United States Court of Appeals
                           FOR THE EIGHTH CIRCUIT
                                    ___________

                                    No. 98-3506
                                    No. 98-3774
                                    No. 98-3925
                                    ___________

City of Bridgeton; City of St. Charles;  *
St. Charles County,                      *
                                         *
       Petitioners,                      *
                                         *
       v.                                * Petitions for Review of the
                                         * Decision of the Federal
FAA; Rodney E. Slater, in his official   * Aviation Administration.
capacity as Secretary of Transportation; *
Jane F. Garvey, in her official capacity *
as FAA Administrator; John E. Turner, *
in his official capacity as FAA Regional *
Administrator,                           *
                                         *
       Respondents.                      *
                                    ___________

                          Submitted: September 15, 1999

                                Filed: April 7, 2000
                                    ___________

Before RICHARD S. ARNOLD, FLOYD R. GIBSON, and LOKEN, Circuit Judges.
                           ___________

LOKEN, Circuit Judge.

      Two cities and a county located to the west of Lambert-St. Louis International
Airport petition for review of a decision by the Federal Aviation Administration
approving and authorizing federal funding for a proposed westward expansion of the
Airport. Judicial review is authorized by the FAA Authorization Act of 1994, 49
U.S.C. § 46110. Petitioners argue the decision violates the National Environmental
Policy Act, 42 U.S.C. §§ 4231 et seq. (NEPA); § 4(f) of the Transportation Act, 49
U.S.C. § 303(c); and the consistency and notice provisions of the Airport and Airway
Improvement Act of 1982, 49 U.S.C. §§ 47106(a)(1) and (c)(1). Having given the
FAA’s decision and the administrative record the thorough, probing, but deferential
review required by Citizens to Preserve Overton Park, Inc. v. Volpe, 

,
414-15 (1971), and other cases, we conclude the FAA’s decision was not “arbitrary,
capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C.
§ 706(2)(A). Accordingly, we deny the petitions for review.

                                   I. Background

      Located some twelve miles northwest of downtown St. Louis, Lambert is one of
the nation’s busiest and most delay-ridden airports. Lambert’s importance to the
National Airspace System is well-documented. In 1996, it ranked fourteenth nationally
in terms of passengers enplaned and deplaned, and eighth in terms of aircraft
operations. All of the nation’s major commercial airlines serve Lambert, and Trans
World Airlines uses Lambert as its primary connecting hub. Lambert is also important
to the region’s economy. Its proximity to St. Louis makes it accessible to the
metropolitan area’s 2,500,000 residents, and Lambert-related activity is estimated to
provide $5 billion annually for the St. Louis region.

       Local planners and the FAA have long explored alternatives for expanding
Lambert’s capacity and reducing air traffic delays. Lambert’s delay problems stem
primarily from its inadequate take-off and landing facilities. For decades, Lambert has
relied on two parallel northwest/southeast runways, denominated 12L/30R and
12R/30L, to handle the bulk of its commercial traffic. Because only 1,300 feet
separate the centerlines of the two runways, Lambert’s air traffic controllers cannot

                                          -2-
conduct independent simultaneous arrival operations in adverse weather conditions.1
Consequently, when bad weather precludes the use of visual flight rules (about six
percent of the year in St. Louis), Lambert becomes not much better than a one-runway
arrival operation, causing average per-operation delays to exceed seventy minutes.
Lambert’s ineffective functioning during inclement weather often disrupts the flow of
air traffic nationwide.

       The City of St. Louis owns Lambert and is the sponsor of the project at issue.
The airport operator, the St. Louis Airport Authority, has been exploring options to
expand runway capacity for over ten years. Between 1989 and 1993, St. Louis
sponsored a comprehensive study resulting in an airport development plan called
Alternative F-4, which called for building four new parallel runways on the site of
runways 12L/30R and 12R/30L. Subsequent analysis revealed, however, that
construction of the new runways would jeopardize Lambert’s ability to function as a
hub and would cost significantly more than originally anticipated. Consequently, St.
Louis again studied numerous options. From nine alternatives representative of this
universe, St. Louis selected a proposal known as Alternative W-1W, which entails the
construction of a new 9,200-foot runway parallel to and to the west of existing runways
12L/30R and 12R/30L. The negative impacts of this expansion -- primarily noise and
the need to acquire property presently devoted to other uses -- will fall most heavily on
the City Bridgeton, which lies immediately to the west of Lambert, and to a lesser
extent on the City of St. Charles and St. Charles County.

      St. Louis applied to the Department of Transportation for federal funding of
Alternative W-1W. The FAA undertook an independent review of the project,
announcing that it would prepare an environmental impact statement for the proposed
Lambert expansion. The FAA spent more than two years analyzing the preferred


      1
       FAA rules currently require a minimum lateral separation of 4,300 feet for such
operations, or 3,400 feet if the airport is equipped with a precision runway monitor.

                                          -3-
expansion plan and several alternatives, holding public meetings and responding to
thousands of public comments. After issuing a Final Environmental Impact Statement
(FEIS) in December 1997, the FAA held meetings with various parties. The agency
received both positive and negative comments about the project and its review process.
On September 30, 1998, the FAA issued its Record of Decision approving federal
assistance for the W-1W project. Bridgeton, St. Charles, and St. Charles County
timely petitioned for judicial review of this final agency action.

                             II. The Record of Decision

        The FAA’s Record of Decision properly includes the findings required by the
relevant federal statutes and regulations, and it concludes with a formal FAA Approval
and Order. But the document also provides a detailed 125-page account of the process
by which the FAA reviewed and ultimately approved the W-1W project. After
recounting the history of Lambert airport and briefly summarizing the development
process, the Record of Decision sets out the purpose and need for the Lambert
expansion to provide “the primary foundation for the identification of reasonable
alternatives and the evaluation of the impacts of the development.” The articulated
purposes of the expansion project are (1) “to effectively and safely accommodate
projected levels of aviation activity at an acceptable level of delay” by increasing
airfield capacity, improving visual flight rules capacity, allowing dual independent
simultaneous arrivals in bad weather conditions, and decreasing delays; (2) to enhance
the National Airspace System by increasing capacity and reducing delays; (3) to
maintain Lambert’s importance to the economic vitality of the St. Louis region; and (4)
to facilitate an airline hub at St. Louis.

