III

      THE FEIS’S PURPOSE AND NEED ANALYSIS DOES NOT COMPLY WITH NEPA

     An EIS’s statement of purpose and need must “briefly specify the underlying purpose and need to which the agency is responding in proposing the alternatives including the proposed action.”  40 C.F.R. 1502.13.

A.   THE FAA’S ANALYSIS OF THE NEED FOR THE PROJECT CONTAINS INACCURATE DATA AND INCORRECT ASSUMPTIONS

     Where an agency’s conclusions are based on false premises or information, its decision is arbitrary and capricious.  Van Abbema v. Fornell, 807 F.2d 633, 639 (7th Cir. 1986).  Accord, Sierra Club v. United States Army Corps of Engineers, 701 F.2d 1011, 1030 (2d Cir. 1983) (an EIS that contains false or inaccurate statements fails to “provide the basis for an informed evaluation or a reasoned decision”).  The use of inaccurate information “may result in approval of a project that otherwise would not have been approved because of its adverse environmental effects.”  Hughes River Watershed Conservancy v. Glickman, supra, 81 F.3d at 446.

     The FAA determined the purpose and need for this project based on four conclusions:  (1) the minimum length of runway needed for FedEx’s operations is 9,000 feet; (2) FedEx requires two runways of this length; (3) the runways must be parallel to allow the use of head-to-head operations (landing from one direction and taking off towards the same direction); and (4) the sorting facility must be situated in the middle of the two runways.  See FEIS, pp. 2-6 to 2-15.*  These conclusions are based on both inaccurate data and unsupported assumptions.

     1.     Minimum Runway Length

     According to the FAA, a particular aircraft will dictate the minimum runway length needed if that aircraft is expected to perform 250 annual departures or 500 annual operations.^[7]  FEIS, p. 2-7.*  This aircraft is known as the “critical aircraft.”  Ibid.  Here, the FAA states that the critical aircraft is a DC-10 that will be making one daily departure from PTIA to San Juan, Puerto Rico.  Ibid.*  Based on the weight of a DC-10 carrying enough fuel to reach San Juan (1,760 nautical miles) and a full load of freight, the FAA concluded that a minimum runway length of 9,000 feet was required.  Ibid.*  Although the FEIS does not discuss the minimum runway length necessary if a DC-10 were not the critical aircraft, the majority of the proposed operations are expected to travel only between 500 and 1,000 nautical miles.  Ibid.*

     The FEIS does not contain FedEx’s schedule of operations and it is not clear that FedEx ever provided it to the FAA.  Instead, it appears that the FAA assumed that, since FedEx would be conducting operations Monday through Friday, the DC-10's trip to San Juan would occur 260 times per year (5 days x 52 weeks), thereby necessitating the 9,000-foot runway.  See FEIS, p. 2-7.*  However, FedEx’s preliminary schedule shows that while five DC-10's will be used, they will fly only Monday through Thursday.^[8]  Ex. 17, p. 2.*  Since only one of these planes will be flying to Puerto Rico (FEIS, p. 2-7*), there will be at most 208 annual departures by DC-10's to that destination.  Under the FAA’s guidelines, this cannot be the “critical aircraft” for determining the minimum runway length.  Accordingly, the FAA arbitrarily concluded that FedEx’s operations require a 9,000-foot runway.

     2.     Need for Two Runways

     Even if a 9,000-foot runway were needed, PTIA already has a runway that is 10,001 feet long (runway 5/23).  FEIS, p. 2-7.*  The evidence does not support the FAA’s conclusion that two 9,000-foot runways are necessary.

     According to the FAA, “a second minimum 9,000-foot runway is needed to prevent interruption of aircraft operations due to runway closures.”  FEIS, p. 2-7.*  The FAA supports this statement with two tables showing that between 1988 and 1999 there have been a total of 790 runway closures for maintenance, amounting to 1,305 hours of closure, and 293 incidents, amounting to 129.47 hours of closure.  FEIS, tables 2.2-1, 2.2-2.*  However, this information is not divided between PTIA’s two runways, so it is not disclosed how often the 10,001-foot runway is closed.  Moreover, the FAA makes no argument (and presents no data) that the closures have caused any delays.  Thus, the data presented in tables 2.2-1 and 2.2-2 do not support the FAA’s contention that a second runway is needed.

     Furthermore, the FAA acknowledges that runway maintenance could be done during the daytime when none of the cargo hub operations would occur.  FEIS, p. 2-8.*  The FAA also admits that routine maintenance could take place on the weekends, particularly weekend nights when there would be minimal regular operations and none of the cargo hub operations.  Ibid.; Ex. 11.*  Thus, the record indicates that maintenance can occur without interrupting the cargo hub’s operations.  As a result, there is no reason to assume that the lack of a second 9,000-foot runway would affect the single flight to Puerto Rico, let alone other FedEx operations.

     3.     The Runways Must Be Parallel

     Even if the FAA had properly concluded that two 9,000-foot runways are required, the record does not support FAA’s conclusion that two parallel runways providing for the use of head-to-head operations is required.  Instead, the record shows that the current, perpendicular configuration would satisfy FedEx’s need (and even more so if existing runway 14/32 were extended to 9,000 feet).

     According to the FAA’s capacity analysis, PTIA’s current configuration has a total capacity of 1,224 operations per day.  This is based on a weighted hourly capacity of 51 per hour (24 x 51).  See FEIS, p. 2-10.*  In 1998, PTIA averaged only 350.76 operations per day.  Id., table 4.2.5-2.^[9]*  By 2006, if this project is built, PTIA is expected to average 418.8 operations per day.  Id., table 5.1.2-4.*  This is only 34 percent of PTIA’s current daily capacity.  By 2019, PTIA is expected to average 575.8 operations per day (id., table 5.1.2-5*), only 47 percent of PTIA’s current daily capacity.

     The extra capacity of the existing airport is even clearer when operations at PTIA are broken down by hour.  PTIA averages only 5.1 operations per hour between 10:00 PM and 7:00 AM (Ex. 13*), when all of FedEx flight operations will occur (FEIS, p. 5-10*).  This means that FedEx would have almost complete access to the existing runways during its operating hours.

     The FAA’s traditional capacity analysis, the Annual Service Volume (“ASV”) calculation, also shows that PTIA’s current configuration would satisfy FedEx’s operational needs.  According to the FEIS, “ASV is the FAA’s standard method of estimating an airport’s 'theoretical' annual operational capacity for the 20-year planning horizon.”  FEIS, p. 2-9.*  Under the FAA’s guidelines, an airport should take action to improve its capacity when it is operating at 80 percent of its ASV.  Id., p. 2-10.*

     Without the cargo hub, the FAA predicts that PTIA’s operations will exceed 80 percent of its ASV in 2019.  FEIS, table 2.2-3.*  With the cargo hub, the FAA predicts that PTIA’s annual operations will exceed 80 percent of its ASV in 2009. Ibid.  Thus, capacity improvements would be required at that time.

     It is questionable whether there is a present need to build a third runway when it is not needed for at least another seven years.  In any event, the FAA’s ASV analysis contains two errors which accelerated by 10 years the date when improvements will be needed.

