Full Opinion

United States Court of Appeals For the First Circuit No. 24–1869 KEVIN BROWN, Individually and as Natural Parent and Next Friend of G.B. and D.B.; CHRISTOPHER BLUNDON, Individually and as Natural Parent and Next Friend of A.B. and Z.B.; THERESA BROWN, Individually and as Natural Parent and Next Friend of G.B. and D.B.; KIMBERLY PEICKER, Individually and as Natural Parent and Next Friend of M.P. and A.P.; MARK PEICKER, Individually and as Natural Parent and Next Friend of M.P. and A.P.; CORRIN WILSON, Individually and as Natural Parent and Next Friend of K.H., L.H., and R.H.; BRIAN HARRIS, Individually and as Natural Parent and Next Friend of K.H., L.H., and R.H.; OLEG GOLTSOV, Individually and as Natural Parent and Next Friend of J.G., Plaintiffs, Appellees, ADAM W. DYER; BRIAN MENDEZ, Individually and as Natural Parent and Personal Representative for M.M.; BRENDA MORSE; JAMES BOLLENGIER; JEAN DOWLING; AMY MENDEZ, Individually and as Natural Parent and Personal Representative for M.M.; BEVERLY VOLNER; JAMES VOLNER; DAWNA WORCESTER; RICHARD SLIDE, Individually and as Natural Parent and Personal Representative for C.S. and M.S.; ERIN SLIDE, Individually and as Natural Parent and Personal Representative for C.S. and M.S.; JONATHAN KILEY, Individually and as Natural Parent and Personal Representative for A.K., K.K., E.K. 1, and E.K. 2; JENNIFER KILEY; APRIL PROVENCHER; PHYLLIS PROVENCHER; RICHARD PROVENCHER; DARLENE DEBLOIS; DARLENE E. DEBLOIS LIVING TRUST; DAVID TRUE; BRIDGET HEARD, Plaintiffs, v. SAINT-GOBAIN PERFORMANCE PLASTICS CORPORATION; GWENAEL BUSNEL, Defendants, Appellants, JOHN DOES 1–5, Defendants. APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF NEW HAMPSHIRE [Hon. Joseph N. Laplante, U.S. District Judge] Before Barron, Chief Judge, Kayatta and Montecalvo, Circuit Judges. Mark S. Cheffo, with whom Bruce W. Felmly, Jeremy T. Walker, McLane Middleton, PA, Douglas E. Fleming III, Bert L. Wolff, Paul A. LaFata, and Dechert LLP were on brief, for appellants. Stephanie A. Douglas, Susan M. McKeever, A. Lane Morrison, and Bush Seyferth PLLC on brief for Product Liability Advisory Council, Inc. and the National Association of Manufacturers as amici curiae in support of appellants. Kevin S. Hannon, with whom Singleton Schreiber, LLP, Morgan A. Hollis, Gottesman & Hollis P.A., John A. Yanchunis, and Morgan & Morgan were on brief, for appellees. Robert S. Peck, Center for Constitutional Litigation, PC, Lori Andrus, Jeffrey R. White, and American Association for Justice on brief for the American Association for Justice as amicus curiae in support of appellees. November 17, 2026 KAYATTA, Circuit Judge. Plaintiff property owners in and around Merrimack, New Hampshire, allege that defendant Saint-Gobain Performance Plastics Corporation released airborne toxic chemicals from its manufacturing facility in Merrimack and that these chemicals then made their way into the municipal water supply and private wells, invading plaintiffs' property and contaminating their drinking water.1 Before the district court, plaintiffs moved to certify a class comprised of property owners who used the municipal water supply and property owners who used private wells within a certain area around the Saint-Gobain facility. The district court allowed the motion in part and denied the motion in part, certifying a class under Federal Rule of Civil Procedure 23(b)(3). In so doing, it found that Saint-Gobain's liability on plaintiffs' negligence, negligent-failure-to-warn, and trespass claims could be determined on a common basis and "predominate[d] over the individualized inquiries pertaining to damages." Brown v. Saint-Gobain Performance Plastics, No. 16-CV-242, 2023 WL 9023158, at *22 (D.N.H. Dec. 29, 2023). Saint-Gobain timely sought permission to appeal, which we granted. 1 Plaintiffs also name as a defendant Gwenael Busnel, general manager of the Saint-Gobain facility during the relevant time. Because plaintiffs assert no claims solely against Busnel, we generally refer to the defendants as "Saint-Gobain" throughout this opinion. - 3 - We now reverse in part and affirm in part. We reverse certification with respect to plaintiffs' negligence and negligent-failure-to-warn claims, and we affirm certification with respect to plaintiffs' trespass claim. I. The facility at the heart of this case -- which we call the "Merrimack Facility" or, simply, the "Facility" -- sits on the left bank of the Merrimack River, in the northeast pocket of the town of Merrimack, surrounded by the municipalities of Bedford, Manchester, Londonderry, and Litchfield. Since the 1980s, the Facility has manufactured chemical- and weather-resistant fiberglass and films under the auspices of Saint-Gobain and its predecessor-in-interest ChemFab Corporation. During that time, the manufacturing processes at the Facility have used poly- and perfluoalkyl substances (PFAS), including