       The Record of Decision next explains how the FAA selected and analyzed
alternatives to the proposed action, noting that the FAA does not “control or direct the
actions and decisions of St. Louis relative to planning,” but has authority to withhold


                                          -4-
funding for and federal approval of the project based upon its own independent review.
Here, the FAA explored alternatives that ranged from “No Action,” to investing in a
high-speed rail line, to constructing a new airport at another location. The Record of
Decision concludes that the “No Action” alternative would not achieve the purposes
and needs of the proposed expansion and indeed would be less desirable than
Alternative W-1W from the standpoint of certain environmental impacts such as air
emissions. Building a new airport elsewhere or a high-speed rail line likewise would
not achieve the project’s purposes and needs.

      The Record of Decision then turns to the task of selecting which runway
expansion alternatives were appropriate for detailed environmental and operational
evaluation in the FEIS. For this task, the FAA employed a three-tiered evaluation
process based upon the aforementioned statement of purpose and need. The Record
of Decision explains the three-tiered analysis (discussed in Part III.A. of this opinion)
and briefly explains how this analysis resulted in detailed consideration of two
expansion alternatives, S-1 and W-1W, and the No Action alternative, X-1.

       The Record of Decision then summarizes the FAA’s detailed consideration of
alternatives W-1W and S-1, which proposed a new parallel runway to the south of the
airport terminal and existing runways. Both alternatives would achieve substantial
annual cost savings, though S-1 would cost more to build. S-1 would be “superior [to
W-1W] from an [airport] operational standpoint,” but the FAA concluded that S-1’s
more severe environmental and socio-economic impacts tilt the balance in favor of W-
1W. S-1 would require the relocation of 9,725 people; W-1W only 5,680 people
(concentrated in the City of Bridgeton). S-1 would force the relocation of 210
businesses; W-1W only 75. S-1 would directly affect fifty-seven acres of parkland; W-
1W only twenty-six acres. After carefully weighing these many factors, the FAA
decided to approve the operationally less desirable W-1W because the agency agreed
with St. Louis it was “the least impacting alternative overall.”


                                          -5-
       Next, the Record of Decision summarizes the twenty-two distinct impacts
associated with implementing the W-1W runway project, and their planned mitigation.
This section discusses the expansion’s impacts on noise levels, air emissions, water
quality, historical and archaeologically significant properties, parks, wetlands, in
addition to various social impacts. Mitigation measures adopted include sound
insulation or sales assistance for homeowners adversely affected by increased noise;
relocation assistance to residents who must move; collection and recycling of aircraft
deicing fluids to prevent runoff into local waterways; and identification of land suitable
for replacing parks and wetlands affected by the project. St. Louis has primary
responsibility for these mitigation efforts, but the FAA will oversee their
implementation.

       Finally, the Record of Decision contains lengthy summaries of the FAA’s efforts
to keep the public apprised of and involved in the process; the agency’s work with
other federal agencies, including the Environmental Protection Agency (EPA), the Air
Force and Navy, the U.S. Army Corps of Engineers, and the Federal Highway
Administration; and the involvement of state and local agencies and concerned private
organizations such as the Air Line Pilots Association and the National Air Traffic
Controllers Association. The Record of Decision summarizes and responds to
concerns and criticisms raised in the many public comments. In the section entitled
“Environmental Issues Raised about the FEIS,” the FAA responds to criticisms leveled
at the agency’s environmental analysis. This section also contains an extensive
discussion of the noise impacts, the mitigation program, and specific complaints.

                                  III. NEPA Issues

      NEPA requires that a federal agency prepare an FEIS whenever it recommends
a “major federal action[]” that will “significantly affect[] the quality of the human
environment.” 42 U.S.C. § 4332(C). NEPA’s requirements are more procedural than


                                           -6-
substantive. Preparation of an FEIS forces the agency to take a “hard look” at
environmental consequences and to inform the public that environmental concerns have
in fact been considered. But NEPA does not mandate particular agency decisions, and
accordingly judicial review under NEPA is limited to ensuring “that the agency has
adequately considered and disclosed the environmental impact of its actions and that
its decision is not arbitrary or capricious.” Baltimore Gas & Elec. Co. v. Natural
Resources Defense Council, Inc., 

, 97-98 (1983); see Robertson v. Methow
Valley Citizens Council, 

, 349 (1989). The reviewing court is not
empowered “to substitute [its] judgment for that of the agency.” Friends of Boundary
Waters Wilderness v. Dombeck, 

, 1128 (8th Cir. 1999).

       A. The Environmental Analysis of Reasonable Alternatives. The principal
issue on appeal is petitioners’ contention that the FAA violated NEPA’s mandate that
the FEIS include a suitable discussion of alternatives to the proposed action. See 42
U.S.C. § 4332(C)(iii). The Council on Environmental Quality’s implementing
regulations provide that an agency’s FEIS must “[r]igorously explore and objectively
evaluate” all reasonable alternatives, but it need only “briefly discuss” the reasons why
other alternatives were eliminated from more detailed study. 40 C.F.R. § 1502.14.
This standard recognizes that a detailed statement of alternatives cannot as a practical
matter “include every alternative device and thought conceivable by the mind of man.”
Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc.,

, 551 (1978). We review an agency’s selection of the alternatives to be
fully discussed in an FEIS under a “rule of reason.” See Boundary 

; accord City of Aurora v. Hunt, 

, 1467 (10th Cir. 1984); Airport
Neighbors Alliance, Inc. v. United States, 

, 432 (10th Cir. 1996).

