Plaintiffs seek to enjoin defendants from proceeding with the construction of a highway project (the "Project") located within the cities of Draper, Sandy and South Jordan in Salt Lake County, Utah. Plaintiffs argue that defendants violated the National Environmental Protection Act (NEPA), 42 U.S.C. § 4332(C) and § 4(f) of the Department of Transportation Act, 49 U.S.C. § 303(c) when they prepared an inadequate Environmental Assessment ("EA") and then proceeded to issue a Finding of No Significant Impact ("FONSI") rather than an Environmental Impact Statement ("EIS"). The Project under review includes several components: the creation of a new freeway interchange at Interstate 15 and 11400 South; the construction of a new bridge over the Jordan River at 11400 South; and the widening and extension of existing 11400 South.1 The proposed expansion and widening of 11400 South will create a new segment of five-lane highway where no road currently exists. A portion of the expanded 11400 South will affect publicly-owned parkland and will require the displacement or demolition of several historic structures.

As currently envisioned, the Project will proceed in two phases. Phase I includes the construction of the 11400 South interchange at I-15 and I-15 improvements. Phase II involves the expansion and extension of 11400 South, including the construction of a new bridge across the Jordan River. While completion of Phase I appears to involve a time horizon of approximately three to four years, there is no definitive timetable established for the completion of Phase II.

The proponents of the Project are the Utah Department of Transportation ("UDOT"), and Draper, Sandy and South Jordan. Due to its potential effects on the environment, parkland and historic structures, the Project implicates both the NEPA, and § 4(f). The Federal Highway Administration ("FHWA") was the federal agency responsible for approving the Project and for preparing the environmental analysis required by NEPA. However, the EA here was initially prepared by Horrocks Engineers under contract with Sandy City and then reviewed and adopted by the FHWA. The Secretary of Transportation was responsible for approving the Project based on the § 4(f) analysis.2

Here the FHWA's combined EA and § 4(f) analysis (EA/4(f)) led to a FONSI. Thus, the FHWA concluded that no EIS was required.

We have identified the following deficiencies with defendants' methodology and conclusions in this case:

1. The EA/4(f)'s consideration of alternatives to the Project is inadequate. Serious consideration was given only to the preferred alternative (i.e., the Project as proposed) and a no-build alternative.

2. The EA/4(f) fails to consider adequately the Project's impacts including cumulative impacts.

3. The EA/4(f) fails to address adequately issues relating to phasing of the Project, particularly given the many-year time frame between the beginning and prospective completion dates for the Project.

4. The § 4(f) analysis failed to satisfy the high burden imposed on projects that make use of a public park and/or historic sites.

5. The EA/4(f) is fatally flawed by its use of vague, unsupported conclusions and inadequate, incomplete analysis.

This court reviews a district court's grant or denial of a preliminary injunction for an abuse of discretion. Prairie Band of Potawatomi Indians v. Pierce, 253 F.3d 1234, 1243 (10th Cir.2001); Kikumura v. Hurley, 242 F.3d 950, 955 (10th Cir.2001). A district court abuses its discretion where it commits a legal error or relies on clearly erroneous factual findings, Prairie Band, 253 F.3d at 1242, or where there is no rational basis in the evidence for its ruling. Utah Licensed Beverage Ass'n v. Leavitt, 256 F.3d 1061, 1065 (10th Cir.2001). We examine the district court's underlying factual findings for clear error, and its legal determinations de novo. Water Keeper Alliance v. United States Dep't of Def., 271 F.3d 21, 30 (1st Cir.2001); see also United States v. Power Eng'g Co., 191 F.3d 1224, 1230 (10th Cir. 1999).

In order to receive a preliminary injunction, the plaintiff must establish the following factors: "(1) a substantial likelihood of prevailing on the merits; (2) irreparable harm unless the injunction is issued; (3) [that] the threatened injury outweighs the harm that the preliminary injunction may cause the opposing party; and (4) [that] the injunction, if issued, will not adversely affect the public interest." Fed. Lands Legal Consortium ex rel. Robart Estate v. United States, 195 F.3d 1190, 1194 (10th Cir.1999). If the plaintiff can establish that the latter three requirements tip strongly in his favor, the test is modified, and the plaintiff may meet the requirement for showing success on the merits by showing "that questions going to the merits are so serious, substantial, difficult, and doubtful as to make the issue ripe for litigation and deserving of more deliberate investigation." Id. at 1195.

Our analysis of likelihood of success on the merits necessarily implicates defendants' compliance with NEPA and § 4(f). At this point, a second layer of review comes into play, because defendants' agency actions are themselves examined under a highly deferential, "arbitrary and capricious" standard. This standard is rooted in the very nature of administrative review.

Judicial review of agency NEPA and § 4(f) decisions is made available through the Administrative Procedure Act ("APA"), 5 U.S.C. § 551 et seq. See All Indian Pueblo Council v. United States, 975 F.2d 1437, 1443 (10th Cir.1992) (NEPA); Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 410, 91 S. Ct. 814, 28 L. Ed. 2d 136 (1971) (§ 4(f)). The APA sets forth a number of standards to be employed by courts reviewing administrative action. See 5 U.S.C. § 706(2). It has long been settled that an agency's decision to issue a FONSI is reviewed under the APA's deferential "arbitrary and capricious" standard. Id. § 706(2)(A); see, e.g., Utah Shared Access Alliance v. United States Forest Serv., 288 F.3d 1205, 1213 (10th Cir.2002); Comm. to Preserve Boomer Lake v. Dep't of Transp., 4 F.3d 1543, 1555 (10th Cir.1993). Although our standard of review of the § 4(f) analysis is somewhat more complex in this case, it generally implicates the arbitrary and capricious standard as well. Overton Park, 401 U.S. at 416, 91 S. Ct. 814.