     First, the ASV calculations are premised on the assumption that PTIA’s weighted hourly capacity is 51.^[10]  This was derived from the assumption that PTIA currently has an hourly capacity of 62 operations during good weather (visual flight rules - “VFR”) and 46 during bad weather (instrument flight rules - “IFR”).  R. 78, p. 3.*  However, in PTIA’s 1998 proposal to solicit FedEx’s business, PTIA stated that its VFR hourly capacity is 76 and its IFR hourly capacity is 59.  Ex. 15, p. AU002166.*  Similarly, PTIA’s December 1994 Master Plan Update states that its VFR hourly capacity is 77 and its IFR hourly capacity is 56.  Ex. 44, p. 5-8.*  Based on the lower values of these figures (76 for VFR and 56 for IFR), PTIA has a weighted hourly capacity of approximately 62.^[11]  This hourly capacity dramatically increases PTIA’s ASV (i.e., its capacity) for all years.  For example, in 2019, the ASV jumps from 198,594 to 241,428.

     Second, the FAA’s estimate of the PTIA’s annual operations with the cargo hub is too high.  According to the FEIS (table 2.2-3*), there will be 169,674 annual operations in 2019 without the cargo hub.  With the hub, this number increases to 202,434.  Ibid.  The difference is due to the assumption that there will be an additional 126 daily FedEx operations, 260 days per year.^[12]  However, as noted above (p. 4), FedEx currently conducts 26 operations and, if the hub is built, will increase that by another 100 (not 126) operations.  Id., table 1.3-1 and n. 2.*  Therefore, the estimate of 2019 operations should have been 195,674.^[13]

     Using the adjusted figures for PTIA’s 2019 annual operations and 2019 ASV, with the development of the hub, the level of PTIA’s annual operations does not reach 80 percent of its ASV until 2019 – ten years later than the date predicted by the FEIS.^[14]  Thus, under the FAA’s standard capacity analysis, improvements with the hub are not needed for another 17 years.

          The FAA admits that the “ASV calculation taken by itself does not demonstrate an immediate need to construct a third runway at PTIA based on capacity.”  FEIS, p. 2-10.*  See also Ex. 16* (the FAA admits that PTIA “is operating at significantly less than its annual capacity”).  However, the FAA maintains that the current, perpendicular configuration is insufficient for this project because it cannot meet the “unique operational characteristics” of FedEx’s operations.  FEIS, p. 2-10.* 

     According to the FEIS (p. 2-9*):

The existing two-runway layout at [PTIA] precludes dual simultaneous independent runway use and efficient use of a head-to-head operation scenario.  Without the ability to support dual simultaneous independent operations and efficient head-to-head operations, [the Airport Authority] concluded that it could not provide the airfield capacity to satisfy air cargo hub requirements.

In other words, FedEx must use head-to-head operations and since the current, perpendicular configuration cannot support such operations, a third runway is necessary.^[15]

     Head-to-head operations are a type of operating scenario where “all landing aircraft arrive from a single direction, and all subsequent departing aircraft take off to that same direction.”  FEIS, p. 2-12.*  This is in contrast to traditional operations where “arriving aircraft land to one direction, and departing aircraft take off to the opposite direction.”  Ibid.  According to the FEIS (ibid.*):

The use of [head-to-head operations] is highly restrictive because it is difficult to integrate other aircraft with other directional requirements into the air traffic pattern.  The head-to-head flow is rarely used at most airports because it does not typically support efficient arrival and departure streams into and out of controlled airspace.  Since mixing head-to-head aircraft flows with “traditional” aircraft flows creates difficulties for air traffic controllers, * * * it is desirable to use either head-to-head or traditional, but not both, at any given time.

     Here, the FEIS says that head-to-head operations are necessary in order to meet FedEx’s critical, minimum departure-time requirements.  FEIS, p. 2-11.*  The FEIS states that FedEx must be able to have all of its aircraft depart within 70 minutes.  Ibid.*  Thus, by 2019, 63 aircraft must be able to depart within 70 minutes.  Ibid.  However, as noted above (pp. 4-5), the FEIS's discussion concerning FedEx’s operations indicates that Phase 2 arrivals will occur between 10:00 PM and 2:00 AM and the departures will occur between 4:00 AM and 7:00 AM.  FEIS, p. 5-10.*  This is also confirmed by FedEx’s preliminary flight schedule which indicates that most flights will occur between 11:00 PM and 5:00 AM.  Ex. 17, p. 2.*  In fact, according to this schedule, the highest number of departures in any given 70-minute period is 16 (2:55 AM to 4:05 AM).  Ibid.  Thus, the record is devoid of any explanation as to why FedEx must be able to have all 63 of its aircraft depart within 70 minutes.

     Moreover, even if such a 70-minute requirement existed, the current configuration could meet it.  As discussed above (pp. 22 - 23), PTIA’s hourly capacity, in good weather, is between 62 and 77 operations and, in bad weather, between 46 and 56 operations.  These numbers translate into 72 and 90 operations per 70 minutes in good weather and 54 and 65 operations per 70 minutes in bad weather.  PTIA experiences good-weather conditions 85 percent of the time.  FEIS, p. 2-10.*  Thus, even if all 63 of FedEx’s departures and arrivals had to occur within 70 minutes, the record shows PTIA can meet the 70-minute requirement in good and bad weather (assuming the higher value for PTIA’s hourly capacity in bad weather).  The worst case scenario would be that PTIA could not meet that requirement 15 percent of the time (assuming the lower value for PTIA’s hourly capacity in bad weather).^[16]

     Similarly, the evidence indicates that the use of head-to-head operations is not an operational requirement.  The FEIS states (p. 2-12*) that, according to FedEx, “it requires the head-to-head scenario to operate efficiently and at full capacity in its unique environment.”  The FAA apparently accepted this statement at face value without any independent assessment.

     FedEx employs a “hub system.”  Pl. Ex. 18, p. AU001949.*  Under this system, FedEx has a “superhub” in Memphis, national hubs in Indianapolis, Fort Worth and Anchorage, and regional hubs in Newark and Oakland.  Id., pp. AU001949-AU001950.*

     In order to conduct head-to-head operations on parallel runways, the runways must be at least 3,400 feet apart.  FEIS, p. 3-22.*  However, maps of the airports containing FedEx’s hubs show that only Memphis and Indianapolis have parallel runways that are sufficiently separated to support head-to-head operations.  See Ex. 19, pp. 1-2, 10-11.*  At Fort-Worth Alliance Airport, the runways are approximately 600 feet apart (Ex. 19, p. 3*); at Newark, the runways are approximately 1,000 feet apart (id., pp. 6-7*); at Oakland, the runways are approximately 1,000 feet apart (id., pp. 8-9*); and at Anchorage, the runways are less than 1,000 feet apart (id., pp. 4-58*).  Thus, at  most, FedEx can operate head-to-head operations at two of its six cargo hubs.  There is no reason to assume that FedEx actually conducts head-to-head operations even at those airports since such operations are “highly restrictive” and rarely used.  See p. 25 above.  Petitioner Gil Happel, a veteran pilot with U.S. Airways, states that “[f]rom my trips to Indianapolis, I know that FedEx seldom uses more than one runway.”  Ex. 32, para. 5.*

     The only discussion in the FEIS supporting the need for head-to-head operations at PTIA involves a Total Airport and Airspace Modeling (“TAAM”) analysis.  See FEIS, p. 2-10 to 2-11;* R. 130.  The TAAM analyzed four alternatives for FedEx’s operations:  (1) maintaining the current configuration but using only runway 5/23; (2) maintaining the existing configuration and using both runways, 5/23 and 14/32; (3) maintaining the current configuration but expanding runway 14/32 to 9,000 feet and using both runways; and (4) building a 9,000-foot runway parallel to runway 5/23.  R. 130, pp. 6-7.^[17]*  In all scenarios, the sorting facility was located in the same place, on the northeastern end of the runway airport, adjacent to runway 5/23.  Ibid.