       Drawing on previous studies, the FAA examined the eight Lambert expansion
alternatives considered by St. Louis, plus a no-action alternative and off-site options
such as expansion of the Scott Air Force Base airport in southern Illinois. The FAA


                                          -7-
used a three-tiered screening process to identify those runway expansion alternatives
that would meet the project’s purpose and need and therefore should be included in the
FEIS’s detailed environmental analysis. It grouped the eight runway alternatives into
“families” of related options based on their direction from the existing runways. The
north family, called N-1, NE-1, and NE-1a, proposed some form of runway expansion
to the north of the existing parallel runways. The west family, W-1W, W-1E, and W-2,
proposed a new parallel runway west of the existing runways, while Alternative S-1
proposed a new parallel runway to the south, across Interstate Highway 70. Alternative
C-1, called the “canted” alternative, resembled the Alternative F-4 plan that St. Louis
had initially rejected; it proposed three new parallel runways on the site of the existing
two.

        To satisfy the FAA’s Tier 1 analysis, an alternative had to fulfill basic
operational goals of the proposed federal action -- increase airfield capacity, reduce
delay time, support independent simultaneous arrival capability in poor weather,
enhance the capacity of the National Airspace System, maintain a passenger hub airport
in St. Louis, and meet the economic goals of the region. Alternative NE-1a and the off-
site options failed this Tier 1 analysis. At Tier 2, the FAA examined whether the
remaining alternatives would meet constructibility and benefit/cost parameters. To be
constructible, an alternative had to be “buildable without an overwhelming amount of
potential construction concerns or interference with the ability to maintain hubbing
activities.” To clear the benefit/cost hurdle, an alternative’s capacity-enhancing
benefits had to outweigh its construction costs. Alternatives N-1, NE-1, and C-1 were
eliminated at Tier 2. At Tier 3, the FAA compared the western family of alternatives
against a broad range of economic, environmental, and operational factors and selected
Alternative W-1W as the best plan from that family. Thus, Alternatives S-1, W-1W,
and the no-action option survived the three-tier screening and were subjected to
detailed environmental analysis in the FEIS.




                                           -8-
       Petitioners contend that all the northern alternatives and a Modified S-1
alternative were “reasonable” within the meaning of 40 C.F.R. § 1502.14(a) and
therefore should have been included in the FEIS’s detailed environmental analysis. To
assess the merits of this contention, we must more closely examine the FAA’s three-
tiered analysis of those alternatives. An alternative that does not accomplish the
purpose of the project in question is unreasonable and does not require detailed
attention in the FEIS. See City of Richfield v. FAA, 

, 907 (8th Cir. 1998).

       1. Alternative N-1. N-1 called for the construction of two new parallel runways
separated by a lateral distance sufficient to support independent simultaneous bad
weather arrivals. In the first stage, N-1 proposed construction of a 9,000 foot runway
1,200 feet north of the existing northern runway (12L/30R), on the site of a McDonnell
Douglas (now Boeing) aircraft factory that would be relocated at a cost of several
hundred million dollars. Next, N-1 proposed what we will term a runway-terminal
shift, which involved decommissioning the southern parallel runway (12R/30L),
constructing a new terminal in its place, demolishing the current terminal, and
constructing a second new runway where the present terminal stands. The end result
would be three parallel runways, with the northernmost and southernmost at least 3,500
feet apart, thereby meeting the FAA’s improved bad weather capacity objective.

        The FAA determined that the cost of Alternative N-1 would exceed the value of
its capacity-related benefits, largely because of the need to relocate the McDonnell
Douglas facility. The FAA rejected N-1after Tier 2 analysis because its costs exceed
its anticipated benefits, and because the need to decommission one runway, demolish
the existing terminal, and build a new terminal make it doubtful Lambert would remain
a hub during construction. Petitioners do not make a lengthy case for N-1, and rightly
so. We conclude the FAA reasonably eliminated Alternative N-1 from detailed
consideration because of its high cost and constructibility concerns. See Airport




                                          -9-


(upholding the FAA’s summary rejection of an alternative
requiring the relocation of two Air Force facilities).

       2. Alternative NE-1. NE-1 avoided the cost of relocating the McDonnell
Douglas facility by shifting the new northern runway 2,000 feet to the southeast.
However, NE-1 included the runway-terminal shift, with its attendant constructibility
problems. The FAA eliminated NE-1after Tier 2 analysis, explaining that its bad
weather capability benefits “would not be realized until 2008 or 2009,” and its
elaborate construction plan “would be disruptive to the operations of the airport to such
a degree that it could create a major interference with hubbing operations for several
years, perhaps even to the extent that airlines would move to another hub location.”
The FAA’s conclusion regarding NE-1 paralleled that of the St. Louis airport
consultant, who also determined that NE-1 presented constructibility concerns.

       Petitioners characterize the FAA’s reasoning as “mere assertions” and
unsupported “speculation” that cannot withstand NEPA scrutiny. This contention
conflicts with the requirement that an agency “briefly discuss” rejected alternatives.
40 C.F.R. § 1502.14(a). By pointing out the delayed capacity benefits and the
likelihood that wholesale reconstruction of the airport terminal would seriously interfere
with hubbing operations, the FAA provided a logical explanation for its decision to
reject NE-1 as an unreasonable alternative. Thus, the FAA’s rejection of NE-1 easily
passes muster under our “rule of reason” standard of review.

       3. Alternative NE-1a. NE-1a is essentially the first phase of NE-1, consisting
of only the addition of the northern runway proposed in NE-1. Thus, NE-1a avoids
both the relocation of the McDonnell Douglas facility and the runway-terminal shift --
the cost and constructibility factors that led the FAA to eliminate N-1 and NE-1.
However, by not relocating the terminal and the more southern of the two existing




                                          -10-
runways, NE-1a fails to provide independent simultaneous arrival capacity in bad
weather. Therefore, the FAA eliminated Alternative NE-1a after Tier 1 analysis
because it would not satisfy a significant purpose of the project.

       In challenging the elimination of NE-1a from detailed environmental analysis,
petitioners argue the FAA “imposed an improperly narrow definition of purpose and
need” by including the independent simultaneous bad weather arrivals requirement.
Petitioners further complain that NE-1a would have increased Lambert capacity and
reduced delay, and the FAA therefore had an inadequate basis for rejecting NE-1a as
a reasonable alternative to W-1W and S-1. We disagree. In the FEIS, the agency
explained why it chose the independent simultaneous arrivals capability as a critical
element of its overall purpose to increase airport capacity and reduce delays:

      In general, runway capacity problems at airports can usually be most
      improved by the construction of another parallel runway. Aircraft delays
      can be reduced by the construction of a parallel runway at a sufficient
      spacing, centerline-to-centerline, that allows for dual simultaneous
      Instrument Flight Rules (IFR) operations, that is, multiple arrivals,
      departures or a combination of both occurring at the same time during
      adverse weather conditions.

This logical criterion has been applied by the FAA in rejecting other airport expansion
alternatives as infeasible. See Airport 

& n.6. Moreover, the
FAA’s general preference for independent simultaneous arrivals capability was
supported in this case by data showing that the average annual delay, based on the
projected level of Lambert operations in 2015, would be 10.4 minutes per operation for
NE-1a, compared to 4.7 minutes per operation for W-1W and 2.5 minutes per operation
for S-1. Because the worst delays typically occur in bad weather, this average annual
data understates the increased delays NE-1a would incur during bad weather. For




                                         -11-
example, a Capacity/Delay Analysis submitted by St. Louis to Congressman James
Talent projected average peak-hour bad weather delays of 78.8 minutes per operation
for Alternative NE-1a, versus 25.5 minutes per operation for Alternative W-1W. This
data led the FAA to conclude that NE-1a is essentially a short-term strategy whose
delay-reducing limitations would require St. Louis to begin planning for additional
runway construction and land acquisition “long before the year 2015.”

        In reviewing the FAA’s selection of FEIS alternatives, we properly look at
whether the agency defined the project’s purpose in terms so unreasonably narrow as
to make the FEIS “a foreordained formality.” Citizens Against Burlington, Inc. v.
Busey, 

, 196 (D.C. Cir.), cert. denied 

(1991); see Simmons
v. U.S. Army Corps of Engineers, 

, 666 (7th Cir. 1997). But we may not
dictate to the FAA how much capacity increase and delay reduction it must accept as
the minimum airport operational improvement which will justify a project of this
expense and magnitude. At the time the FAA selected the alternatives to be given
detailed analysis in the FEIS, its extensive experience in airport operations and the data
evaluating the possible alternatives for expanding Lambert’s operational capacity
justified the requirement that the project include independent simultaneous IFR arrivals
capacity in bad weather. Therefore, the elimination of NE-1a satisfies our rule-of-
reason alternatives review.2


      2
        Petitioners also challenge the FAA’s conclusion that W-1W will allow for
simultaneous independent bad weather arrivals, and they criticize the agency for failing
to perform a so-called “real-time” study to determine whether W-1W will operate as
projected. These arguments invite us to second guess the substantive merits of the
FAA’s decision to approve Alternative W-1W. Such contentions are not cognizable
under NEPA. See Stryckers Bay Neighborhood Council v. Karlen, 

, 227-
28 (1980) (per curiam). To the extent we have jurisdiction under the Administrative
Procedure Act to review the merits of the FAA’s decision, we conclude the agency
action was not arbitrary or capricious. See Boundary 

.



                                          -12-
        4. Alternative Modified S-1. In late 1996, the Air Line Pilots Association
submitted a revised version of a prior Modified S-1 plan. This version shifted the new
southern runway proposed by S-1 substantially to the west. The Association argued
this new alternative was superior to either S-1 or W-1W, and urged the FAA to include
Modified S-1 in the detailed FEIS environmental analysis. After the Association and
FAA exchanged drawings and correspondence clarifying the location of the new
runway proposed in Modified S-1, the FAA rejected this alternative for detailed
analysis because “it is not better than the alternative we are including in the EIS (S-1).
The Association compares their newest proposal to S-1 when it suits them and
compares it to W-1W when it suits them. If we were to take any of the alternatives and
begin to shift their location slightly, then we could be faced with an infinite number of
alternatives to analyze.” On appeal, St. Charles County argues that the FAA was
capable of fleshing out the specifics of the Modified S-1 plan and was required to do
so. We disagree. An agency is not required to consider alternatives that are so lacking
in specificity as to be remote or speculative. See Olmstead Citizens for a Better
Community v. United States, 

, 209 (8th Cir. 1986). Here, the Association
provided some details for its eleventh-hour modified alternative. FAA took a hard look
at that information and concluded that Modified S-1 did not warrant detailed analysis
in the FEIS because it was not a better alternative than S-1, the southern alternative that
had been selected for detailed analysis. This was a reasonable basis for excluding
Modified S-1 from the FEIS.

       To summarize, NEPA requires that federal agencies take a “hard look” at a
contemplated major federal action’s environmental consequences prior to taking that
action. See Baltimore Gas & 

. In this case, after reasonably
selecting the alternatives to be analyzed in detail, the FAA prepared an FEIS
comprising several hundred pages of findings, analysis, and correspondence with
interested parties and government agencies. The FEIS devotes more than two hundred




                                           -13-
pages to a comparison of the environmental impacts of Alternatives W-1W, S-1, and
X-1, the three alternatives considered in detail. In the Record of Decision, the FAA
briefly discussed why the other runway expansion alternatives were eliminated from
detailed consideration, and it specifically addressed twenty-two distinct kinds of
impacts. NEPA requires no more.3

       B. Evaluation of Noise Impacts. The St. Charles petitioners argue the FEIS is
inadequate because it fails to address the adverse impact of the increased aircraft
overflight noise that Alternative W-1W will entail, particularly on the City of St.
Charles’s historic properties and activities. We disagree.

       NEPA requires that the FEIS’s detailed environmental analysis of Alternative W-
1W include a discussion of adverse environmental consequences and of measures to
mitigate those consequences. See 40 C.F.R. §§ 1502.14-.16. Applying the
methodology approved in its Part 150 Noise Compatibility Program, which we discuss
in greater detail in Part IV of this opinion, the FAA determined that the St. Charles
petitioners (City and County) lie outside the sixty-five decibel (65 dB) noise contour,
the point at which aircraft noise level is considered compatible with all land uses.
Having determined that St. Charles will be below 60 dB, including all noise from the
expanded airport, the FAA concluded that “further evaluations of aviation noise




      3
         After the FAA and the EPA exchanged comments concerning the FEIS, the EPA
advised: “With the additional clarification provided in your letter of April 9, 1998, we
better understand how FAA conducted the tiered alternatives screening, and believe
that the analysis of alternatives, based on the particular purpose and needs identified
for this project, represents an adequate screening of the alternatives consistent with the
requirements of NEPA.”




