motor Co. v. Allen, 600 F.3d 813, 815-16 (7th Cir. 2010)); Sher v. Raytheon, 419 Fed.Appx. 887, 890-91 (11th Cir. 2011) (unpublished); ⁶³ see also Hendrix, ex rel. G.P. v. Evenflo Co., 609 F.3d 1183, 1194 (11th Cir. 2010) (requiring a " rigorous" Daubert analysis). Under this " full" Daubert analysis, if the expert information is necessary to establish a Rule 23 requirement, the district court, as the trier of fact for class certification, " must conclusively rule on any challenge to the expert's qualifications or submissions prior to ruling on a class certification motion." Sher, 419 Fed.Appx. at 890-91 (quoting Am. Honda, 600 F.3d at 815-16) (also noting, " Tough questions must be faced and squarely decided," quoting Am. Honda, 600 F.3d at 817). The Eleventh Circuit determined in Sher that a district court errs as a matter of law " by not sufficiently evaluating and weighing conflicting expert testimony" because the plaintiff must prove more than a prima facie case under Rule 23 for class certification.⁶⁴ Id. Therefore, where an expert opinion necessary to class certification is challenged, such as in this case, the Court must perform a full Daubert analysis.

63 While unpublished opinions are not considered binding, they may be considered as persuasive authority. See 11th Cir. R. 36-2; see also United States v. Futrell, 209 F.3d 1286, 1289 (11th Cir. 2000).

64 Some courts have noted that in the class certification context, the Daubert analysis may be narrower than that conducted for purposes of trial because " the inquiry is limited to whether or not the expert reports are admissible to establish the requirements of Rule 23." Braggs v. Dunn, 317 F.R.D. 634, 2016 WL 6917203, at *5 (M.D. Ala. 2016) (quoting Fort Worth Employees' Ret. Fund v. J.P. Morgan Chase & Co., 301 F.R.D. 116, 126 (S.D.N.Y. 2014)). The Eleventh Circuit, however, has not expounded further on the practical application of the " full Daubert " standard in the class certification context in a case involving a bifurcated discovery procedure, such as in this case. The Eighth Circuit applies a " focused Daubert" standard to scrutinize expert testimony necessary to class certification issues, not necessarily requiring a conclusive determination of admissibility at trial. See In re Zurn Pex Plumbing Prod. Liab. Litig., 644 F.3d 604, 614 (8th Cir. 2011). The Third Circuit recently found it unnecessary to determine whether differences exist between the " full" Daubert standard articulated in the Seventh Circuit (adopted by the Eleventh) and the standard applied in the Eighth Circuit, stating that under either articulation, courts limit the Daubert inquiry to expert testimony necessary to prove the requirements of Rule 23. In re Blood Reagents Antitrust Litig., 783 F.3d 183, 188 n.8 (3d Cir. 2015) (declining to examine whether there might be some variation between the Seventh and Eighth Circuit formulations of the relevant Daubert inquiry, noting only that a plaintiff cannot rely on expert testimony to establish class certification without satisfying Daubert ). This Court likewise finds no reason to speculate on the potential differences among the circuits on this matter but instead will endeavor to apply the " full" Daubert analysis as set forth in American Honda, which the Eleventh Circuit has twice cited with approval in Sher and Local 703.

          Under Rule 702 of the Federal Rules of Evidence, a qualified witness may offer expert opinion testimony if:

(a) the expert's scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue;

(b) the testimony is based on sufficient facts or data;

(c) the testimony is the product of reliable principles and methods; and

(d) the expert has reliably applied the principles and methods to the facts of the case. Fed.R.Evid. 702. Daubert describes the district court's gatekeeping function, as requiring " a preliminary assessment" of whether the proposed expert's reasoning or methodology " is scientifically valid" as applied to the facts at issue.⁶⁵ 509 U.S. at 592-93. This " gatekeeper" role is designed to ensure that " speculative, unreliable expert testimony does not reach the jury." Kilpatrick v. Breg, Inc., 613 F.3d 1329, 1335 (11th Cir. 2010) (citing Daubert, 509 U.S. at 597, 589); Rink, 400 F.3d at 1291. While " general acceptance" of a theory " is not a necessary precondition to the admissibility of scientific evidence," the Supreme Court has explained that the trial judge's task is to ensure that an expert's opinion " rests on a reliable foundation and is relevant to the task at hand." Daubert, 509 U.S. at 597.

65 This gatekeeping function applies to both scientific and experience-based expert testimony. See Kumho Tire Co. v. Carmichael, 526 U.S. 137, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999).

          In the Eleventh Circuit, this gatekeeping function involves a " rigorous three-part inquiry," requiring the district court to consider whether:

(1) the expert is qualified to testify competently regarding the matters he intends to address; (2) the methodology by which the expert reaches his conclusions is sufficiently reliable as determined by the sort of inquiry mandated in Daubert ; and (3) the testimony assists the trier of fact, through the application of scientific, technical, or specialized expertise, to understand the evidence or to determine a fact in issue.

Hendrix, 609 F.3d at 1194 (quoting United States v. Frazier, 387 F.3d 1244, 1260 (11th Cir. 2004)).

         Regarding the first factor, an expert " may be qualified in various ways." Seamon v. Remington Arms Co., LLC, 813 F.3d 983, 988 (11th Cir. 2016) (internal quotation omitted). The Court considers the witness's scientific training, education, and experience. See id.

          Regarding the second factor-reliability as described by Daubert -the Court's analysis is guided by a nonexhaustive list of considerations, including (1) whether the expert's scientific method, technique, or theory can be objectively tested; (2) whether it has been subjected to peer review and publication; (3) the known or potential rate of error of the method; (4) the existence of standards and controls; and (5) whether the method, technique, or theory has been generally accepted in the scientific community.⁶⁶ See Kumho Tire, 526 U.S. at 149-50; Daubert, 509 U.S. at 592-95; Hendrix, 609 F.3d at 1194. The court is afforded " considerable leeway" in deciding reliability in any given context. Kumho Tire, 526 U.S. at 152. The inquiry is not whether the testimony is " scientifically correct" but whether, " by a preponderance of the evidence, it is reliable." Allison, 184 F.3d at 1312. The Court in Daubert explained that the adjective " scientific" in Rule 702 " implies a grounding in the methods and procedures of science." 509 U.S. at 590. " [E]xperts commonly extrapolate from existing data," but where the opinion evidence is connected to existing data by nothing more than " the ipse dixit of the expert," the district court is free to " conclude that there is simply too great an analytical gap between the data and the opinion proffered" for reliability. General Elec. Co. v. Joiner, 522 U.S. 136, 146, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997); Hendrix, 609 F.3d at 1194. Also, Daubert cautions that the " knowledge" requirement which qualifies an expert under Rule 702 " connotes more than subjective belief or unsupported speculation." 509 U.S. at 590.

66 The Court must strike a balance between the type of " uncompromising 'general acceptance' test" that was disavowed in Daubert and the need to reach a legal resolution based on the application of reliable scientific evidence. See 509 U.S. at 596.

          The third factor, helpfulness, bears on relevance and " fit." Seamon, 813 F.3d at 988. Relevance means the evidence has a scientific connection to the facts of the case such that it " logically advances a material aspect of the proposing party's case." Allison, 184 F.3d at 1312 (internal quotation marks omitted). While the relaxed standard of relevance requires only that the evidence makes a fact more or less probable, an expert opinion with no " valid scientific connection to the pertinent inquiry" should be excluded for lack of " fit." Seamon, 813 F.3d at 988 (internal quotations omitted). That is, " there is no fit where a large analytical leap must be made between the facts and the opinion." Chapman v. Procter & Gamble Distrib., LLC, 766 F.3d 1296, 1306 (11th Cir. 2014) (internal quotations omitted).

         1. Motions to Exclude CALUX Methodology, Dr. Flowers, Dr. Sawyer

         Defendants move to exclude the opinions of Dr. Flowers and Dr. Sawyer as unreliable to the extent they relied on the CLAUX measurements to establish a class boundary and to define a classwide health risk.⁶⁷ More particularly, Defendants argue that CALUX is not a reliable scientific method for accurately quantifying chlorinated TCDD-TEQ values or for assessing health risks, and therefore, all CALUX-based opinions should be excluded. Plaintiffs' experts admitted at the hearing that the CALUX method has some limitations. Nonetheless, Plaintiffs maintain that the method is reliable for a comparative analysis, i.e., comparing CALUX data against other CALUX data, which is what Dr. Flowers did to establish a class boundary. Plaintiffs also assert that Dr. Sawyer's CALUX-based health risk opinion is reliable because he also considered the GC/MS results.