     The FEIS claims that the TAAM analysis was intended to answer whether “the existing two runway airfield layout at PTIA [could] accommodate the nighttime operational requirements associated with the proposed air cargo hub.”  FEIS, p. 2-11.*  However, the TAAM analyzed only which configuration would result in the least amount of delay for FedEx and therefore would save FedEx the most money.  R. 130, p. 8.*  The TAAM concluded that “Alternative 4 gives the best results as far as on time planning factors are concerned.”  Id., p. 9.*  See also id., pp. 41-42.*

     The TAAM never addresses whether the other alternatives have sufficient capacity to enable FedEx to conduct all of its operations within 70 minutes, let alone within 3 or 4 hours.  The TAAM also does not indicate that any of the other alternatives would be cost-prohibitive.  Moreover, since the proposed sorting facility is located at the northeastern end of the airport in all of the alternatives, rather than at the southeastern end where it could be located (see FEIS, pp. 3-26 to 3-27*), the third alternative would unnecessarily require FedEx’s aircraft to taxi over a mile and one half to reach runway 14/32.  R. 130, p. 7.*  Nonetheless, despite this configuration, the TAAM shows that this alternative will result in only slightly more delays than the parallel runway-configuration, especially with respect to departure delay, which is the critical component to FedEx’s alleged 70-minute requirement.  Id., pp. 41-42.^[18]*

     In sum, the FAA’s standard method (ASV calculation) of evaluating whether an airport is ripe for expansion indicates that PTIA need not do so until 2019.  Therefore, in order to attempt to show that parallel, 9,000-foot runways are presently necessary, the FAA  deviated from this method and imposed a 70-minute departure requirement for all aircraft.  However, the record shows that the FAA's conclusion relied on misleading data and made incorrect assumptions.  Since the FAA deviated from its traditional method of determining whether capacity improvements are required, petitioners submit that it is particularly important that this Court require the FAA to do an accurate analysis.

     Moreover, the FAA has shown only that head-to-head operations on parallel runways would save the most time and money for FedEx.  Petitioners have no reason to doubt that that is true since it is what FedEx wants and FedEx has stated that, without it, it will look to other airports to develop its Mid-Atlantic Hub.  R. 163, p. 3.*  However, where the project also involves the expenditure of $108,000,000 in federal taxpayer dollars, taking numerous private properties, severe harm from noise to nearby residents, noxious emissions, and destruction of 23.8 acres of wetlands and 13,917 linear feet of streams (ROD, p. 66*), NEPA requires the agency to weigh the economic benefit of the project against the environmental and other costs.  Hughes River Watershed Conservancy v. Johnson, supra, 165 F.3d at 289.  This mandates a thorough, detailed analysis that is provided to the public in the EIS.  Natural Resources Defense Council, Inc. v. Morton, 458 F.2d 827, 833 (D.C. Cir. 1972).  The FAA’s failure to provide accurate information in the FEIS concerning the need for the project violates NEPA.

B.   THE FAA DEFINED THE PURPOSE OF THE PROJECT TOO NARROWLY

     The definition of the project’s purpose necessarily dictates the range of alternatives that the agency will consider.  Therefore, “an agency cannot define its objectives in unreasonably narrow terms.”  Carmel-by-the-Sea v. DOT, 123 F.3d 1142, 1155 (9th Cir. 1997).  The purpose of this project should have been defined as developing a Mid-Atlantic cargo hub.  That is what FedEx sought to do when it solicited bids from airports in North and South Carolina.  See pp. 7-8 above.

     However, according to the FAA, the “purpose and need for the proposed project is to develop facilities at PTIA that would provide airside, landside, and surface transportation improvements to support the development and efficient operation of an overnight, express air cargo hub facility at PTIA.”  FEIS, p. 2-16.*  As we will discuss below, this narrow view of the project, the construction of a hub at PTIA, allowed the FAA to eliminate from serious consideration reasonable and feasible alternatives at other sites.

       IV

THE FEIS’S ANALYSIS OF ALTERNATIVES DOES NOT

COMPLY WITH NEPA OR THE AAIA

     An EIS must contain a detailed statement of the “alternatives to the proposed action.”  42 U.S.C. 4332(B)(iii).  CEQ’s regulations provide (40 C.F.R. 1502.14):

This section is the heart of the environmental impact statement.  * * *  [I]t should present the environmental impacts of the proposal and the alternatives in comparative form, thus sharply defining the issues and providing a clear basis for choice among options by the decisionmaker and the public.  In this section agencies shall:

(a)     Rigorously explore and objectively evaluate all reasonable alternatives * * *,

(b)     Devote substantial treatment to each alternative considered in detail including the proposed action so that reviewers may evaluate their comparative merits; * * *.

     “The range of alternatives that an agency must consider is not infinite, of course, but it does include all ‘feasible’ or ‘reasonable’ alternatives to the proposed action.”  Grapevine v. DOT, 17 F.3d 1502, 1506 (D.C. Cir. 1994).  The “existence of a viable but unexamined alternative renders an [EIS] inadequate” (internal quotations and citation omitted).  Idaho Conservation League v. Mumma, 956 F.2d 1508, 1519 (9th Cir. 1992).  Accord, Dubois v. USDA, supra, 102 F.3d at 1287.

     Petitioners submit that the FAA’s alternatives analysis does not comply with NEPA and the AAIA because the FAA failed to consider in detail even a single off-site alternative and because the FAA failed to consider, except for the No-Action Alternative, any on-site alternative that did not involve the construction of a second, parallel 9,000-foot runway.

     The FEIS’s alternatives analysis has two levels.  At the first level, the FAA eliminated any alternative that would not meet “the purpose and need criteria.” See FEIS, p. 3-3 to 3-8.*  At the second level, it considered the constructability, costs, and environmental impacts of the remaining alternatives. 

     Five build alternatives survived the first-level:  Alternatives W1-A1 (the preferred alternative), W2-A, W3-A, N-D, and N-E.  FEIS, p. 3-34.*  These five alternatives also survived the second level and therefore were considered in detail along with the No-Action Alternative.  Id., p. 3-41.*  The only differences among these alternatives are the direction of the parallel, 9,000-foot runways and the location of the sorting facility.  See id., figure 3.3.2-1, pp. 3-22 to 3-26.*  Thus, the FEIS did not seriously consider any build-alternative other than the construction of a new 9,000-foot runway at PTIA.

