Integrating Aviation and Passenger Rail Planning (2015)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

80 Introduction and Structure Chapter 6 reviews key issues for integrating air and rail in the Midwest of the United States. These issues are presented in four parts. Part One examines Chicago’s O’Hare Airport (ORD) as a case study in the sheer difficulty of providing a “seamless con- nection” from a long-distance rail station to a dispersed set of airside concourse piers that are already built, in operation, and unlikely to be moved. Unlike the luxury of having a “greenfield” setting, planners at ORD must deal with the extensive infra- structure already in place, which constrains the realistic options for locating a long-distance train station on the site, and with institutional partners concerned about capital costs. Part Two reviews recent proposals for long-distance rail improvements that influence air in various ways in light of the site planning constraints described in the opening section; Part Two also helps understand the demand characteristics that are discussed in the succeeding section. Part Three documents the demand for travel between the Chicago area and 300-mile radius of the surrounding area. The possible role for Midwest HSR to serve in a competitive mode is examined first by exploring the possible scale of diversion of under-300-mile trips away from air to HSR. For the comple- mentary role, the section first examines the logic of HSR ser- vices serving as a feeder mode to longer distance flights; then, it examines the logic of HSR serving to divert ground access trips to the airports away from the automobile to rail. Part Four reviews the results of interviews with air and rail managers in the Midwest about the adequacy of existing tools and resources to undertake complex, multimodal planning and decision making. Part One: Rail at Chicago O’Hare— Dealing with Uncertain Futures Introduction Recent events in the planning of Chicago O’Hare (ORD) airport illustrate the challenge of retrofitting an existing, largely decentralized airport to allow for high-quality trans- fer from long-distance rail to long-distance air. In Chapter 3, Newark Liberty International Airport (EWR) was seen as an example of an airport where major improvements to rail connections may be possible. There, terminal reconstruction will take place with or without the context of a rail strategy. With the option open for a higher quality Automated People- Mover (APM) connector, the possibility exists for either send- ing regional service (PATH) directly to the passenger terminals; alternatively, the airport APM can be directly taken to the rail location with the greatest amount of train connections. Later, in Chapter 7, the San Diego Airport (SAN) is examined, where highly promising site planning options to connect the air- port with long-distance rail are possible, and where planners can choose from among a set of good options for intermodal connections. Here, by contrast, the case of ORD is presented where there are no easy or simple options to connect the air terminals with long-distance rail—some form of compromise will be needed. This is largely because of the decentralized nature of the ter- minals. The airport currently has four terminals, named Ter- minals 1, 2, 3, and 5 (Figure 6-2). Planning has proceeded for several years on a possible fifth terminal, named Terminal 7 (Figure 6-1). For the airport site planner, designing a good connection between a long-distance rail station and the gates in airports at Tampa, Orlando, Atlanta, Pittsburgh, or Denver would be comparatively easy, as each of these airports consoli- dates almost all landside passenger processing at one point. In theory, if the long-distance rail station could be located in or close to such a dominant landside terminal, the problem would be solved. Connecting a single long-distance rail station to all the gates at ORD would be harder, but it is theoretically possible. Figures 6-3 and 6-4 show the location of a new “West Termi- nal Area” developed by the planners of the ORD Moderniza- tion Program planning process at the Chicago Department of Aviation, where Terminal 7 is the westernmost structure. The planning process documented that, in the long term, new C H A P T E R 6 Air and Rail in the Midwest

81 Locaon of proposed intermodal terminal, now deleted from airport plan Locaon of proposed development area at exisng rail staon Figure 6-1. Possible locations for Air/Rail transfer at Chicago O’Hare International Airport. Imagery © 2100 DigitalGlobe, Sanborn, USDA Farm Service Agency, GeoEye, US Geological Survey, Map Data © 2011 Google. Figure 6-2. The existing O’Hare terminals. Source: O’Hare Airport.

82 Figure 6-3. Schematic Diagram of people mover connecting West Terminal area with all concourses. Redrawn from diagram supplied by Chicago Department of Aviation, O’Hare Modernization Program. Image © 2014 DigitalGlobe, Landsat, U.S. Geological Survey, USDA Farm Service Agency. Map Data, © 2014 Google. Figure 6.4. In the O’Hare Preferred Development Plan, the people-mover was to extend from the West Terminal Area to Terminals 1 and 2. Source: Chicago Department of Aviation, O’Hare International Airport Master Plan, Preferred Development Plan (Annotations and emphasis for people-mover alignment added). gates could be located to the west of the present terminals; planners had worked closely with the Illinois Department of Transportation in the planning of new ramps from the high- way system to the new Western Terminal. The long-distance rail station would have been located here, along with a second CTA transit station. Advocates of integrated HSR in the Midwest were strong supporters of the Terminal 7 concept, and lobbied for its development as a key intermodal terminal, served by the list of transportation modes shown in Figure 6-5 (Midwest High-Speed Rail Association 2011). Connecting to the Airport Terminals The plan was ambitious and expensive. As diagrammed in Figure 6-3, all airport users could access the new Terminal 7 check-in; they would then be connected to the airside con- courses of some of the four existing terminals.