67 Defendants do not challenge the qualifications of Dr. Flowers or Dr. Sawyer, and thus the Court has no need to consider them.

         Considering Daubert's nonexclusive list of factors bearing on reliability, see Kumho Tire, 526 U.S. at 149-50; Daubert, 509 U.S. at 592-95; Hendrix, 609 F.3d at 1194, the Court finds for the reasons stated below that, although the CALUX method is not inherently unreliable, its purpose is limited, and in this case, Plaintiffs have extended the method beyond its scientific validity. The record includes peer-reviewed scientific literature confirming that the scientific community, including the EPA, accepts CALUX as a screening method to aid in the detection of dioxin contamination, see, e.g., Pla. Ex. 10, 12, 13, 15, which means it is useful for estimating total dioxin and dioxin-like compounds in soils and sediments, for the purpose of identifying locations for more sensitive GC/MS testing, and for relative comparisons of " dioxin-like" activity. Dr. Uhler explained that CALUX and other rapid screening tests were developed and used by the EPA for the purpose of quickly and inexpensively detecting and estimating the presence of chemical contamination in a particular area in order to provide information about which areas to target for further investigation.⁶⁸ The undisputed expert testimony on both sides of this case confirms that the CALUX bioassay responds to all dioxins (chlorinated and brominated) as well as other " dioxin-like" chemicals but cannot distinguish between PCDD/F and PBDD/F.⁶⁹

68 Although Plaintiffs have filed a motion in limine to exclude certain opinions of Dr. Uhler, they have not sought to exclude his opinions regarding CALUX.

69 The one exception is the dioxin-like chemical, PCB, which can be extracted from the analysis, as occurred with Plaintiffs' 2014 samples. Regardless, PCB is not relevant here; the chemical of concern in this case is PCP-derived chlorinated dioxin (PCDD/F), which CALUX cannot distinguish from other dioxins ( i.e., PBDD/F) and dioxin-like chemicals.

          The Supreme Court recognized in Daubert that " scientific validity for one purpose is not necessarily scientific validity for other, unrelated purposes." 509 U.S. at 591. Here, although CALUX is a useful means of estimating the toxicity of dioxin and dioxin-like compounds, the record indisputably confirms that it is not valid for making the quantitative determinations as to specific congeners of PCP-derived PCDD/F necessary to identify whether the contaminant is attributable to the Koppers Site. It is well established on this record that GC/MS testing is the " gold standard" for quantification of PCDD/F. Although Daubert does not demand the best or most persuasive scientific method, it insists on a reliable method. Daubert, 509 U.S. 591-92; see also Braggs v. Dunn, 317 F.R.D. 634, 646 (M.D. Ala. 2016) (stating Daubert and Kumho Tire require that " an expert need not use the best method of evaluation, only a reliable one." (internal quotations omitted)). Notably, although the EPA used the CALUX screening method concurrently with GC/MS testing in its 2012 floor dust sampling, neither the EPA nor the FDOH relied on CALUX results for quantifying PCDD/F, assessing health risks, or determining the source of PCDD/F contamination in any individual home in Gainesville, due to the variability seen among the CALUX results themselves and between the results of CALUX and GC/MS, i.e., the CALUX method's tendency to overstate toxicity. See Pla. Ex. 2 (Dr. Cline's Technical Report); Pla. Ex. 5 (FDOH 2014 Health Consultation); Def. Ex. 74 (EPA 2013 Indoor Dust Study Data Report).

         More specifically, all of the experts agree that the CALUX method is incapable of reliably sorting out accurate quantifications of chlorinated TCDD-TEQ values from brominated TCDD-TEQ values. Proof of this is in the record. The FDOH report noted wide disparity in the Plaintiffs' 2010 indoor dust test results between samples analyzed concurrently under GC/MS and CALUX; the CALUX TEQ values were anywhere between 13% and 1700% higher than the GC/MS TCDD-TEQ values. See Pla. Ex. 5, at 4. Dr. Cline explained that these disparities were largely due to the presence of PBDD/F from in-home sources. Dr. Uhler also explained that this lack of uniformity in the difference between CALUX and GC/MS results makes any direct comparison between them based on an average of the differences inherently unreliable. See Def. Ex. 139. Also, Dr. Cline also noted that no data validation protocol exists for CALUX and that, after reviewing the results of the EPA's 2012 dust testing, her confidence in the CALUX data was low, so much so that the CALUX results were considered unusable. See Pla. Ex. 2.

         Plaintiffs argue that the CALUX method as used in Dr. Flowers's spatial distribution analysis, in which he compared CALUX results with other CALUX results to pinpoint a source, is reliable. See Abrams v. CIBA Specialty Chems. Corp., No. 8-68-WS-B, 2010 WL 779273 (S.D. Ala. Mar. 2, 2010) (finding a spatial analysis reliable where DDT was identified). However, even in this context, CALUX paints with too broad a stroke. Because CALUX cannot accurately quantify chlorinated TCDD-TEQ values, Dr. Flowers was left to infer the presence of PCDD/F and level of TCDD-TEQ by the CALUX-TEQ estimates. In Abrams, by contrast, where a similar spatial analysis was used, the chemical of concern DDT-was clearly identified in the testing.⁷⁰ The Eleventh Circuit has cautioned that an assumption about chemical structure without scientific evidence is not reliable. See McClain v. Metabolife Int'l, Inc., 401 F.3d 1233, 1246 (11th Cir. 2005) (a toxic tort case). Moreover, as Dr. Uhler explained, the record in this case reflects that the more accurate GC/MS results did not show a comparable " ramp up" in relation to the Koppers Site.

70 The court in Abrams found that expert testimony using site-data to estimate the background concentration level of DDT and finding diminishing concentration levels of DDT " as one moves away from the plant," similar to the content of Dr. Flowers's testimony, was reliable under Daubert. See Abrams v. CIBA Specialty Chem. Corp., No. 08-68-WS-B, 2010 WL 779273 (S.D. Ala. Mar. 2, 2010). There, however, the expert was considering testing that showed specific levels of DDT, whereas here, the CALUX measurements do not provide specific TCDD-TEQ values for PCDD/F.

         Practically every expert testified that dioxins are ubiquitous in an urban environment and that in this case, numerous potentially contributing sources exist even within each home, making it important to identify and eliminate other potential sources in order to pinpoint the source of PCDD/F. Dr. Flowers admitted that he saw variability in the data ( i.e., CALUX-TEQ values different from one home to the next), suggesting that other sources could be impacting the data. Additionally, he admitted that PBDD/F also could be included in the CALUX measurements. When he was asked whether the CALUX indoor dust numbers he analyzed included brominated dioxins (PBDD/F) he remarked, " Maybe, maybe not." But he did nothing to exclude potential in-home sources of PCDD/F or PBDD/F and instead simply disregarded them and relied solely on the " ramp up" in the general CALUX toxicity values, which he said increased closer to the Koppers Site, as confirmation that Koppers is a dominant source. This failure to fully consider and account for obvious alternative explanations also weighs against admissibility. See also Fed.R.Civ.P. 702, Advisory Committee Notes (2000 Amends.).

         All of this goes to show that the CALUX screening method, which results in one combined total dioxin measurement, including PCDD/F, PBDD/F, and other dioxin-like chemicals, cannot be used to reliably quantify a true measurement of TCDD-TEQ alone. Because of this limitation, the CALUX results are skewed, misleading, and overall unreliable for quantification or pinpointing the source of PCDD/F. See Coleman v. Union Carbide Corp., 2013 WL 5461855, at *33 (S.D. W.V. Sept. 30, 2013) (noting the potential for expert testimony to be " both powerful and quite misleading" (quoting PBM Prods., LLC v. Mead Johnson & Co., 639 F.3d 111, 123 (4th Cir. 2011)).