A.   THE NARROWLY DEFINED PURPOSE RESULTED IN THE IMPROPER ELIMINATION OF REASONABLE, OFF-SITE ALTERNATIVES

     As noted above, the FAA defined the purpose of this project as developing a cargo hub at PTIA.  Nonetheless, the FEIS claims that “off-site * * * alternatives were developed and explored which could meet the purpose and need for this project * * *.”  FEIS, p. 3-1.*  Despite this claim, when the FAA screened these alternatives under the purpose and need criteria, every off-site alternative was summarily rejected because it was not PTIA.

     The five airports which submitted proposals to serve as FedEx’s Mid-Atlantic hub, namely Global Transpark, Raleigh-Durham International Airport, Charlotte International Airport, Greenville-Spartanburg International Airport, and Columbia International Airport, constitute obvious off-site alternatives.  The FEIS briefly addresses them but concludes that none meets the purpose and need criteria, because each airport “does not locate the hub at PTIA.”  FEIS, pp. 3-18 to 3-20.*

     The Global Transpark (“GTP”) is envisioned to be “a facility where industry assembles, stockpiles, and delivers goods for quick distribution through a huge airport facility that is linked by interstate quality highways to the rest of the Southeast.”  Ex. 23, p. 2.*  GTP’s Master Plan calls for two parallel runways of 11,500 and 13,000 feet.  Ex. 22, p. 6.*  This facility was located at Kinston because of the “need for a commercial distribution hub in the Eastern United States that can reach overnight more than 60 percent of the nation’s population * * *.”  Id., p. 3.*  Thus, this airport was developed for the very purpose of providing a site for operations such as those proposed by FedEx.

     The FAA rejected GTP for two reasons:  (1) it is not PTIA and (2) it is 30 miles from a major highway.  FEIS, p. 3-18.*  If the first reason is valid, there clearly is no point in discussing this alternative, or any off-site alternative.  With respect to the FAA’s second reason, while GTP may be 30 miles from an interstate highway, it is located approximately 5 miles from U.S. Highway 70, a National Highway System road.  Such a road is “important to the nation’s economy, defense and mobility.”  Ex. 21, p. 1.*  By the end of 2001, the State was expected to have completed the “Crescent Connector” in order to allow direct access from U.S. 70 to GTP.  Ex. 22, p. 8.*  “[T]his corridor will feature multi-lane access, separated turning lanes with upgraded intersections, grade separated railroad crossings and signal systems to handle increased traffic flows.”  Id., p. 9.*  Moreover, GTP involved the expenditure of over $100,000,000 in planning and development by a “team of nationally and internationally recognized professionals.”  Id., p. 6; Ex. 23, p. 2.*  They undoubtedly considered the surrounding road system in their efforts to create the ideal facility for integrating “air, rail, road, and nearby sea transportation.”  Ex. 22, p. 1.*  Accordingly, petitioners submit that GTP was a viable alternative that the FAA was required by NEPA and the AAIA to consider.

     Similarly, the FAA improperly dismissed from detailed consideration Raleigh-Durham International Airport (“RDU”).  RDU currently has parallel runways.  FEIS, p. 3-18.*  The runways are 10,000 feet and 7,500 feet long and they are sufficiently spaced to allow for head-to-head operations.  Ibid.

     The FAA dismissed RDU for three reasons:  (1) it is not PTIA; (2) it does not have redundant 9,000-foot runways; and (3) there is no room to place a sorting facility between the two runways without relocating existing facilities.  FEIS, p. 3-18.*  None of these reasons justifies exclusion from detailed consideration.

     In order to have redundant 9,000-foot runways, RDU would only have to extend its 7,500-foot runway by 1,500 feet, obviously a much smaller project than building a completely new runway as is being proposed at PTIA.  In addition, as we showed above (pp. 18-19), it is not clear that even a single 9,000-foot runway is necessary, let alone two 9,000-foot runways.  It also is not clear that the sorting facility must be located between the two runways since FedEx’s superhub in Memphis does not have that configuration.  Ex. 19, p. 2.*  Moreover, the FEIS does not discuss how extensive a project it would be to relocate some of the existing facilities at RDU and put FedEx’s facility in between the runways.

     Charlotte/Douglas International Airport (“CLT”) currently has two parallel runways of 8,845 and 10,000 feet and a crosswind runway.  FEIS, p. 3-20.*  The parallel runways are sufficiently spaced to permit head-to-head operations.  Ibid.  In addition, the FAA has approved the construction of a third 9,000-foot parallel runway.  Ex. 24, p. 2.*  Thus, CLT will soon have three parallel runways, two of which are over 9,000 feet and one is nearly 9,000 feet.

     The FAA did not consider CLT in detail because:  (1) it is not PTIA and (2) CLT’s existing aircraft operations and noise abatement requirements would conflict with FedEx’s head-to-head operations.  FEIS, pp. 3-20 to 3-21.*

     The FAA does not provide any evidence to support its claim that FedEx’s nighttime operations would conflict with existing aircraft operations or with noise abatement requirements.  FedEx manages to operate effectively at numerous other airports that have competing aircraft operations.  See p. 27 above.       Moreover, following the events of September 11, 2001, and the decline in demand for air travel, CLT halted construction on its improvements.  Ex. 25, p. 2.*  This could mean that there would be a new runway available almost exclusively for FedEx operations.  Therefore, even if a conflict with existing aircraft would have previously been a factor to consider, there is no reason to assume that this still exists.

     Greenville-Spartanburg International Airport (“GSP”) is located only one mile from a major interstate and currently has one runway of 11,000 feet.  FEIS, p. 3-19.*  The FEIS summarily dismisses GSP because:  (1) it is not PTIA; (2) it would require the construction of a parallel runway; (3) it would require the relocation of many facilities; and (4) it does not have enough space in the event that FedEx’s operations grow.  Ibid.  The construction of a parallel runway obviously does not distinguish GSP from PTIA.  The FEIS also does not discuss how extensive a project it would be to relocate some of the existing facilities, except to say that it depends on the location of the sorting facility.  Ibid.  Finally, the FEIS states only that “the proposed site for the air cargo hub is too small,” without providing any more information, and does not indicate whether any site other than the one GSP proposed to FedEx would be feasible.  Ibid.

     Columbia Metropolitan Airport (“CAE”) has two runways, 8,000 and 8,602 feet long, and is located within one mile of a major interstate.  FEIS, p. 3-19.*  The FEIS rejects CAE because:  (1) it is not PTIA and (2) competing aircraft operations would be inconsistent with head-to-head operations.  However, as noted above (pp. 27-28), the FAA's assumption that head-to-head operations are essential is erroneous since other FedEx hubs do not operate head-to-head.

