Loughlin, Thomas P. v. United States

393 F.3d 155, 364 U.S. App. D.C. 132, 34 Envtl. L. Rep. (Envtl. Law Inst.) 20158, 2004 U.S. App. LEXIS 26524

• Court: Court of Appeals for the D.C. Circuit
• Decided: December 21, 2004
• Docket: 03-5284, 03-5286
• Status: Published

Opinion

Opinion for the Court filed by Circuit Judge EDWARDS.

HARRY T. EDWARDS, Circuit Judge.

The Federal Tort Claims Act (“FTCA”) gives district courts jurisdiction over civil actions on claims against the United States (the “Government”) for money damages for injury or loss of property, or personal injury or death, caused by the negligent or wrongful act or omission of any employee of the Government while acting within the scope of his or her employment, under circumstances where the Government, if a private person, would be liable to the claimant under the law of the place where the act or omission occurred. See 28 U.S.C. §§ 1346(b), 2671-2680 (2000). This waiver of sovereign immunity does not extend to claims against the United States “based upon an act or omission of an employee of the Government, exercising due care, in the execution of a statute or regulation ... or based upon the exercise or performance or the failure to exercise or perform a discretionary function.” 28 U.S.C. § 2680(a).

This case involves an FTCA action brought by appellants — American University (“AU”); Glenbrook Limited Partnership, Lawrence N. Brandt, Inc., Lawrence N. Brandt, and Robert Brant (collectively “Glenbrook-Brandt”); Thomas P. Lough-lin and Kathi Loughlin, individually and on behalf of their children; Patricia Gillum; and Camille Saum — for the Government’s alleged negligence in (1) burying dangerous, munitions and toxic chemicals on property leased from AU in the Spring Valley area of the District of Columbia around the time of World War I, (2) failing to issue warnings about the buried munitions and chemicals and the resulting dangerous conditions, and (8) failing to investigate and remedy the hazards and contamination it caused. Gillum and Saum initially filed their FTCA and local law claims in D.C. Superior Court. Their local law actions claimed that AU was liable to the plaintiffs under District of Columbia law, because the dangerous munitions and toxic chemicals on AU’s property, and the hazardous conditions resulting therefrom, caused injuries to neighboring property owners. AU removed these actions to the District Court under 28 U.S.C. § 1441(b)-(c) (2000). The Loughlins filed both their FTCA action and supplemental local law claims similar to those filed by Gillum and Saum in District Court. The District Court invoked its supplemental jurisdiction over all local law claims against AU under 28 U.S.C. § 1367(a) (2000).

AU filed a motion under Rule 12(b)(6) of the Federal Rules of Civil Procedure to dismiss the local law actions for failure to state a claim. The District Court denied this motion and wrote a lengthy opinion suggesting that the Loughlins, Gillum, and Saum had stated a cause of action against AU under District of Columbia law. See Loughlin v. United States, 209 F.Supp.2d 165 (D.D.C.2002) (“Loughlin I”). The District Court turned to the FTCA matter after rendering a judgment on the supplemental action. The trial court first allowed the parties jurisdictional discovery limited to the existence of rules, regulations, or. directives that might pertain to the first part of the discretionary function exception. The District Court then granted the Government’s motion to dismiss with prejudice the FTCA actions under the discretionary function exception. See Loughlin v. United States, 286 F.Supp.2d 1, 3 (D.D.C.2003) (“Loughlin IF). Having found that it lacked subject matter jurisdiction under the FTCA, the District Court dismissed all remaining claims without prejudice. Id. at 30. The FTCA claimants appeal the dismissal of their claims, as well as the limited scope of the trial court’s discovery orders. AU separately appeals the District Court’s denial of its motion to dismiss the supplemental action against the University. In the alternative, AU asserts that, if the actions resting on District of Columbia law are moot, then the District Court’s decision on the non-federal claims should be vacated.

We affirm the District Court’s dismissal of the FTCA claims under the discretionary function exception. Although the trial court’s framework for discovery was misguided, we nonetheless find that the parties had a full and fair opportunity to determine the relevant jurisdictional facts and the District Court had an adequate record upon which to rest its judgment. Finally, we vacate the District Court’s decision denying AU’s motion to dismiss. The District Court had*no subject matter jurisdiction over the FTCA action. Therefore, the trial court had no supplemental jurisdiction under § 1367(a) to entertain non-federal claims. Accordingly, because it should not have reached the merits of the negligence claims under District of Columbia law, the District Court’s views on local law are a nullity and must be vacated.

I. Background

In April 1917, at the invitation of AU, the United States Army leased grounds from the University and gave its Corps of Engineers (“Corps”) exclusive control over the property. Later that year, the Bureau of Mines established the American University Experiment Station (“AUES”) in or7 der to consolidate its chemical weapons research. When AUES was transferred from civilian control to the War Department’s newly formed Gas Service, it became central to the Gas Service’s Research Division, which used the experiment station to develop, manufacture, and test myriad chemical weapons. In order -to simulate battlefield conditions, gas weapons were tested in trenches, bunkers, and pits created on the property. Loughlin II, 286 F.Supp.2d at 3-4.

Shortly after the war’s end, AUES was disbanded and the Army'transferred personnel and equipment to other bases. It is undisputed, however, that some munitions and chemical warfare materials remained buried in Spring. Valley, either as,a result of weapons testing or deliberate burial. Id. at 4. In March 1920, the Army signed an agreement pledging to restore the buildings and grounds to the condition they were in when the Government took: control of the property. This agreement appears to have been superceded, however, by a subsequent agreement, dated June 21, 1920, in which the University agreed to release the Government from its obligation to restore the property in exchange for the transfer of title to certain buildings erected by the Army. Id. The Army nevertheless performed some salvage and restoration work before leaving AUES; some contaminated .structures were burned and others were boarded up.and surrounded with fencing. Id. at 4 n. 4.