         Defendants argue that Dr. Sawyer's CALUX-derived risk benchmark of 57.08 ppt CALUX-TEQ is also unreliable because he determined background based on a combination of results from floor dust samples and living dust samples ( i.e., he combined the Plaintiffs' CALUX-TEQ with the EPA's CALUX-TEQ). The Court agrees.⁷¹ Regardless of whether the results are CALUX-TEQ or GC/MS TCDD TEQ, the experts, and even Dr. Sawyer himself, drew distinctions between floor dust and living quarters dust, describing them as very different media. For the most part, the experts made a point to carefully delineate whether they were discussing floor dust or indoor living quarters dust when considering TCDD-TEQ levels. Dr. Sawyer's testimony, however, was internally inconsistent about this, in that in some circumstances he stated that floor dust is not comparable to indoor living quarters dust but then inexplicably found them comparable in other instances.⁷² Dr. Sawyer's only explanation for combining the average toxicity values of different media to determine the background was that the averages were statistically the same. But, he did not similarly combine the near-Site results (floor dust and indoor living quarters dust) because he found that those two sample sets were not statistically the same. Even assuming this is a valid reason, statistically speaking, Dr. Sawyer skewed the results by combining the two media to determine background but not doing so when comparing background to the near-Site CALUX-TEQs. His explanation of " statistical significance" between the sample sets as the sole basis to exclude the EPA's near-Site floor dust values is disingenuous because, had Dr. Sawyer combined the EPA's near-Site floor dust CALUX-TEQ (which overall were lower than Plaintiffs' results), as he did for the background, this would have seriously undermined his conclusion that the near-Site toxicity values were " significantly" elevated over background. See supra Note 42. The Court finds that this inconsistent use of toxicity values for different media ( i.e., floor dust and indoor living quarters dust) lacks scientific reliability. See McClain, 401 F.3d at 1245 (stating " any step that renders the analysis unreliable under the Daubert factors renders the expert's testimony inadmissible" (quoting Amorgianos v. Nat'l R.R. Passenger Corp., 303 F.3d 256, 267 (2d Cir. 2002)).

71 Defendants also argue that Dr. Sawyer's background benchmark is nothing more than impermissible ipse dixit. The Court does not agree that the calculation is ipse dixit. Dr. Sawyer used an accepted statistical analysis, used also by the EPA, to define background by virtue of the 95% upper confidence limit.

72 Dr. Sawyer combined the results of floor dust and indoor living quarters dust to calculate the CALUX background, but he drew a stark distinction between them later in his testimony. When it was pointed out that only two of the EPA's CALUX samples from the Stephen Foster neighborhood (within the proposed class and close to the Koppers Site) exceeded his CALUX risk benchmark, Dr. Sawyer dismissed the EPA's CALUX results as not comparable, because they were " floor dirt, [not] atmospheric deposited dust in the household living quarters. It's a different matrix." He also later explained that he did not use the EPA's GC/MS or CALUX results in his risk assessment because, he explained, the EPA " used floor dirt rather than living quarter dust." And yet, to " confirm" his CALUX risk assessment through the GC/MS data, Dr. Sawyer compared the average of Plaintiffs' within-class indoor living quarters dust (TCDD-TEQ) against the EPA's floor dust average (TCDD-TEQ) from background samples and insisted that this showed a significant elevation (confirming the CALUX data), and then he used the Plaintiffs' GC/MS average in the risk assessment. It appears that Dr. Sawyer had no trouble combining the results of two different media if doing so produced a result consistent with his theory, but he was quick to distinguish between them when the results did not support his conclusions.

         As for Dr. Sawyer's risk assessment analysis, the Court finds it unreliable to the extent he used CALUX data because a risk assessment analysis depends on quantifying the toxicity of the contaminant at issue, which, as the Court has already said is beyond the capability of CALUX. The FDOH and Dr. Cline both stated they could not reliably assess the health risk to residents in the Stephen Foster neighborhood based on the CALUX screening results for dust samples because the variability within the data was too great. See Pla. Ex. 5, at 4; Pla. Ex. 2. Even Dr. Sawyer withdrew his reliance on CALUX at the hearing, admitting that CALUX measurements in general are higher than GC/MS and stating that " there's no question" in this case that the results were inflated due to the presence of PBDD/F. Again, as the Eleventh Circuit has acknowledged, " even small differences in chemical structure can sometimes make very large differences in the type of toxic response that is produced." McClain, 401 F.3d at 1246 (internal quotations omitted).

         Dr. Sawyer tried to save his CALUX-based risk analysis by validating the results using GC/MS, but he did little to explain his validation process. According to Dr. Sawyer, he considered the Plaintiffs' 14 samples of living quarters dust collected within the proposed class area that were analyzed using GC/MS and found that the average of the GC/MS TCDD-TEQ value was 58.1 ppt TCDD-TEQ. In Dr. Sawyer's view, because this number was close to, and slightly above, his background benchmark of 57.01 ppt CALUX-TEQ, the GC/MS results would not change his risk assessment calculation. He also noted that the GC/MS " average" was significantly elevated over the EPA's GC/MS background (the geometric mean of floor dust), and he said this " fully confirmed" the CALUX background benchmark and the reliability of his risk assessment calculation. The Court disagrees.

         There is no scientific explanation for Dr. Sawyer's opinion that the " average" of the 14 GC/MS living quarters dust samples collected within the proposed class boundary validates the background benchmark. He never established a GC/MS background of living quarters dust to compare against this GC/MS class area average. Although he compared the GC/MS class area average against the EPA's background, this was not a valid comparison since the EPA tested floor dust, which as noted earlier, even Dr. Sawyer said was not comparable to living quarters dust.⁷³ See supra Note 71. Dr. Sawyer did not perform any side-by-side comparison of individual sample results to illustrate how, or whether, the individual GC/MS results were consistent or even comparable to the individual CALUX results, except to compare these two non-comparable " averages." With this limited set of GC/MS samples and no valid GC/MS background for comparison, Dr. Sawyer cannot reliably estimate whether the GC/MS results within the proposed class are elevated over background.⁷⁴ In sum, Dr. Sawyer's last-minute attempt to validate the CALUX numbers by merely saying that they are validated is classic ipse dixit. See Joiner, 522 U.S. at 146.

73 For instance, Dr. Sawyer drew a clear distinction when comparing the CALUX results. When he was asked how many of the EPA's 2012 CALUX results exceeded his CALUX risk benchmark, Dr. Sawyer answered, " Not many, but this [the EPA's test] was floor dirt, it wasn't atmospheric deposited dust in the household living quarters. Its' a different matrix." In fact, the EPA's CALUX results showed no elevation over background, and Dr. Sawyer admitted that only two of the EPA's CALUX samples exceeded his CALUX benchmark. He discounted the significance of this due to the difference between floor dust and indoor dust.

74 The Court notes that there is caselaw indicating that, regardless of any health risk, an individual may have a claim for trespass or nuisance based on the presence of a contaminant on his property caused by the defendant if the level of the contaminant is significantly elevated over the level that exists in background homes unaffected by the defendant. See Abrams v. CIBA Specialty Chem. Corp., 663 F.Supp.2d 1243, 1257-58 & n.30 (S.D. Ala. 2009) (discussing Alabama law). However, Dr. Sawyer's testimony does not reliably show the necessary elevation over background because of his failure to compare the near-Site GC/MS indoor living quarters dust test results with an urban background of GC/MS indoor living quarters dust test results.

         Defendants further argue that, even considering Dr. Sawyer's risk calculation as appropriately based on GC/MS data, Dr. Sawyer offered no scientific support for his opinion that a 4.09 x 10^-5 statistical increased risk of cancer represents a lifetime cancer risk that is " excessive and persistent," requiring remediation. The Court agrees. Dr. Sawyer concedes there is no applicable regulatory level defining the cancer risk from PCDD/F in indoor dust, and he did not disagree with the EPA's and FDOH's site-specific determination that the increased risk (even by his calculation) is in the FDOH's " very low" category. Most importantly, he never explained how a " very low" statistical risk of cancer presents an " excessive and persistent" health risk to the Plaintiffs. This conclusion is also mere ipse dixit.