     In short, the FAA rejected each of these alternatives without significant analysis simply because it is not PTIA.  Other reasons were given, but these reasons are presented in cursory fashion and are not persuasive, at least in the absence of analysis.  In fact, FedEx implicitly recognized that these airports are reasonable alternatives because it stated that it would be forced to consider them if PTIA does not build the proposed, parallel runway.  R. 163, p. 3.*

     The FAA’s contractor and the FAA informed FedEx that they were required to analyze off-site alternatives and they therefore requested an explanation as to why these other airports were not viable.  R. 153, p. 2;  R. 156, p. 2.*  FedEx  responded by setting forth the factors it considered in the selection process and saying that it rated PTIA the highest in most of them.  R. 163, pp. 6-7.*  FedEx refused to provide its own analysis of these airports to the FAA on the ground that the information was confidential.  R. 145, p. 2.*  However, NEPA imposes an independent obligation on the FAA to review this information.  See 40 C.F.R. 1506.5(a)(“The agency shall independently evaluate the information submitted and shall be responsible for its accuracy”).  Here, the FAA simply accepted FedEx’s conclusion at face value.  Consequently, the FAA’s rejection of these alternatives was arbitrary and capricious.

     Essentially, the issue here is whether NEPA requires the FAA to analyze only which is the best way to configure PTIA to carry out FedEx's wishes or whether NEPA requires the FAA to analyze other reasonable and feasible alternatives that may accomplish the “underlying purpose and need to which the agency is responding * * *.”  40 C.F.R. 1502.13.  Petitioners submit that based on the express statutory purpose and the CEQ regulations, NEPA requires the FAA to look beyond the application presented to it and analyze alternatives that can accomplish the underlying purpose, namely, to develop a Mid-Atlantic cargo hub.  The total failure of the FEIS to conduct any significant analysis of these alternatives violated probably the most important requirement of NEPA.  See Simmons v. United States Army Corps of Engineers, 120 F.3d 664, 670 (7th Cir. 1997) (by “focusing on the single-source idea [for a water supply], the Corps never looked at an entire category of reasonable alternatives and thereby ruined its environmental impact statement”); Van Abbema v. Fornell, supra, 807 F.2d at 638 (“the evaluation of ‘alternatives’ mandated by NEPA is to be an evaluation of alternative means to accomplish the general goal of an action; it is not an evaluation of the alternative means by which a particular applicant can reach his goals” (emphasis in original)); Sierra Club v. Marsh, 714 F. Supp. 539, 577 (D. Me. 1989) (“A project’s principal goals must override the stated preferences of the applicant for purposes of NEPA's ‘reasonable alternatives’ analysis”); California ex rel. Van De Kamp v. Marsh, 687 F. Supp. 495, 499 (N.D. Cal. 1988) (by failing to “evaluate  the alternative of relocating some or all of the air cargo project at nearby airports,” the Corps violated NEPA); 46 Fed. Reg. 18,026, 18,027 (“Reasonable alternatives include those that are practical or feasible from the technical and economic standpoint and using common sense, rather than simply desirable from the standpoint of the applicant”) (emphasis in original).  But see Citizens Against Burlington, Inc. v. Busey, 938 F.2d 190, 198-199 (D.C. Cir. 1991), certiorari denied, 502 U.S. 994 (1991).

B.   THE NARROWLY DEFINED PURPOSE RESULTED IN THE IMPROPER ELIMINATION OF REASONABLE, ON-SITE ALTERNATIVES

     The FAA did not analyze in detail whether the cargo hub could be constructed at PTIA based on its perpendicular configuration and extending runway 14/32 to 9,000 feet.  The FEIS states that these alternatives (X-A through X-E) would partially meet the purpose and need criteria because they would locate the project at PTIA, provide redundant, 9,000-foot runways, and provide a sorting facility that was rectangular in shape and on 300 acres.  FEIS, pp. 3-28 to 3-29.*  However, the FEIS dismissed all of them from detailed consideration because they would not provide for the ability to conduct head-to-head operations or locate the sorting facility between parallel runways.  Id., p. 3-29.*  The FEIS then explained, referencing its previous purpose and need analysis, that these two criteria are essential.

     However, as shown above (pp. 27-28), the use of head-to-head operations on parallel runways does not exist at four of FedEx's six other hubs, it is not clear that FedEx performs such operations at the two hubs where they are feasible, and the only map of the other hubs (Memphis) that depicts the location of the sorting facility shows that it is not located between the two parallel runways.  Therefore, these reasons do not justify the FAA’s dismissal of these alternatives from detailed consideration.

     In fact, the perpendicular alternatives are similar to the third alternative analyzed in the TAAM, which was found to be only slightly worse than the preferred alternative in terms of saving FedEx time.  See p. 29 above.  Moreover, unlike the TAAM’s third alternative, Alternative X-E locates the sorting facility at the southeastern end of the airport and therefore allows direct access to both runways which would save more time.  See FEIS, figure 3.3.2-1.*  However, the FEIS does not examine this benefit or even note the difference between this alternative and the third one in the TAAM analysis.

     Even if modifying the present configuration would not meet all of FedEx’s requirements, the FAA still should have analyzed at least one such alternative since NEPA requires the FAA to consider reasonable alternatives that meet only some of the project’s goals.  North Buckhead Civic Ass’n v. Skinner, 903 F.2d 1533, 1542 (11th Cir. 1990); Natural Resources Defense Council, Inc. v. Callaway, 524 F.2d 79, 93 (2d Cir. 1975).  Accord, Dubois v. USDA, supra, 102 F.2d at 1288 (the use of artificial ponds for snowmaking, instead of a natural pond, “is not so facially implausible that it can be dismissed out of hand”); Sierra Club v. Marsh, supra, 714 F. Supp. at 586-587 (alternatives analysis did not comply with NEPA because agency considered only a two-berth cargo terminal that was expandable to six berths and not a two-berth terminal that was partially expandable to less than six berths).

        V

      THE FEIS’S NOISE ANALYSIS DOES NOT COMPLY WITH NEPA OR THE AAIA

     An EIS must discuss “any adverse environmental effects which cannot be avoided should the proposal be implemented * * *.”  40 C.F.R. 1502.16.  The harm caused by significant noise levels must therefore be considered in an EIS.

     Sound is often expressed in decibels (“dBA”).  Ex. 29, p. 5.*  When the overall noise level increases by 10 dBA, a person will perceive a doubling in the degree of loudness.  Id., p. 10.*  See also FEIS, App. D, p. 2.*  Where one source of sound produces a specific noise level (70 dBA for example), adding a second source that also produces 70 dBA of noise will increase the overall noise level by 3 dBA.  Ex. 29, p. 7.*

     The FEIS presents three types of measurements for disclosing the noise impacts from FedEx’s operations:  the Day-Night Sound Level (“DNL”),^[19] the Equivalent Sound Level for 9 hours (10:00 PM to 7:00 AM) (“Leq(9)”), and the Sound Exposure Level (“SEL”).  FEIS, pp. 5-4 to 5-7.*

     The “DNL is the metric commonly used to determine impact from aircraft operations and is the principal metric used by the FAA in determining the potential noise impacts of air development actions.”  FEIS, p. 5-7.*  It shows the noise level of an average, annual day.  ROD, vol. 2, p. 18.*  In other words, it is the average noise level of a 24-hour period, measured over 365 days.  Ibid.; FEIS, p. 5-7.*  It is measured by adding all of the sound exposure during the daytime (7:00 AM - 10:00 PM) and all of the sound exposure during the nighttime (10:00 PM - 7:00 AM).  FEIS, p. 5-7.*  Ten dBA are added to the nighttime noises to reflect the fact that since “community background noise typically decreases about 10 dBA at night, nighttime noise events sound louder because there is less ambient noise.”  Ibid.