                                          -14-
impacts, such as speech interference and sleep deprivation effects, in St. Charles were
not deemed necessary for the FEIS.”

       St. Charles County disputes the FAA’s noise measurement findings, arguing its
own week-long study in early 1996 showed an average day-night noise level of 63.6
dB. The FAA’s standard measures yearly not weekly average noise levels. The
agency, not a reviewing court, “is entrusted with the responsibility of considering the
various modes of scientific evaluation and theory and choosing the one appropriate for
the given circumstances.” Sierra Club v. United States DOT, 

, 129 (D.C.
Cir. 1985). We decline to second-guess the FAA’s noise level findings.

       Similarly, the St. Charles petitioners fault the FEIS for failing to measure whether
there will be aircraft-induced speech interference with historic events in St. Charles,
such as Goldenrod Showboat programs and festivals in Frontier Park. Petitioners argue
the FEIS was required to consider these single-event noise impacts.4 However, courts
have consistently upheld the FAA’s discretion to choose its cumulative noise impact
methodology instead of single-event noise analysis. See Morongo Band of Mission
Indians v. FAA, 

, 578-79 (9th Cir. 1998), and cases cited. Congress has
instructed the agency to establish a single system of noise measurement capable of
uniform application. See 49 U.S.C. § 47502. The FAA’s methodology was developed
with the assistance of federal environmental agencies and has been upheld in numerous
court decisions such as Seattle Comm. Council Fed. v. FAA, 

, 833 (9th



      4
        As petitioners note, the Environmental Protection Agency initially urged the
FAA to measure single-event noise impacts both inside and outside the 65 dB contour.
But after the FAA responded, defending its “comprehensive analysis of noise impacts,”
the EPA replied that its comments “should not be viewed as questioning whether the
FEIS met the spirit, intent, and requirements of [NEPA].”



                                           -15-
Cir. 1992). We conclude the FAA’s analysis of noise impacts and their mitigation
complied with NEPA.

                        IV. Transportation Act § 4(f) Issues

       A. Section 4(f) of the Transportation Act, 49 U.S.C. § 303(c), requires that the
FAA take certain measures if it determines that a transportation project will “use”
natural and historic resources protected by the statute. The St. Charles petitioners
argue that the FAA’s decision violates § 4(f) because the noise impacts of Alternative
W-1W will constitute a use of St. Charles’s unique Historic District. The FAA rejected
this contention because the Goldenrod Showboat and Frontier Park lie outside the 65
dB noise contour and therefore will not be significantly impacted by the project.

        Although noise can constitute a “use” for § 4(f) purposes, no use occurs “where
an action will have only an insignificant effect on the existing use of protected lands.”
Allison v. Deparment of Transp., 

, 1028 (D.C. Cir. 1990). The FAA’s
guidelines for determining significant noise impact are found in the Part 150 Noise
Compatibility Program, 14 C.F.R. §§ 150 et seq. These guidelines were developed to
satisfy the mandate that the agency “establish a single system of measuring noise” that
can be uniformly applied at airports and surrounding areas. 49 U.S.C. § 47502(1). In
brief, the regulations require airport operators to measure in decibels the yearly day-
night average noise level (symbolized as Ldn) at various points near the airport.5 The
operator drafts a noise contour map, connecting points of equal noise exposure. The
regulations require contour lines marking 65, 70, and 75 Ldn and anticipate that
additional contour lines may be appropriate. See 14 C.F.R. § 150, app. A, part B,


      5
       The Ldn analysis adds a 10 dB penalty to events measured between the hours
of 10 p.m. and 7 a.m. because nighttime noise is particularly irksome.



                                          -16-
§ A150.101. The operator identifies the uses to which land within the 65 Ldn contour
is put and uses Part 150’s land use compatibility table to determine whether each land
use is compatible with the noise level to which it will be exposed. For instance, the
table provides that an Ldn of 65-70 dB is incompatible with the operation of a school.

        The land use compatibility table provides that all the listed categories of use are
compatible with an Ldn below 65 dB “without restrictions.” The “implicit conclusion”
is that an average noise level below Ldn 65 dB would not amount to a “use” of any
property whose protected use is fairly encompassed by the compatibility table. 

; accord City of Grapevine v. Department of Transp., 

,
1507-08 (D.C. Cir), cert. denied., 

(1994); Communities, Inc. v. Busey,

, 623-25 (6th Cir.), cert. denied, 

(1992). Therein lies the
basis of the St. Charles attack. The FAA’s reliance on its Part 150 analysis is flawed,
petitioners argue, because the St. Charles Historic District is a protected use unlike any
of the uses set forth in the Part 150 compatibility table. There is support for this theory
in 

, where the court concluded that the impact of noise on the
kinds of land uses listed in Part 150 “bears little or no relation to the effect of noise”
on a wildlife refuge, and further support in dicta in Grapevine, where the court posited
that the Ldn 65 dB standard might not adequately protect an historic village “preserved
specifically in order to convey the atmosphere of rural life in an earlier (and presumably
a quieter) 

.

       The problem with the theory in this case is that St. Charles’s Historic District is
not a wildlife refuge or a rustic village. Its uses do find reasonable parallels in the Part
150 compatibility table, for example, the subcategories for auditoriums, concert halls,
outdoor music shells, amphitheaters, nature exhibits, zoos, parks, resorts, and golf
courses. The FAA does not deny that the Historic District’s cultural events, musical
festivals, and historic floating theater are worthy of § 4(f) protection. But the agency




                                           -17-
notes that the District is located just off Interstate 70, permits automobile traffic on its
one main street, and includes numerous bars, restaurants, and retail shops. We
conclude the FAA was not arbitrary and capricious in applying its Part 150 analysis to
determinine that Alternative W-1W will not constitute a § 4(f) use of the St. Charles
Historic District.

       B. The City of Bridgeton raises different § 4(f) issues, arguing the FAA violated
the statute in approving Alternative W-1W because of the parks and historic properties
in Bridgeton that will be destroyed or constructively used by this airport expansion.

        1. Section 4(f)(1) provides that the FAA may not approve a transportation
project that will use a public park or historic site unless there is no “prudent and
feasible alternative to using that land.” 49 U.S.C. § 303(c)(1). Bridgeton attacks the
FAA’s reliance on its NEPA alternatives analysis in concluding that Alternatives N-1,
NE-1, NE-1a, and C-1 are not prudent and feasible alternatives to the use of protected
§ 4(f) resources which Alternative W-1W will entail. In reviewing this contention, we
must determine whether the Secretary “reasonably believed that . . . there are no
feasible alternatives or that alternatives do involve unique problems.” Applying the
arbitrary and capricious standard of review, we “must consider whether the decision
was based on a consideration of the relevant factors and whether there has been a clear
error of judgment.” Overton 

.

       The alternatives analyses under NEPA and § 4(f) are not precisely the same --
“an alternative is imprudent under section 4(f)(1) if it does not meet the transportation
needs of a project.” 

. However, the analyses are similar
in that “an alternative that does not effectuate the project’s purposes is, by definition,
unreasonable,” and need not be evaluated in detail under § 4(f). Ringsred v. Dole, 

, 1304 (8th Cir. 1987). In this case, as we have explained, the FAA




                                           -18-
eliminated Alternatives N-1, NE-1, NE-1a, and Alternative C-1 because they would not
meet the project’s transportation purposes and needs -- increased capacity, reduced
delays, and continued hubbing. Thus, the agency reasonably concluded these were not
feasible and prudent alternatives for § 4(f) purposes.

       2. The FAA determined that both of the feasible and prudent alternatives, S-1
and W-1W, will use § 4(f) resources. In this situation, a project must “include[] 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)(2). In reviewing
an agency’s choice among feasible and prudent alternatives, we again apply the
arbitrary and capricious standard of review. See Concerned Citizens Alliance, Inc. v.
Slater, 

, 694 (3rd Cir. 1999).

       Bridgeton argues the FAA was arbitrary and capricious in selecting W-1W over
S-1 because S-1 is the less harmful alternative. We disagree for a number of reasons.
First, while § 4(f)(2) surely requires the FAA to make a comparative § 4(f) analysis
when there are multiple feasible alternatives, we doubt whether the statute mandates
a rigid least-harm standard in airport expansion cases. Such a reading might well
conflict with the congressional mandate “that airport construction and improvement
projects that increase the capacity of facilities to accommodate passenger and cargo
traffic be undertaken to the maximum feasible extent so that safety and efficiency
increase and delays decrease.” 49 U.S.C. § 47101(a)(7).

       Second, the FAA did an extensive comparative § 4(f) analysis and concluded
that Alternative W-1W will have less impact on parks and wetlands than Alternative
S-1 (36 acres versus 68 acres directly affected). Bridgeton quarrels unpersuasively
with these findings, arguing the FAA gave parks in the communities affected by S-1 too
much protection by presuming that all parks are significant despite the absence of a




                                         -19-
declaration to that effect by the relevant local authority referred to in 49 U.S.C.
§ 303(c). This contention is without merit. It is perfectly appropriate, indeed
environmentally sound, for the FAA to presume that “land used as a public park is . . .
‘significant’ unless explicitly determined otherwise.” Arlington Coalition on
Transportation v. Volpe, 

, 1335-36 (4th Cir.), cert. denied, 

(1972) (reversing the Secretary for presuming otherwise).

       Bridgeton criticizes the FAA for treating § 4(f) resources as fungible, thereby
placing undue emphasis on easily replaceable neighborhood parks and not enough
emphasis on the five irreplaceable historic sites that lie in the path of Alternative W-
1W’s construction. The various studies in the record confirm that W-1W would affect
historic sites in Bridgeton, while S-1 would have a greater impact on parks in other
communities. However, the FAA did far more than merely total up the acres and make
a superficial § 4(f)(2) decision. The agency catalogued in detail the nature of each
property that would be affected under both alternatives. It discussed each site’s
location, its size, its function, its significance, the activities associated with it, and the
degree to which it would be adversely affected. And most significantly, the agency
developed detailed plans to avoid, reduce, or mitigate adverse § 4(f) impacts, including,
with regard to historic properties, a Memorandum of Agreement with interested parties
such as the Missouri State Historic Preservation Officer and the Advisory Council on
Historic Preservation. (Bridgeton declined to enter into that Agreement.)

       FAA selected W-1W as its preferred alternative, despite S-1’s operational
advantages, in large part because W-1W was environmentally superior and had “less
severe impacts on resources protected under special purpose laws,” including § 4(f).
Our review of the record reveals that the FAA based this decision on the relevant
factors and an appropriate § 4(f) analysis. The decision was not arbitrary or capricious.




                                            -20-
                 V. Airport and Airway Improvement Act Issues

       A. Consistency Requirement. The Improvement Act provides that the FAA
may only approve an application for federal funding of an airport project if “the project
is consistent with plans (existing at the time the project is approved) of public agencies
authorized by the State in which the airport is located to plan for the development of
the area surrounding the airport.” 49 U.S.C. § 47106(a)(1). Alternative W-1W is not
consistent with Bridgeton’s development plans. Therefore, Bridgeton argues, the
project may not go forward because the statute mandates that an airport expansion plan
be consistent with the development plans of every public agency that has planning
authority over any portion of the land surrounding the airport. We disagree.

       In making its consistency determination, the FAA relied primarily on the
endorsement of the East-West Gateway Coordinating Council, a regional planning
commission charged with “making a comprehensive plan for the development of the
region.” Reliance on the approval of such an overseeing agency was upheld in
Suburban O’Hare Comm’n v. Dole, 

, 199 (7th Cir.), cert. denied, 

(1986). Suburban O’Hare was decided under a predecessor statute which
required that the project be “reasonably consistent” with local agency plans. Pub. L.
No. 88-280, § 8(e), 78 Stat. 158, 160-61 (1964). The legislative history of this earlier
provision explicitly recognized that planning inconsistencies are inevitable, that the
statute was “not intended to require that all Federal aid be withheld until all local area
development plans are in harmony,” and that in cases of conflict “the granting of aid
would depend upon the Administrator’s appraisal of all relevant facts.” H. REP. NO.
88-1002, reprinted in 1964 U.S.C.C.A.N. 2063, 2067-68.