Under NEPA regulations, an agency undertaking an action is required to determine whether its proposal is one that normally requires or normally does not require an EIS. 40 C.F.R. § 1501.4(a). If the answer to this question is not clear-cut, the agency should prepare an EA. Id. § 1501.4(b).3 If the agency determines, based on the EA, that no EIS is needed because its action would not significantly affect the environment, it may then prepare a FONSI. Id. §§ 1501.4(e), 1508.13. Otherwise, it must prepare an EIS.

When we review an EA/FONSI to determine whether an EIS should have been prepared, we must determine whether the agency acted arbitrarily and capriciously in concluding that the proposed action "will not have a significant effect on the human environment." Thus, our review of an EA/FONSI has a substantive component as well as a component of determining whether the agency followed procedural prerequisites. If the plaintiffs can demonstrate substantively that the FHWA's conclusion of non-significant effect on the environment represents a "clear error of judgment," then that conclusion must be reversed. Utah Shared Access Alliance, 288 F.3d at 1213.

As will be seen, here plaintiffs have demonstrated both procedural deficiencies in the EA as well as demonstrating that the Project will have a significant impact on the environment. Defendants propose to construct a five-lane highway through a park where no road currently exists. This five-lane highway will bisect the park, require the construction of a new bridge across the Jordan River, require the demolition or removal of several historic structures, and triple noise levels in portions of the park. By the year 2020, it is expected to carry at least 34,000 cars per day.

Furthermore, the record establishes here that the defendants prejudged the NEPA issues. This prejudgment diminishes the deference owed to the federal defendants in our review of their decision to issue a FONSI rather than an EIS. The addendum to the Engineering Services Agreement between Sandy City and Horrocks Engineers (Horrocks), a consultant on the Project, required Horrocks to distribute an EA no later than October 25, 1999, and further stipulated that a FONSI be signed and distributed by FHWA by February 25, 2000. In other words, Horrocks, the consultant employed to prepare the draft EA/4(f) for FHWA's review, was contractually obligated to prepare a FONSI and to have it approved, signed and distributed by FHWA by a date certain. The decision whether to prepare a FONSI should be based on the EA, of course, not the other way around. See 40 C.F.R. § 1501.4(c), (e). Horrocks thus had an inherent, contractually-created bias in favor of issuance of a FONSI rather than preparation of an EIS. See, e.g., Metcalf v. Daley, 214 F.3d 1135, 1143-44 (9th Cir. 2000).

We discern two ways in which this prejudgment is attributed to the federal decisionmakers in this case, who were not themselves parties to the Engineering Services Agreement. First, the district court found, and the record establishes, that "FHWA was involved throughout the NEPA process," Davis, 148 F.Supp.2d at 1218, thus implicating FHWA directly in Horrocks' "rush to judgment" on the environmental issues. The minutes of a meeting held October 12, 1999, between representatives of Sandy City, Horrocks and the federal defendants, indicate that "FHWA wants to advance the environmental documents and not issue a FONSI until they are checked against the new WFRC model."4 Significantly, this meeting occurred before the EA had been submitted to FHWA for formal review and approval. Additionally, the record contains a memo to UDOT's project manager dated August 7, 2000, including FHWA's recommendation to proceed with public review of the revised EA/4(f) document, subject to certain revisions. FHWA wanted the FONSI, which had already been prepared, removed "from the document to be circulated for public review." Id. This suggests that a FONSI had already been prepared prior to an evidently pro forma public opportunity to comment on the revised EA. FHWA's own regulations suggest that a FONSI not be prepared until FHWA has reviewed the public's comments concerning the EA. 23 C.F.R. § 771.121(a).

Second, the record establishes that the FHWA failed to conduct a sufficient independent review of Horrocks' work to insulate itself from the biases toward a FONSI reflected in Horrocks' draft EA. The district court concluded that FHWA made an independent investigation before issuing a FONSI. Its conclusion rests on two facts: (1) that FHWA was involved throughout the NEPA process, and (2) that it rejected the initial EA/4(f) document prepared by Horrocks. Davis, 148 F.Supp.2d at 1218.

However, as will be seen, the EA/4(f) prepared in this case is insufficient in many particulars. Although FHWA rejected an initial draft of the EA/4(f), this did not result in the correction of the many problems plaintiffs raise with the FONSI. This is true even though many of these problems were brought to FHWA's attention. Defendants employed a law firm to assist with draft responses to comments received concerning the EA/4(f) documentation. The law firm wrote a memorandum to defendants which is highly critical of the EA/4(f), echoing many of the complaints plaintiffs have raised in this appeal. For example, it notes that "[t]he Project EA addresses alternatives to the Project in a conclusory manner, and addresses only the preferred alternative and the no project alternative as alternatives subject to full analysis." It further notes that the EA does not adequately address phasing issues, in spite of frequent public comments on the issue. The memo warns that the failure to discuss alternatives adequately "appears to pose an appreciable litigation risk." The FHWA received a copy of the very critical evaluation of the EA and even agreed with many of the comments. Yet, in spite of this dire warning and critical appraisal of the deficiencies of the EA, FHWA did not fix the problems identified with the EA.