     In analyzing the DNL, the FAA generated three noise contours:  one depicting the areas that will experience 75 dBA and higher (the 75 DNL contour), 70 dBA and higher (the 70 DNL contour), and 65 dBA and higher (the 65 DNL contour).  See FEIS, figure 6.2.1-4.*

     All persons living within the 70 DNL contour can have their land purchased by the government.  ROD, p. 45.*  Here, that means 53 homes and 126 people.  FEIS, table 5.1.3-3.*  All persons living between the 65 and 70 DNL contours are eligible for sound insulation.  ROD, p. 46.*  This includes 209 homes and 503 people.  FEIS, table 5.1.3-3.*

     In addition, under FAA guidelines, when there is an increase of 1.5 dBA within the 65 DNL contour, the change is considered significant and mitigation must be considered for those homes.  FEIS, p. 5-15.*  If such an increase occurs, the FAA will also consider mitigation for homes located within the 60 DNL contour if there is an increase of 3 dBA.  Ibid.  Here, 163 homes and 284 people situated within the 65 DNL contour will experience a change of 1.5 dBA, and 452 homes and 1,096 people situated within the 60 DNL contour will experience a change of 3.0 dBA.  FEIS, table 5.1.3-7.*  However, the FEIS says that the change within the 60 DNL contour is not a significant impact and the data are provided only “for informational purposes.”  Id., n. 2.* Therefore, no mitigation has been offered to homeowners within the 60 DNL contour regardless of whether they experience a change of 3 dBA or more.  See ROD, p. 46.*

     The Leq(9) is similar to the DNL except that it is the average sound level of all of the noise events that occur between 10:00 PM and 7:00 AM.  FEIS, p. 5-6.*  An Leq can be done for time periods other than nine hours.  The FAA used an Leq(9) because it “approximates the time span of FedEx operations at night.”^[20]  FEIS, p. 5-7.*

     While DNL and Leq(9) are cumulative noise metrics in that they average noise over time, the SEL measures a single event.  It is used to analyze disturbances with sleep and interferences with speech.  FEIS, p. 5-5.*  During a single noise event, sound will vary with time.  Ibid.  However, a single measurement describes “only one dimension of the noise,” i.e., how loud was it.  Ibid.  Two events, each with identical noise levels, may generate different sound exposures: one very short and one that remains loud for an extended period of time.  Ibid.  The latter is perceived to be much more disturbing, yet a single measurement will not reveal this difference.  Ibid. 

     The SEL is designed to provide a comparison of such different sound exposures.  FEIS, p. 5-5.*  It treats the exposure as if it occurred in one second, i.e., “all of the sound that occurs over the entire time of the event is assumed to occur in one second.”  FEIS, p. 5-5.*  As a result, louder events have higher SEL’s than quieter ones and longer events have higher SEL’s than shorter ones.  Ibid.

A.   THE NOISE ANALYSES ARBITRARILY ASSUMED THAT ALL OF FEDEX’S OPERATIONS WOULD OCCUR TO AND FROM THE SOUTHWEST 95 PERCENT OF THE TIME

     In performing all three noise analyses, the FAA assumed that FedEx would operate head-to-head to and from the southwest 95 percent of the time, i.e. 95 percent of all aircraft would land from and take off towards the southwest.  FEIS, p. 5-11.*  This is dramatically different from the current operations at PTIA which occur to and from the southwest 54 percent of the time and to and from the northeast 46 percent of the time. Id., p. 2-10.*  The 95-percent assumption “results in unusual configurations to the noise contours * * *”  (id., p. 5-11*), indicating that there will be nearly no impact on the 2,500 people living directly off the northeastern end of the new runway (runway 5L), some as close as one-half mile (see FEIS, figure 6.2.4-1*).

     There are serious problems with the FAA’s assumption.

     First, the record contains no commitment from FedEx that it actually intends to operate in this manner and neither the Airport Authority nor the FAA has imposed any such requirement.  Instead, the FAA assumed that 95 percent of FedEx’s operations will be to and from the southwest because, in doing so, FedEx would be landing towards and taking off from the sorting facility, which “reduces time on the ground by minimizing taxi times by aircraft.”  FEIS, p. 5-9.*  In other words, the FAA believes that it is more efficient to operate that way; therefore, it assumed FedEx would do so.

     The record contains evidence that FedEx insisted on having unrestricted access to PTIA’s runways.  In a letter to the Airport Authority, FedEx stated that, given “the strategic nature of the Mid-Atlantic Hub, FedEx will require assurance we can operate unrestricted at [PTIA] on a 24 hour per day 7 days per week basis and will not be subject to curfews or other interruptions to air or ground traffic and hub operations.”  Ex. 30, p. 2.*  In response to this demand, the Airport Authority stated that it would “not impose any restrictions” on FedEx, but “FedEx would be required to land from a southwest direction, and take off in a southwest direction, whenever possible, in order to minimize the noise impact of FedEx’s flights over residential developments” (emphasis added).  Ex. 31, p. 4.*  However, this standard is not included in any contract or other document binding on FedEx.

     Second, there is no reason simply to assume that FedEx will operate 95 percent of the time to and from the southwest since there are good reasons for it not to do so.  Unless there is a tailwind,^[21] it is often more efficient “for planes to take off in the direction of their destination, even if that requires that the plane taxi down to the other end of the runway.”  Ex. 33, para. 4.*  This saves fuel, reduces engine wear, and saves flight time since the aircraft will travel a shorter distance to its destination.  Ibid.; Ex. 1, p. 11.* 

     In fact, the very analysis on which the FAA relied to show that head-to-head operations are the most efficient, the TAAM analysis, assumed that operations occurred, almost evenly, to and from the northeast and southwest.  Specifically, the TAAM assumed that, between midnight and 3:44 AM, all operations occurred to and from the southwest and, between 3:45 AM and 06:55 AM, all operations occurred to and from the northeast.^[22]  R. 130, p. 7.*

     Third, several factors may prevent FedEx from conducting head-to-head operations.  Competing air traffic from private aircraft^[23] or from late-arriving commercial aircraft may make head-to-head operations impossible.  See Ex. 32, para. 3; Ex. 33, para. 3.*  In addition, the 95-percent assumption is premised on the fact that winds moving southwest will be greater than 10 knots (11 miles per hour) only 5 percent of the time.^[24]  FEIS, p. 5-11.*  However, when there is a 7-to-9-knot wind moving southwest, a pilot may choose for safety reasons to take off into the wind to the northeast.  See Ex. 33, paras. 3-4; Ex. 1, p. 5.*

B.   THE NOISE ANALYSES UNDERESTIMATE THE LIKELY IMPACTS

     Even if the 95-percent assumption were reasonable, the FAA's noise analyses substantially underestimate the noise impacts from FedEx’s operations.