83 level of facility service: immediate check-in at the “rail” sta- tion, followed by the kind of post-security, on-airport con- nections operated successfully in airports worldwide. In terms of the high-quality European services described in Chapter 2, this Chicago concept would be similar in concept to Zurich, where highly centralized check-in is located at/near the rail station, followed by security and airside people-mover access to some concourses. It would be more integrated than solu- tions in Paris CDG and Dusseldorf, where major air terminals must be accessed by landside people-movers, which deliver the passengers to various locations external to security clear- ance. The Chicago Terminal 7 site planning concept would be superior to that of Frankfurt, where no APM system connects directly with the long-distance rail terminal; there, rail riders must walk to Terminal One, where the APM connects Termi- nal 2 with both airside (secure) and landside (non-secure) shuttle services. What Happened? The reaction of the airlines to the creation of a new Western Terminal at ORD was predictable. The major airlines took the City of Chicago to court to kill the project, and, in March 2011, they prevailed. Their position was that the full program was too expensive. As of the summer of 2012, the website for the Chicago O’Hare Modernization Program (OMP) still refers to a planned Western Terminal. The website states simply, “The OMP will also bring a new Western Terminal facility with more airline gates and parking. The Western Terminal will be connected to O’Hare’s main terminal core by an APM system” (O’Hare Air- port 2012). Interviews with the Chicago Department of Aviation made clear that the airport plan does not include the additional ter- minal or the additional gates designed in the modernization plan. The legal agreement clarifies that the Western Terminal is no longer part of the airport’s plan, but leaves certain issues unresolved. Plans had progressed for new roadway access to a major parking facility, and these plans will have to be reas- sessed. Given the increase in the throughput of the runway system, the issue of more gates undoubtedly will be addressed at some time, and in some way. During the interview process, the Research Team was told that not having more gates was strongly in the interests of the dominant airlines, who did not want any more gate capacity for new airlines. A New Site for a Long-distance Rail Station? However, the airport planners already had a second (and much lower cost) option available for siting a rail station for a future higher speed rail program. Looking at the location labeled “Economy Parking F” in Figure 6-2 (and marked with As shown in Figure 6-3, the inside-security, airside APM would directly serve a new midfield airside passenger con- course and proceed under the existing runway. At that point, the APM system would have a tunnel station with connections to both Concourse B and Concourse C of the Terminal 1 com- plex (United Airlines). The airside APM could continue to the International Terminal with direct services to midpoints in the airside piers of Terminals 2 and 3, labeled as Concourses G, H, K, and L in Figure 6-2. In the airport’s Master Plan’s Preferred Development Plan, the people-mover operates from Terminal 7 to Terminals 1 and 2, as shown in Figure 6-4. Because the air passenger would alight the APM in the mid- dle of the existing airside concourse piers, it would be essen- tial for baggage check-in for all carriers to be located at the new Terminal 7. Because some airlines would oppose dividing their check-in services over several locations, some form of common or pooled services would be required. In addition, a baggage pick-up strategy would need to be found. Strictly from the vantage point of this study of site plan- ning options for long-distance rail at airports, the full scheme developed in the Modernization Program offered an optimal Figure 6-5. Role of Terminal 7 in the Chicago intermodal system. Source: Midwest High Speed Rail Association.