On this record it is apparent that FHWA failed to take sufficient steps to insulate the final resulting documentation from Horrocks' tainted analysis. See Assocs. Working for Aurora's Res. Env't v. Colo. Dept. of Transp., 153 F.3d 1122, 1129 (10th Cir.1998) (discussing the kind of independent review that might insulate the reviewing agency from the biases of the preparer of the underlying EA).

With regard to our review under § 4(f) of the Department of Transportation Act, there are some special rules pertaining to the standards we apply. That is because § 4(f) conditions approval of a transportation project through a park upon a finding 1) that there is no prudent and feasible alternative to using the land and 2) that the project includes all possible planning to minimize harm to the park. In the first step, the reviewing court must decide whether the Secretary of Transportation acted within the scope of his authority. Boomer Lake, 4 F.3d at 1549. This inquiry requires the district court to examine whether "the Secretary could have reasonably believed that in this case there are no feasible alternatives or that alternatives do involve unique problems." Id. (quotation omitted). The review of this issue centers on whether "there is no prudent and feasible alternative" to construction of the project over land protected by § 4(f). 49 U.S.C. § 303(c). Section 4(f) "requires the problems encountered by proposed alternatives to be `truly unusual' or [to] `reach extraordinary magnitudes' if parkland is taken." Boomer Lake, 4 F.3d at 1550 (quotation omitted).

In the second step, the reviewing court examines whether the Secretary's decision was arbitrary, capricious, an abuse of discretion or otherwise not in accordance with law. Id. at 1549. Here, the court asks whether the Secretary's decision was based on a consideration of the relevant factors and whether the Secretary made a clear error of judgment. Id. In assessing this factor, the reviewing court must conduct a careful and searching inquiry into the facts; however, once it becomes satisfied that the Secretary took a "hard look" at the relevant factors, the court should not substitute its judgment for that of the agency. Id. at 1551.

Finally, the reviewing court asks whether the Secretary followed the necessary procedural requirements in reaching his decision. Id. at 1549. This court reviews the district court's conclusions on all three of these issues de novo, according no deference to the district court's conclusions. Id.

II. Preliminary Injunction Factors of Harm to Plaintiffs; Balancing of Harm to Defendants; and Public Interest.

Having set out in some detail the standard by which we review both the decision of the district court and the underlying agency decisions, we move now to the merits of this appeal. As mentioned earlier, we examine four factors to determine whether plaintiffs have demonstrated their entitlement to a preliminary injunction: harm to the plaintiffs; harm to the defendants; the public interest; and plaintiffs' likelihood of success on the merits.

The district court concluded that plaintiffs had failed to demonstrate that construction of the Project would cause irreparable harm to their environmental interests. Davis, 148 F.Supp.2d at 1221. The district court relied on the insufficiency of plaintiffs' specific, tangible assertions of harm and the unlikelihood of their success on the procedural aspects of their claim.

In mandating compliance with NEPA's procedural requirements as a means of safeguarding against environmental harms, Congress has presumptively determined that the failure to comply with NEPA has detrimental consequences for the environment.5 See 42 U.S.C. § 4321 (congressional declaration of purpose); cf. Sierra Club v. Marsh, 872 F.2d 497, 500 (1st Cir.1989) (stating harm NEPA seeks to prevent is complete when agency makes decision without considering information NEPA seeks to place before decision-maker and public); Comm. to Save the Rio Hondo v. Lucero, 102 F.3d 445, 448-49 (10th Cir.1996) ("The injury of an increased risk of harm due to an agency's uninformed decision is precisely the type of injury the National Environmental Policy Act was designed to prevent."). For this reason, we hold that harm to the environment may be presumed when an agency fails to comply with the required NEPA procedure.6

Plaintiffs must still make a specific showing that the environmental harm results in irreparable injury to their specific environmental interests. They have done so here. Plaintiffs' property will be directly impacted by this Project and the size and scope of this Project supports a conclusion that the injury is significant. Amoco Prod. Co. v. Village of Gambell, 480 U.S. 531, 544, 107 S. Ct. 1396, 94 L. Ed. 2d 542 (1987). See Marsh, 872 F.2d at 504 ("The difficulty of stopping a bureaucratic steam roller, once started ... seems to us ... a perfectly proper factor for a district court to take into account ... on a motion for preliminary injunction.").7 Thus, plaintiffs have established irreparable harm under NEPA.

The substantive harm contemplated by § 4(f) is the actual harm to parkland or historic sites that will occur if the Secretary of Transportation does not (1) consider every prudent and feasible alternative to using the land, and (2) make all possible plans to minimize the harm, if use is required. Plaintiffs have shown adequate proximity to and use of the parkland sufficient to establish injury flowing from degradation of the parkland. The proposed extension of 11400 South will make use of public parkland and historic sites thus causing the sort of harm that is protected against by § 4(f).