         The Court also agrees that Dr. Sawyer's statistical health risk calculation, no matter how reliable,⁷⁵ is unhelpful to the class certification issues in this case. Statistical risk assessments, as opposed to clinical risk assessments, are largely used for regulatory purposes to protect public health and therefore are " of limited utility" in proving a classwide tort because they " do not provide information about actual risk or causation." Rhodes v. E.I. du Pont de Nemours and Co., 253 F.R.D. 365, 377 (S.D.W.Va. 2008) (finding statistical risk unhelpful and denying class certification in a toxic tort medical monitoring case). Only actual injury can be the basis for a tort. See generally, Eagle-Picher Indus., Inc. v. Cox, 481 So.2d 517 (Fla. 3rd DCA 1985) (plaintiff could not recover for a risk of cancer alone after exposure to asbestos). Notably, defense expert Dr. Scott Phillips testified that epidemiological studies of the class area over the past 30 years have shown no increase in cancer rates, and no expert contradicted this.⁷⁶ Dr. Sawyer's statistical risk ignores the actual cancer studies done in the area. Plaintiffs offered no evidence as to why a statistically increased risk is helpful in a class action where the record shows that actual or clinical risk would vary from person to person because the exposure level in each home within the class may be different (the Plaintiffs' GC/MS results showed a home with 9.9 ppt TCDD-TEQ very close to a home with 193 ppt TCDD-TEQ within the proposed class boundary), and existing epidemiological studies show that in fact there has been no increase in the cancer rates in the area in nearly 30 years. Dr. Sawyer's health risk opinion therefore will also be excluded as unhelpful to the class certification issues.

75 Dr. Sawyer's difference of opinion that the EPA and FDOH used the wrong ingestion rate is a battle of the experts. The Court does not question Dr. Sawyer's use of an ingestion rate contained in the EPA's own Handbook.

76 Dr. Sawyer only took some issue with the fact that the study encompassed a broader area than the proposed class boundary.

         2. Motion to Exclude Dr. Wade

         Defendants move to exclude Dr. Wade's opinions on grounds of qualifications and reliability. More specifically, Defendants argue that Dr. Wade is not a statistician and his application of Principal Component Analysis (PCA) in this case is not reliable.

         Dr. Wade is a forensic organic geochemist with over 36 years of post-doctorate work experience. He has worked as project manager for a variety of research programs, specializing in petroleum pollution, and has provided chemical forensics services, involving hydrocarbon fingerprinting analysis. Dr. Wade's education included courses in statistics, and he testified that as a forensic chemist, he frequently uses statistics to analyze data related to contamination. The Court finds that Dr. Wade is qualified to render opinions on chemical fingerprinting. Also, despite the fact that he is not a statistician, he regularly applies statistics as a forensic geochemist, and thus the Court finds him qualified to testify regarding the use of statistics in chemical fingerprinting analysis.

         Dr. Wade's report states summarily that he applied PCA. Dr. Wade explained at the hearing that the PCA " output" is very simple. To illustrate, Dr. Wade drew an analogy between PCA and throwing a bag of M& M candies. He explained that when you throw M& Ms, the candies are like data points and the data is examined based on their colors/characteristics. According to Dr. Wade, one can examine the output points and the " colors" to determine whether there is a relationship between them. Although Dr. Wade stated he had considered data related to the 17 congeners and had calculated the sum of all individual congeners, he gave no detailed analysis as to how the congeners were similar to or different from PCP-derived PCDD/F. He also said he " probably" would have looked at EPA data from the Site in the 1980s " at some point." Dr. Wade said he found a relationship in the data, but he also noted " other things going on besides a dominant source." He determined that the Koppers Site was a major source of PCDD/Fs in the area, but when asked about whether he observed confounders close to the Koppers Site, Dr. Wade testified, " I didn't do that level of analysis." Dr. Wade concluded only that, based on trends he observed in the data, " [t]here is a relation in the dioxin data for finger printing. It is not random."

         Defendants do not challenge the scientific acceptance or reliability of PCA methodology but argue that Dr. Wade's failure to complete the necessary second step of the analysis renders his opinion unreliable. The Court agrees. Dr. Uhler explained that distribution of the 17 congeners can only be identified for comparison when the factor score plot is read together with the factor loading chart so that a determination can be made as to which variables in the congeners are causing the similarities or differences in the data. He criticized Dr. Wade's analysis for not explaining the two independent but necessary steps of the PCA analysis. Dr. Uhler stated he does not use the same software program that Dr. Wade used for the PCA but that in his experience, PCA software automatically produces both the three-dimensional plots plus the factor loading chart showing the actual numerical output of the loadings. He said, " I can't imagine a program that doesn't produce both pieces of output." Dr. Uhler said only the factor score chart was appended to Dr. Wade's report, and he stressed that without interpreting both sets of information, it is not possible to reliably evaluate the factor score plots or determine which congeners are creating the similarities and the differences within the PCA analysis, which is critical for distinguishing among sources.

         Plaintiffs argue that Dr. Wade used all available data, normalized the data for analysis in an accepted manner, and followed the methodology using widely available software, but this is not reflected in Dr. Wade's abbreviated report or his testimony. The Court agrees with Defendants that Dr. Wade failed to explain how he had evaluated the factor score plots in relation to the factor loading chart, if he had done so at all, and also failed to provide any detail as to the 17 congeners and how they compared with the Koppers Site, other than to say he had calculated their sums. Instead, Dr. Wade expected that the Defendants and the Court should find the explanation in his voluminous file, but the hearing was his opportunity to explain, and he did not.⁷⁷ Dr. Wade's method suffers from the impermissible " black box" syndrome, where " data is fed at one end and . . . an answer emerges at the other, and the jury cannot see how the pieces fit together or how the data drives the conclusion." Open Text S.A. v. Box, Inc., Case No. 13-cv-04910-JD, 2015 WL 349197, at *6 (N.D. Ca. Jan. 23, 2015) (internal citation omitted). Without showing his completion of both of the PCA steps, that is, how his examination of the chart and the specific congeners leads to his conclusion that there is a relation in the congener data, the analysis is incomplete and scientifically unreliable. " [T]he court must undertake an independent analysis of each step in the logic leading to the expert's conclusions; if the analysis is deemed unreliable at any step the expert's entire opinion must be excluded." Hendrix v. Evenflo Co., Inc., 255 F.R.D. 568, 578 (N.D. Fla. 2009), aff'd., 609 F.3d 1183 (11th Cir. 2010). The need for an expert to complete and explain his methodology is at the heart of reliability.⁷⁸

77 Although an expert presenting a summary report to the court " in the form of conclusory statements devoid of factual or analytical support is simply not enough" under Daubert, the deficiency can be cured by a full evidentiary hearing held at the court's discretion, giving the proponent of the testimony the opportunity to present additional evidence to meet their burden of showing that the opinion was based on more than " the ipse dixit of the expert." Cook ex rel. Estate of Tessier v. Sheriff of Monroe Cty., Fla., 402 F.3d 1092, 1113 (11th Cir. 2005) (internal quotations omitted) (recognizing that a Daubert hearing may be helpful in complicated cases involving multiple expert witnesses and can provide an additional opportunity to demonstrate admissibility).

78 The scientific method must be a " disciplined, methodical" process and " the accuracy of the model depends upon the rigor applied in the input gathering process. An in-depth data investigation, a searching historical analysis, an excruciating attention to detail, and a methodology designed to wring error out of the process seems especially apropos here, where one is attempting to recreate decades of emissions, plume movements, and particle depositions." Coleman v. Union Carbide Corp., CIV. A. 2:11-0366, 2013 WL 5461855, at *24 (S.D.W.Va. Sept. 30, 2013). The analyses of Dr. Flowers, Dr. Sawyer, and Dr. Wade do not come close to approaching this standard.