perfluorooctanoic acid (PFOA).2 PFAS are human-made chemicals commonly used in the manufacture of many commercial products. Studies suggest that exposure to PFAS, including PFOA, may lead to adverse health effects for humans. PFOA is a persistent chemical that does not break down and can accumulate 2 We follow the district court's lead and largely refer to the substances at issue as PFOA. See Brown, 2023 WL 9023158, at *2 n.3. The distinctions between PFAS generally, PFOA specifically, and other subtypes of PFAS do not affect the resolution of this appeal. - 4 - both in the environment and in the human body. Particulate PFOA used in manufacturing may be released into the air via stack emissions, transported by wind, deposited onto the ground, and then dissolved into groundwater, contaminating the drinking water of nearby communities. That is precisely what plaintiffs allege happened here. On March 4, 2016, the New Hampshire Department of Environmental Services (NHDES) issued a press release announcing that Saint-Gobain had detected PFOA in water samples taken from faucets at its Merrimack Facility. As a result, NHDES worked with Saint-Gobain and the Merrimack Village District Water Works (the "District Water Works") to investigate the presence of PFAS in drinking water in Merrimack. The investigation focused on two sources of water in the area: the District Water Works, which supplies water to 25,000 customers in Merrimack; and private wells. Ultimately, the investigation found PFOA contamination exceeding regulatory standards in groundwater around the Merrimack Facility. The level of PFOA varied at different wells, with concentrations generally decreasing as distance from the Merrimack Facility increased. Overall, within the District Water Works, PFOA contamination varied from "non-detect" to 141 parts per trillion ("ppt"); 45% of results showed contamination at 20 ppt or - 5 - greater, with 4% of results at 70 ppt or greater.3 In private wells, PFOA concentrations ranged from undetectable to 1600 ppt, with 70% of results at 20 ppt or greater and 34% of results at 70 ppt or greater. Concurrently with the NHDES investigation, Saint-Gobain hired Barr Engineering Company "to evaluate and simulate transport mechanisms of PFOA, released by historical air emissions at [the Facility] that may have resulted in the presence of PFOA in soil and groundwater in the vicinity of th[e] facility." To do so, Barr calculated air emissions of PFOA from the Facility from 2004 to 2015, and used the data from 2004 to estimate PFOA emissions from 1986 to 2003, a period for which data was not available. Barr then used four separate models to simulate the four phases of potential PFOA transport from the Facility to local groundwater. First, Barr used the AERMOD Modeling System, the dispersion model preferred by the U.S. Environmental Protection Agency, to simulate air dispersion and deposition on the ground, thereby "estimat[ing] annual PFOA ground surface deposition at each model node." Second, Barr used the Soil–Water Balance model to calculate groundwater infiltration rates for PFOA deposited on the ground. Third, Barr simulated the "leaching and retention of PFOA in the unsaturated zone" using the MODFLOW-NWT and MT3D-USGS codes. Finally, Barr 3 New Hampshire sets its "Ambient Groundwater Quality Standard" for PFOA at 70 ppt. - 6 - used the MT3D-USGS code to simulate PFOA transport in groundwater and the MODFLOW-NWT code to simulate groundwater flow. Barr then compared these simulation results with measurements of PFOA in soil and groundwater in the study area to determine whether those measurements could be associated with the Merrimack Facility. In a report released in June 2017 and updated in September 2018 (the "Barr Report"), Barr concluded that "air deposition of emitted PFOA from the [Merrimack F]acility may have contributed to the observed PFOA in groundwater in portions of the study area located in close proximity to the [Merrimack F]acility." However, the Barr Report also noted "areas where air deposition from the [Merrimack F]acility does not explain the presence or reported concentration of PFAS," and posited "numerous other likely sources of PFAS in the study area," which it did not simulate for purposes of the report. In 2018, under a consent decree in an action brought by the State of New Hampshire, Saint-Gobain agreed to pay certain remediation costs for select contaminated properties in Bedford, Litchfield, Manchester, and Merrimack. The consent decree did not contain any admission of wrongdoing or liability on Saint-Gobain's part. II. Plaintiffs began this putative class action in New Hampshire state court in May 2016. Saint-Gobain removed the action - 7 - to federal court, and the case followed a labyrinthine path of pretrial proceedings for the next several years. We summarize here only the relevant details of the case as it now stands. A. Named