In Amoco Production Co., 480 U.S. at 544, 107 S. Ct. 1396 (1987), the Supreme Court gave some guidance to evaluating harm connected with violations of substantive environmental statutes. It stated that substantive "[e]nvironmental injury, by its nature, can seldom be adequately remedied by money damages and is often permanent or at least of long duration, i.e., irreparable." Amoco, 480 U.S. at 545, 107 S. Ct. 1396. That is certainly the case here. The § 4(f) statement concedes that construction of the Project will impair the aesthetic attributes associated with the Jordan River Parkway and "will disrupt the natural setting and feeling of the park." Noise levels are expected to increase at least ten decibels in both Willow Creek Park and Jordan River Parkway and perhaps as much as twenty decibels at times in Jordan River Parkway. This harm is irreparable in the sense that it cannot adequately be remedied by nonequitable forms of relief. See Sierra Club v. Hodel, 848 F.2d 1068, 1097 (10th Cir. 1988), overruled on other grounds by Vill. of Los Ranchos de Albuquerque v. Marsh, 956 F.2d 970, 972 (10th Cir.1992).8

Plaintiffs' property will be directly impacted by these environmental harms, and accordingly plaintiffs have also established irreparable harm under § 4(f) of the Department of Transportation Act.

We must next balance the irreparable harms we have identified against the harm to defendants if the preliminary injunction is granted. Defendants allege that significant financial penalties will be incurred by UDOT if the Project is delayed. Defendants rely on the affidavit of Angelo Papastamos, UDOT's Senior Project Manager. This affidavit describes costs that will be incurred by UDOT based on the delays experienced and anticipated by this litigation. However, it appears that many of these costs may be self-inflicted. As we have previously concluded, the state entities involved in this case have "jumped the gun" on the environmental issues by entering into contractual obligations that anticipated a pro forma result. In this sense, the state defendants are largely responsible for their own harm. Cf. Pappan Enters., Inc. v. Hardee's Food Sys., Inc., 143 F.3d 800, 806 (3d Cir.1998) (stating, in context of franchise contract dispute, that "[t]he self-inflicted nature of [defendant's] harm ... weighs in favor of granting preliminary injunctive relief").

On this record, it appears that the environmental harms from proceeding with the Project without adequate NEPA review outweigh the legitimately incurred costs to defendants resulting from an injunction.

Conflicting public interest values are involved here. Obviously, the public interest does not favor payment of construction-related penalties from the public fisc, and it does favor the construction of much-needed highways. On the other hand, plaintiffs assert a strong public interest in NEPA compliance.

This is not a case in which "any adverse effect upon the environment would appear to be negligible as compared with the obvious need for correcting a dangerous highway intersection." Pub. Interest Research Group of Mich. v. Brinegar, 517 F.2d 917, 918 (6th Cir.1975). First, the environmental dangers at stake in this case are serious, particularly where the effect on parkland and historic structures protected by § 4(f) is concerned. Second, the proposed highway construction has not yet begun, and so we are not confronted with equities in favor of completion of a partially-completed project.

We conclude that the public interest associated with completion of the Project must yield to the obligation to construct the Project in compliance with the relevant environmental laws. On balance, the public interest favors compliance with NEPA and § 4(f).

If plaintiffs have made a "strong showing" on the last three factors, the showing on the merits of their claim could be relaxed. While plaintiffs have satisfied the last three requirements for obtaining a preliminary injunction, we do not believe that this showing is sufficiently strong to entitle them to rely on the relaxed standard on the merits. Plaintiffs must therefore satisfy the traditional first criterion in full, by showing that they are likely to succeed on the merits of their claims.

We now arrive at the heart of our analysis. We must consider whether plaintiffs have shown that they are likely to succeed on the merits of their NEPA and § 4(f) claims. We begin with an overview of each statute, before discussing the specific statutory problems at issue here.

We have already provided a basic overview of the NEPA process. As mentioned, the basic issue is whether the Project will have a significant effect on the human environment such that an EIS, rather than a FONSI, should have been prepared. 42 U.S.C. § 4332(2)(C); 40 C.F.R. § 1508.13. Plaintiffs have the burden of showing that FHWA's decision not to prepare an EIS was arbitrary and capricious. Boomer Lake, 4 F.3d at 1555.

In this case, FHWA's own regulations implementing NEPA assist plaintiffs in meeting their burden of proof. The Council on Environmental Quality (CEQ) regulations require agencies to determine whether a project is one that "normally requires" an EIS. 40 C.F.R. § 1501.4(a)(1). Pursuant to this requirement, FHWA has adopted a regulation providing examples of actions that normally require an EIS. One of the actions that the FHWA determined "normally required an EIS" is "[a] highway project of four or more lanes on a new location." 23 C.F.R. § 771.115(a)(2). That, of course, precisely describes a portion of the Project: the expansion of 11400 South from 700 West to 1300 West.

We have held, it is true, that regulations of this type do not inevitably create a mandatory duty to prepare an EIS. Boomer Lake, 4 F.3d at 1554-55. The fact that there is no mandatory duty, however, does not mean that this regulation has no effect at all. This FHWA regulation presumes that an EIS will normally be prepared for highway projects of four or more lanes on a new location, thereby imposing on the FHWA the burden of establishing why that presumption should not apply in this particular case. If FHWA arbitrarily and capriciously failed to follow its own regulation, its decision must be reversed.

The Secretary [of Transportation] may approve a transportation program or project ... requiring the use of publicly owned land of a public park, recreation area, or wildlife and waterfowl refuge of national, State, or local significance, or land of an historic site of national, State, or local significance (as determined by the Federal, State, or local officials having jurisdiction over the park, area, refuge, or site) only if —

(1) there is no prudent and feasible alternative to using that land; and

(2) the program or project includes all possible planning to minimize harm to the park, recreation area, wildlife and waterfowl refuge, or historic site resulting from the use.