      In the 1994 recodification of this provision, Congress omitted the word
“reasonably,” a change described as eliminating surplus. See 49 U.S.C.A. § 47106




                                          -21-
Historical and Statutory Notes, at p.660 (West 1997). No intent to change the effect
of the statute was clearly expressed. See Finley v. United States, 

, 554
(1989). Therefore, based upon the earlier legislative history and the decision in
Suburban O’Hare, we reject Bridgeton’s contention that the interests of the National
Airspace System may be held hostage to the conflicting plans of a single local
community, when the relevant metropolitan or regional planning agency or agencies
have approved. The FAA’s consistency determination was not arbitrary or capricious.6

        B. Sponsor certification. St. Charles County asserts that the FAA’s Record
of Decision violates the requirement that the project sponsor certify it has advised “the
communities in which the project is located . . . they have the right to petition the
Secretary.” 49 U.S.C. § 47106(c)(1)(A)(ii). However, the W-1W project will not
extend into St. Charles County, so the County has no standing to raise this claim. See
Bennett v. Spear, 

, 1161-62 (1997). St. Louis did inform Bridgeton of
its right to petition and informed the FAA that it had provided such notice.

       The petitions for review are denied. The FAA’s motion to strike the City of
Bridgeton’s January 18, 2000, submission under Eighth Circuit Rule 28(j), and the City
of Bridgeton’s motion for leave to file a supplemental brief, are denied.

RICHARD S. ARNOLD, Circuit Judge, dissenting.

       Both the Federal Aviation Administration and this Court make an entirely
plausible case for the choice of runway W-1W as the best way to expand Lambert
International Airport at St. Louis, so far as air-transportation considerations are


      6
       The question whether Bridgeton’s zoning regulations apply to the Lambert
expansion project is currently pending before a Missouri appellate court. This litigation
may affect the ability of St. Louis as airport operator to implement the project, but it
does not affect the FAA’s decision to approve Alternative W-1W under federal law.


                                          -22-
concerned. Those considerations are primarily the province of the FAA, not the courts,
and our authority to review the substance of the FAA's decision is narrow. In fact, so
far as the National Environmental Policy Act of 1969, 42 U.S.C. §§ 4321 et seq.
(NEPA), is concerned, we apparently are powerless to review the substance of the
agency's decision. See, e.g., Vermont Yankee Nuclear Power Corp. v. Natural
Resources Defense Council, 

, 558 (1978). But there is much more to
NEPA than this. The central requirement of the statute is procedural. It does not
mandate particular results. But it does require that agencies considering major federal
actions that will significantly affect the human environment go through a thorough
process. All of the considerations, including environmental ones, that lead to the
agency's decision are to be thoroughly explored and exposed to the scrutiny of the
Congress and the public.

       With respect, I believe that both the FAA and this Court have misconceived the
nature and purpose of the NEPA process in this case, and I therefore feel constrained
to dissent. My reasons may be briefly summarized.

       At the center of the NEPA process is a comparison by the agency of all of the
reasonably available alternatives for solution to the particular problem confronting it
– here, the decision to expand capacity and reduce delay at the airport in St. Louis.
The statute requires that the agency's decision be thoroughly explored in an
Environmental Impact Statement (EIS), 42 U.S.C. § 4332(2)(C). A key part of this
statement is the evaluation of all reasonable alternatives. See 42 U.S.C.
§ 4332(2)(C)(iii); 40 C.F.R. § 1502.14(a) (regulations of the Council on Environmental
Quality). But the statute does not stop there. Entirely apart from its provisions
regarding the preparation of an EIS, the statute expressly requires a focus on
alternatives to a proposed action, including different ways of accomplishing the purpose
of the action. See 42 U.S.C. § 4332(2)(E). In addition, the statute requires that
economic costs, which are, relatively speaking, quantifiable, be compared with
environmental costs, which may seem to some "soft" or difficult to describe. Despite


                                         -23-
the difficulties, the agency must make an effort to quantify all of the costs and benefits,
economic and environmental, which it takes into account. See 42 U.S.C. § 4332(2)(B).
And all of this must be done not grudgingly, or in a perfunctory way, but "to the fullest
extent possible." 42 U.S.C. § 4332(2).

       I do not question for a minute the good faith and expert judgment of the FAA.
But I believe the agency has violated NEPA by excluding from detailed consideration
certain alternatives to its preferred solution. I begin by quoting the agency's own
description of its process:

              The DEIS alternatives evaluation utilized a three-tiered
              evaluation process that concentrated on the purpose and
              need for the proposed project. The first tier evaluated
              whether the various alternatives met the purpose and need
              criteria established in Section 2.0 of the DEIS. Alternatives
              that satisfied these criteria were retained for evaluation
              under the second tier of analysis. The second tier evaluated
              the "constructability" . . . and the benefit/cost ratio (BCR) of
              the alternatives . . .. Alternatives that met these criteria
              were retained for evaluation under the third tier of analysis.
              The third tier evaluated multiple specific criteria relating to
              operational efficiency (taxi times, delay times), cost per
              passenger (lower costs vs. higher costs) and environmental
              impacts (noise, land use, social, etc.).


Record of Decision at 14.

        The agency's description of its own process is clear. And, it should be said,
there is nothing inherently illegal or irrational about the three-tier analytical approach
it has used in this case. If, after considering all alternatives, the agency determines that
certain proposed actions will not meet the purpose and need it has in mind, it need not
give those alternatives further detailed consideration by comparing environmental and


                                           -24-
non-environmental costs and benefits as among them. Here, however, it seems to me
that the agency has omitted at least one reasonable alternative, referred to by the parties
as NE-1a, from detailed consideration. If this alternative is a reasonable one, the
agency has erred in discarding it before reaching the third tier of its analysis. This
question, in turn, depends on how the purpose and need that the agency is seeking to
meet are defined.