         Additionally, even assuming Dr. Wade's PCA methodology is reliable, the Court finds his report and testimony woefully inadequate. While " Daubert is not a mechanism for grading expert reports," and infirmities in reports can be explored by deposition or a hearing, see Abrams v. CIBA Specialties Chem. Corp., No. 08-68-WS-B, 2010 WL 779273, at * 5 n.11 (S.D. Ala. Mar. 2, 2010), the expert's opinion must be supported by " the same level of intellectual rigor that characterizes the practice of an expert in the relevant field." Kumho Tire Co. v. Carmichael, 526 U.S. 137, 152, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). Dr. Wade maintained that he was only asked to determine whether there was a relationship in the data between on-Site and off-Site samples, but this reflects the limited helpfulness of his testimony, which contained only a vague conclusion that there is a " relation in the data." His analysis did not distinguish between dust or soil samples, did not distinguish as to location except as to on-Site versus off-Site, and provided no quantification of PCDD/F toxicity levels. While the testimony is minimally relevant to determining the source of contamination, the limitation inherent in the vague conclusion cannot be used to support class certification. Therefore, his testimony must be excluded as unhelpful as well. See Cook ex rel. Estate of Tessier v. Sheriff of Monroe Cty., Fla., 402 F.3d 1092, 1111 (11th Cir. 2005) ( courts may exclude expert testimony that is vague or not adequately explained).

         3. Motions to Exclude Dr. Exner and Dr. Kilpatrick

         Defendants also move to exclude the expert opinions of Dr. Jurgen Exner, regarding the cost of remediation, and Dr. John Kilpatrick, regarding classwide stigma damages. The Court will deny these motions as moot because, for reasons stated below, Dr. Exner's and Dr. Kilpatrick's opinions fail under the Court's class analysis, even assuming they are scientifically reliable and admissible.

         4. Motion to Exclude Dr. Uhler

         Plaintiffs move to exclude the chemical fingerprinting testimony of defense expert, Dr. Allen Uhler, insofar his fingerprinting analysis is used to oppose class certification. Dr. Uhler's credentials are not challenged.

         Dr. Uhler testified that there are a variety of viable chemical fingerprinting methodologies, and he chose the Diagnostic Ratio Analysis because he found it to be the most straight forward methodology for fingerprinting the PCP-derived chlorinated dioxin in this case. According to Dr. Uhler, Diagnostic Ratio Analysis is useful for identifying patterns in proportions or ratios between congeners necessary for discriminating among possible dioxin sources, and he supplemented his analysis with radar plotting, which is a graphical representation technique.⁷⁹ He noted that the PCDD/Fs found off-Site (within the proposed class boundary) had different diagnostic ratio values from PCP-derived PCDD/F, which he said could only be explained by the influences of other in-home or nearby sources. He also discussed many sources that can contribute to dioxin levels and give rise to localized areas of higher PCDD/F toxicity values.⁸⁰

79 Dr. Uhler explained the difference between the Diagnostic Ratio Analysis he applied and PCA analysis performed by Dr. Wade, stating that Diagnostic Ratio Analysis uses one or more pairs of individual congeners to calculate a value that can be compared (graphically in this case) against other samples, whereas PCA uses all congener variables at once to determine a mathematical algorithm to show which congeners are driving the variance between two samples. He finds PCA to be a valuable analysis but not a statistical one.

80 As one example, he noted that the National Institute of Health has documented higher concentrations of PCDD/F in soil near roadways. Also, various in-home sources can impact results.

         Plaintiffs argue that Diagnostic Ratio Analysis is not supported by peer reviewed literature and is unreliable. The Court disagrees. The diagnostic ratio method is an accepted manner of displaying dioxin-related data, as discussed in numerous scientific articles cited in Dr. Uhler's report and discussed and admitted at the hearing.⁸¹ See e.g., Def. Exs. 137, 141-153. Dr. Uhler fully explained his methodology and applied it to data collected in this case. Plaintiffs concede that the use of radar plots is standard in forensic data analysis. The Court finds Dr. Uhler's methodology reliable. Plaintiffs' argument that Dr. Uhler considered too few samples ( i.e., only furan samples) goes to weight, not admissibility. See Quiet Technology DC-8, Inc. v. Hurel Dubois UK Ltd., 326 F.3d 1333, 1345 (11th Cir. 2003).

81 Dr. Cline discussed the diagnostic ratio method as well, explaining that examining specific ratios is helpful in seeing relationships and patterns between chemical samples.

         Plaintiffs also argue that Dr. Uhler's opinion regarding other potential sources of dioxin, such as volcano fallout, is ipse dixit and lacking in analysis of the actual impact of potential alternate sources in Gainesville on any individual home.⁸² The Court disagrees. Dr. Uhler discussed a list of potential regional and property specific sources that could potentially contribute to dioxins in the class area, not intending to identify the specific source in each home. He explained that the discussion in his report of volcanic fallout was intended merely to describe that the general global background includes dioxins, not to pinpoint volcanoes as a source in this case.

82 Plaintiffs also argue that Dr. Uhler's opinion as to class size is mere ipse dixit because he was not asked to define the class. The Court need not address this argument because, to the extent Dr. Uhler offered any opinion on class size, the Court does not rely on it.

         5. Motion to Exclude Dr. Cline

         Plaintiffs move to exclude certain testimony and opinions of Dr. Cline. Dr. Cline, however, is a fact witness and both sides have relied on portions of her testimony and reports in this case. She was involved in the Koppers Site investigation and cleanup, first on behalf of the City and later on behalf of Protect Gainesville's Citizens, Inc., the EPA, and FDOH Dust Work Group. She prepared reports for the EPA and spoke at public meeting throughout the years, informing the community regarding technical aspects of the contamination testing campaigns, analysis, and cleanup efforts. Her testimony and opinions are not subject to Rule 702 or Daubert.

         6. Motion to Exclude Dr. Libicki

         Plaintiffs move to exclude the opinion of Dr. Libicki that vegetation (most notably trees) on and bordering the Koppers Site acted as a barrier to the transmission of dioxin-contaminated dust from the Site and quantitatively reduced emissions of the dioxin-contaminated dust into the proposed class area. Plaintiffs concede that Dr. Libicki, as a Stanford University-educated chemist with significant work experience related to the fate and transport of chemical pollutants, is qualified. Plaintiffs argue only that Dr. Libicki identified a methodology for quantifying these alleged reductions in dioxin-contaminated dust emissions but conceded she did not follow that methodology.

         Dr. Libicki testified that it is important for experts attempting to track the impact and source of dioxin contamination to understand and account for the impact of the local vegetation, which she observed was not uniform throughout the proposed class area. She identified the following methodology for determining the impact of trees on dust transport: (1) identify the types of trees on or bordering the site during the relevant time periods, (2) use that information to calculate the so-called " leaf area index" (" LAI" ) for the trees, and (3) and use wind-speed data to quantify the degree to which the leaves on the trees would act as a barrier to wind-driven, dioxin-contaminated dust. Plaintiffs assert that although Dr. Libicki identified the types of trees, purported to identify a leaf area index, and reviewed the historic meteorological data for the Koppers Site, she never actually quantified any reductions attributed to the presence of vegetation. Defendants state that Dr. Libicki was not asked to offer a quantitative opinion but rather a qualitative opinion, and her opinion, based on her observation, experience and historical data, is that the presence of a vegetative barrier along the western and southern boundaries of the Koppers Site would have reduced any airborne transport of dioxin from the Site, and should be admitted. The Court finds that Plaintiffs' arguments more appropriately go to weight rather than admissibility because Defendants have no burden to quantify a precise reduction of airborne transmission from trees. Dr. Libicki's opinion will not be excluded.

         7. Motion to Exclude Michael Corn

         Plaintiffs move to exclude the testimony of Michael Corn, who similarly testified that the presence of trees bordering the Koppers Site quantitatively reduced the emissions of dioxin dust from the Site. He also said that fugitive dust particles on the Koppers Site would travel at most only a few hundred feet, relying on an EPA document that Plaintiffs assert applies to larger dust particles and does not address smaller particles. Plaintiffs argue that Corn's opinion was merely a " personal" opinion based on ipse dixit, not a quantitative analysis. The Court rejects these arguments. Corn clearly explained at the hearing that he was expressing his professional opinion. Corn has over 40 years of experience as an environmental engineer and was fully qualified to testify regarding the impact of vegetation on the emissions and transport of dust from the Site. His testimony that leaf cover generally reduces the transmission of dust is based on that experience and the extent to which it is supported or not by an EPA document goes to weight, not admissibility.