plaintiffs are all occupants or owners of residential property in or around Merrimack. They allege that Saint-Gobain's airborne emissions of PFOA from the Merrimack Facility intruded onto their property and contaminated their household wells and water supplies. Asserting claims for trespass, nuisance, negligence, and negligent failure to warn,4 plaintiffs seek damages for the diminished value and marketability of their property, the costs of remediation and mitigation, the loss of use and enjoyment of their property, and annoyance, discomfort, and inconvenience.5 In December 2020, plaintiffs filed the motion for class certification that gives rise to this appeal. In it, plaintiffs sought to represent a class of property owners pursuing "damages 4 For simplicity, we refer to plaintiffs' negligence and negligent-failure-to-warn claims together as their "negligence- based claims." 5 Plaintiffs also originally included a medical-monitoring claim for the increased risk of illness resulting from exposure to PFOA. On certification from the district court, the Supreme Court of New Hampshire held that New Hampshire does not recognize a cause of action or remedy for medical monitoring. Brown v. Saint-Gobain Performance Plastics Corp., 300 A.3d 949, 950 (N.H. 2023). As a result, the district court mooted plaintiffs' medical-monitoring claims. - 8 - for harm to their real property interests caused by the contamination" from Saint-Gobain's PFOA emissions, to wit, "economic losses for diminution in value and cost of replacement water," as well as "non-economic claims for interference with property rights." Plaintiffs defined the putative class as: All persons who on or after March 4, 2016 own or owned residential properties with private wells in the Private Well Property Owners Class Geographic Area or residential properties in the Merrimack Village District Water Works (MVDWW) Class Geographic Area which are supplied household water by [the District Water Works] . . . . The Private Well Property Owners Class Geographic Area includes: In Bedford and Merrimack, the geographic area west of the Merrimack River within three (3.0) miles of the property boundary of the Saint-Gobain Site; in Litchfield, the Geographic area bounded by the Merrimack River on the west, Cummings Drive on the South, extended east to the Merrimack River and west to the Londonderry Town line, and the Londonderry Town line on the East and the City of Manchester on the North and East . . . and the geographic area in Manchester bounded by Raymond Wieczorek Drive on the North . . . . Although plaintiffs' pleading subdivided this class into two "subclasses" -- property owners who used private wells and property owners who used the District Water Works supply -- they make no argument that either subclass might be certified even if - 9 - the class as a whole is not. Like the parties and the district court, we therefore train our attention on the class as a whole. The parties litigated plaintiffs' class-certification motion for three years, with each side submitting expert reports to support its position. Because much of the parties' briefing on appeal focuses on what those expert reports do and do not show, we briefly summarize the relevant reports here. Plaintiffs' expert David Sullivan, a meteorologist, submitted two reports. His reports "evaluate[d] the air quality and airborne deposition pathway for [PFAS] in relation to the class geographic area and the observed PFAS contamination in groundwater." Sullivan did not conduct any modeling of PFAS deposition himself but rather relied on a site visit and existing documentation and data, including the Barr Report's modeling, well sampling results, and meteorological data. Based on this evidence, Sullivan concluded that "dispersion and deposition of air pollutants from the [Merrimack Facility] have occurred throughout and beyond the boundaries of the class geographic area," and that "[t]he methodology of the study of the dispersion and deposition of pollutants from the [Facility] is common to the class geographic area." In deposition testimony, Sullivan further opined that, although previous investigations may not have detected PFOA in some well samples, PFOA "has fallen on every property within the class geographic area." - 10 - Plaintiffs' expert Dr. Hyeong-Moo Shin was at the relevant time an Assistant Professor in the Department of Earth and Environmental Sciences at the University of Texas at Arlington. Plaintiffs asked Dr. Shin to "examine whether the properties defined in the lawsuit were contaminated with [PFOA] due to releases from the [Merrimack Facility]." Dr. Shin concluded that "[a]ir deposition of emitted PFAS from the [Merrimack Facility] has contributed to the observed PFAS in groundwater in the class area." Dr. Shin acknowledged that other sources may have contributed to contamination within the class area but opined that "[o]ther likely sources of PFAS listed in the Barr [Report] cannot explain patterns of observed PFOA in groundwater in close proximity to the [Merrimack Facility] and widespread contamination of PFOA in the properties defined in the lawsuit." Plaintiffs' expert Dr. Russell Detwiler, an Associate Professor in the Department of Civil and Environmental Engineering at the University of California, Irvine, "examine[d] the migration of PFOA in soil and groundwater related to the [Merrimack] Facility." Dr. Detwiler reviewed and re-ran several of the Barr Report's models, calibrating the models to the measured results from the District Water Works wells. His report suggests that the Barr Report may have underpredicted the PFOA concentrations in groundwater. - 11 - Christopher Baggett, a licensed engineer from engineering firm Wright-Pierce's Water Treatment Practice Group, also provided plaintiffs with a report modeling "the transport of PFOA through the [District Water Works]" over time. This report "provide[d] reasonable estimates of contaminant levels at various locations" in the District Water Works. Finally, plaintiffs presented two reports on the economic damages from PFOA contamination. The first -- prepared by economists Jeffrey Carr, Robert Chase, and Nathan Masse -- calculated the costs associated with switching from a contaminated private well to the District Water Works, concluding that "the present value sum of additional utility costs to residential properties in the Towns of Litchfield, Bedford, and Merrimack is approximately $2,626,494." Dr. Randall Bell, a real estate economist and appraiser, calculated the overall diminution in value of properties in the class area due to PFOA contamination, finding that the total loss was not less than $577,935,862. Saint-Gobain countered with several experts of its own. Principally, Dr. Michael Mobile, a civil engineer, provided a pair of reports rebutting the opinions of Sullivan, Dr. Shin, and Dr. Detwiler. Dr. Mobile acknowledged that the Merrimack Facility "is a source of PFOA measured in groundwater at certain locations within the proposed Private Well Class Geographic Area." But Dr. Mobile pointed out that plaintiffs' experts did not rule out - 12 - other potential sources of PFOA and criticized their reliance on the Barr Report, which he characterized as limited to "generally assess[ing] the potential chemical transport pathway" of PFOA. Dr. Sorab Panday and John A. Connor, engineers at GSI Environmental Inc., also provided a pair of reports addressing "the nature and extent of PFOA . . . within the water sources and distribution systems for the water supply district and private well users" and rebutting the modeling conducted by plaintiffs' experts, including Dr. Detwiler and Baggett. The reports concluded that "[g]iven the inevitable variability of PFOA . . . levels in the water delivered to individual water users, determination of their level or duration of exposure requires an individual analysis and cannot be generalized across the [District Water Works] service area, pressure zones, neighborhoods, or among private well users." Further, according to these reports, "[t]he actual water quality measurements that are available do not tell us when PFOA arrived at the [District Water Works] wells, when the concentrations peaked, or at what concentrations." Finally, atmospheric scientist Lyle Chinkin also reviewed the reports of Sullivan, Dr. Shin, and Dr. Detwiler. Chinkin, like Dr. Mobile, criticized plaintiffs' experts' reliance on the Barr Report, averring that the AERMOD model used to simulate air dispersion and deposition on the ground "is not able to accurately predict PFOA deposition at any specific location." - 13 - Additionally, Chinkin identified another fabric coating facility south of Merrimack, local landfills, and the Manchester Airport as potential alternative sources of PFOA found in the class areas, though he did not conduct any specific analyses tied to these alternative sources. Overall, Chinkin opined that "[g]iven the number of potential PFAS sources and the complexities of PFAS fate and transport, a robust, property-by-property analysis is required to reliably determine the sources contributing to PFAS concentrations on any given property." The district court deemed these reports largely admissible, see Brown v. Saint-Gobain Performance Plastics Corp., No. 16-CV-242, 2023 WL 9016496, at *4–17 (D.N.H. Dec. 29, 2023), and relied on its review of the reports in its certification decision, see Brown, 2023 WL 9023158, at *5. B. In December 2023, the district court granted in part and denied in part plaintiffs' motion for class certification. Brown, 2023 WL 9023158, at *22. The court's ruling had three relevant parts. First, as to plaintiffs' nuisance claim, the court denied certification, concluding that claim "necessarily turn[ed] on individual assessments of the harm experienced by each putative class member, making class treatment of these issues - 14 - unmanageable." Id. at *2. No party here challenges that conclusion. Second, the court certified "the class as to liability issues under the trespass, negligence, and negligent failure to warn claims." Id. at *1. According to the court, liability on those claims was "predicated on predominantly common issues" including "the defendants' actions in emitting toxic chemicals and their subsequent efforts to investigate, mitigate, and warn the potentially affected population of the emissions; the hazardous nature of the chemicals emitted; and the geographical scope and foreseeability of groundwater contamination." Id. Third, as to damages on the three surviving claims, the court concluded that "damages issues cannot be determined on a class-wide basis." Id. at *2. As the court explained, damages would "vary based on factors including the extent of contamination attributable to Saint-Gobain in the groundwater sources for individual property owners, each individual's duration of residence at the relevant property, the effects and timing of mitigation efforts, and specific inconveniences and discomfort experienced by each class member." Id. These individualized issues "preclud[ed] certification as to damages." Id. at *20. Nevertheless, the court remained convinced that "several core liability issues can be proven on a class-wide basis, and these common issues predominate over the individualized inquiries - 15 - pertaining to damages." Id. at *22. Accordingly, the court decided to bifurcate trial proceedings: The first phase would "focus on liability determinations" and "be managed as a class action," while the second phase would address damages. Id. The district court did not specify how it would manage the second phase, stating only that "[f]urther discovery and findings from the liability phase will inform the format of the damages phase." Id. Saint-Gobain timely petitioned us for permission to appeal the district court's certification decision pursuant to Federal Rule of Civil Procedure 23(f). Saint-Gobain then moved in the district court for a stay of proceedings pending our resolution of the petition. The district court denied that motion. Ultimately, we granted Saint-Gobain's petition, and this appeal followed. III. We review the district court's class-certification decision for abuse of discretion. In re Asacol Antitrust Litig., 907 F.3d 42, 51 (1st Cir. 2018). "Within this ambit, we review pure issues of law de novo and 'fact-dominated' issues for clear error." Id. (quoting In re New Motor Vehicles Canadian Exp. Antitrust Litig., 522 F.3d 6, 17 (1st Cir. 2008)). To certify a class action, "the district court must undertake a 'rigorous analysis'" to determine whether plaintiffs - 16 - satisfy the requirements of Federal Rule of Civil Procedure 23. In re Nexium Antitrust Litig., 777 F.3d 9, 17 (1st Cir. 2015) (quoting Comcast Corp. v. Behrend, 569 U.S. 27, 33 (2013)). First, plaintiffs must meet the four threshold prerequisites of Rule 23(a): numerosity, commonality, typicality, and adequacy of representation. Id. Second, plaintiffs must show that the proposed class falls into one of the three class types provided by Rule 23(b). Comcast, 569 U.S. at 33 ("The party [seeking class certification] must . . . satisfy through evidentiary proof at least one of the provisions of Rule 23(b)."). Here, the district court certified the class pursuant to Rule 23(b)(3), which allows for class treatment where "questions of law or fact common to class members predominate over any questions affecting only individual members," and "a class action is superior to other available methods for fairly and efficiently adjudicating the controversy." These are referred to as Rule 23(b)(3)'s "predominance" and "superiority" prongs, respectively. IV. A. Saint-Gobain's initial critique of the district court's certification order portrays the order as a sua sponte misuse of Rule 23(c)(4). That Rule states: "When appropriate, an action may be brought or maintained as a class action with respect to particular issues." According to Saint-Gobain, "[t]he district - 17 - court erred by reading Rule 23(c)(4) to override Rule 23(b)(3)'s predominance and superiority requirements" -- creating a liability-only issue class where plaintiffs' claims as a whole could not be certified. Had the district court done this, it would raise a question subject to significant debate: Where a party seeks class certification pursuant to Rule 23(b)(3) but is unable to satisfy that provision's requirements with respect to an entire cause of action, may a court nevertheless certify for class treatment only the subset of issues relevant to that cause of action that do satisfy Rule 23(b)(3)? Contrast Castano v. Am. Tobacco Co., 84 F.3d 734, 745 n.21 (5th Cir. 1996) ("The proper interpretation of the interaction between subdivisions (b)(3) and (c)(4) is that a cause of action, as a whole, must satisfy the predominance requirement of (b)(3) and that (c)(4) is a housekeeping rule that allows courts to sever the common issues for a class trial."), with Jacks v. DirectSat USA, LLC, 118 F.4th 888, 898 (7th Cir. 2024) (holding that "a district court can certify an issue under Rule 23(c)(4) so long as the resolution of that issue is driven predominantly by common questions" and "certifying the proposed issues would be the most practical and efficient way to resolve the litigation"); see also Laura J. Hines, The Unruly Class Action, 82 Geo. Wash. L. Rev. 718, 725–29 (2014) (summarizing debate); - 18 - Myriam Gilles & Gary Friedman, The Issue Class Revolution, 101 B.U. L. Rev. 133, 137–41 (2021) (same). But the district court did no such thing. Rather, the court found Rule 23(a)'s four prerequisites and Rule 23(b)(3)'s predominance and superiority requirements satisfied as to the three certified claims -- negligence, negligent failure to warn, and trespass -- as a whole.6 Granted, in doing so, the court did divide the elements of those claims into liability issues -- which it found could be resolved class-wide -- and damages issues -- which it found could not. Brown, 2023 WL 9023158, at *16–19, *22. But the court then concluded that the "core liability issues" "predominate[d] over the individualized inquiries pertaining to damages." Id. at *22. The only time the court explicitly invoked Rule 23(c)(4) was in response to Saint-Gobain's stay motion, as the source of its "authority to bifurcate liability and damages when certifying a class without an affirmative request from a party to do so." In other words, the court found the three certified claims satisfied Rule 23(a) and (b), then proposed to use (c)(4) to treat on a common basis fewer than all the issues 6 Saint-Gobain does not argue that the district court erred in analyzing the class-action criteria claim by claim, rather than across the entire lawsuit, and we have previously used a claim- by-claim approach. See Smilow v. Sw. Bell Mobile Sys., Inc., 323 F.3d 32, 38–43 (1st Cir. 2003). - 19 - raised by those claims (i.e., those issues identified as common in the court's (b)(3) analysis).7 In summary, Saint-Gobain's Rule 23(c)(4) argument rests on a misreading of the district court's certification order. That being so, we have no occasion now to decide whether the district court could have used Rule 23(c)(4) to certify a part of a cause of action without finding that the action as a whole satisfied one of the Rule 23(b) alternatives. B. We turn our attention, then, to the more familiar issue raised in class litigation: whether the district court acted within its discretion in finding that the requirements of Rule 23(b)(3) were satisfied. We start -- and end -- with Rule 23(b)(3)'s predominance requirement. "The predominance inquiry asks whether the common, aggregation-enabling, issues in the case are more prevalent or important than the non-common, aggregation-defeating, individual issues." Tyson Foods, Inc. v. Bouaphakeo, 577 U.S. 442, 453 (2016) (citation modified). "[A] common question is one where the same evidence will suffice for each member to make a prima facie showing or the issue is susceptible to generalized, class-wide proof." 7 Rule 23(b)(3) itself plainly anticipates that some cases may present both predominating common issues and non-predominating individual issues, and (c)(4) makes clear that the court can maintain those predominating common issues as a class action. - 20 - Id. (citation modified). An individual question, in contrast, is one for which "members of a proposed class will need to present evidence that varies from member to member." Id. (citation omitted). Although predominance requires that common issues dominate the adjudication of the action, "the need for some individualized determinations at the liability and damages stage does not defeat class certification." Nexium, 777 F.3d at 21; see also Amgen Inc. v. Conn. Ret. Plans & Tr. Funds, 568 U.S. 455, 469 (2013) ("Rule 23(b)(3) . . . does not require a plaintiff seeking class certification to prove that each element of her claim is susceptible to classwide proof." (citation modified)). Ultimately, "[t]he aim of the predominance inquiry is to test whether any dissimilarity among the claims of class members can be dealt with in a manner that is not 'inefficient or unfair.'" Asacol, 907 F.3d at 51 (quoting Amgen, 568 U.S. at 469). As we have explained: Inefficiency can be pictured as a line of thousands of class members waiting their turn to offer testimony and evidence on individual issues. Unfairness is equally well pictured as an attempt to eliminate inefficiency by presuming to do away with the rights a party would customarily have to raise plausible individual challenges on those issues. Id. at 51–52. - 21 - Here, the district court found that, as to plaintiffs' negligence-based and trespass claims, common issues of liability predominated over individualized issues of damages. Brown, 2023 WL 9023158, at *22. Saint-Gobain insists the district court erred in this conclusion. In the end, we agree in part with Saint-Gobain: The district court erred in its predominance conclusion with respect to the negligence-based claims, but not the trespass claim. We consider the two sets of claims separately. 