49 U.S.C. § 303(c). There are two types of protected properties at issue here: parkland and historic sites.

Two publicly owned parks are planned within the APE: the Jordan River Parkway and Willow Creek Park. The Jordan River Parkway is currently owned partly by two private landowners and partly by the Utah Department of Natural Resources, Division of Parks and Recreation. The Willow Creek Park is included in the Draper City Master Plan, Id. at 7800, but it is currently owned by a private landowner in the vicinity of the proposed action. Id.

In order to qualify as § 4(f) property, parkland must be "publicly owned." Under these facts, Willow Creek Park does not qualify, in that it is owned by a private landowner. That portion of Jordan River Parkway which is owned by a private landowner is similarly disqualified. The fact that these properties may eventually be transferred to public ownership does not qualify them as § 4(f) properties. See Nat'l Wildlife Fed'n v. Coleman, 529 F.2d 359, 370 (5th Cir.1976). The remaining portion of Jordan River Parkway, however, currently owned by the State of Utah and designated as a public park, does qualify as § 4(f) property.9

Plaintiffs assert that the Project will affect the publicly-owned portion of the Jordan River Parkway. Our review of the record indicates that a part of the publicly-owned portion of this parkland will be directly impacted and likely even directly taken by the extension of 11400 South across the Jordan River. The FHWA believed the direct and cumulative impacts on publicly-owned parkland were sufficiently serious to apply a § 4(f) analysis to the park, and we will follow its lead on this point.

Historic structures qualify for § 4(f) consideration if they are either on the National Register of Historic Places, or eligible for inclusion on the National Register. 23 C.F.R. § 771.135(e). There are 29 historic properties located within the 11400 South APE; however, four of these are not recommended for inclusion on the National Register. The remaining 25 properties are subject to § 4(f) analysis.

Once the property is determined to be § 4(f) property, the Secretary of Transportation can approve the use of such publicly-owned parkland or historic sites for highway construction only if "there is no prudent and feasible alternative to using that land," 49 U.S.C. § 303(c)(1), and only after the Secretary has performed "all possible planning" in the project "to minimize harm to the park ... or historic site resulting from the use." Id. § 303(c)(2).

C. Specific problems with defendants' 4(f) and EA analysis under NEPA

The EA/4(f) prepared in this case is inadequate for three reasons. First, it fails to consider reasonable alternatives to the Project. Second, it contains an inadequate discussion of the Project's impacts. Finally, the treatment of the issues is too vague, incomplete and inadequate to allow the decision-maker to decide if an EIS is required.

Plaintiffs argue that the EA/4(f) defines the scope of the Project so narrowly as to eliminate reasonable alternatives. Plaintiffs had argued for consideration of alternatives that would avoid altogether the construction of an additional crossing over the Jordan River at 11400 South. The district court, discussing this issue as it relates to the § 4(f) analysis, stated:

Because the articulated purpose and need for the Project is, in part, to provide another crossing point of the Jordan River, such an alternative [avoiding another Jordan River crossing] would be per se unfeasible and imprudent because any alternative addressing the purpose and needs articulated for the Project would necessarily have to cross the Jordan River Parkway.

Davis, 148 F.Supp.2d at 1219.

While it is true that defendants could reject alternatives that did not meet the purpose and need of the project, Boomer Lake, 4 F.3d at 1550, they could not define the project so narrowly that it foreclosed a reasonable consideration of alternatives. Colo. Envtl. Coalition v. Dombeck, 185 F.3d 1162, 1174-75 (10th Cir.1999); Simmons v. United States Army Corps of Eng'rs, 120 F.3d 664, 669 (7th Cir.1997). Although the scope of the Project certainly contemplates additional road capacity across the Jordan River, we do not believe that a fair reading of the Project purposes and needs requires that this additional capacity necessarily be achieved by extending 11400 South across the Jordan River.

The EA/4(f) lists the following purposes for the Project:

"Maintain I-15 mainline capacity, decrease existing and future congestion on I-15, and improve the level of service at adjacent interchanges by improving interchange spacing and eliminating the existing weave maneuver between the 12300 South interchange and the State Street off-ramp."

"Improve the functionality of the 11400 South corridor as an important local and regional travel corridor as identified in the local and regional master plans of Sandy, Draper, South Jordan, WFRC, and UDOT."

"Enhance access and mobility throughout the project area so that sufficient access opportunities are available for all properties and developments."

"Help accommodate the regional traffic demand for east-west travel across the southern end of the Salt Lake Valley."

"Improve the geometry and operation of the 11400 South/State Street intersection so that the efficiency of the intersection is increased, traffic flow is improved, travel time and other delays are decreased, and the frequency and lengths of backups is reduced."

Further, if the Project did narrowly express its purposes and needs as requiring a new crossing across the Jordan River at 11400 South, we would conclude that such a narrow definition of Project needs would violate NEPA given the more general overarching objective of improving traffic flow in the area. In Simmons v. United States Army Corps of Eng'rs, 120 F.3d 664, 669 (7th Cir.1997), the Seventh Circuit observed that:

the `purpose' of a project is a slippery concept, susceptible of no hard-and-fast definitions. One obvious way for an agency to slip past the structures of NEPA is to contrive a purpose so slender as to define competing `reasonable alternatives' out of consideration (and even out of existence). The federal courts cannot condone an agency's frustration of Congressional will.