         8. Motion to Exclude Dr. Phillips

         Plaintiffs also move to exclude the report and testimony of Dr. Phillips as speculative because he failed to perform his own risk assessment. Defendants have no burden in connection with the motion for class certification. Dr. Phillips criticized Dr. Sawyer's use of a regulatory risk analysis and his conclusion that there is an excessive and persistent risk of cancer. Although Dr. Phillips acknowledged that there is a low statistical risk from the reportedly elevated dioxin levels, he maintained there is no clinically significant cancer risk in this case and that the issue is too individualized to be appropriate for class certification. Dr. Phillips was not required to perform a statistical risk assessment calculation to offer these opinions, which are based on relevant epidemiological studies, exposure assessments, and FDOH health consultations and reports for an area that includes the proposed class. The Court finds no grounds for excluding Dr. Phillips's testimony.

         B. Class Certification Rule 23 Standards

         Plaintiffs are pursuing class certification on claims of negligence,⁸³ trespass,⁸⁴ and nuisance⁸⁵ on the ground that there is a 66% probability that any home within the class area is contaminated with PCP-derived chlorinated dioxin in living quarters dust, attributable to the Koppers Site, and consequently, there is an increased risk of cancer classwide, resulting in a need for classwide remediation. They also seek stigma damages for the diminished value of their properties.⁸⁶

83 In Florida, negligence requires proof of four elements: (1) duty of care; (2) breach; (3) legal or proximate causation; and (4) actual damages. Wallace v. Dean, 3 So.3d 1035, 1047 (Fla. 2009); Williams v. Davis, 974 So.2d 1052, 1056 (Fla. 2007).

84 Florida law states that trespass to real property is " an injury to or use of the land of another, by one who has no right or authority." Brown v. Solary, 37 Fla. 102, 112, 19 So. 161 (1896) ; see also Glen v. Club Mediterranee, S.A., 450 F.3d 1251, 1254 n.1 (11th Cir. 2006) (quoting Guin v. City of Riviera Beach, 388 So.2d 604, 606 (Fla. 4th DCA 1980)).

85 A nuisance under Florida law is " [a]nything which annoys or disturbs one in the free use, possession, or enjoyment of his property, or which renders its ordinary use or occupation physically uncomfortable." Jones v. Trawick, 75 So.2d 785, 787 (Fla. 1954); Knowles v. Cent. Allapattae Properties, Inc., 145 Fla. 123, 198 So. 819, 822 (Fla. 1940). " As to the nuisance claim, such an action is dependent upon an interference with the plaintiff's health, comfort, safety, or proprietary rights." State ex rel. Pettengill v. Copelan, 466 So.2d 1133, 1135 (Fla. 1st DCA 1985). Under Florida common law, " a tort plaintiff seeking to recover for economic harm caused by pollution or contamination need not own property that is itself polluted or contaminated." Adinolfe v. United Techs. Corp., 768 F.3d 1161, 1178 (11th Cir. 2014) (discussing nuisance and negligence claims) (citing Curd v. Mosaic Fertilizer, LLC, 39 So.3d 1216 (Fla. 2010)).

86 The Second Amended Complaint, ECF No. 238, was filed in 2013 before any off-Site soil remediation had occurred. In the Second Amended Complaint, the Plaintiffs raised a claim under Florida's Water Quality Assurance Act, Fla. Stat. § 376.30, et seq., as well as common law claims for strict liability, nuisance, negligence, trespass, and medical monitoring. Plaintiffs do not seek class certification on the statutory claim and at the hearing, Plaintiffs withdrew the claim for classwide medical monitoring. Nonetheless, Plaintiffs continue to argue that there is an excessive and persistent classwide health risk that requires remediation.

          Class actions are an exception to the rule that litigation is ordinarily conducted only on behalf of individually named parties. See Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 348, 131 S.Ct. 2541, 180 L.Ed.2d 374 (2011). The district court has discretion to certify a class if, after " a rigorous analysis," the court is satisfied that the requirements of Rule 23 of the Federal Rules of Civil Procedure are met. See Comcast Corp. v. Behrend, 133 S.Ct. 1426, 1432, 185 L.Ed.2d 515 (2013) (citing Dukes, 564 U.S. at 351); Carriuolo v. Gen. Motors, Co., 823 F.3d 977, 981 (11th Cir. 2016). The party seeking to certify the class bears the burden to establish the requirements of Rule 23 by a preponderance of the evidence, and remaining doubts are resolved against class certification. See Brown v. Electrolux Home Prods., Inc., 817 F.3d 1225, 1233 (11th Cir. 2016). The district court must resolve all conflicts in the evidence necessary to make the requisite determinations under Rule 23. See Comcast Corp. v. Behrend, 133 S.Ct. 1426, 1432, 185 L.Ed.2d 515 (2013) (noting Rule 23 is not a mere pleading standard and it may be necessary for the court to " probe behind the pleadings" to determine the certification issue); Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 178, 94 S.Ct. 2140, 40 L.Ed.2d 732 (1974) (the class action inquiry is not whether the plaintiff has stated a cause of action or will prevail on the merits but whether Rule 23 is satisfied); Sher, 419 Fed.Appx. at 891 (" [A] district court must make the necessary factual and legal inquiries and decide all relevant contested issues prior to certification." ). " Although the trial court should not determine the merits of the plaintiffs' claim at the class certification stage, the trial court can and should consider the merits of the case to the degree necessary to determine whether the requirements of Rule 23 will be satisfied." See Valley Drug Co. v. Geneva Pharm., Inc., 350 F.3d 1181, 1188 n. 15 (11th Cir. 2003); see also Amgen Inc. v. Connecticut Retirement Plans and Trust Funds, 568 U.S. 455, 133 S.Ct. 1184, 1194-95, 185 L.Ed.2d 308 (2013) (cautioning that, although a " class-certification analysis must be 'rigorous' and may 'entail some overlap with the merits of the plaintiff's underlying claim,'" quoting Dukes, 564 U.S. at 351, there is " no license to engage in free-ranging merits inquiries at the certification stage" ) (securities fraud).

          Before a class is certified, the plaintiff must establish all four prerequisites of Rule 23(a) plus at least one of the alternatives listed in Rule 23(b). See Valley Drug, 350 F.3d at 1188 (" Failure to establish any one of these four factors and at least one of the alternative requirements of Rule 23(b) precludes class certification." ). The Rule 23(a) prerequisites of " numerosity, commonality, typicality, and adequacy of representation," Little, 691 F.3d at 1304, require Plaintiffs to show, and the Court to find, by a preponderance of the evidence that:

(1) the class is so numerous that joinder of all members is impracticable; (2) there are questions of law or fact common to the class;

(3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and

(4) the representative parties will fairly and adequately protect the interests of the class.

         Fed. R. Civ. P. 23(a). Additionally, because Plaintiffs in this case seek certification under Rule 23(b)(3),⁸⁷ they must prove that:

87 Although the Second Amended Complaint references class certification under Rule 23(b)(1) and (b)(2), the Motion for Class Certification advances only Rule 23(b)(3), and therefore, the Court finds that any alternate grounds under Rule 23 have been abandoned.

questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.

Fed. R. Civ. P. 23(b)(3).

          In addition to the Rule 23 requirements, at least one named plaintiff seeking class certification must have standing. See Prado-Steiman v. Bush, 221 F.3d 1266, 1279-80 (11th Cir. 2000). The plaintiffs must also demonstrate that the proposed class is " adequately defined and clearly ascertainable." Carriuolo, 823 F.3d at 984 (quoting Little v. T-Mobile USA, Inc., 691 F.3d 1302, 1304 (11th Cir. 2012)). The district court has discretion to redefine the class to cure a deficiency in the proposed class, see Benefield v. Int'l Paper Co., 270 F.R.D. 640, 645 (M.D. Ala. 2010), but the Plaintiffs " must propose an administratively feasible method by which class members can be identified," Karhu v. Vital Pharms., Inc., 621 Fed.Appx. 945, 947 (11th Cir. 2015) (unpublished).

