United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT
Argued December 14, 2020 Decided May 7, 2021
No. 20-5035
PAVEMENT COATINGS TECHNOLOGY COUNCIL, APPELLANT
v.
UNITED STATES GEOLOGICAL SURVEY, APPELLEE
Appeal from the United States District Court for the District of Columbia (No. 1:14-cv-01200)
Lawrence S. Ebner argued the cause for appellant. With him on the briefs was David A. Kanter.
Joshua M. Koppel, Attorney, U.S. Department of Justice, argued the cause for appellee. With him on the brief were Ethan P. Davis, Acting Assistant Attorney General, at the time the brief was filed, and Mark B. Stern, Attorney. Paul Cirino, Trial Attorney, and R. Craig Lawrence, Assistant U.S. Attorney, entered appearances.
Before: ROGERS, MILLETT and WILKINS, Circuit Judges.
Opinion for the Court filed by Circuit Judge WILKINS. 2
WILKINS, Circuit Judge: The U.S. Geological Survey (“USGS” or “Survey”) is the nation’s largest water, earth, and biological science agency. Housed within the Department of the Interior, it has no regulatory mandate and is instead relied upon to provide unbiased and policy-neutral information to decision-makers at the local, state, and federal level. The Survey and its scientists regularly publish this research in agency reports and scientific journals. Two studies, and the records that underlie them, are at issue in this Freedom of Information Act (“FOIA”) appeal.
With a note on the special role of summary judgment in FOIA cases, we reverse and remand to the District Court with respect to the model runs withheld under Exemption Five for further proceedings consistent with this opinion. We affirm the District Court’s ruling as to Exemption Six.
I.
A.
Requestor Pavement Coatings Technology Council (“PCTC”) is a trade association for producers of refined coal tar-based sealant. Coal tar sealant prolongs asphalt’s service life by protecting it from degradation caused by sunlight, water, and leaked oil or gasoline. It has also been identified as a major source of polycyclic aromatic hydrocarbons (“PAHs”). A. 18. There are different types of PAHs, sixteen of which are considered priority pollutants by the Environmental Protection Agency (“EPA”). See 40 C.F.R. Part 423, App. A; see also A. 18–19. PAHs are toxic, mutagenic, teratogenic, and some are probable human carcinogens. A. 19. In the late 1990s, scientists at Respondent USGS noted an upward trend in PAH levels in sediment recovered from urban lakes. 3 USGS Research Hydrologists Barbara J. Mahler and Peter Van Metre undertook several studies to identify the sources of PAHs in urban environments. Two of those studies— “Contribution of PAHs from coal tar pavement sealcoat and other sources to 40 U.S. lakes” (“urban lakes study”) and “Coal-tar-based pavement sealcoat: An unrecognized source of PAH to settled house dust” (“house dust study”)—and their records are at issue here. Both studies identified coal tar sealant as a major source of urban PAHs. USGS has released all of the raw data underlying both studies.
Mahler and Van Metre produced computer modeling input and output data to analyze this raw data. As described by Van Metre, “[m]odeling is a broad term that generally means to develop a mathematical model of some natural process” by relating one or more variables (e.g., urban land use) to the occurrence of other variables (e.g., the concentration of a pollutant in local streams) in order to better understand how the environment works. A. 23. Complicated models require calibration, achieved by adjusting or replacing variables and parameters that control the model in order to test how well the model represents the natural process studied. In the urban lakes study, Mahler and Van Metre sought to determine what PAH sources—vehicle emissions, power plant emissions, or coal tar sealant, among other sources—contributed to the PAHs they measured in lake sediment samples collected across the United States.
To do so, Mahler and Van Metre tested three widely used source receptor models developed by the EPA. A source receptor model attempts to isolate the source of contaminants identified in a sample. Testing the three models with different combinations of sediment samples and PAHs revealed that the EPA’s “contaminant mass balance” (“CMB”) model was the preferred choice. The CMB model, like the raw data it 4 crunched, is publicly available. See United States Environmental Protection Agency, Chemical Mass Balance (CMB) Model, https://www.epa.gov/scram/chemical-mass- balance-cmb-model (last visited Apr. 23, 2021). Mahler and Van Metre did not publish the “exploratory decision-making process” behind their choice to use the CMB model because such a comparative modeling analysis “would be a major undertaking” distinct from their goal of “using a given model to evaluate sources in the environment.” A. 24.