     The noise analysis was performed using the FAA’s Integrated Noise Model (“INM”).  FEIS, p. 5-4.*  The INM normally assumes that there is a headwind.  Ex. 1, p. 7; Ex. 2, p. 28.*  Here, there will undoubtedly be numerous operations with a tailwind if 95 percent of all operations occur to and from the southwest.  Ibid.  Whenever there is a headwind for landings, there must be a tailwind for takeoffs and vice-versa. 

     An aircraft that takes off with a tailwind will remain at a lower altitude longer since as it will have less air speed, which it needs to climb.  Ex. 2, p. 28.*  Therefore, the aircraft will be closer to the ground and the noise will be greater.  Id., pp. 28-29.*  However, the FEIS does not indicate that it adjusted the INM to account for this fact and the FAA did not respond to petitioners’ comment regarding this problem.  See ROD, “FP0021," p. 28.*  Therefore, the FEIS has significantly underestimated the noise impacts.

     Similarly, it is not clear that the INM factored in the weight of the cargo.  “It is highly unusual for an FAA-sponsored FEIS not to state clearly what weights and/or trip lengths are used in the application of the” INM.  Ex. 1, p. 6.*  However, here, the FEIS did not provide that information.

     FedEx’s aircraft will transport freight which is “much denser than passengers and baggage.”  Ex. 1, p. 5.*  “Heavier aircraft lift off further down the runway and climb more slowly than lighter aircraft.”  Ibid.  A fully loaded cargo plane can create 5 to 8 more dBA of noise than a passenger aircraft.  Id., pp. 5-6.*  Since it appears that the FEIS did not factor in the weight of the cargo, it significantly underestimated the noise that will be generated.  In response to this comment, the FAA stated only that its analysis considered the different weights of the types of aircraft FedEx will use.  ROD, p. 16.*  Notably, the FAA did not say that it considered the heaviness of the freight.

     The noise analyses also substantially underestimated the impacts from FedEx’s operations because they were produced from an outdated INM.  The FEIS states (p. 5-4*) that it used version 5.2a of the INM which was “the latest version of the model at the time the EIS was started.”  This is not accurate.  The FAA released version 6.0 in September 1999.  Ex. 2, Attachment 8.*  The FAA released the draft EIS in April 2000, seven months after version 6.0 was released.  While there may not have been time to run version 6.0 in the draft, there was more than enough time to do so in the final EIS.^[25]  Nonetheless, in response to this comment, the FAA maintained that the “FEIS uses the most recent version of the” INM.  ROD, p. 16.*

     “Accurate scientific analysis * * * [is] essential to implementing NEPA.”  40 C.F.R. 1500.1(b).  The recent version of the INM more accurately analyzes “the propagation of sound from planes on the ground.”  Ex. 2, p. 27.*  When an aircraft is rolling down a runway prior to takeoff, it emits sound from its engines.  Ibid.  Since the FAA assumed that 95 percent of the takeoffs would occur to the southwest, nearly all of the noise emitted from the engines will begin at the northeastern end and be propagated in that direction.  This is near the densely populated area which the FEIS indicates will receive almost no noise impact.  See FEIS, figure 6.2.1-4.*  The FAA’s failure to use the most recent version of the INM resulted in the inadequate disclosure of noise effects and violated NEPA.  See North Carolina Alliance for Transportation Reform v. DOT, 151 F. Supp.2d 661, 696 (M.D. N.C. 2001) (use of outdated air emissions model “prevented decision-makers and the public from more fully understanding the [project’s] effect on air quality” and therefore violated NEPA”).

C.   THE NOISE ANALYSES DID NOT TAKE INTO ACCOUNT FEDEX’S UNIQUE OPERATIONS

     Finally, the noise analyses should have been adjusted to account for FedEx’s unique operations.  As noted above, the FAA’s traditional noise analysis (the DNL) measures an airport’s operations over 365 days per year, 24 hours per day.  It “is most useful when the noise does not vary much from day to day * * *.”  Ex. 2, p. 13.*  Here, the only proposed change to PTIA’s operations will occur on 260 days each year and only between 10:00 PM and 7:00 AM.  Therefore, the DNL dilutes the consequences from these operations by averaging them with the airports daytime and weekend operations. 

     For example, the predicted noise level from one takeoff over point 25, which is off the northeastern end of the new runway (see FEIS, figure 5.1.3-1*), is sufficient to generate a 65.6 DNL.  Ex. 1, pp. 2-3.*  However, the actual 65 DNL contour does not extend nearly that far because the measurement is averaged over 24 hours and over 1 year.  See FEIS, figure 6.2.1-4.*  Nonetheless, the residents located in that area “will at times be exposed to extreme noise for short periods unlike anything they have experienced or would have ever expected to experience.”  Ex. 2, p. 40.*  Overall, petitioners’ expert estimates that the traditional DNL analysis understates the actual noise from FedEx’s operations by 4.3 dBA.  Ex. 1, p. 2.*

     In justifying the need for this project, the FAA deviated from its traditional capacity analysis because of FedEx’s “unique operational characteristics.”  See p. 24 above.  Based on those same characteristics, it was equally necessary for the FAA to deviate from its traditional noise analysis.  If FedEx truly must conduct all of its operations within 70 minutes in the middle of the night, the FAA should have disclosed to the public the noise that will be generated during that 70-minute period.  At the very least, the FAA should have disclosed the noise that will be generated during FedEx’s stated operational periods for arrivals and departures.

     This was especially important because Greensboro is not densely populated and therefore has less background noise levels than those assumed in the FAA’s DNL analysis.  The level of the DNL in a “residential area is closely related to the population density averaged over an area of at least a few square miles.”  Ex. 2, p. 22.*  As noted above, under the FAA’s DNL model, noise levels that do not exceed 65 dBA are not considered significant.  However, an area with a DNL 65 dBA is normally found in a community of approximately 20,000 people per square mile.  Ibid.  Greensboro’s population density is approximately 2,500 people per square mile.  Ibid.  A DNL 55 dBA is normal in Greensboro and the nighttime noise level would therefore be approximately 45 dBA.  Ibid.; FEIS, p. 5-7.*

     If FedEx’s operations increase the nighttime noise level by 10 dBA to an overall level of 55 dBA, the background noise level of the community will have doubled.  See p. 43 above.  Here, one of the ten sites used for predicting effects from the operations shows that there is a 10.4 dBA change but the overall noise level does not exceed 65 dBA.  See FEIS, tables 5.1.7-2, 5.1.7-3; Ex. 2, p. 52.*  However, the number of people or homes impacted is not disclosed under the FAA’s DNL analysis since it considers impacts only when the overall noise level exceeds 65 dBA.  See Ex. 2, pp. 15-16.* 

     The FEIS’s two other analyses, the Leq(9) and the SEL, do not compensate for the deficiencies in the DNL analysis.