         In light of the Court's Daubert rulings excluding the opinions of Dr. Sawyer, Dr. Flowers, and Dr. Wade, the class analysis amounts to a hollow exercise. Although the proposed class representatives have standing,⁸⁸ Plaintiffs are unable to meet their burden for class certification. Without their experts, Plaintiffs cannot establish an ascertainable class or prove every element necessary under Rule 23, i.e., numerosity, typicality, commonality or the predominance of class issues over individual issues. Specifically, without Dr. Flowers's testimony, Plaintiffs cannot establish a class boundary or show causation classwide. Similarly, without Dr. Wade's chemical fingerprinting analysis, they have no means of showing that the Koppers Site is responsible for increased chlorinated dioxin contamination in indoor dust in the proposed class area. Absent Dr. Sawyer's testimony, there is no background benchmark by which to gauge whether homes within the proposed class are contaminated above background for purposes of establishing a class, proving classwide liability, or identifying a classwide health risk. In addition, although the GC/MS evidence is reliable, the record of GC/MS data in this case is insufficient alone to prove the elements necessary for class certification. Thus, the motion for class certification must be denied.

88 " [A]ny analysis of class certification must begin with the issue of standing." Griffin v. Dugger, 823 F.2d 1476, 1482 (11th Cir. 1987). At least one named plaintiff must have standing in the constitutional sense to represent the class, which requires a showing of a redressable injury that is fairly traceable to the defendant's conduct. See id. However, " mere apprehension of speculative future injury" does not meet Article III's standing requirement. Fisher v. Ciba Specialty Chems. Corp., 238 F.R.D. 273, 298 (S.D. Ala. 2006) (quoting LaBauve v. Olin Corp., 231 F.R.D. 632, 650 (S.D. Ala. 2005)). Standing is not challenged in this case, but the Court has considered the issue. There is evidence that at least one proposed class representative, Day, has shown by the EPA's GC/MS test results that her rental home, which is within the proposed class area, contains PCDD/F contamination in floor dust in an amount that is elevated over the average background levels. The EPA's testing shows a geometric mean of 13.6 ppt TCDD-TEQ for background homes whereas Day's rental home tested at 44.9 ppt TCDD-TEQ. The Eleventh Circuit has stated that " exposure to a toxic substance constitutes sufficient injury in fact to give a plaintiff standing to sue in federal court." Adinolfe v. United Techs. Corp., 768 F.3d 1161, 1172 (11th Cir. 2014) (quoting Carlough v. Amchem Prods, Inc., 834 F.Supp. 1437, 1454 (E.D. Pa. 1993)). This sufficiently demonstrates an injury in the constitutional sense for at least one named Plaintiff.

         Alternatively, even if the Court erred in its analysis on any Daubert motion, in whole or in part, class certification would still be denied. Therefore, the Court will address some of the class issues in the alternative for this purpose, assuming the admissibility of Plaintiffs' expert evidence.

         1. Ascertainability

         " An identifiable class exists if its members can be ascertained by reference to objective criteria." Bussey v. Macon Cty. Greyhound Park, Inc., 562 Fed.Appx. 782, 787 (11th Cir. 2014) (quoting Fogarazzo v. Lehman Bros., Inc., 263 F.R.D. 90, 97 (S.D.N.Y.2009)); see also Manual for Complex Litigation (Fourth) § 21.222 (2004) (class definition must be " precise, objective, and presently ascertainable . . . by reference to objective criteria" ). The class definition is stated as:

[T]hat area in the City of Gainesville, Florida within in an area surrounding the Koppers Site as depicted in [Pla. Ex. 131]. The Proposed Class Area is a polygon with nine vertices as shown in Figure 1 [Pla. Ex. 131]. Straight line segments connect the vertices to form a closed polygon with a finite area. Coordinates for each vertex are given in Table 1 referenced to the 1983 North American Datum (NAD83) using the Universal Trans-mercator projection (UTM) for Zone 17N, which includes Gainesville, Florida.

The Proposed Class includes all current propert[y] owners of single family residential properties within the Class Affected Area.

ECF No. 778; see also Pla. Ex. 131. This definition is clearly over-inclusive, even considering Plaintiffs' CALUX evidence, and there is no means of re-defining an appropriate class on this record. According to Plaintiffs, each named putative Plaintiff is similarly affected by the toxic dioxins that have migrated from the Koppers Site onto their properties. The proposed class definition, however, does not mention contamination in indoor dust but purports to include all property owners in this geographic area, regardless of whether the property is contaminated. More importantly, because the data reflects that chlorinated dioxin was identified in all samples (background and within the proposed class) Plaintiffs must establish a class boundary in which contamination from PCP-derived chlorinated dioxin (PCDD/F) exceeds the upper limit of background TCDD-TEQ values and ultimately must prove that the specific contamination in their homes is attributable to the Koppers Site in order to prevail. The record shows that not all 7,000 residential properties within the existing proposed class boundary are similarly contaminated above Dr. Sawyer's CALUX-TEQ background. Plaintiffs' own expert, Dr. Flowers, opined that there is only a 66% probability that any home within the proposed geographical class boundary is contaminated in excess of Dr. Sawyer's CALUX-based benchmark. To the extent Dr. Sawyer is relying instead on the GC/MS average of TCDD-TEQ values within the proposed class as confirming the CALUX-derived benchmark, the GC/MS data itself also shows varying TCDD-TEQ values, such that not all homes within the proposed class will exceed Dr. Sawyer's benchmark. Dr. Sawyer provided no analysis of background GC/MS TCDD-TEQs but assumed from the average of the near-Site TEQs that the background would be appreciably less, and thus there is no scientifically established method for re-drawing the geographical boundary of the proposed class. Thus, while the Court has discretion to redraw the class boundary or redefine the class as residential property owners whose properties contain chlorinated dioxin (PCDD/F) dust contamination in excess of background levels, there is no objective means for determining which property owners are within that class short of establishing a valid GC/MS background for living quarters dust and testing every individual home.⁸⁹ Thus, Plaintiffs have not proven an ascertainable class.

89 If Plaintiffs are asserting that the class is ascertainable because of PCDD/F in soil, the class is still overly broad in light of the remedy sought. The record shows that soil remediation has been carried out in the limited portion of the Stephen Foster neighborhood where TEQs were the highest and there is no basis for establishing a class based on soil contamination. Also, although Day appears to claim stigma damages for one of her homes without showing indoor dust PCDD/F contamination levels exceeding any background benchmark, Plaintiffs have not argued that a subclass of residential property owners who have only suffered stigma damages should be established. Even if they had, identifying a proximity subclass with objective criteria could not be accomplished without individual testing.

                    2. Commonality

         " [F]or purposes of Rule 23(a)(2) even a single common question will do." Carriuolo, 823 F.3d at 984 (quoting Dukes, 564 U.S. at 359); Williams v. Mohawk Indus., Inc., 568 F.3d 1350, 1355 (11th Cir. 2009) (" Commonality requires that there be at least one issue whose resolution will affect all or a significant number of the putative class members." ). Plaintiffs need not demonstrate that each element of their claim is " susceptible to classwide proof." Amgen Inc. v. Connecticut Ret. Plans and Trust Funds, 568 U.S. 455, 133 S.Ct. 1184, 1196, 185 L.Ed.2d 308 (2013). However, the common question must be one that has a classwide answer or resolution, " which means that determination of its truth or falsity will resolve an issue that is central to the validity of each one of the claims in one stroke." Carriuolo, 823 F.3d at 984. Also, the class members must have all suffered the same injury. See Dukes, 564 U.S. at 350 (quoting Gen. Telephone Co. of the Southwest v. Falcon, 457 U.S. 147, 157, 102 S.Ct. 2364, 72 L.Ed.2d 740 (1982)).

         Plaintiffs argue that commonality is satisfied by several common questions of liability and damages, such as, whether the proposed class members and their properties are " all similarly affected by the chlorinated dioxin," whether the chlorinated dioxin migrated from the Koppers Site onto Plaintiffs' properties due to negligence, whether there is a classwide method of remediation, and whether property values in the class are affected by the contamination. The Court agrees these are common questions, but raising common questions, " even in droves," does not mean that class certification is appropriate; what matters instead is " the capacity of a classwide proceeding to generate common answers." Dukes, 564 U.S. at 350 (quoting Nagareda, Class Certification in the Age of Aggregate Proof, 84 N.Y.U.L.Rev. 97, 132 (2009)). The Supreme Court has stated that " [d]issimilarities within the proposed class are what have the potential to impede the generation of common answers." Id. (quoting Nagareda, Class Certification in the Age of Aggregate Proof, 84 N.Y.U.L.Rev. 97, 132 (2009)). On the other hand, commonality also is recognized as imposing a " low hurdle." Williams, 568 F.3d at 1356.