Id. at 666. The record here suggests potentially viable alternatives of expanding east-west traffic capacity across the Jordan River at 9800 South, 10600 South, and 12300 South. Thus, if the purposes and needs of the Project were so narrowly construed as to mandate the extra capacity only at 11400 South, we would conclude that such a narrow definition would be contrary to the mandates of NEPA.

A properly-drafted EA must include a discussion of appropriate alternatives to the proposed project. 42 U.S.C. § 4332(2)(E); 40 C.F.R. § 1508.9(b). "An agency decision concerning which alternatives to consider is necessarily bound by a rule of reason and practicality." Airport Neighbors Alliance, Inc. v. United States, 90 F.3d 426, 432 (10th Cir.1996) (quotation omitted).

With even more force, § 4(f) requires the Secretary of Transportation to consider all "prudent and feasible alternatives." 49 U.S.C. § 303(c)(1). "Whether an alternative is `prudent' ... involves a common sense balancing of practical concerns, but § 4(f) requires the problems encountered by proposed alternatives to be truly unusual or to reach extraordinary magnitudes if parkland is taken. Nevertheless, an alternative that does not solve existing or future traffic problems ... may properly be rejected as imprudent." Assocs. Working for Aurora's Res. Env't, 153 F.3d at 1131 (quotation and citations omitted).

The § 4(f) document studied only two primary alternatives: a "no build" alternative, and the proposed alternative. The document concluded that the "no build" alternative was the only prudent alternative that would not impact § 4(f) properties within the APE. Naturally, the "no build" alternative was then rejected because it did not meet the purpose and need for the Project. Defendants summarily rejected, without a hard look, secondary avoidance alternatives such as "minor alignment shifts, a reduced typical section, and retaining structures." The EA similarly rejected without a hard look the Transportation System Management (TSM), mass transit, and various build alternatives, simply concluding that each, by itself, would not meet the purpose and need of the Project or was otherwise unfeasible. Defendants were then left with only two alternatives: the "Proposed Action Alternative" (the Project) and a no build alternative. Again, the no build alternative was rejected in favor of the proposed alternative.

Thus, in both the EA and § 4(f), only two alternatives were given serious consideration: build the Project as conceptualized, or adopt a "no build" approach. This summary treatment of alternatives must be measured against the standards in 42 U.S.C. § 4332(2)(E) and 40 C.F.R. § 1508.9(b) (requiring the agency to study, develop and discuss appropriate alternatives and to briefly describe those alternatives). Plaintiffs contend that this analysis was deficient for several reasons:

a. Additional Jordan River crossing

Plaintiffs argue that defendants should have given more detailed consideration to alternatives involving the extension or expansion of other roads that cross the Jordan River. In particular, they claim the crossings at 12300 and 10600 could be expanded as an alternative to constructing a new crossing at 11400 South. The EA summarily rejected various piecemeal options involving added lanes to other highways in the area. Each option was rejected because standing alone it would not meet the purpose and need of the Project.

FHWA relies on several traffic studies that purport to show that reasonable traffic volume can only be achieved by creating at least four lanes on 11400 South. Those studies, however, provide an insufficient basis for failing to consider additional lane configurations at existing crossings or other locations as a means of meeting the project's purpose and need.

Those studies, and particularly the Fehr & Peers report to USDOT dated September 7, 1999, show the obvious that construction of a four lane crossing over the Jordan River at 11400 South will increase traffic capacity in the area and will improve traffic flows. But that is not the test for whether alternatives should be studied in an EA. Tellingly, that Report also reveals that substantial traffic benefits could alternatively be obtained by expanding capacity over the Jordan River at 12300 South and 10600 South and by constructing a new crossing at 9800 South. Nothing in that report nor anywhere else in the record that we have seen justifies a conclusion that these alternatives are not practical, reasonable, and perhaps in some instances even preferable to a crossing at 11400 South. The Fehr & Peers report points out that these alternatives "must be further evaluated ... on [their] own merits" in order to make an adequate environmental appraisal of them, but it was precisely that further analysis that was eschewed in this EA.

Alternatives need not be studied if they are "remote, speculative ... impractical or ineffective." Airport Neighbors Alliance, Inc., 90 F.3d at 432 (quotation omitted). However, particularly in light of § 4(f)'s stringent mandates, we have seen nothing in the record to justify the failure of the EA to study the alternatives, separately or in combination, of expanded capacity over the Jordan River at 12300 South and 10600 South and a new crossing at 9800 South.

b. Alternative routing of 11400 South

The § 4(f) statement concludes that, once it is given that 11400 South will be extended across the Jordan River, there are no feasible and prudent alternative alignments of 11400 South that would avoid the Jordan River Parkway and Willow Creek Park altogether. However, those conclusions are so vague and unspecific as to be little more than platitudes. Defendants are required to use "all possible planning" to minimize the effect on parkland. 49 U.S.C. § 303(c)(2). The § 4(f) statement does not discuss whether the proposed alignment and bridge across the Jordan River have been placed at the least environmentally-sensitive location in the Jordan River Parkway, or whether an alignment shift north or south might minimize the effect of noise and other environmental effects on the Parkway.