     In the Leq(9) analysis, the FEIS provides hardly any information.  The entire analysis consists of a table containing generic information about the Leq(9)’s numerical results and cursory statements concerning that table.  For example, the entire discussion of the Leq(9)’s Phase 1^[26] analysis of the preferred alternative is (FEIS, p. 5-24*):

Table 5.1.3-4a presents the nighttime equivalent or Leq(9) noise levels at the specific points for Alternative W1-A1.  Under Phase 1, Alternative W1-A1 would result in Leq(9) noise levels at the specific points that vary from 28.2 to 69.2 dBA.  Noise levels would decrease up to 17.2 dBA at some points, while increasing up to 11.7 dBA at others when compared to the No-Action.[^[27]]

     A noise impact analysis should allow a reader “to look at locations on a map with appropriate graphics and see the change, either directly or by comparing two maps showing the before and after conditions.”  Ex. 2, pp. 31-32.*  The Leq(9) analysis does not do this.  There are no noise contours or other maps provided which would disclose where noise changes will occur.  No information is provided as to how many people, homes, and businesses will be affected.  Therefore, the cursory information contained in the FEIS’s Leq(9) analysis does not adequately disclose the impacts from FedEx’s operations.  See Hughes River Watershed Conservancy v. Glickman, supra, 81 F.3d at 445-446 (agency’s cursory analysis of zebra mussel infestation violated NEPA).

     In addition, the Leq(9) does not add the 10 dBA penalty in order to account for the fact that nighttime noise levels have dropped by that amount.  See Ex. 1, p. 3.*  Therefore, even the numeric results in the table understate the consequences from FedEx’s operations.

     According to EPA, the Leq(9) analysis “does not provide a totally accurate view of the additional noise situation, i.e., noisy takeoff and landing events occurring during a relatively short late-night and early-morning timeframe.”  FEIS, vol. 5, App. O, “DF0004,” p. 7.*  EPA explained that while “the current information/data is correct for the average daily situation, [the DNL and Leq(9)] understate[] the noise exposure for an actual operational day * * *.”  Ibid.

     In the SEL analysis, the FAA generated SEL 100 dBA noise contours and informed the public that no more than 10 percent of adults who are exposed to an arrival or departure producing an SEL of 100 dBA “are likely to be awakened.”  FEIS, p. 5-5.*

     However, the SEL analysis does not attempt to provide additional information beyond the DNL.  The FEIS explains that the “100 dBA SEL contour was selected because it results in a similar degree (or percentage) of affected populations as the DNL 65 dBA noise contour.”  FEIS, p. 5-5.*  Thus, the FEIS admits that the use of the SEL 100 dBA was not intended to disclose consequences that might be overlooked from the use of the DNL analysis.

     If the FAA had used a lower SEL than 100 dBA, the number of people awakened would be less than 10 percent.^[28]  Ex. 2, p. 6.*  However, the population affected would have been much higher because the SEL contour would extend much farther out from the runway.  Ibid.  Thus, the total affected population would probably be much greater.  Id., p. 25.*  These impacts were not disclosed since those affected individuals do not lie within the 65 DNL contour.  Id., p. 38; FEIS, p. 5-5.*

       VI

THE FEIS FAILED TO ANALYZE WHETHER THERE WOULD

BE ANY IMPACTS FROM TOXIC AIR EMISSIONS

     Airports are among the largest single source emitters of toxic air pollutants due to an array of emission sources associated with their operation (i.e., aircraft, motor vehicles, ground support equipment and stationary plant operations).  R. 318 (Piazza DEIS Comments, pp. 2-3*).  Accordingly, in order to disclose the possible effects from toxic air emissions from this project, the FEIS should have done an emissions inventory and then performed a health risk assessment.  Id., p. 4.*

     According the FAA, it did not do this because based “on information available in the scientific literature, there is no known ‘cause-and-effect’ between toxic air emissions from planes and health problems around airports.”  FEIS, App. O, part 2, Response 7-3.*  The FEIS does not indicate to what scientific literature it is referring, and petitioners set forth substantial scientific literature to the contrary, which the FAA did not refute.  See Ex. 4, pp. 1-4; ROD, p. 27.*  For example, EPA conducted a health risk assessment in 1993 which reported that emissions from aircraft engines were responsible for approximately 10.5 percent of the cancer cases around Chicago’s Midway Airport.  Ex. 4, pp. 1-2.*  The FAA itself has participated in studies of human health risks from toxic air emissions related to operations at the Santa Monica Airport and at the Los Angeles Airport.  Id., p. 2.*  The former found “cancer risks for individuals living in proximity of the airport to readily exceed the U.S. EPA’s ‘ample margin of safety’ threshold of one in one million.”  Ibid.  The latter stated that impacts from increased emissions of toxic air pollutants “may include increased cancer risks and non-cancer health hazards” to people “working, living, recreating, or attending school on or near the airport.”  Ibid.  Accordingly, the FAA’s failure to analyze the potential risks from emissions of toxic air pollutants associated with this project violated NEPA and the AAIA.

      VII

THE FEIS FAILED TO ANALYZE ADEQUATELY THE

INDUCED-GROWTH EFFECTS FROM THIS PROJECT

     An EIS’s discussion of environmental effects must include the project’s indirect effects and their significance.  40 C.F.R. 1502.16(b).  Indirect effects are defined as reasonably foreseeable effects and may include “growth inducing effects and other effects related to induced changes in the pattern of land use, population density or growth rate, and related effects on air and water and other natural systems, including ecosystems.”  40 C.F.R. 1508.8(b).

     In discussing the merits of this project, the FEIS estimated that it will create 16,308 new jobs by 2019, including 2,650 people directly employed by FedEx and the rest by new businesses that start because of increased cargo capacity in the area.  FEIS, p. 5-103, table 5.4.3-4.*  The creation of these new jobs is expected to result in 10,872 new households and 25,006 new persons in the area.  Id.,p. 5-103.*

     While the FEIS addresses the benefits of this growth, i.e. the fiscal advantages that this growth will provide for both the regional and national economy, it does not discuss its adverse environmental impacts.  For example, the FEIS examines increases in demand for potable water and for the disposal and treatment of wastewater, but this analysis is based only on the population increases directly associated with the FedEx facility and its employees.  See, e.g., FEIS, p. 5-148.*  The FEIS does not address water-quality impacts that will result from the additional 13,500 jobs and new households that will come to the area.  Since CEQ’s regulations require such an analysis, the FEIS does not comply with NEPA and the AAIA.

CONCLUSION

     For the foregoing reasons, petitioners request that the Court order the FAA to rescind the ROD and prepare a new draft EIS which is consistent with NEPA and the AAIA.

  REQUEST FOR ORAL ARGUMENT

     Counsel for petitioners respectfully request that the Court hold oral argument on these issues.

                                   Respectfully submitted,

                                   BRUCE J. TERRIS

                                   DEMIAN A. SCHANE

                                   Terris, Pravlik & Millian, LLP

                                   1121 12th Street, N.W.

                                   Washington, D.C.  20005-4632

                                   (202) 682-2100

                                   Counsel for Petitioners

May 6, 2002