         Despite Plaintiff's identification of numerous common questions in this case, the Court finds that most of them would not provide a common answer. Only one is needed for class certification, however. The question of whether Defendants negligently released PCP-derived PCDD/F into the surrounding community is the only identifiable common question in this case.

         3. Predominance

         The predominance inquiry of Rule 23(b)(3) " is far more demanding" than the commonality inquiry. Carriuolo, 823 F.3d at 985 (quoting Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 623-24, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997)); Williams, 568 F.3d at 1357. Under this standard, " the issues in the class action that are subject to generalized proof, and thus applicable to the class as a whole, must predominate over those issues that are subject only to individualized proof." " Common issues can predominate only if they have a direct impact on every class member's effort to establish liability that is more substantial than the impact of individualized issues in resolving the claim or claims of each class member." Carriuolo, 823 F.3d at 985; see also quoting Klay v. Humana, Inc., 382 F.3d 1241, 1255 (11th Cir. 2004), abrogated in part on other grounds by Bridge v. Phoenix Bond & Indem. Co., 553 U.S. 639, 128 S.Ct. 2131, 170 L.Ed.2d 1012 (2008). Additionally, Rule 23(b)(3) also requires proof that " a class action is superior to other available methods for fairly and efficiently adjudicating the controversy." Fed.R.Civ.P. 23(b)(3).

         Even assuming that Plaintiffs can prove the common question that negligence occurred at the Koppers Site and caused off-Site migration of PCDD/F into indoor living quarters dust, proof of any Plaintiffs' claim of negligence, nuisance or trespass " will require distinctly case-by-case inquiries into the facts," Jackson v. Motel 6 Multipurpose, Inc., 130 F.3d 999, 1006 (11th Cir. 1997). Considering Plaintiffs' experts' opinions as reliable, their testimony does not establish that the homes within the proposed class are similarly contaminated, and in fact Dr. Flowers admitted that not every home within the proposed class is contaminated above Dr. Sawyer's benchmark. Even considering the GC/MS evidence, there is no established upper confidence limit of GC/MS background to compare with living quarters dust to determine if contamination within the class area exceeds background. The data and testimony of numerous experts, namely Dr. Uhler, Dr. Libicki, and Michael Corn, convincingly show that indoor living dust contamination is not consistent throughout the proposed class due to a variety of factors, including numerous in-home factors, which necessitates testing each home individually to determine whether each home has contamination levels above background and whether the contamination can be attributed to the Koppers Site. Given the variability in the CALUX measurements, and in the GC/MS TCDD-TEQs, there is no basis for assuming that each home within the boundary of this proposed class is contaminated in excess of background levels.

         Each Plaintiff will also have to prove that the PCDD/F in their home is attributable to the Koppers Site. Although Dr. Flowers identified a " ramp up" of dioxin levels in the CALUX-TEQ values and the Koppers Site is recognized as a dominant source, no such " ramp up" exists in the GC/MS TCDD-TEQ values, and Dr. Flowers, Dr. Sawyer, and Dr. Wade all admitted that variabilities and confounders in the data exist. Dr. Wade offered only a vague conclusion that a relationship exists between the off-Site data and the Koppers Site, which tells little or nothing about classwide causation. Dr. Uhler, Dr. Cline, Dr. Libicki, and Corn all agree, and Plaintiffs' experts do not seriously dispute, that multiple variables, such as wind, storms, and vegetation, affected the transport of contaminated dust from the Koppers Site. The alternate sources must be accounted for and ruled out in each home in order to prove that the excess PCDD/F in any given home is in fact attributable to the Koppers Site. Thus, a property-by-property inquiry is inevitable to determine the connection between in-home dust contamination and a particular source with regard to any particular piece of property in the proposed class area. See Fisher v. Ciba Specialty Chems. Corp., 238 F.R.D. 273, 275 & n.70 (S.D. Ala. 2006) (noting expert testimony about dust transport tells little about the source or quantity of contamination on any particular parcel of property).

         It is also obvious that individual re-testing of all homes will be necessary in light of the now-completed soil remediation in the Stephen Foster neighborhood. Plaintiffs under took no post-remediation testing, and the limited re-testing performed by the Dust Work Group showed that TCDD-TEQ values in floor dust had returned to background levels in four of the five homes re-tested after soil remediation and ordinary cleaning in the homes. The one increase noted was attributed to the use of a wood burning stove inside that home. This fact further confirms the individualized nature of the proof necessary in this case.

         Similarly, because the record shows varying chlorinated dioxin levels from home to home, any health risk assessment must be undertaken on a plaintiff-to-plaintiff basis, considering a long list of individual factors. The statistical health risk Dr. Sawyer identified cannot apply equally to all class members. See Rhodes, 253 F.R.D. at 377 (stating, a " risk assessment cannot and does not support an opinion that each individual class member has experienced a significantly increased risk of disease" ).

         Damages likewise will be dominated by individual factors. Assuming the admissibility of Dr. Exner's and Dr. Kilpatrick's testimony, the common methods they are riddled with individual inquiries. First, and foremost, their opinions are predicated on an assumption that each home was similarly contaminated, which the record does not support, and even Plaintiffs concede that not every home will require remediation. Also, some of the putative class members have already received payment for indoor cleaning, but it is unknown whether the money was used for remediation (Day testified that she did not use the money because it was not enough for proper cleaning). Further, Dr. Exner's common remediation method would involve numerous individual factors as to those homes determined to need remediation. For example, although a set cost per foot relative to the square footage of the house could be established as a baseline, as to any individual home, the method and cost will vary depending on the need to repeat the cleaning process, the number of walls and surfaces in the home, whether there is an attic, and the size of the residence. Regarding stigma damages, Dr. Kilpatrick's common method for mass appraisal will similarly need to account for numerous individual characteristics of the homes and surrounding neighborhoods, which are not uniform throughout the proposed class. Moreover, Dr. Kilpatrick has not yet run his AVM model, not even for the class representatives, and thus, there is no evidence that any home in the proposed class area has suffered a diminution in property value in this case, let alone classwide diminution in value.

         In conclusion, and without question, the individual issues in this case will predominate over class issues. As decided in other contamination cases, " whether a plaintiff's property is contaminated, the source of such contamination, the extent of such contamination, the cause and timing of harm, and the resulting damage measured in diminution of property value, are all questions that will require plaintiff-by-plaintiff scrutiny." See LaBauve v. Olin Corp., 231 F.R.D. 632, 673 (S.D. Ala. 2005) (explaining the difficulty of showing predominance where classwide proof would not satisfy the elements of trespass and nuisance in a case involving mercury dust contamination). This case is no exception. Thus, even assuming the reliability of Plaintiffs' experts' opinions, the Court concludes class certification is not appropriate.

         Accordingly:

         1. Defendants' Motion to Exclude Plaintiffs' Expert George C. Flowers, Ph.D. and William R. Sawyer, Ph.D. Reliance on CALUX as Unreliable, ECF No. 729, is GRANTED, and Defendants' Motions to Exclude Plaintiffs' Expert George C. Flowers, Ph.D., ECF No. 728, and William R. Sawyer, Ph.D., ECF No. 727, are GRANTED.

         2. Defendants' Motion to Exclude Class Certification Expert Michael J. Wade, ECF No. 723 is GRANTED.

         3. Defendants' Motions to Exclude Plaintiffs' Expert John A. Kilpatrick, ECF No. 725 and Jurgen H. Exner, ECF No. 726, is MOOT.

         4. Plaintiffs' Motions in Limine to Exclude Testimony and Opinions of Allen Uhler, Ph.D., ECF No. 716, Patricia Cline, ECF No. 717, Sherry Libicki, ECF No. 719, Michael Corn, ECF No. 721, and Scott Phillips, M.D., ECF No. 724, are DENIED.

         5. Plaintiffs' Motion for Class Certification, ECF No. 778, is DENIED

         6. The parties are directed to confer and submit an amended joint report within twenty-one (21) days advising the Court as to their discovery and scheduling needs going forward without the class action allegation.

         DONE AND ORDERED.