Calibrating the CMB model required Mahler and Van Metre to make several choices: they selected which lakes to include and, from each lake, which samples. They decided how to process that data before entering it into the model (for example, by using the logarithm of the actual sample values or tweaking how sample uncertainty was estimated). They chose which types of PAHs to source. And they chose which sources to include, since there are statistical limits on how many sources the model can consider in a run. For example, it made sense to the researchers to include coke-oven emissions for lakes near Chicago, where coke ovens are still in operation, but not for lakes in Washington State or Florida, where coke-ovens are not.
The USGS scientists also ran combinations of variables that the researchers believed were highly unlikely to occur in the real world to see how the model would respond to adjustments to those variables. Other runs allowed them to “investigate alternative hypotheses for the causes of . . . PAH contamination.” A. 24. The scientists also shifted parameters for various runs, including instructing the model to run a certain number of iterations before arriving at a solution. Given these possible combinations of inputs, “the possible outcomes of a particular model—even if used on the same data set—are virtually limitless.” A. 26. Van Metre explained that 5 the process of selecting “samples, source profiles, PAHs and modeling parameters reflects the working thoughts of a scientist as he or she attempts to make sense of the data presented.” A. 26. Mahler described the process as “not dissimilar to a writer trying out different combinations of words or paragraphs in a draft document in an effort to create the most logical sequence.” A. 14.
From the more than 200 model runs tested, Van Metre and Mahler chose four representative runs “on the basis of good quantitative and qualitative model performance.” A. 27. These four runs were “in general agreement with the vast majority of the 200 models tested,” though there was “considerable variability in those results that could be exploited to make it appear” that USGS overstated coal tar sealant’s role as a source. A. 27. The record does not disclose whether peer reviewers—either within USGS or engaged by the publishing journals—reviewed the 196 model runs that were not selected for publication. At oral argument, counsel for USGS suggested that approving officials within the agency “may have seen some [of the 196 model runs] in the preliminary drafts,” but conceded that neither scientist declared that they selected the four chosen model runs to inform the ultimate decision- maker’s decision whether to publish the urban lakes study. Oral Arg. Tr. 16:22–17:9; 27:19–28:12; see also 21:13–17.
Requestor PCTC engages experts who use the Survey’s data to conduct in-depth post-publication peer reviews “to better understand data that has been collected about products made by PCTC members.” To test the soundness of the urban lakes study’s methods and conclusions, PCTC seeks the computer modeling input and output data Mahler and Van Metre produced while analyzing their raw data. PCTC takes issue with USGS’s decision not to publish the exploratory decision-making process, namely, the inputs and parameters 6 for each run. PCTC believes that Mahler and Van Metre manipulated the model inputs to ensure that outputs identified coal tar sealants as the source. Mahler and Van Metre allegedly achieved this result by using unweathered PAH source profiles for the non-coal tar sealant sources and weathered samples for the coal tar sealant source. According to PCTC, since “all combustion sources of PAHs have similar PAH profiles and all weather to similar, undistinguishable weathered profiles, Mahler and Van Metre have rigged the model to guarantee that most PAHs in the environment (which are all weathered) look like what they have called [coal tar sealant].” A. 120. PCTC points to this alleged manipulation to illustrate why it “must be provided with the model runs it has requested to both attempt to replicate the reasoning behind the work of Mahler and Van Metre, and to defend itself against attempts to ban or restrict the sale and use of [coal tar sealant].” A. 120. Van Metre’s affidavit stated that providing PCTC with the model runs would “give [it] yet another tool to try and confuse the public and discredit our work.” A. 27.
PCTC’s Executive Director, Anne LeHuray, testified in her affidavit that concealing scientific processes is unsound scientific methodology. She claimed that a “full account of the work performed” is necessary to attempt to reproduce Mahler and Van Metre’s research and pointed to the scientific journal Nature’s conditions of publication, which require authors to “make materials, data, code, and associated protocols promptly available to readers without undue qualifications,” to support the argument that these model runs would usually be revealed to the scientific community and the public. A. 129, 168.