As planned, the Project will also result in the complete use of five historic sites, with impacts on fifteen others. The § 4(f) document considers and rejects a number of minor alignment shifts in an effort to avoid these impacts. This discussion of alignment shifts, while site-specific, is qualitatively insufficient to address § 4(f) concerns. For example, certain alignment shifts are rejected in the § 4(f) evaluation because they may impact other structures; however, no discussion is presented of the comparative historic value of these other structures nor is there any quantitative comparison of the impact of various orientations of 11400 South even assuming it is expanded at some point over the Jordan River. This discussion of alignment alternatives is so vague and non-specific as to be essentially meaningless.

c. Cumulative alternatives

In addition to the possibility of expanding existing roads, options involving Transportation System Management (TSM) and mass transit were eliminated from study under the EA. The EA/4(f) rejected these options because, standing alone, they would not meet the purpose and need of the Project. However, no effort was made to consider TSM and mass transit together and/or in conjunction with alternative road expansion as a means of meeting Project goals.10 This represents one of the most egregious shortfalls of the EA. According to the various reports in the record, TSM could significantly contribute to traffic management in the area and mass transit in any number of iterations is apparently under active consideration in this area by a number of jurisdictions involved. Although mass transit in the area is apparently not a sure thing, neither is there anything in the record to establish that it is such a "remote, speculative, impractical or ineffective" alternative that it did not need to be studied as a viable alternative in the EA. There are no cost studies, cost/benefit analyses or other barriers advanced that would warrant a conclusion that mass transit alternatives are unreasonable, standing alone or in conjunction with other alternatives.

Again, because of the need of FHWA to engage in "all possible planning," the failure of the EA carefully to consider mass transportation, particularly in combination with other alternatives, is not justified. "[A]ll possible planning" requires consideration of whether reasonable resources currently planned for extension of 11400 South across the Jordan River might instead be effectively directed toward expansion of mass transit and other traffic management strategies that could adequately meet the needs and purpose of the Project in a way that avoids the adverse impact on parkland or historic structures.

d. Conclusion

The EA/4(f) does not contain an adequate discussion of alternatives. Many alternatives were improperly rejected because, standing alone, they did not meet the purpose and need of the Project. Cumulative options, however, were not given adequate study. Alternatives were dismissed in a conclusory and perfunctory manner that do not support a conclusion that it was unreasonable to consider them as viable alternatives in the EA. As a result, only two alternatives were studied in detail: the no build alternative, and the preferred alternative. FHWA acted arbitrarily and capriciously in approving an EA/4(f) that does not provide an adequate discussion of Project alternatives.

We conclude that the EA/FONSI discussion of the impacts expected to result from the Project is also inadequate.

a. Induced growth

The problem of induced growth arises fairly frequently in NEPA cases. See Daniel R. Mandelker, NEPA Law and Litigation § 8.07[5] (2d ed.2001). NEPA requires agencies to consider the growth-inducing effects of proposed actions. 40 C.F.R. § 1508.8(b); City of Carmel-by-the-Sea v. United States Dep't of Transp., 123 F.3d 1142, 1162 (9th Cir.1997). A conclusory statement that growth will increase with or without the project, or that development is inevitable, is insufficient; the agency must provide an adequate discussion of growth-inducing impacts. Laguna Greenbelt, Inc. v. United States Dep't of Transp., 42 F.3d 517, 526 (9th Cir.1994).

Here, the EA/4(f) finds that "[i]ndirect and cumulative impacts of the [Project] to growth will be minimal because they do not include the impacts of development, which will occur with or without the [Project]." Admin. Rec. at 7786 (emphasis added). The EA/4(f) acknowledges, however, that "the rate of development on lands east of the Jordan River may increase as a result of the project." Id.11 The district court accepted the agency's conclusion that the Project would not significantly increase the rate of growth.

However, in a letter dated September 19, 2000, commenting on the draft EA/4(f), the Environmental Protection Agency (EPA) opined that "[e]nhanced transportation facilities will generate or enhance economic activity and development," and that "related federal, state, and private actions" may result in significant environmental impact. The EPA concluded that "because all direct impacts may not have been identified and assessed ... [it] believes there is not a reasonable basis for a finding of no significant impact (FONSI) for the preferred alternative." It does not appear these concerns were ever adequately addressed in the EA. The EPA's viewpoint on this issue is undeniably relevant. While it is true that NEPA "requires agencies preparing environmental impact statements to consider and respond to the comments of other agencies, not to agree with them," Custer County Action Ass'n, 256 F.3d at 1038, it is also true that a reviewing court "may properly be skeptical as to whether an EIS's conclusions have a substantial basis in fact if the responsible agency has apparently ignored the conflicting views of other agencies having pertinent expertise." Sierra Club v. United States Army Corps of Eng'rs, 701 F.2d 1011, 1030 (2d Cir.1983).

Defendants did provide a graphic analysis of socio-economic growth in the area bounded by State Street, 10200 South, 12600 South and 4800 West and the Provo Reservoir Canal during the period from 1970 through the present, extrapolated through 2020. See id. at 7329. This table shows that continued growth in this area is anticipated through 2020, albeit at a lesser rate than has existed since 1990. The graph, however, contains no discussion or comparison of the local effects in the area directly impacted by this Project of induced growth caused by the extension of 11400 South as compared to a no-build alternative or the use of other alternatives.