In 1986, when AU embarked on plans to build a new athletic facility, the University discovered a 1921 article in The American University Courier, which reported that the Army had buried munitions on or near the University campus during World War I. The University notified the Army, which conducted document reviews and scoured the site with metal detectors, but did not uncover conclusive evidence of any buried munitions. Id. at 4. In 1990, American University sold property to Glenbrook-Brandt, which planned to construct two houses, at what is today 4825 and 4835 Glenbrook Road. In the course of these construction projects, workers uncovered old laboratory equipment and possible chemical contaminants. They also experienced severe physical reactions to the site that required emergency hospital care. Id. at 5. Glenbrook-Brandt informed the University, which retained an industrial hygiene consulting firm to investigate. The firm identified a herbicide, Silvex, in the soil, which it explained could irritate the senses, but was not a hazardous substance. Id.

Around this time, workers excavating land approximately one mile from the Glenbrook-Brandt property discovered an underground munitions bunker. That project’s developer contacted the Army, which commenced investigations that lasted until 1995 and unearthed live and spent munitions and chemical warfare-related materials from the World War I era. These events gave rise to separate litigation against the United States by the owner/developer of that property. Id. at 5 & n. 6 (citing W.C. & A.N. Miller Cos. v. United States, 963 F.Supp. 1231 (D.D.C.1997)).

In January 1994, the Army, now immersed in a comprehensive investigation to locate buried weapons, sought and received permission from Glenbrook-Brandt to access its properties and sample the soil. Id. at 5. In February 1994, Thomas and Kathi Loughlin tendered a purchase offer to buy the property at 4825 Glen-brook Road. Glenbrook-Brandt disclosed to the Loughlins the recent developments, and the Loughlins hired an independent testing organization to sample the soil and evaluate potential environmental hazards. This independent firm found no contamination from hazardous substances. Id. at 5-6. On March 21, 1994, the Loughlins contracted to purchase 4825 Glenbrook Road.

Meanwhile, on March 9, 1994, the Corps collected soil samples from the Glenbrook-Brandt properties. Soil and groundwater samples were also collected from other “points of interest” throughout the area. In June 1995, the Defense Department issued its final Record of Decision, which concluded that no further action was necessary with respect to the removal operation in Spring Valley. At this time, the Army had .removed from the area 141 pieces of ordnance, 43 of which were suspected of being chemical weapons. Id. at 5.

The Environmental Protection Agency (“EPA”) also conducted soil sampling at 4825 Glenbrook Road during this time. EPA collected seven samples on March 11, 1994, one of which revealed heightened levels of arsenic. Id. at 6. Both the Loughlins and Glenbrook-Brandt allege that they lacked knowledge of this abnormal result until it was disclosed to them by the Corps in early February 1999. Id. Indeed, in January 1995, the Corps issued a letter to the Loughlins and other Spring Valley residents, which stated that the soil samples had not revealed chemical agents or explosives and that no hazard to human health or to the environment existed as a result of the Army’s activities at AUES. Id.

In June 1996, however, workers at the site of the AU President’s residence at 4835 Glenbrook Road, next door to the Loughlins’ home, suffered reactions to odors and fumes that burned their eyes. These workers unearthed laboratory glassware and broken bottles filled with chemicals. The University hired Apex Environmental, Inc., which conducted soil samples that confirmed the existence of a contaminated area approximately 12 feet in diameter and up to two feet deep; the contamination included arsenic. Id. These events led the District of Columbia to conduct its own investigation, which also found elevated levels of arsenic and other toxic substances. Id.

In February 1998, the Corps conducted a geophysical survey of the Korean Ambassador’s residence at 4801 Glenbrook Road, which also abuts 4825 Glenbrook Road, and found two potential burial pits. The Corps informed Spring Valley residents that it would investigate whether additional chemical warfare-related materials existed at 4801 Glenbrook Road. A 75 millimeter projectile was discovered buried six inches deep in the Ambassador’s property in February 1999. Id. at 6-7.

In December 1998, the Corps also began further investigation of the Loughlins’ property “ ‘to confirm the absence of buried munitions, remnants thereof, and associated material.’ ” Id. at 7 (quoting United States’ Stat. of Material Facts Not in Dispute ¶ 70). On June 9, 1999, the Corps collected 22 soil samples. All but four of the samples contained elevated levels of arsenic. Id. Based on the June 1999 samples, the Corps concluded that there was an “ ‘unacceptable hazard’ ” from arsenic on the properties at 4801 and 4825 Glen-brook Road. Id. The Corps informed the Loughlins, who were forced to permanently evacuate their home. Since 1996, Army investigations in Spring Valley unearthed 667 pieces of ordnance, including chemical munitions and bottles of chemicals. These events spawned several lawsuits. Id.

This case presents FTCA claims brought by appellants against the Government. The Loughlins, Gillum, and Saum also brought negligence claims against AU, id.; Loughlin I, 209 F.Supp.2d at 167, which brought cross-claims against the Government under the FTCA, see Br. of AU, Loughlins, Glenbrook-Brandt, Saum, and Gillum (“FTCA Appellants’ Br.”) ii-iii. The Gillum and Saum actions were initially filed in Superior Court and then removed to federal court under 28 U.S.C. § 1441(b)-(c). The Loughlins filed both their FTCA action and their supplemental local law claims in District Court. The District Court exercised its supplemental jurisdiction over the local law claims under 28 U.S.C. § 1367. On March 13, 2002, AU moved to dismiss the local law claims pursuant to Rule 12(b)(6). On June 3, 2002, the District Court denied AU’s motion to dismiss, finding that AU owed a legal duty under the applicable District of Columbia law. See Loughlin I, 209 F.Supp.2d at 167.