PCTC also seeks location information for dust samples collected as part of the urban house dust study. USGS scientists collected samples from twenty-three Austin, Texas residences in mid-2008. USGS labeled each sample with a 7 sample ID and created sample sheets linking the sample IDs and concentrations of PAHs found in each associated sample. Participants filled out a questionnaire and answered follow-up questions to identify factors that may have affected PAH concentrations in their house dust. The questions asked participants to disclose how many adults and children lived in the residence, whether they smoked, how often they left their homes and how long they were gone, their eating and cooking habits, and whether and how long their windows were kept open. Volunteers were told their samples would be used only for the purposes of the study and that personally identifiable information would remain confidential.
USGS produced all sample sheets and questionnaires (with the volunteer’s name and address, and sample and site IDs, redacted) and a “means by which to match the responses to the results of the sample analysis.” A. 55, 148. However, PCTC seeks the location information associated with each sample in order to “conduct a study (or a post-publication peer review of the USGS’ [sic] dust study) which examines the potential role alternative (non-[coal tar sealant]) factors—known as confounders—may have contributed to the PAH levels found in the USGS dust study.” A. 171. PCTC argues that it must be provided with volunteers’ addresses because “[w]ithout the addresses of locations used in the USGS dust study, PCTC will not be able to construct a study design that adequately replicates the USGS dust study.” A. 172.
B.
In 2011, PCTC filed a FOIA request seeking release of USGS records relating to the agency’s coal tar sealant studies. PCTC sought eight years of documents, including correspondence regarding coal tar sealant and all documents relating to studies or publications about coal tar sealant, as well 8 as all lab data for coal tar sealant-related research. USGS produced 52,000 pages of records, including all raw data collected for both the urban lakes and house dust studies, but withheld the modeling data and personally identifiable information relevant to this appeal. USGS withheld the model runs under FOIA’s Exemption Five on the ground that “[r]elease of the exploratory analysis would inhibit the ability to freely explore and analyze data without concern for external criticism.” A. 50. It withheld the house dust study participants’ personal information under FOIA’s Exemption Six because “[r]elease would constitute a clearly unwarranted invasion of personal privacy” and “would not serve a public interest because the pertinent scientific data associated in this category of records is already released.” A. 55–56.
After PCTC filed its complaint in 2014, the parties submitted cross-motions for summary judgment. The District Court ruled on those motions on November 13, 2019, issuing an order granting USGS’s motion for summary judgment and denying PCTC’s cross-motion. PCTC timely appealed. The District Court exercised jurisdiction under 5 U.S.C. § 552(a)(4)(B) and 28 U.S.C. § 1331. We have jurisdiction under 28 U.S.C. § 1291.
II.
As we have often observed, “FOIA protects the basic right of the public to be informed about what their government is up to.” Hall & Assocs. v. EPA, 956 F.3d 621, 624 (D.C. Cir. 2020) (quoting Competitive Enter. Inst. v. Office of Sci. & Tech. Policy, 827 F.3d 145, 150 (D.C. Cir. 2016) (internal quotations omitted)). FOIA, 5 U.S.C. § 552, requires agencies to disclose records upon request, unless they fall within one of nine exemptions. U.S. Fish & Wildlife Serv. v. Sierra Club, Inc., — U.S. —, 141 S. Ct. 777, 785 (2021). We construe these 9 exemptions narrowly. Milner v. Dep’t of Navy, 562 U.S. 562, 565 (2011).
We review the District Court’s decision on summary judgment in a FOIA case de novo. Hall & Assocs., 956 F.3d at 629. Summary judgment is appropriate only “if the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Id. (quoting FED. R. CIV. P. 56(a)); see also Petroleum Info. Corp. v. U.S. Dep’t of Interior, 976 F.2d 1429, 1433 (D.C. Cir. 1992) (“This circuit applies in FOIA cases the same standard of appellate review applicable generally to summary judgments.” (citing Wash. Post Co. v. Dep’t of Health and Hum. Servs., 865 F.2d 320, 325–26 & n.8 (D.C. Cir. 1989))). “In the FOIA context this requires that we ascertain whether the agency has sustained its burden of demonstrating that the documents requested are . . . exempt from disclosure.” Am. C.L. Union v. U.S. Dep’t of Justice, 655 F.3d 1, 5 (D.C. Cir. 2011) (quoting Gallant v. NLRB, 26 F.3d 168, 171 (D.C. Cir. 1994) and citing 5 U.S.C. § 552(a)(4)(B) (stating that “the burden is on the agency to sustain its action”)). At this stage, “the inferences to be drawn from the underlying facts . . . must be viewed in the light most favorable to the party opposing the motion,” in this case, requestor PCTC. Judicial Watch, Inc. v. U.S. Secret Serv., 726 F.3d 208, 215 (D.C. Cir. 2013) (quoting Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587 (1986) (internal quotations omitted)).
USGS seeks to withhold some 196 unpublished model runs under Exemption Five. Exemption Five excludes from FOIA’s disclosure requirements “inter-agency or intra-agency memorandums or letters that would not be available . . . in litigation with the agency.” 5 U.S.C. § 552(b)(5). “A form of executive privilege, the deliberative process privilege covers deliberative, pre-decisional communications within the 10 Executive Branch,” Nat’l Sec. Archive v. CIA, 752 F.3d 460, 462 (D.C. Cir. 2014), and was “intended to protect not simply deliberative material, but also the deliberative process of agencies,” Montrose Chem. Corp. of Cal. v. Train, 491 F.2d 63, 71 (D.C. Cir. 1974). To qualify for withholding, information must be both pre-decisional and deliberative. Petroleum Info. Corp., 976 F.2d at 1434.
With respect to the urban lakes study, USGS stumbles at both hurdles. The agency first failed to introduce any evidence establishing what role the requested model runs played in its decision to publish the urban lakes study. Second, we find no evidence on this record that disclosing the model runs would expose the Survey’s decision-making process “in such a way as to discourage candid discussion within the agency and thereby undermine the agency’s ability to perform its functions.” See Dudman Commc’ns Corp. v. Dep’t of Air Force, 815 F.2d 1565, 1568 (D.C. Cir. 1987).
USGS failed to carry its burden to show that the model runs are pre-decisional. In order to establish that government documents are pre-decisional, “the agency has the burden of establishing what deliberative process is involved, and the role played by the documents in issue in the course of that process.” Coastal States Gas Corp. v. Dep’t of Energy, 617 F.2d 854, 868 (D.C. Cir. 1980); see also Paisley v. CIA, 712 F.2d 686, 698 (D.C. Cir. 1983), vacated in part on other grounds, 724 F.2d 201 (D.C. Cir. 1984). Put simply, a pre-decisional record is one “prepared in order to assist an agency decisionmaker in arriving at his decision.” Petroleum Info. Corp., 976 F.2d at 1434 (quoting Renegotiation Bd. v. Grumman Aircraft, 421 U.S. 168, 184 (1975)). The government says that the relevant agency decision to which the model runs are assertedly pre- 11 decisional is the Survey’s decision to adopt and publish the urban lakes study in its own name. USGS Br. 15–16. PCTC accepts that as the relevant decision for purposes of Exemption Five. See PCTC Br. 22. But we are faced with a record devoid of evidence that any decision-maker at USGS considered the discarded model runs in determining whether and in what form to publish the urban lakes study. The agency bears the burden of explaining its decision-making process, but we have no declaration that does so. Indeed, counsel for the agency conceded as much at oral argument. See Oral Arg. Tr. 21:12– 22:1 (“I don’t believe that what the approving officials look at is in the record.”). Counsel is correct.
All we are told is that “approving officials may have seen some” of the discarded runs in preliminary drafts, id. at 17:1–2 (emphasis added), but this claim was made only at oral argument and is unsupported by the record evidence. USGS chose to identify its decision as the decision to publish, but it did not explain how that decision was made, aside from averring that the process included peer review. A. 52–53. Yet USGS failed to explain in detail whether the model runs were shared with peer reviewers and what role, if any, they played in the peer review process. USGS instead conflates the deliberative process of coming to a reliable scientific result with the approving officials’ decision to publish the urban lakes study. See Oral Arg. Tr. 20:20–25. The Survey is widely respected because it publishes reliable scientific research, but, for FOIA purposes, the decision to publish a paper and the underlying scientific determination are not one and the same. Without more, we cannot find that USGS has carried its burden to explain the model runs’ role in its decision-making process.
The “more” we are looking for requires USGS to establish how its decision to publish the urban lakes study was reached; what information was shared with reviewers, internal and 12 external; whether drafts reviewed by agency officials making the publication decision included the underlying model run data; and how the exploratory data runs influenced the decision to publish or the form the final publication would take. Without this information, USGS has “failed to supply us with even the minimal information necessary to make a determination.” See Coastal States, 617 F.2d at 861. We remind USGS that “the burden is on them to establish their right to withhold information from the public and they must supply the courts with sufficient information to allow us to make a reasoned determination that they were correct.” Id.