       The agency has defined the purpose and need of the project as increasing
capacity and reducing delay at Lambert. See Appendix of the Petitioner City of
Bridgeton, Volume 2, p. 317 (hereinafter cited as App.). NE-1a meets this goal. It
would permit a substantial increase in airfield capacity and would substantially reduce
projected delay, when compared with the alternative of doing nothing. App. 2:380.
FAA itself states that NE-1a does meet the project's goals with respect to the National
Airspace System and does meet the hubbing and economic goals of the region. 

380. NE-1a was rejected solely, so far as the EIS reveals, because it did not
provide independent simultaneous IFR arrival capability. Certainly this is a desirable
goal, and a desirable feature of any solution selected. But it is only one way of
achieving increased capacity and reduced delay. As the FAA itself says, independent
simultaneous instrument-flight-rules arrival capability is "one of many indicia of a
capacity-building alternative." FAA Brief 62. There are many ways of achieving the
goal. Independent simultaneous capability is one such way, and it is desirable. The
fact that NE-1a lacks this feature, if indeed it does, might be a reason for ultimately
rejecting the alternative, but it is not a reason for not comparing NE-1a with W-1W on
a detailed basis at the third and last tier of the agency's analysis.

       The FAA now says that although NE-1a would increase capacity and reduce
delay, it would not accomplish these goals so well as W-1W. This may be true, but it
is only one part of the NEPA analysis. Suppose, for the sake of argument, that W-1W
will increase capacity and reduce delay better than NE-1a. There are still
environmental costs to consider and to place in the balance, in particular, the fact that


                                           -25-
W-1W will require the relocation of over 5,000 people in the City of Bridgeton, while
NE-1a would require the relocation of almost no one, except people who are already
going to be relocated for reasons independent of this particular project. Remember,
when we talk about "environmental" considerations here, we are not talking about snail
darters (however valuable they may be). We are talking about human beings, their
homes and businesses. It seems to me entirely possible that the human and economic
cost of moving so many people might be great enough to outweigh the increased
transportation benefits of W-1W over NE-1a. But, as the EIS is presently drafted, we
do not know the answer to that question, because the FAA never reached it.

       Is the fact (assuming for the sake of argument that it is a fact) that NE-1a is
inferior, purely as a matter of air-transportation policy, to W-1W sufficient to justify
the exclusion of NE-1a from detailed analysis? I think not. This is exactly the sort of
thing that NEPA was designed for. The possibility that some substantive benefits of
a project might not be great enough to justify its environmental costs is exactly the
point of NEPA. In the end, the agency will make this decision, but it is allowed to do
so, under the statute, only after a detailed explanation of alternatives. North Buckhead
Civic Association v. Skinner, 

, 1542-43 (11th Cir. 1990), is in point here.
In discussing what alternatives an agency must consider, the Court said:

             [A]lternatives that would only partly meet the goals of the
             project may allow the decision maker to conclude that
             meeting part of the goal with less environmental impact may
             be worth the trade off with a preferred alternative that has
             greater environmental impact.

To be sure, the facts of the case are different, but the principle underlying
it is sound and applies here. Only reasonable alternatives need to be given detailed
consideration, but an alternative can be reasonable even if the solution preferred by the
agency is clearly superior in substantive terms. E.g., Surfrider Foundation v. Dalton,


                                          -26-


, 1326 (S.D. Cal. 1998), aff'd per curiam, 

(9th Cir.
1999).

       For these reasons, I conclude that the EIS prepared by the agency in this case did
not comply with the law, and I would therefore remand the case to the agency for
further proceedings.

       Additionally, I cannot agree that the FAA's consistency determination satisfied
the Improvement Act. The Act prevents the FAA from approving a project unless it is
consistent with the plans "of public agencies authorized by the State in which the
airport is located to plan for the development of the area surrounding the airport." 49
U.S.C. § 47106(a)(1). The proposed expansion will lie almost entirely within the
boundaries of the City of Bridgeton. The FAA acknowledges that Bridgeton has
adopted a development plan diametrically opposed to the W1-W alternative.
Nevertheless, the FAA argues that W1-W is reasonably consistent with local
development plans because the East-West Gateway Coordinating Council, a regional
planning commission, has endorsed the project. But the Council's development plans
are "solely advisory" under Missouri law; its plans exist "solely to aid" local
government officials in the development of their binding plans. Mo. Ann. Stat.
§§ 251.300, 251.350 (2000). A local government is not required to adopt the Council's
development plan. Mo Ann. Stat. § 251.370 (2000). In this case, the City of Bridgeton
most emphatically rejected the Council's plan.

       It distorts the plain language of the Improvement Act to find consistency under
these conditions. W1-W conflicts with the only non-advisory development plan
covering the affected area. It also distorts Missouri law to use the Council's advisory
plan as grounds for ignoring a municipality's binding zoning laws. The Court
analogizes this case with Suburban O'Hare Comm'n v. Dole, 

, 199 (7th
Cir. 1986), where a regional planning commission also approved an airport project.
But the Suburban court noted only that the agreement of a regional commission


                                          -27-
provided support for a finding of consistency. 

no per se rule.
Moreover, in that case, the commission's plan was unopposed; no local municipality
had adopted zoning rules directly contrary to the proposed project.

       The Court worries that unless Bridgeton's development plan is ignored, the
"interests of the National Airspace System" would be "held hostage to the conflicting
plans of a single local community" even though the "relevant" planning agency has
approved the plan. Ante, at 22. But under Missouri law the East-West Gateway
Coordinating Council is not the relevant planning agency. Bridgeton is. Morever, it
is Congress, not Bridgeton, that has ordered the FAA to condition project approval on
consistency with local planning. In attempting to block W1-W by implementing its
own contrary development plans, Bridgeton was only acting in accordance with the
balance of national and local interests established by Congress, not holding the FAA
"hostage."

      A true copy.

             Attest:

                CLERK, U. S. COURT OF APPEALS, EIGHTH CIRCUIT.




                                        -28-