On September 20, 2002, the Government filed a motion to dismiss or, in the alternative, for summary judgment against all FTCA claimants (appellants) under the discretionary function exception. See United States District Court for the District of Columbia, Civil Docket for Case No. l:02-cv-00294-ESH, reprinted in Joint Appendix (“J.A.”) 20, 28. Appellants argued that the 'motion was premature, because jurisdictional discovery was necessary. The District Court permitted jurisdictional discovery limited to the existence of mandatory directives relevant to the first part of the discretionary function, exception. See Loughlin II, 286 F.Supp.2d at 7. The District Court subsequently granted the Government’s motion .to dismiss under the discretionary function ex ception, dismissing all claims against the United States with prejudice, and all remaining claims, including those against AU, without prejudice. See id. at 30.

Appellants appeal the dismissal of their case under the discretionary function exception and the court’s restrictions on jurisdictional discovery. See FTCA Appellants’ Br. 10-41. AU separately appeals the District Court’s denial of its motion to dismiss under Rule 12(b)(6), seeking reversal on the merits with respect to Saum if we find jurisdiction under the FTCA, or vacatur if we affirm the applicability of the discretionary function exception. See Appellant AU’s (“AU”) Br. 9-11. Prior to oral argument, we ordered, sua sponte, that counsel be prepared to address the jurisdictional basis for this court to exercise appellate review and the effect that a determination that diversity jurisdiction exists for the case between Saum and American University would have on AU’s request for vacatur. See Loughlin v. United States, No. 03-5284, Order (D.C.Cir. Oct. 15, 2004) (“Order of 10/15/04”).

II. Analysis

The FTCA vests the district courts with jurisdiction over civil damages claims against the United States

for injury or loss of property, or personal injury or death caused by the negligent or wrongful act or omission of any employee of the Government ... if a private person[] would be liable to the claimant in accordance with the law of the place where the act or omission occurred.

28 U.S.C. § 1346(b)(1). This waiver of sovereign immunity is limited in part by 28 U.S.C. § 2680(a), which insulates the Government from suits regarding

[a]ny claim based upon an act or omission of an employee of the Government, exercising due care, in the execution of a statute or regulation, whether or not such statute or regulation be valid, or based upon the exercise or performance or the failure to exercise or perform a discretionary function or duty on the part of a federal agency or an employee of the Government, whether or not the discretion involved be abused.

28 U.S.C. § 2680(a). The second clause of § 2680(a) “marks the boundary between Congress’ willingness to impose tort liability upon the United States and its desire to protect certain governmental activities from exposure to suit by private individuals.” United States v. S.A. Empresa de Viacao Aerea Rio Grandense (Varig Airlines), 467 U.S. 797, 808, 104 S.Ct. 2755, 2761-62, 81 L.Ed.2d 660 (1984).

A. The Discretionary Function Exception

The discretionary function exception is a barrier to subject matter jurisdiction. See Cope v. Scott, 45 F.3d 445, 448 (D.C.Cir.1995). A district court thus has no authority to address the merits of claims allegedly arising under the FTCA in cases in which the plaintiff is unable to overcome this jurisdictional barrier. Because federal jurisdiction determinations are purely legal, we review de novo the District Court’s judgment on the applicability of the discretionary function exception. See Macharia v. United States, 334 F.3d 61, 64 (D.C.Cir.2003), cert. denied, 540 U.S. 1149, 124 S.Ct. 1146, 157 L.Ed.2d 1042 (2004) (dismissal pursuant to Fed. R. Crv. P. 12(b)(1) reviewed de novo)-, Cope, 45 F.3d at 450 (rulings on application of the discretionary function exception reviewed de novo). A complaint may be dismissed on jurisdictional grounds only if “it appears beyond doubt that the plaintiff can prove no set of facts in support of his claim which would entitle him to relief.” Conley v. Gibson, 355 U.S. 41, 45-46, 78 S.Ct. 99, 101-02, 2 L.Ed.2d 80 (1957).

The Supreme Court has established a two-step test to determine whether a governmental act or omission falls within the ambit of the discretionary function exception. See Cope, 45 F.3d at 448 (citing United States v. Gaubert, 499 U.S. 315, 111 S.Ct. 1267, 113 L.Ed.2d 335 (1991)). These two prongs track the language of § 2680(a). The first asks whether a “ ‘federal statute, regulation, or policy specifically prescribes a course of action for an employee to follow....”’ Gaubert, 499 U.S. at 322, 111 S.Ct. at 1273 (quoting Berkovitz v. United States, 486 U.S. 531, 536, 108 S.Ct. 1954, 1958-59, 100 L.Ed.2d 531 (1988)). If such a binding directive exists, then “the employee has no rightful option but to adhere to the directive.” Berkovitz, 486 U.S. at 536, 108 S.Ct. at 1958-59. Failure to abide by such directives opens the United States to suit under the FTCA.

In the absence of such specific directives and where the “challenged conduct involves an element of judgment,” id., the second step determines whether the challenged discretionary act or omission is “of the nature and quality that Congress intended to shield from tort liability.” Varig Airlines, 467 U.S. at 813, 104 S.Ct. at 2764. The Supreme Court has explained that “[bjecaus'e the purpose of the exception is to prevent judicial second-guessing of legislative and administrative decisions grounded in social, economic, and political policy through the medium of an action in tort, ... the exception protects only governmental actions and decisions based on considerations of public policy.” Gaubert, 499 U.S. at 323, 111 S.Ct. at 1273-74 (internal quotation marks and citations omitted). “What matters i$ not what the decisionmaker was thinking, but [rather] the type of decision being challenged _” Cope, 45 F.3d at 449.

Appellants principally contend that the Government failed to warn them of the buried munitions and toxic chemicals and of the soil contamination that these materials purportedly created. They argue that the District Court erred when it concluded that these claims are covered by the discretionary function exception. See FTCA Appellants’ Br. 12-34. .

Appellants first submit that the Government’s decision not to warn of the buried material violated binding directives. See FTCA Appellants’ Br. 23-34. If mandatory directives so constrained the Government’s discretion, the discretionary function exception would not protect a failure to warn. We agree with the District Court, however, that appellants have failed to cite any regulations or policies that prescribed a nondiscretionary duty to warn. We therefore affirm the District Court’s ruling on prong one of the discretionary function test for the reasons enunciated by the District Court in its exhaustive opinion. See Loughlin II, 286 F.Supp.2d at 9-19.

Although appellants point to myriad documents from the World War I era, they fail to identify a specific and binding directive that (1) addressed AUES or .research facilities in general, and (2) proscribed the burial of munitions or required that such burials be marked. As the District Court explained, appellants point to a 1919 shipping order that addressed AUES but did not discuss buried munitions, and a Gas Warfare Bulletin that proscribed burial in most cases and required markings where burial was permitted, but gives no indication that it applied to activities at AUES. See id. at 9-14.

Appellants also fail to identify specific and binding directives that pertain to their negligence claims for the post-1986 time period, ie., after the Government had begun to investigate the potential existence of submerged chemical warfare materials and the risk of contamination. See id. at 14-19. We reject, moreover, appellants’ suggestion that the Government had adopted an unwritten or de facto policy to inform the community during this time period. As the District Court concluded:

While ... documents do suggest that the Army wished to communicate information to the public about the progress of its investigation, it reads far too much into them to argue that they demonstrate an official, irrevocable commitment to alert the public to every development in the search for buried munitions, or to every potential danger that was discovered.

Id. at 29. Thus, appellants have failed to identify relevant binding directives or policies, the violation of which would bring this case outside of § 2680(a).

We turn, then, to the second prong of the discretionary function exception. Appellants argue that the Government’s failure to warn of the buried munitions and of the resulting contamination was not a decision susceptible to public policy considerations. See FTCA Appellants’ Br. 12-23. First, appellants seem to urge that the Government’s decision to bury the munitions without disclosing their burial during World War I or its immediate aftermath is not the type of decision susceptible to public policy considerations. This argument is untenable. The decision whether to warn of these burials in the immediate aftermath of the War was “fraught with ... public policy considerations,” Cope, 45 F.3d at 449 (internal quotation marks omitted). As the District Court observed, it required balancing “competing concerns of secrecy and safety, national security and public health.” Loughlin II, 286 F.Supp.2d at 23. Indeed, at oral argument, appellants’ counsel conceded that national security concerns are greater when the decision whether to warn is contemporaneous with the war effort. See Recording of Oral Argument at 3:17-:28.

Appellants contend, however, that because the Government’s duty to warn was ongoing, concerns that were salient in wartime cannot determine the nature of the Government’s decision not to warn over the subsequent 80 years. See FTCA Appellants’ Br. 22; Recording of Oral Argument at 2:30-:49. The argument is without merit. Appellants’ position assumes that the Government was required consistently and regularly to revisit its initial decision not to warn and re-balance the relevant factors. A judicially constructed requirement to rethink particular decisions not to warn on a regular basis for over 80 years would constitute precisely the “judicial second-guessing” that the discretionary function exception was intended to displace. Gaubert, 499 U.S. at 323, 111 S.Ct. at 1273-74 (internal quotation marks omitted). It would insert the courts into prioritization and resource allocation decisions that implicate serious political, social, and economic considerations. Instead, in order for us to consider whether the decision not to warn was susceptible to public policy considerations in subsequent years, appellants must identify circumstances in which the Government itself would have revisited the decision not to warn.

Appellants point to two such events. The first is a finding by EPA that one of seven soil samples collected in 1994 from the 4825 Glenbrook Road property revealed heightened levels of arsenic. See FTCA Appellants’ Br. 22. Appellants allege that they were not informed of this result until 1999. Second, appellants allege that they were not informed of a 1993 Draft Field Sampling Plan, which identified as “points of interest” two munitions pits adjacent to the 4825 Glenbrook property. See id.; see also Engineering-Science, Inc., 1 Field Sampling Plan for Hazardous and Toxic Waste, Spring Valley Project, Washington, D.C. (on behalf of the U.S. Army Corps of Engineers), 6/30/93 (“Draft Field Sampling Plan”), J.A. 2857, 2878. Appellants submit that the Government’s knowledge of specific hazards (the abnormal soil sample and the points of interest in the Draft Field Sampling Plan) renders the decision whether to warn void of public policy considerations. See FTCA Appellants’ Br. 22.

Appellants’ arguments on these two points fail for the reasons provided by the District Court:

In deciding whether this information [on the abnormal soil sample result] should be made public, the agency would have had to weigh several factors, including the reliability of the test, the significance of one unusual result, the possibility of unnecessarily alarming Spring Valley residents should the danger have ultimately proved unfounded, and whether further testing should be done before this data was revealed. Conducting this delicate balance is a matter that calls for discretion of the sort that the FTCA shields from judicial second-guessing. ...

Similarly, with respect to the pits, while there were two shell pits ... identified in a Draft Field Sampling Plan prepared by the Corps in 1993 ..., no information existed at the time that the pits actually contained munitions.... [T]he existence of munitions pits — which were constructed to test, not bury, explosives — does not necessarily indicate the presence of buried contaminants. Certainly, the Plan which claimants suggest was withheld from them does not indicate that weapons or other toxic materials were actually in the ground. As such, the Army’s decision whether to announce this limited information, with its highly speculative relationship to actual risk, is subject to the same policy considerations that attended upon EPA’s decisions whether to go public with its soil test results.

Loughlin II, 286 F.Supp.2d at 28-29 (internal quotation marks and citations omitted). The District Court’s judgment is consistent with this court’s decision in Wells v. United States, 851 F.2d 1471 (D.C.Cir.1988), where we held that the discretionary function exception barred claims that EPA negligently failed to inform residents neighboring three lead smelters of potential risks and to provide a timely remedy for the hazards. See id. at 1472. Wells concluded that EPA’s decision to engage in further study to determine the appropriate lead testing level was based on public policy considerations, including the socio-political and economic implications of recognizing an action level in one situation that could not be consistently applied. See id. at 1473, 1477.

Appellants contend, however, that Cope v. Scott, not Wells, controls this case. In Cope, we held that prong two of the discretionary function exception did not extend to the Government’s decision not to post a warning sign on a particular strip of road. See Cope, 45 F.3d at 451. We rejected the Government’s purported public policy concern of aesthetics, given the numerous signs on the same stretch of road. See id. at 452. Appellants submit that just as Cope held that the decision whether to post warning signs in suitable locations was not susceptible to public policy considerations once other warning signs were placed on the same strip of road, the Government’s decision in 1999 to warn Spring Valley residents of contamination reveals that public policy considerations did not determine the Government’s earlier deci sions not to issue such warnings. Thus, appellants argue, the only remaining questions are what the Government knew and when, which turn on negligence law, not public policy. See Recording of Oral Argument at 13:25-:52, This argument cannot carry the day for appellants.

•The geography gap in Cope (warning sign at one location but not at another) is not analogous to the time gap here (warning issued in 1999 but not earlier). In Cope, the presence of no less than 23 signs on the same strip of road was probative of the nature of the decision to place an additional warning sign, because it demonstrated that the Government was not concerned with preserving a pristine view on the particular stretch of road. See Cope, 45 F.3d at 452 (“[T]he Park Service has chosen to manage the road in a manner more amenable to commuting through nature than communing with it.”). In contrast, the Government’s decision to warn Spring Valley residents of contamination in 1999 sheds no light on whether the prior decision not to warn was “ ‘susceptible to policy judgment’ and involve[d] an exercise of ‘political, social, [or] economic judgment.’ ” Cope, 45 F.3d at 448 (quoting Gaubert, 499 U.S. at 325, 111 S.Ct. at 1274; Varig Airlines, 467 U.S. at 820, 104 S.Ct. 2755) (second alteration in original). That a Government agent in 1999 made a different decision is irrelevant. See, e.g., Allen v. United States, 816 F.2d 1417, 1424 (10th Cir.1987) (“However erroneous or misguided [the government’s] deliberations may seem today, it is not the place of the judicial branch [through the FTCA] to now question them.”). The passage of time gave the Government more information to digest, an opportunity to re-weigh the political, social, and economic considerations, and occasion to make a new policy judgment. The earlier judgment was no less a matter of policy because the later judgment was arguably better informed.

In other words, the nature of the decision whether to warn in 1999 was different from the nature of the decision whether to warn prior to 1999 because of intervening discoveries that rendered the risk to public health substantially less speculative. The burial pit on the property of the Korean Ambassador, which abuts the Loughlins’ property, was not discovered until 1998-1999. When this pit was finally located, it was found to contain submerged munitions. Loughlin II, 286 F.Supp.2d at 28 n. 27. In addition, 18 of 22 soil samples taken in June 1999 revealed elevated levels of arsenic. Id. at 7. Thus, the warnings that were issued in 1999 followed the discovery of a new burial pit and extensive new soil sampling results. They do not shed light on the nature of the earlier decisions not to warn.

B. The Discovery Orders

Before leaving the FTCA appeal, we turn to appellants’ challenge to the District Court’s discovery orders. Appellants submit that the District Court abused its discretion by limiting discovery to the' existence of binding directives under prong one of the discretionary' function exception. See FTCA Appellants’ Br. 34-4Ó. We review the District Court’s discovery rulings for abuse of discretion. See Macharia, 334 F.3d at 64. Although the District Court’s partition between “prong-one discovery” and “prong-two discovery” is misguided, we find no basis for reversal.

Because the prong-two inquiry looks to the type of decision whether to warn, irrespective of considerations that factored into the actual decision, see Cope, 45 F.3d at 449, it may often be the case that discovery is unnecessary to determine whether prong two of the discretionary function exception extends to any particular act or omission. There are situations, however, where the factual predicate is critical to an accurate analysis of the nature of the decision made. In Cope, for example, we rejected the Government’s purported policy justification because we considered it in light of the factual context: The government’s alleged aesthetic considerations were undermined by the presence of multiple signs on the same stretch of road. See Cope, 45 F.3d at 452.

The District Court therefore erred in suggesting that our “[cjircuit law permits discovery as to prong 1 only.” Tr. of Status Conference of 12/16/02, J.A. 1728, 1796. Ignatiev v. United States, 238 F.3d 464 (D.C.Cir.2001), on which the District Court relied, see Tr. of Mots. Hearing of 1/13/03, J.A. 1826,1834, is not on point. In Ignatiev, FTCA claimants alleged that the Secret Service “likely had internal objectives or policies that created the requisite mandatory obligation” to take certain safety precautions, which the claimants alleged had not been taken. 238 F.3d at 466. We held that the lower court erred in not permitting discovery on the existence of such directives, noting that “the only discovery necessary to establish jurisdiction pertains ... to the existence vel non of internal governmental policies ....” Id. at 467. Circumscribing discovery in this way was consistent with the Ignatiev appellants’ claim. Even in Ignatiev, however, we recognized that where “facts [are] necessary to establish jurisdiction,” plaintiffs must be afforded the “opportunity for discovery of [such] facts ... prior to” the granting of a motion to dismiss for lack of subject matter jurisdiction. Id.

Jurisdictional discovery on the data that was available when a particular decision was made may be especially salient in the failure-to-warn context, where the speculativeness of the risk affects the nature of the decision not to disclose. Indeed, the District Court relied on such facts when it concluded that the discretionary function exception applies to appellants’ failure-to-warn claim. See Loughlin II, 286 F.Supp.2d at 28 (“[W]hile there were two shell pits ... identified in a Draft Field Sampling Plan prepared by the Corps in 1993 ..., no information existed at the' time that the pits actually contained munitions.”). When such facts are probative of the nature of the decision itself, jurisdictional discovery may be illuminating. No- bright line rule confines discovery to prong one of the discretionary function exception. The suggestion, moreover, that one can neatly partition “prong-one discovery” from “prong-two discovery” is misleading. Evidence that sheds light on the type of decision made is not necessarily distinct from evidence on binding directives that would be responsive to prong one. The search for binding directives may — and here, did — uncover important facts pertaining to the nature of the decisions at issue.

Although the District Court misconstrued our case law on jurisdictional discovery, the discovery orders nonetheless afforded the parties a full and fair opportunity to pursue relevant information, which, in turn, allowed the District Court to make a just finding on the speculativeness of the risk. This occurred because, as it turned out in this case, the so-called prong-one discovery reasonably embraced the available data on potential hazards. The 1993 Draft Field Sampling Plan identified pits and trenches as “potential burial areas,” and presented “sampling activities [to] be used to evaluate the possible presence of chemical agents or chemical agent-related contamination and explosives or explosive-related contamination.” Draft Field Sampling Plan at J.A. 2901. The Draft Field Sampling Plan itself, therefore, undermines the suggestion that the Government had specific knowledge of buried munitions or contamination in the area. A Final Remedial Investigation Evaluation Report explains that the location of “point of interest 24,” the munitions pit that was eventually discovered on the Korean Ambassador’s property, was revised when aerial and supporting photographs were reevaluated as a result of a 1997 evaluation of the earlier remedial investigation. See U.S. Army Engineering and Support Center, U.S. Army Corps of Engineers & Parsons Engineering Science, Inc., Final Remedial Investigation Evaluation Report, Operation Safe Removal Formerly Used Defense Site, Washington, D.C., 1/8/98, J.A. 1972, 2078. Thus, the documents cited by appellants reveal an involved investigation and Government actors struggling to define the scope of the potential hazard.

To the extent that appellants seek additional discovery relevant to the prong-two inquiry, they have failed to particularize their requests. In their briefs and when pressed at oral argument, appellants consistently failed to articulate precisely what information, pertaining to the nature of the decision whether to warn, they had been denied. See Recording of Oral Argument at 32:17-34:59. Given the sprawling record already compiled in this case and appellants’ failure to particularize additional .discovery requests, we conclude that the District Court’s discovery orders reasonably provided the trial court and this court with the necessary facts to evaluate the nature of the challenged decisions.

C. Disposition of the District Court Opinion Denying American University’s Motion to Dismiss under Rule 12(b)(6).

The only remaining question is whether the District Court’s decision addressing the local law claims can survive the court’s subsequent determination that it lacked subject matter jurisdiction over the federal claims. We hold that it cannot.

The District Court considered claims against AU pursuant to its supplemental jurisdiction under 28 U.S.'C. § 1367. Section 1367(a) provides:

[I]n any civil action of which the district courts have original jurisdiction, the district courts shall have supplemental jurisdiction over all other claims that are so related to claims in the action within such original jurisdiction that they form part of the same case or controversy under Article III of the United States Constitution....

28 U.S.C. § 1367(a). The District Court exercised supplemental jurisdiction over the local law claims against AU based on its original jurisdiction over the FTCA claims against the Government. On March 13, 2002, AU moved to dismiss the local law claims pursuant to Rule 12(b)(6). The District Court denied AU’s motion to dismiss, because it found that AU owed a legal duty under the applicable District of Columbia laws. See Loughlin I, 209 F.Supp.2d at 167-69. AU seeks vacatur of the court’s decision, because the District Court had no authority to exercise supplemental jurisdiction and render a decision against AU in the absence of subject matter jurisdiction in the FTCA action. AU also fears that litigants will attempt to invoke the opinion in other proceedings. AU’s concern is well founded. The opinion has been cited as authoritative by other plaintiffs suing AU in Superior Court. See Pls.’s Opp’n to Def. The American University’s Mot. to Dismiss, Jack v. Am. Univ., No. 03-CA-004659, 2003 WL 22520495 (D.C.Super.Ct.), filed 9/4/03, reprinted in Appellee Saum’s Br. Add. 61, 84 (“All three of these arguments [against a legal duty owed by AU to Spring Valley residents] have already been rejected by the federal court in denying AU’s motion to dismiss the Spring Valley personal injury cases in the case of Loughlin v. United States, 209 F.Supp.2d 165 (D.D.C.2002) (applying D.C. law).”).

Appellee Saum contests AU’s standing to seek vacatur. See 10/17/04 Letter filed by Appellee Saum, in response to Order of 10/15/04 (“Letter from Saum of 10/17/04”). Saum confuses standing with mootness, however. Although Article Ill’s case-or-controversy requirement under-girds both our standing and mootness jurisprudence, the two justiciability doctrines differ in critical respects. See Friends of the Earth, Inc. v. Laidlaw Envtl. Servs. (TOC), Inc., 528 U.S. 167, 180, 120 S.Ct. 693, 703-04, 145 L.Ed.2d 610 (2000). As the Court has noted, its “repeated statements that the doctrine of mootness can be described as ‘the doctrine of standing set in a time frame’ ” may have created an impression of greater overlap than actually exists between the two doctrines. Id. at 189-91, 120 S.Ct. at 708-10 (internal citations omitted). In fact,

[s]tanding doctrine functions to ensure, among other things, that the scarce resources of the federal courts are devoted to those disputes in which the parties have a concrete stake. In contrast, by the time mootness is an issue, the case has been brought and litigated, often (as here) for years.

Id. at 191, 120 S.Ct. at 709-10. In other words, “[mjootness doctrine encompasses the circumstances that destroy the justicia-bility of a suit previously suitable for determination.” 13A ChaRles Alan Wright, Arthur R. Miller & Edward H. Cooper, Federal Practioe And Procedure § 3533 (2d ed. 1984).

The issue here is not whether AU lacked standing; indeed, AU was the defendant who removed cases involving local law claims to federal court, so its standing was never an issue. Rather, the issues here are whether the local law claims against AU, which rested solely on the District Court’s supplemental jurisdiction, were rendered moot due to the court’s determination that it lacked subject matter jurisdiction over the FTCA case, and, if so, whether the District Court’s decision addressing the merits of the non-federal claims must be vacated.

Saum contends that AU has no “standing to appeal,” because the non-federal claims are no longer before the District Court. Letter from Saum of 10/17/04. In furtherance of this argument, Saum suggests that the District Court’s opinion on the merits should not be disturbed, because it was issued against AU before the court decided that it had no supplemental jurisdiction. See Appellee Saum’s Br. 4-5. According to Saum, this confluence of circumstances bars AU from seeking vacatur of the District Court’s decision on appeal. See id. These contentions are meritless in light of the Supreme Court’s decision in Arizonans for Official English v. Arizona, 520 U.S. 43, 117 S.Ct. 1055, 137 L.Ed.2d 170 (1997).

The action in Arizonans for Official English was mooted just after the district court had rendered a judgment in plaintiffs favor, but before the court of appeals had heard the appeal. The Ninth Circuit declined to find the case moot and upheld the judgment for the plaintiff. Before the Supreme Court, plaintiff-respondent urged that the District Court judgment should not be upset because it was entered before the mooting event occurred and was not properly appealed.” Id. at 72, 117 S.Ct. at 1071. While expressing “grave doubts” over whether the named petitioners had standing to pursue appellate review, id. at 66, 117 S.Ct. at 1068, the Supreme Court nonetheless entertained the appeal, held that the case was moot, and concluded that “vacatur down the line” was required, id. at 75, 117 S.Ct. at 1072-73. In reaching this result, the Court stressed that:

“[E]very federal appellate court has a special obligation to satisfy itself not only of its own jurisdiction, but also that of the lower courts in a cause under review .... And if the record discloses that the lower court was without jurisdiction this court will notice the defect .... [When the lower federal court] lack[s] jurisdiction, we have jurisdiction on appeal, not of the merits but merely for the purpose of correcting the error of the lower court in entertaining the suit.”

Id. at 73, 117 S.Ct. at 1071-72 (quoting Bender v. Williamsport Area Sch. Dist., 475 U.S. 534, 106 S.Ct. 1326, 89 L.Ed.2d 501 (1986) (internal quotation marks and citations omitted)) (alterations in original).

Saum appears to submit that the District Court in this case had subject matter jurisdiction over the local law negligence claims at the time when it ruled on AU’s motion to dismiss, irrespective of whether it ultimately lacked such jurisdiction. See Appellee Saum’s Br. 4-5. There is no merit to this contention. As we held in Tuck v. Pan American Health Organization, 668 F.2d 547 (D.C.Cir.1981), subject matter jurisdiction “is, of necessity, the first issue for an Article III court,” for “[t]he federal courts are courts of limited jurisdiction, and they lack the power to presume the existence of- jurisdiction in order to dispose of a case on any other grounds.” Id. at 549. In the absence of subject matter jurisdiction, the District Court’s finding that AU owed Spring Valley residents a legal duty constitutes little more than an impermissible advisory opinion. Cf . Steel Co. v. Citizens for a Better Env’t, 523 U.S. 83, 101, 118 S.Ct. 1003, 1016, 140 L.Ed.2d 210 (1998) (“Hypothetical jurisdiction produces nothing' more than a hypothetical judgment — which comes to the same thing as an advisory opinion, disapproved by this Court from the beginning.”).

In addition, as Arizonans for Official English indicates, an appellate court may act sua sponte to vacate a trial court decision if it determines that the lower court lacked jurisdiction due to mootness. This is precisely what happened in Flynt v. Weinberger, 762 F.2d 134 (D.C.Cir.1985) (per curiam), where the District Court dismissed plaintiffs action as moot, but offered an opinion on the merits. On appeal, we affirmed the dismissal of the case on grounds of mootness and sua sponte vacated the District Court’s opinion. The decision in Flynt holds

that the district court, while purporting to dismiss the case for lack of jurisdiction, improperly considered and offered judgments on the underlying merits of the dispute....