USGS also failed to prove beyond dispute that the model runs are deliberative. Our deliberativeness inquiry “focuse[s] on whether disclosure of the requested material would tend to discourage candid discussion within an agency.” Petroleum Info. Corp., 976 F.2d at 1434 (internal quotations omitted). We ask whether the document is “so candid or personal in nature that public disclosure is likely in the future to stifle honest and frank communication,” Coastal States, 617 F.2d at 866, “and thereby undermine the agency’s ability to perform its functions,” Dudman, 815 F.2d at 1568. USGS failed to establish how or why disclosure of the model runs would chill scientists’ use of exploratory model runs in the future or impact the accuracy or efficiency of the Survey’s operations.
The agency’s affidavits contain no explicit statement that disclosure will harm the agency’s decision-making. At oral argument agency counsel could not point to a similar claim, nor did we find one in the record. Cf. Oral Arg. Tr. 24:9–10 (USGS’s counsel conceded he was “not sure if” the agency affidavits “explicitly say that [release] would change the way they do their business”). We find only claims that releasing the 13 model runs will enable criticism of USGS. A. 27, 49. But criticism is not a recognized harm against which the deliberative process privilege is intended to protect. Granted, USGS argues that release could cause public confusion, and we have acknowledged misperception of agency positions as a ground for withholding deliberative materials. See Judicial Watch, Inc. v. Dep’t of Def., 847 F.3d 735, 739 (D.C. Cir. 2017). But the privilege’s “‘ultimate aim’ is to ‘prevent injury to the quality of agency decisions.’” Petroleum Info. Corp., 976 F.2d at 1434 (quoting NLRB v. Sears, Roebuck & Co., 421 U.S. 132, 151 (1975)). And it is also clear from our—and Supreme Court—precedent that we are obligated to construe the exemption narrowly and focus on whether disclosure will harm intra-agency candor and efficiency. Id. at 1434–35; see also Milner, 562 U.S at 565 (“[FOIA’s] exemptions are explicitly made exclusive, and must be narrowly construed.” (internal citations and quotations omitted)); U.S. Fish & Wildlife Serv., 141 S. Ct. at 785 (“To encourage candor, which improves agency decisionmaking, the privilege blunts the chilling effect that accompanies the prospect of disclosure.”). USGS does not say it will and does not explain how, if these model runs are disclosed, scientists will cease to conduct model runs in the future or do them differently.
To the contrary, PCTC introduced evidence that disclosure of the model runs will not impede or impair the scientists’ work going forward. The Director of PCTC introduced, in her affidavit, an unrebutted excerpt from Nature Magazine’s conditions of publication, which require authors to “make materials, data, code, and associated protocols promptly available to readers without undue qualifications.” A. 168. Although it is not clear that other scientific journals, or even Nature, would expect the authors to release all 200 model runs, the Survey cannot meet its burden of justifying the categorical withholding of all unpublished model runs given this 14 unrebutted evidence. And at oral argument the Survey’s attorney admitted that performing preliminary model runs “is quite standard.” Oral Arg. Tr. 22:23–24. Consequently, we hesitate to presume that USGS scientists will be discouraged from performing these standard calibrations in the future in a way that would harm the agency’s decision-making. For the same reason, it is not obvious that disclosure would result in the same harm to agency decision-making as in Dudman, 815 F.2d at 1569 (finding that publication of draft agency history would discourage the candid exchange of ideas), and Russell v. Dep’t of the Air Force, 682 F.2d 1045, 1048 (D.C. Cir. 1982) (same).
Nor do we find these model runs exactly analogous to peer review comments. See Formaldehyde Inst. v. Dep’t of Health & Human Servs., 889 F.2d 1118, 1120, 1123–25 (D.C. Cir. 1989). In Formaldehyde we found reviewers’ comments that contained “advice, constructive criticism and guidance with respect to revision” of a scientific study submitted for publication protected by the privilege where the “undisputed factual record clearly establishe[d] that [agency] personnel . . . must regularly rely on the comments of expert scientists to help them evaluate the readiness of agency work for publication.” Id. at 1124–25. That record also contained “unrefuted evidence” that release of reviewers’ comments “would very likely have a chilling effect on either the candor of potential reviewers of government-submitted articles or on the ability of the government to have its work considered for review at all.” Id. at 1125. Those findings supported the inference that “a government author is likely to be less willing to submit her work to a refereed journal at all if critical reviews could come to light somewhere down the line” resulting “in the publication of inferior work in (presumably) inferior and less widely circulated journals.” Id. Because there is no evidence here that scientists will cease to run thorough and searching 15 exploratory analyses, we cannot follow the same chain of inferences we did in Formaldehyde, where we could easily see the detrimental effects of disclosure.