Defendants also point to a map in the record that shows regional development density as of the year 2000. That map, however, merely confirms plaintiffs' claim that the 11400 South corridor remains in large part an island of open space in a sea of development. Defendants' refusal to study the possibility that the relatively unspoiled nature of this local area might be due, at least in part, to the present lack of a major roadway through it is arbitrary and capricious, and the district court abused its discretion in ignoring this factor in its analysis.

b. Phasing

Plaintiffs claim that the EA does not adequately discuss the significant impacts associated with phasing the Project. "Phasing" refers to environmental impacts that occur because a project will not be constructed all at once.12 The Project here is to be constructed in two phases: the I-15 interchange will be constructed before 11400 South is expanded and extended across the Jordan River. Plaintiffs identify two significant impacts from phasing the project: (1) for an extended period, persons living along the proposed expansion of 11400 South may suffer from pollution, noise and safety impacts as the result of living with a planned, but unconstructed, five-lane highway project made necessary by the first phase of the Project; and (2) it is possible that the second phase of the Project will never be completed because of permitting issues, resulting in many environmental problems caused by a major interchange that dumps traffic onto an unimproved two-lane road.

The district court found that FHWA and the Secretary of Transportation had taken a "hard look" into the phasing question and had determined that the impacts from phasing would not be significant. Davis, 148 F.Supp.2d at 1209 n. 3. This determination appears to rest on a single traffic study. This study deals only with traffic volumes on 700 West. It does not assess impacts to persons who live along 11400 South during the time period after the interchange has been constructed, but before the extension and expansion have been completed, a time period which may span a decade or more.13 The potentially significant impacts from phasing have not been adequately studied in the EA. See 40 C.F.R. § 1508.8(b).

c. Noise

The EA/4(f) discloses a significant, and in some cases dramatic, increase in noise levels in the vicinity of the Project. The average noise level in the Project area is currently approximately 57 dBA. This level is defined in the UDOT noise abatement policy as an appropriate noise level for "[l]ands on which serenity and quiet are of extraordinary significance and serve an important public need." In other words, the Project area right now is a serene island of quiet. The Jordan River Parkway is even more placid, with noise levels ranging from 45 to 50 dBA.

Once the Project is constructed, noise levels are expected to jump to over 65 dBA in an area affecting 119 residences, churches, parks and businesses.14 Sound levels in the Jordan River Parkway are expected to jump from ten to twenty decibels, to 55 to 70 dBA. Since each ten decibel increase is equivalent to a doubling of the noise volume, this means the noise level in portions of the park may quadruple.

The EA/4(f) lists a number of proposed mitigation measures that might be used to decrease the effect of the noise pollution.15 The district court found these measures persuasive on the question of significant impact. Davis, 148 F.Supp.2d at 1217. However, most of the measures considered were rejected in the EA/4(f), and the remaining, recommended measures are speculative without any basis for concluding they will occur.

Defendants rejected the option of slowing down traffic on the road, even through the parkland section, because the Project's purpose and need requires a speed of at least 40 to 45 mph. Horizontal and/or vertical alignment shifts were rejected as impractical, as were landscaping and berms. Noise walls are mentioned for part of the roadway, but merely as a recommendation: "a final decision on the installation of the abatement measures will be made upon completion of the project design and the public involvement process." It was noted in the EA/4(f) that the noise walls are quite costly ($850,000) and the defendants made no commitment to build them.16

Mitigation measures may be relied upon to make a finding of no significant impact only if they are imposed by statute or regulation, or submitted by an applicant or agency as part of the original proposal. As a general rule, the regulations contemplate that agencies should use a broad approach in defining significance and should not rely on the possibility of mitigation as an excuse to avoid the EIS requirement.

Forty Most Asked Questions Concerning CEQ's National Environmental Policy Act Regulations ("Forty Questions"), 46 Fed.Reg. 18,026, 18,038 (Council on Envtl. Quality 1981) (emphasis added).17 In short, the EA/4(f) makes no firm commitment to any noise mitigation measures. We disagree with the district court's conclusion that the increase in noise levels associated with the Project is not a significant impact.

d. Cumulative impacts

The CEQ regulations require agencies to discuss the cumulative impacts of a project as part of the environmental analysis. 40 C.F.R. § 1508.7. The EA does not provide an adequate discussion of the cumulative impacts of the Project on the human environment. These cumulative impacts may be significant. This five-lane highway project, of a type that normally requires an EIS, bisects two parks, requires the demolition or moving of numerous historic structures and will affect others, may quadruple noise levels in one of the parks, will increase traffic to 34,000 cars per day, and will require the construction of a new bridge over the Jordan River. On the record before us, it appears that the cumulative environmental impact of the Project is significant and FHWA's contrary conclusion represents a clear error of judgment; at the least, the EA as presently drafted fails to provide an adequate basis for a FONSI conclusion that the Project "will not have a significant effect on the human environment."18

Defendants contend that plaintiffs should be required to post a bond in order to obtain a preliminary injunction. The amount of a bond is a matter to be resolved by the district court on remand. Fed.R.Civ.P. 65(c). Ordinarily, where a party is seeking to vindicate the public interest served by NEPA, a minimal bond amount should be considered. See, e.g., Friends of the Earth, Inc. v. Brinegar, 518 F.2d 322, 322-23 (9th Cir.1975). Here, plaintiffs are private individuals rather than a public interest organization. Nevertheless, plaintiffs' strong showing on the merits and the defendants' apparent prejudgment to proceed prematurely with the Project before the required environmental studies were considered suggests that a large bond should not be required.

The order of the United States District Court for the District of Utah denying a preliminary injunction is REVERSED, and the case is REMANDED for entry of a preliminary injunction barring further road construction pending resolution of this case on the merits. NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE