NORTH CAROLINA WILDLIFE FEDERATION; Clean Air Carolina; Yadkin Riverkeeper, Plaintiffs-Appellants,
v.
NORTH CAROLINA DEPARTMENT OF TRANSPORTATION; Eugene Conti, Secretary of the North Carolina Department of Transportation; Federal Highway Administration; John F. Sullivan, Division Administrator, Federal Highway Administration, Defendants-Appellees.

*597 ARGUED: Frank S. Holleman, III, Southern Environmental Law Center, Chapel Hill, North Carolina, for Appellants. Seth Morgan Wood, Office of the United States Attorney, Raleigh, North Carolina; Scott Thomas Slusser, North Carolina Department of Justice, Raleigh, North Carolina, for Appellees. ON BRIEF: Chandra T. Taylor, Kimberley Hunter, J. David Farren, Southern Environmental Law Center, Chapel Hill, North Carolina, for Appellants. Roy Cooper, Attorney *598 General, Raleigh, North Carolina, for Appellees North Carolina Department of Transportation and Eugene Conti, Secretary of the North Carolina Department of Transportation; Thomas G. Walker, United States Attorney, Raleigh, North Carolina, for Appellees Federal Highway Administration and John F. Sullivan, Division Administrator, Federal Highway Administration.

Before MOTZ, SHEDD, and AGEE, Circuit Judges.

Vacated and remanded by published opinion. Judge MOTZ wrote the opinion, in which Judge SHEDD and Judge AGEE joined.

OPINION

DIANA GRIBBON MOTZ, Circuit Judge:

The North Carolina Department of Transportation and the Federal Highway Administration (collectively "the Agencies") recently approved construction of a new twenty-mile toll road in North Carolina linking Mecklenburg and Union Counties—the Monroe Connector Bypass. Seeking to enjoin construction of the toll road, the North Carolina Wildlife Federation, Clean Air Carolina, and Yadkin Riverkeeper (collectively "the Conservation Groups") filed this suit, contending that the process by which the Agencies approved the road violated the National Environmental Protection Act ("NEPA"). The district court granted summary judgment to the Agencies. The Conservation Groups now appeal. Because the Agencies failed to disclose critical assumptions underlying their decision to build the road and instead provided the public with incorrect information, they did indeed violate NEPA. Accordingly, we must vacate the judgment of the district court and remand for further proceedings consistent with this opinion.

I.

We begin by recounting the undisputed facts and the procedural history of this case.

A.

In 2007, after a number of failed attempts, the North Carolina Department of Transportation again proposed construction of the Monroe Connector Bypass (hereinafter "the Monroe Connector") and, under the Federal Highway Administration's supervision, commenced the environmental assessment process required by NEPA.

The Agencies began by creating a Statement of Purpose and Need, which evaluated the region's existing transportation network, including an in-depth analysis of U.S. Highway 74. That road serves as the "primary transportation connection between Union County, the fastest growing county in North Carolina, and Mecklenburg County/City of Charlotte, the economic hub of the region," and "also serves as an important commercial corridor." The Agencies determined that U.S. 74's "[a]verage travel speeds range from approximately 20 to 30 miles per hour during the peak hour" with "one-third of the intersections operating at an unacceptable Level of Service." The Statement of Purpose and Need concluded that U.S. 74 suffers from "[c]apacity [d]eficiencies" and does not allow for "high-speed regional travel."

To remedy these shortcomings, the Agencies proposed developing "a facility that allows for safe, reliable, high-speed regional travel in the U.S. 74 Corridor . . . while maintaining access to properties along existing U.S. 74."

*599 In evaluating possibilities that might meet these goals, the Agencies created an Environmental Impact Statement (hereinafter "Impact Statement"). The Agencies began with a draft Impact Statement, and after taking public comment, published a final Impact Statement. See 40 C.F.R. §§ 1502.9(a), 1503.1(a), 1503.4.

The draft Impact Statement analyzed a wide variety of proposals, among them the Agencies' preferred choice—the Monroe Connector. After two initial screenings, the Agencies eliminated all but three categories of proposals: (1) improve U.S. 74, (2) construct the Monroe Connector, and (3) a hybrid—combining the Monroe Connector with improvements to U.S. 74. Within these categories, the Agencies developed twenty-five "preliminary study alternatives." They analyzed these preliminary study alternatives to determine which ones "should be carried forward" as "detailed study alternatives" to receive greater analysis. After considering economic impacts and traffic projections, the Agencies "carried forward" sixteen "build" alternatives for detailed study.^[1] These "build" alternatives consisted of nearly identical paths for constructing the Monroe Connector.

The Agencies also carried forward a "no action" alternative—a scenario "without major improvements." The draft Impact Statement and the final Impact Statement explained that the purpose of the "no action" or "no build" alternative was "to provide a baseline for comparison" with the "build" alternatives. For example, in the draft Impact Statement, the Agencies compared the year-2035 traffic projections for the "build" alternatives against the year-2035 traffic projections for the "no build" baseline. Similarly, in the final Impact Statement the Agencies compared the indirect and cumulative environmental effects of the "build" alternatives with the "no build" baseline. Given its widespread use the accuracy of the "no build" baseline was critically important.

The Agencies created the "no build" baseline using information from a local planning organization. They relied on the Mecklenburg-Union Metropolitan Planning Organization ("MUMPO") to develop socioeconomic data based on its Regional Travel Demand Model. This model projects "population, household, and employment figures" for the region using a two-step process. First, a "top-down" analysis of "census and employment projection data. . . at the county [level]" calculated "the maximum number of households, population and employment." Appellees' Br. at 15-16. Second, a "bottom-up" process allocated the top-down projected growth throughout the Monroe Connector's future land use study area. Id.

The bottom-up process divided the area into 571 "traffic analysis zones." MUMPO's model then calculated projected growth for each of these zones using a number of "land development factors": developable and redevelopable residential land, population change, water availability, sewer availability, expert predicted growth, municipal growth policy, and travel time to employment. A zone's "travel time to employment" depended on MUMPO's anticipated roadway network for the region. This anticipated roadway network included the proposed Monroe Connector. Thus, although the Agencies used MUMPO's *600 projections as the "no build" baseline, part of MUMPO's data actually assumed construction of the Monroe Connector. By using MUMPO's data, therefore, the Agencies incorporated "build" assumptions into the "no build" baseline.

Throughout the NEPA process, public commentators repeatedly asked the Agencies whether the "no build" baseline in fact assumed construction of the Monroe Connector. In responding to these comments, the Agencies either failed to address the underlying issue or incorrectly stated that the Monroe Connector was not factored into the "no build" baseline.

For example, when the Agencies published their estimate that the 2035 "build" traffic volume would be less than the 2035 "no build" baseline traffic volume, the Conservation Groups queried whether this "implausibl[e]" conclusion might be the result of the "no build" data actually "assum[ing] that the Monroe Connector" would be built. Without responding to the Conservation Groups' underlying concerns regarding the accuracy of the "no build" baseline, the Agencies simply issued an errata table lowering the 2035 "no build" traffic projection baseline to below the "build" levels. The Agencies offered no explanation as to the source of the error and instead summarily stated that "the 2035 No-Build Alternative [traffic] forecast was inadvertently overestimated," and assured the public that "all other conclusions and discussions remain valid."

Similarly, when the Agencies concluded that the indirect and cumulative environmental effects of the Monroe Connector were minimal compared to the "no build" baseline, the United States Fish and Wildlife Service requested "further clarification regarding the basis for the No-build scenario." "Specifically," the Fish and Wildlife Service asked "if MUMPO's [projections] are the basis for the no-build scenario and [if] they contain the [Monroe Connector], how is this a true characterization of no-build?" The Agencies responded with a memorandum drafted by their consultant, Baker Engineering, concluding that "the methodology for determining the No-Build scenario. . . is appropriate and defensible." Baker expressly stated, however, that "the MUMPO . . . projections do not account for the Monroe Connector[ ]." (emphasis added).

Unsatisfied with this response, the Fish and Wildlife Service asked whether the Agencies were "doubly sure about th[eir] assumption." The Agencies subsequently asked Baker to contact MUMPO and local officials to ask whether they "would agree with [its] assumption that the[ ] ["no build" baseline] forecasts represent a future scenario without the Monroe Connector." After receiving generally confirming responses, Baker assured the Fish and Wildlife Service. In light of this assurance, although the Fish and Wildlife Service "continued to be concerned about the level of impacts," it issued its Endangered Species Act concurrence, thereby removing a major impediment to the Monroe Connector.

On August 27, 2010, the Agencies issued their Record of Decision—the final component of the NEPA process. See 40 C.F.R. § 1505.2. The Conservation Groups again asked whether the underlying data represented a true "no build" scenario. The Record of Decision responded with additional denials, stating that the "socioeconomic forecasts for the No Build Scenario did not include the Monroe Connector." (emphasis added). The Record of Decision ultimately selected a site for construction of the Monroe Connector at an approximate cost of $800 million.

B.

On November 2, 2010, the Conservation Groups filed this action seeking to enjoin *601 construction of the Monroe Connector. After the district court denied the Conservation Groups' motions for a preliminary injunction and to complete and supplement the record, the parties filed cross-motions for summary judgment.

The Conservation Groups asserted that the Agencies violated NEPA by: "(1) failing to analyze the environmental impacts of the Monroe Connector[ ]; (2) conducting a flawed analysis of alternatives; and (3) presenting materially false and misleading information to other agencies and to the public." N.C. Wildlife Fed'n v. N.C. Dep't of Transp., 2011 WL 5042075, at *1 (E.D.N.C. Oct. 24, 2011). In response, for the first time, the Agencies admitted that the "no build" baseline data did, indeed, assume the existence of the Monroe Connector, but the Agencies nonetheless contended that they were entitled to summary judgment. See id. at *4, *9. The district court agreed with the Agencies, reasoning that their "use of and reliance on [MUMPO's] data was reasonable and did not violate defendants' NEPA obligations." Id. at *10. Accordingly, the district court granted summary judgment to the Agencies.

The Conservation Groups timely noted this appeal.

II.

NEPA claims are subject to judicial review under the Administrative Procedure Act, which permits a reviewing court to set aside an agency action if the action was "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." 5 U.S.C. § 706(2)(A); see Dep't of Transp. v. Pub. Citizen, 541 U.S. 752, 763, 124 S. Ct. 2204, 159 L. Ed. 2d 60 (2004); Marsh v. Or. Natural Res. Council, 490 U.S. 360, 375-76, 109 S. Ct. 1851, 104 L. Ed. 2d 377 (1989). "This inquiry must `be searching and careful,' but `the ultimate standard of review is a narrow one.'" Marsh, 490 U.S. at 378, 109 S. Ct. 1851 (quoting Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416, 91 S. Ct. 814, 28 L. Ed. 2d 136 (1971)). A reviewing court must ensure that the agency has "`examine[d] the relevant data and articulate[d] a satisfactory explanation for its action,'" F.C.C. v. Fox Television Stations, Inc., 556 U.S. 502, 513, 129 S. Ct. 1800, 173 L. Ed. 2d 738 (2009) (quoting Motor Vehicle Mfrs. Ass'n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S. Ct. 2856, 77 L. Ed. 2d 443 (1983)), and must not reduce itself to a "rubber-stamp" of agency action. Fed. Mar. Comm'n v. Seatrain Lines, Inc., 411 U.S. 726, 745-46, 93 S. Ct. 1773, 36 L. Ed. 2d 620 (1973); see also Ohio Valley Envtl. Coal. v. Aracoma Coal Co., 556 F.3d 177, 192 (4th Cir.2009).

The NEPA process includes a range of "`action-forcing' procedures that require ... agencies [to] take a `hard look' at environmental consequences [of a proposed action] and [to] provide for broad dissemination of relevant environmental information." Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 350, 109 S. Ct. 1835, 104 L. Ed. 2d 351 (1989) (quotation marks and citation omitted). This process does not mandate particular substantive results, but "merely prohibits uninformed—rather than unwise — agency action." Id. at 351, 109 S. Ct. 1835; see also Pub. Citizen, 541 U.S. at 756-57, 124 S. Ct. 2204 ("NEPA imposes only procedural requirements"; not "particular results"). "By so focusing agency attention, NEPA ensures that the agency will not act on incomplete information, only to regret its decision after it is too late to correct. Similarly, the broad dissemination of information mandated by NEPA permits the public and other government agencies to react to the effects of a proposed action at *602 a meaningful time." Marsh, 490 U.S. at 371, 109 S. Ct. 1851 (citing Robertson, 490 U.S. at 349, 109 S. Ct. 1835).

"At the heart of NEPA is a requirement" that for every "major Federal action[] significantly affecting the quality of the human environment," the agency involved must prepare "a detailed" environmental impact statement. Pub. Citizen, 541 U.S. at 757, 124 S. Ct. 2204 (quoting 42 U.S.C. § 4332(2)(C)). This statement must assess, inter alia, (1) "alternatives to the proposed action," and (2) "the environmental impact of the proposed action." Id.; see also Shenandoah Valley Network v. Capka, 669 F.3d 194, 196 (4th Cir.2012); Nat'l Audubon Soc'y v. Dep't of the Navy, 422 F.3d 174, 185 (4th Cir.2005).

An agency's assessment of alternatives to the proposed action "sharply defin[es] the issues and provid[es] a clear basis for choice among options by the decisionmaker and the public." 40 C.F.R. § 1502.14. Agencies must "[r]igorously explore and objectively evaluate all reasonable alternatives." Id. § 1502.14(a). Although agencies have discretion to identify the range of "reasonable" alternatives, they must "include the alternative of no action." Id. § 1502.14(c)-(d); see also Theodore Roosevelt Conserv. P'ship v. Salazar, 661 F.3d 66, 72 (D.C.Cir.2011).

An agency's assessment of environmental impacts, in turn, is the "scientific and analytic basis for the comparison[ ]" of alternatives. 40 C.F.R. § 1502.16. As part of this analysis, agencies must measure the indirect and cumulative environmental effects of proposed actions. See id. § 1502.16(a)-(b). Conclusory statements that the indirect and cumulative effects will be minimal or that such effects are inevitable are insufficient under NEPA. See Ctr. for Biological Diversity v. U.S. Dep't of Interior, 623 F.3d 633, 642-43 (9th Cir.2010); Davis v. Mineta, 302 F.3d 1104, 1122-23 (10th Cir.2002); see also State Farm, 463 U.S. at 43, 103 S. Ct. 2856.

III.

In this case, there is no dispute that the Monroe Connector constitutes a "major Federal action" under NEPA, which requires an environmental impact statement. The Conservation Groups argue that the Agencies failed to assess alternatives or measure the indirect and cumulative effects of the Monroe Connector because of a fundamental inaccuracy in the "no build" baseline. Specifically, the Conservation Groups maintain that, in calculating the "no-build" baseline, the Agencies relied on data that assumed that the Monroe Connector existed. By doing so, the Agencies assertedly conflated the "no build" and "build" scenarios, making it impossible to accurately isolate and assess the environmental impacts of the Monroe Connector. Moreover, the Conservation Groups claim that the Agencies compounded this error by refusing to acknowledge it to the public, and instead, maintaining throughout the administrative process that the "no build" data did not assume the existence of the Monroe Connector.

The Agencies concede much of the Conservation Groups' argument. At the district court, the Agencies acknowledged that MUMPO's data assumed the existence of the Monroe Connector. See N.C. Wildlife Fed'n, 2011 WL 5042075, at *4, *9. Before us, the Agencies also conceded that this fact came to their attention during the administrative process. See Oral Argument at 17:32, 18:45, 26:50, N.C. Wildlife Fed'n v. N.C. Dep't of Transp., 677 F.3d 596 (4th Cir. May 3, 2012) (No. 11-2210), available at http://coop.ca4. uscourts.gov/OAarchive/mp3/11-2210-20120321.mp3. The Agencies additionally *603 admitted that they publicly (and erroneously) denied this fact throughout the administrative process. See Oral Argument at 30:20, 35:18, 37:07, 38:47; N.C. Wildlife Fed'n, 2011 WL 5042075, at *5. Nonetheless, the Agencies maintain that because they "conducted a thorough analysis of the environmental impacts" of the Monroe Connector and "accepted comments from the public," we should defer to their expertise. Appellees' Br. at 29.

What the Agencies would have us ignore is that NEPA procedures emphasize clarity and transparency of process over particular substantive outcomes. See Pub. Citizen, 541 U.S. at 756-57, 124 S. Ct. 2204; Robertson, 490 U.S. at 350-51, 109 S. Ct. 1835; see also Or. Natural Desert Ass'n v. Bureau of Land Mgmt., 625 F.3d 1092, 1121 n. 24 (9th Cir.2010) ("Clarity is at a premium in NEPA because the statute... is a democratic decisionmaking tool. ..."). Accordingly, agencies violate NEPA when they fail to disclose that their analysis contains incomplete information. See N.M. ex rel. Richardson v. Bureau of Land Mgmt., 565 F.3d 683, 708 (10th Cir. 2009); Native Ecosystems Council v. U.S. Forest Serv., 418 F.3d 953, 964 (9th Cir. 2005); Sierra Club v. U.S. Army Corps of Eng'rs, 701 F.2d 1011, 1030 (2d Cir.1983); see also State Farm, 463 U.S. at 43, 103 S. Ct. 2856 (holding that an agency acts arbitrarily and capriciously when it fails to "examine the relevant data and articulate a satisfactory explanation for its action including a rational connection between the facts found and the choice made") (internal quotation marks omitted). Such required "up-front disclosures [include] relevant shortcomings in the data or models." Lands Council v. Powell, 395 F.3d 1019, 1032 (9th Cir.2005); see 40 C.F.R. § 1502.22 (An agency "shall make clear" if there is "incomplete or unavailable information" in an environmental impact statement.). Here, the Agencies not only failed to disclose the assumptions underlying MUMPO's data, but provided the public with erroneous information.

The very purpose of public issuance of an environmental impact statement is to "provid[e] a springboard for public comment." Pub. Citizen, 541 U.S. at 768, 124 S. Ct. 2204 (alteration in original). In this case, however, the Agencies' responses to the public comments contravened that purpose. In commenting, the Fish and Wildlife Service and a number of private parties, including the Conservation Groups, repeatedly raised questions regarding the "no build" baseline. But, rather than take these opportunities to make a "candid acknowledgment" of what they knew to be the truth, Nat'l Audubon Soc'y, 422 F.3d at 185, the Agencies maintained that the "no build" data did not include the Monroe Connector.

This mischaracterization related to a critical aspect of the NEPA process—the accuracy of the "no build" baseline. NEPA requires that an agency's alternatives analysis include a "no build" alternative. 40 C.F.R. § 1502.14(d). "Without [accurate baseline] data, an agency cannot carefully consider information about significant environment impacts ... resulting in an arbitrary and capricious decision." See N. Plains Res. Council, Inc. v. Surface Transp. Bd., 668 F.3d 1067, 1085 (9th Cir. 2011). Accordingly, courts not infrequently find NEPA violations when an agency miscalculates the "no build" baseline or when the baseline assumes the existence of a proposed project. See, e.g., Friends of Yosemite Valley v. Kempthorne, 520 F.3d 1024, 1037-38 (9th Cir.2008);^[2]N.C. Alliance for Transp. Reform, Inc. v. U.S. Dep't of Transp., 151 F. Supp. 2d 661, 690 (M.D.N.C.2001).

*604 In an attempt to avoid this outcome, the Agencies contend that although they provided the Fish and Wildlife Service, the Conservation Groups, and the public incorrect information regarding the "no build" data, their subsequent admissions during this litigation cured these missteps. They cite no support for this proposition, and we have found none. To accept the Agencies' argument would amount to acceptance of "post hoc rationalizations for agency action," State Farm, 463 U.S. at 50, 103 S. Ct. 2856, rather than "judg[ing] the propriety of [the] action solely by the grounds invoked by the agency." SEC v. Chenery Corp., 332 U.S. 194, 196, 67 S. Ct. 1575, 91 L. Ed. 1995 (1947).

This we cannot do. For "[i]t is well-established that an agency's action must be upheld, if at all, on the basis articulated by the agency itself." State Farm, 463 U.S. at 50, 103 S. Ct. 2856 (citing Chenery, 332 U.S. at 196, 67 S. Ct. 1575). The "basis articulated by the agency" is the administrative record, not subsequent litigation rationalizations. See O'Reilly v. U.S. Army Corps of Eng'rs, 477 F.3d 225, 238-39 (5th Cir.2007); Hall v. U.S. Envt'l Prot. Agency, 273 F.3d 1146, 1161 (9th Cir.2001). Our focus on the administrative record in this case is particularly appropriate given that NEPA emphasizes the importance of an open and public environmental assessment process. See Nat'l Audubon Soc'y, 422 F.3d at 184. NEPA "guarantees that the relevant information will be made available to the larger audience that may also play a role in both the decisionmaking process and the implementation of that decision." Robertson, 490 U.S. at 349, 109 S. Ct. 1835.^[3]

The Agencies now admit that the administrative record mischaracterizes the "no build" data. Such an acknowledgment made during litigation does not change the fact that the NEPA process itself relied on those mischaracterizations. For example, the Record of Decision—the "final agency action," Natural Desert, 625 F.3d at 1118, and conclusive statement of "what the [agency's] decision was," 40 C.F.R. § 1505.2(a)—incorrectly states that the "socioeconomic forecasts for the No Build Scenario did not include the Monroe Connector." Litigation admissions cannot change this.

When relevant information "is not available during the [impact statement] process and is not available to the public for comment[,]... the [impact statement] process cannot serve its larger informational role, *605 and the public is deprived of [its] opportunity to play a role in the decision-making process." N. Plains, 668 F.3d at 1085 (citing Robertson, 490 U.S. at 349, 109 S. Ct. 1835). Accordingly, we reject the Agencies' argument that their after-the-fact disclosures assuage the harms incurred during the NEPA process.^[4]

In sum, although we need not and do not decide whether NEPA permits the Agencies to use MUMPO's data in this case, we do hold that by doing so without disclosing the data's underlying assumptions and by falsely responding to public concerns, the Agencies failed to take the required "`hard look' at environmental consequences." Shenandoah Valley, 669 F.3d at 196. We therefore vacate the judgment of the district court and remand so that the Agencies and the public can fully (and publicly) evaluate the "no build" data.^[5]

IV.

For the reasons discussed above, we vacate the judgment of the district court and remand for further proceedings consistent with this opinion.

VACATED AND REMANDED

NOTES

[1] The Agencies eliminated the hybrid category after determining that the necessary improvements "would have a significant adverse impact on businesses and the economy of Union County." They eliminated the "improve U.S. 74" category after projecting that by 2035, daily traffic volume on an improved U.S. 74 would far exceed traffic volume on U.S. 74 and the Monroe Connector combined.

[2] Laguna Greenbelt v. U.S. Department of Transportation, 42 F.3d 517 (9th Cir. 1994), on which the Agencies heavily rely, not only predates Friends of Yosemite, but in fact provides no support for the Agencies' arguments. There, the court rejected the argument that relying on "local planning documents that assume[d] the existence of the tollroad" violated NEPA, but only because the "record show[ed] that 98.5% of all land in the project's `area of benefit' [wa]s already accounted for by either existing or committed land uses not contingent on construction of the [project]." Id. at 525. The case at hand is markedly different. The record here is devoid of any evidence establishing that the region is developmentally saturated such that a major toll road will have no appreciable environmental impact.

[3] The Agencies have steadfastly opposed the Conservation Groups' attempts to supplement the administrative record with contemporaneous emails between agency officials, assertedly evidencing that the Agencies acted in bad faith. In doing so, the Agencies have repeatedly argued that the certified administrative record provides the complete basis for judicial review. See Appellees' Br. at 55. Though we need not decide whether the district court abused its discretion in denying the Conservation Groups' motion to supplement the administrative record with these emails, we do note the incongruity of the Agencies' position: the Agencies contend that the Conservation Groups should not be permitted to add to the administrative record emails sent before the record closed, while at the same time urging us to give controlling weight to their litigation concessions, not made until well after the administrative record closed.

[4] The Agencies briefly raise two other arguments. First, they contend that through Baker they adequately responded to concerns about the propriety of the "no build" data. But in response to queries from public and private groups as to the propriety of the "no build" data, Baker undertook two patently inadequate steps. One concluded that the Monroe Connector was not in the "no build" baseline, the other did not focus at all on the potential error.

Second, the Agencies maintain that the Conservation Groups "overstate" the importance of the error involved in including the Monroe Connector in the "no build" baseline. The Agencies acknowledge that the error affected one of the factors in MUMPO's model—travel time to employment. In general, areas with reduced travel time to employment tend to experience greater population growth and development, which can substantially impact the local environment. But by including the Monroe Connector in the "no build" baseline's travel time to employment, the Agencies assumed that the new road would not decrease travel time to employment, and thus would not cause development with attendant environmental impacts. To minimize this mistake, the Agencies argue that the model incorporated other, untainted factors. In doing so, the Agencies ignore their own record evidence of the importance of travel time to employment. Not only did this factor receive disproportionate weight in the model, but the Agencies repeatedly noted that decreased travel time to employment often spurs development. Accordingly, the administrative record does not demonstrate the irrelevance of travel time to employment. To assume this fact would overstep our limited scope of review. See Chenery, 332 U.S. at 196, 67 S. Ct. 1575.

[5] The Conservation Groups point to a number of other instances where the Agencies assertedly failed to comply with NEPA's requirements. See, e.g., 40 C.F.R. § 1502.24. We need not address these contentions because on remand, when the Agencies reevaluate the Impact Statement, they will have an opportunity to provide full public disclosure and all necessary explanations of their process.