1. "Considering whether 'questions of law or fact common to class members predominate' begins, of course, with the elements of the underlying cause of action." Erica P. John Fund, Inc. v. Halliburton Co., 563 U.S. 804, 809 (2011) (quoting Fed. R. Civ. P. 23(b)(3)). Plaintiffs' negligence claim, arising under New Hampshire law,8 requires plaintiffs to "establish that the defendant owed a duty to the plaintiff, breached that duty, and that the breach proximately caused the claimed injury." Est. of Joshua T. v. State, 840 A.2d 768, 771 (N.H. 2003). Plaintiffs' negligent-failure-to-warn claim follows roughly the same contours, with the caveat that the duty to warn arises "only so far as there 8 No party disputes that New Hampshire law governs plaintiffs' state-law negligence, negligent-failure-to-warn, and trespass claims. See Forbes v. BB&S Acquisition Corp., 22 F.4th 22, 25 (1st Cir. 2021) (noting that, in diversity suits, "we apply state substantive law as that law has been applied by the state's highest court" (citation modified)). - 22 - is reason to apprehend danger to those not instructed." Trombly v. H.P. Hood & Sons, 146 A. 815, 815–16 (N.H. 1929) (citation omitted). Generally, then, we can describe the liability elements of plaintiffs' negligence–based claims as comprising duty, breach, injury, and causation. The district court found that both the existence of a duty and Saint-Gobain's breach of that duty turned on common questions susceptible to class-wide proof. Brown, 2023 WL 9023158, at *17. We see no error in those findings. "In general, anyone who does an affirmative act is under a duty to others to exercise the care of a reasonable person to protect them against an unreasonable risk of harm to them arising out of the act." Coan v. N.H. Dep't of Env't Servs., 8 A.3d 109, 115 (N.H. 2010) (citation modified). Thus, the existence of a duty depends on "the likelihood and magnitude of the risk" created by a defendant's conduct. Thibeault v. Campbell, 622 A.2d 212, 214 (N.H. 1993) (citation omitted). Accordingly, evidence of duty and breach will focus on Saint-Gobain's conduct, not the individual circumstances of each class member. In other words, "the same evidence will suffice for each member to make a prima facie showing" of duty and breach, rendering both common questions. Tyson Foods, 577 U.S. at 453 (citation omitted). More problematic is the need to prove the third element of plaintiffs' negligence–based claims: injury. Plaintiffs - 23 - propose to prove injury by presenting expert testimony to prove the geographic scope of Saint-Gobain's groundwater contamination and then limiting the class to owners of property within the scope of contamination. To this end, their experts produced a map delineating the geographic scope of the contamination. Their testimony, in brief, is that all properties within this area experienced PFOA contamination from Saint-Gobain's Facility. When called upon by the district court to explain how it intends to refute this class-wide proof of contamination, Saint-Gobain put forward the testimony of its experts challenging the accuracy of the models and assumptions plaintiffs' experts relied upon to prove the geographic scope of contamination. Saint-Gobain also pointed to evidence of other sources of PFOA contamination within the class area. None of its experts attempted a property–by–property refutation of plaintiffs' claims. The district court found that, as framed by the parties, the geographic scope of the contamination was a common issue. If the factfinder determines that plaintiffs' experts are correct, all of their properties suffered at least some contamination. If the factfinder rejects that evidence, all class members lose given that plaintiffs offer no individual property-specific proof. A favorable finding for plaintiffs on the scope of contamination would therefore suffice to prove at least some intrusion of Saint-Gobain's PFOA on all class members' properties. - 24 - But for the two negligence–based claims, that mere intrusion is not enough to establish liability. Rather, plaintiffs must show that the intrusion caused harm to that property. See Restatement (Second) of Torts § 165 (1965) (providing that negligently causing an object to enter another's land gives rise to liability "if, but only if, . . . the presence of the thing . . . upon the land causes harm to the land [or] to the possessor"). As the district court observ