Where a controversy has become moot, it is the duty of the appellate court to clear the path for future reliti-gation of the issues raised. Accordingly, while we affirm the dismissal of this particular complaint on the ground that it is moot, we vacate the opinion of the district court.

Id. at 135-36 (citing United States v. Munsingwear, 340 U.S. 36, 40, 71 S.Ct. 104, 107, 95 L.Ed. 36 (1950)). This holding is on point here. As in Flynt, the District Court in this case had no jurisdiction to render a judgment on the merits of the matters before it. Therefore, the decision of the District Court must be vacated to “elear[ ] the path for future relitigation by eliminating a judgment [AU] was stopped from opposing on direct review.” Arizonans for Official English, 520 U.S. at 71, 117 S.Ct. at 1070-71 (internal quotation marks omitted). We “utilize[ ] vacatur ‘to prevent a judgment, unreviewable because of mootness, from spawning any legal consequences,’ ” irrespective of whether such consequences are imminent or “remote.” Am. Family Life Assurance Co. v. FCC, 129 F.3d 625, 631 (D.C.Cir.1997) (quoting Munsingwear, 340 U.S. at 41, 71 S.Ct. at 107).

In this case, had the District Court properly considered the FTCA action before addressing the claims that rested solely on supplemental jurisdiction it would have been clear beyond the slightest doubt that Saum had neither subject matter jurisdiction nor standing to seek a federal court ruling on the merits of her local law claims against AU. Saum’s suit against AU would have been dismissed for want of subject matter jurisdiction with no decision from the District Court on the merits. Saum surely cannot avoid vacatur merely because the District Court considered the FTCA and local law actions in the wrong order.

Finally, having foregone the opportunity to cross-appeal on the issue of subject matter jurisdiction, Saum now argues for the first time that the District Court in fact had jurisdiction over the claims against AU under 28 U.S.C. § 1332(a) (2000) (diversity of citizenship). See Ap-pellee Saum’s Br. 6. This claim comes too late and barren of the requirements necessary to establish diversity jurisdiction.

In her brief to this court, Saum acknowledges that “[njeither of the parties presented diversity as the basis for jurisdiction in the district court. Indeed, the University, as the removing defendant, was precluded from asserting diversity in support of its removal notice under 28 U.S.C. § 1441(b).” Appellee Saum’s Br. 6 n.2. Saum adds that she “was not required to amend her complaint after the removal and set forth basis for federal court jurisdiction.” Id. This argument misses the point. AU’s removal rested in part on an assumption that the District Court had supplemental jurisdiction over the non-federal claims under 28 U.S.C. § 1367(a). When it became clear that the court lacked supplemental jurisdiction, Saum had no grounds upon which to pursue her claims against AU unless she amended her complaint to establish diversity jurisdiction. She never did this.

“Because federal courts are of limited jurisdiction, there is a presumption against the existence of diversity jurisdiction. Accordingly, the party seeking the exercise of diversity jurisdiction bears the burden of pleading the citizenship of each and every party to the action.” Naartex Consulting Corp. v. Watt, 722 F.2d 779, 792 (D.C.Cir.1983) (internal citations omitted). Saum has never pleaded the facts necessary to establish diversity. In a supplemental submission to this court, Saum does nothing to meet her burden. Rather, relying on District of Columbia ex rel. American Combustion, Inc. v. Transamerica Insurance Co., 797 F.2d 1041, 1044 (D.C.Cir.1986), Saum merely argues that diversity -jurisdiction may be found and exercised in the court of appeals where the facts establishing such jurisdiction appear in the record. See Letter from Saum of 10/17/04. Saum misconstrues Transamerica.

In Transamerica, we stated that “a party who has not proved, or even alleged, that diversity exists [may] amend his pleadings even as late as on appeal,” particularly when the amendment is uncontested. Id. at 1044 (discussing 28 U.S.C. § 1653 (1982) (“Defective allegations of jurisdiction may be amended, upon terms, in the trial or appellate courts.”)). Unlike the appellee in Transamerica, Saum has not moved to amend her complaint to aver diversity jurisdiction. And she has presented no rationale that would warrant a departure from this requirement. See, e.g., Wolfe v. Marsh, 846 F.2d 782, 785 n. 4 (D.C.Cir.1988) (per curiam). Nor has Saum cross-appealed on the jurisdictional issue. Diversity jurisdiction must be pleaded by thp party claiming it. Jurisdictional pleading must not be reduced to a mere afterthought years after the commencement, of the action. The court cannot .act without jurisdiction and it is the complaining party’s burden to plead it. The question whether diversity exists here has not been conceded by AU, and it has not been properly presented by Saum.

There is a further problem with Saum’s position. Even if we were to find diversity jurisdiction, there would be no appropriate redress. The District Court dismissed Saum’s action for lack of subject matter jurisdiction, and Saum has never appealed that judgment. Any decision by this court finding diversity would be merely advisory. We therefore decline Saum’s belated invitation to determine whether there is an adequate basis for diversity jurisdiction.

In sum, the District Court had no subject matter jurisdiction over the FTCA action. The court had no supplemental jurisdiction under § 1367(a) to entertain any of the non-federal claims. Accordingly, we hold that, because it should not have reached the merits of the negligence claims under District of Columbia law,.the District Court’s decision against AU on local law claims is a nullity’ and must be vacated.

III. "Conclusion

The District Court’s dismissal of appellants’ FTCA claims under the discretionary function exception is hereby affirmed. The District Court’s decision in Loughlin v. United States, 209 F.Supp.2d 165 (D.D.C.2002), which purports to address the local law claims, is hereby vacated.

So ordered.