We find sufficient uncertainty about whether this type of data is ordinarily disclosed, or whether there is an expectation that it will not be disclosed, or what impact it would have if is disclosed, to rule for USGS at the summary judgment stage.
C.
The absence of evidence establishing that the requested model runs are protected from disclosure amounts to the agency’s failure to establish that it is entitled to judgment as a matter of law. See FED. R. CIV. P. 56(a). We leave to the District Court the decision how to proceed. Generally speaking, discovery is rarely appropriate in FOIA cases, In re Clinton, 970 F.3d 357, 364 (D.C. Cir.), on reh’g, 973 F.3d 106 (D.C. Cir. 2020), cert. denied sub nom. Jud. Watch, Inc. v. Clinton, No. 20-1051, 2021 WL 1163766 (U.S. Mar. 29, 2021), and the preferred approach, if possible, is to resolve the lawsuit without discovery and by summary judgment. If the District Court believes that it can resolve this case for one side or the other with supplemental affidavits and further summary judgment briefing, it should do so. This strikes us as the prudent course where it allows the District Court to resolve a contested FOIA request most efficiently. FED. R. CIV. P. 1 (“[These rules] should be construed, administered, and employed by the court and the parties to secure the just, speedy, and inexpensive determination of every action and proceeding.”).
Of course, summary judgment is not barred merely because “the parties’ affidavits disagree on the probable consequences of a disclosure,” Alyeska Pipeline Serv. Co. v. EPA, 856 F.2d 309, 313 (D.C. Cir. 1988), but if the undisputed 16 material facts “are susceptible to divergent inferences bearing upon an issue critical to the disposition of the case, summary judgment is not available,” id. at 314. In such an instance, the matter can be resolved by an adjudicatory proceeding tailored to the factual dispute in the case, such as a “paper trial” on a stipulated evidentiary record with findings of fact and conclusions of law set forth pursuant to Federal Rule of Civil Procedure 52. See Wash. Post Co., 865 F.2d at 326; Sears, Roebuck & Co. v. Gen. Servs. Admin., 553 F.2d 1378, 1382–83 (D.C. Cir. 1977).
III.
We affirm the District Court’s grant of summary judgment to USGS with respect to the sampling location information collected for the 2010 house dust study. Exemption Six allows withholding of “personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy.” 5 U.S.C. § 552(b)(6). It aims “to protect individuals from the injury and embarrassment that can result from the unnecessary disclosure of personal information.” Dep’t of State v. Wash. Post Co., 456 U.S. 595, 599 (1982).
The study participants have a greater than de minimis privacy interest in their addresses, household compositions, smoking and cooking habits, and the extensive personal details included in the questionnaires. See U.S. Dep’t of Def. v. FLRA, 510 U.S. 487, 500–01 (1994). And releasing their addresses serves no cognizable public interest because it would shed no additional light on the Survey’s “operations or activities,” since USGS has already produced the questionnaires and a “means by which to match [participants’] responses to the results of the sample analysis.” See Nat’l Ass’n of Home Builders v. Norton, 309 F.3d 26, 33–34 (D.C. Cir. 2002) (internal quotations 17 omitted); A. 55, 148. PCTC thus has all the data it needs to replicate the USGS scientists’ analysis of how participants’ habits may have impacted the concentration of coal tar sealant in their homes. And in any case, PCTC’s desire to identically replicate the study using the addresses is foreclosed given that coal tar sealant has been outlawed in the relevant area since 2006.
Since “something, even a modest privacy interest, outweighs nothing every time,” the District Court correctly found the study participants’ personal information properly withheld under Exemption Six. Nat’l Ass’n of Retired Fed. Emps. v. Horner, 879 F.2d 873, 879 (D.C. Cir. 1989).
IV.
For the foregoing reasons, we reverse and remand to the District Court PCTC’s claims regarding the urban lakes model runs withheld under Exemption Five, and affirm its decision to withhold the house dust study location information under Exemption Six.
So ordered.