84 train rider could gain good access to the International Ter- minal, with continuing APM service to Terminals 3, 2, and 1 within the main terminal complex. While often called a “two seat ride,” such a system would be comparable in quality with landside people-mover concepts in Paris and Dusseldorf. In this concept, the passenger transferring from rail would remain outside security until arriving at existing passenger terminal facilities. Site Planning Concept #3: Connect to Network at a Central Place: Connect with the Best Consolidated Rail Transfer Point Possible In this option, the need for any new rail line with long- distance through service stopping at ORD is avoided, saving a considerable amount of travel time for the riders on the corridor between Chicago and Milwaukee, generally consis- tent with the rail operator business case arguments made in Chapter 3. In such an option, the City of Chicago would seek partners to build a high-quality shuttle between ORD and Union Station. The strategy would quickly (and as seamlessly as possible) get ORD users to the point in the region’s trans- portation system with the most connecting train options. Thus, the transferring air passenger would have a good con- nection to the major rail terminal from which higher speed trains would radiate out in four directions, rather than one. Importantly, this is the option currently being most aggres- sively pursued by the City of Chicago, which (under the previous mayor) issued a Request for Proposal (RFP) for international consortia interested in providing such service at their own financial risk. Left unstated in that RFP is just how such a service would interface with the airport itself. With the demise of the Ter- minal 7 concept, this leaves the Economy Parking F location as the most probable terminus for this local train option, given the cost of any rail tunnel that might take the train to the Inter- national Terminal and on to the central terminal complex— a route already covered by the existing people-mover and its extension. Direct rail service only to the Economy Parking station would create, in effect, a “three-seat ride” from long- distance rail, to airport rail shuttle, to APM connecting to land- side terminals. Part Two: The Range of Higher Speed Rail Options in the Midwest A conclusive analysis of the subject of the role of airports in the developing HSR system in the Midwest is impossible because there is no consensus about the long-term HSR system to be developed in the Midwest. Multi-state rail planning—examining a series of spokes emanating out of the hub at Union Station in Chicago—is divided between an arrow on Figure 6-1), the Chicago Aviation Department has already commenced the process to develop the site into a mixed-use complex. The site development package shows an extension of the existing landside airport people-mover into the new site, close to the existing, but infrequently used, Metra rail station. Thus, it is possible to locate a larger rail station at this loca- tion. At this ITC, there would be no security clearance, as the rest of the APM system is totally outside of security. If any baggage check-in were attempted, it would most likely utilize trucks and tugs, but not the APM system; its stations in the terminal area are not well integrated into the passenger ter- minals. Similar truck delivery systems were used for the entire Newark baggage check-in at the rail station; these systems were also used for the on-airport segment of the Heathrow Express service, before both of these baggage systems were abandoned because of a lack of customer use. The options examined in ORD planning to date can be reviewed in terms of the site planning categories presented in Chapter 2. Site Planning Considerations Consistent with the typology developed earlier in this report, the site planning options at the Chicago O’Hare International Airport can be reviewed in terms of an international perspec- tive. A review of international airport stations suggested three major categories for airport rail station solutions: Site Planning Concept #1: Full Integration at Airport: Reroute the Long-distance Rail Line to Go to a Point from Which the Air Traveler Can Walk from the Train to the Check-in Terminal As discussed, the full conceptual plan developed for the O’Hare Terminal 7 scheme would have been comparable in quality with the most expensive of the European examples of transfer efficiency. In fact, the contemplated people-mover tunnel to the midpoints of existing airside pier concourses would provide a higher quality of retrofit of existing facilities than has been attempted elsewhere. Site Planning Concept #2: Shuttle to Rail Alignment: Build a Separate Airport Rail Station on the Rail Alignment at a Point As Close As Possible to the Airport Passenger Terminal Complex If the HSR connection is made by connecting the existing airport people-mover to the existing Metra station, shown on the upper right hand corner of Figure 6-1, a shuttle to rail alignment would result, as exemplified by the present station connection at Newark. With a single transfer, a high-speed

85 more than 80 groups, including a number of states, train operators, and train constructors, had sent letters describing their interest in being part of the development of high-speed train travel in the United States. The first major proposal for the Midwest was prepared by SNCF (Societe National de Chemin de Fer, the French National Railway) in September 2009. Because the US DOT policies have been encouraging private sector proposals, the exact legal status of the proposals is unclear. According to The Transport Politic (Freemark 2009), SNCF submitted a 1,000-page document detailing efforts in four corridors; one corridor was the Midwest. According to this online source, the first phase of rail investments for the Midwest would extend from Milwaukee to Detroit, via a bypass around Chicago, Fort Wayne, and Toledo by 2018, with a link to Cleveland opening by 2020. The full system would include new con- nections from Chicago to St. Louis; Chicago to Cincinnati; and Milwaukee to Minneapolis. SNCF predicts full opera- tion by 2023, though further links along the Ohio 3C (Cleve- land, Columbus, and Cincinnati) corridor and to Kansas City, Pittsburgh, and Toronto could be considered for future development. SNCF expects that the system would more than cover operations costs, allowing the network’s revenues to be used to repay some of the initial construction costs. The public would subsidize 54% of the $68.5 billion total cost of right- of-way, construction, and train sets. Benefits from reduced car and air travel, however, are expected to offset 150% of the government investment in construction costs over a period of just 15 years of operation. By the calculations of SNCF, potential journey times would connect Chicago and St. Louis in 1 hour 44 minutes; Minneapolis and Chicago in 2 hours 42 minutes; Chicago and Detroit in 1 hour 53 minutes; and Indianapolis and Detroit in 2 hours 52 minutes. The first phase would attract 15.8 million passengers per year by 2022; the completed sys- tem would serve 42.3 million passengers by 2028. Although the SNCF vision has no legal standing as a “master plan” for HSR, its ambitious treatment of major corridors, like the routing of Chicago to Detroit service via Toledo to get the most appropriate alignment for super high-speed service, has evidently had a great effect upon the planning efforts of others. The Midwest High-Speed Rail Association Plan In the spring of 2010, an ambitious proposal was released by the Midwest High-Speed Rail Association (MHSRA), which represents the highest level of consensus yet about the form of a “220”-mph system for the Midwest. The key ele- ments of the plan are represented in Figure 6-6 with the lines designated in the maps as “220 mph.” those pursuing a near-term system with maximum speeds of 110 mph, and those advocating early investment in a “220”-mph system for the same geographic area. The two orientations occasionally conflict. The near-term program is being developed by the Midwest Regional Rail Initiative, which is run mainly by state departments of transportation; the more ambitious high-speed program is advocated by the Midwest High-Speed Rail Association, a private group funded by private donations. Midwest Regional Rail Initiative (MWRRI) The Midwest Regional Rail Initiative (MWRRI) is a coopera- tive, multi-agency effort that began in 1996 and involves nine Midwest states (Indiana, Illinois, Iowa, Michigan, Minnesota, Missouri, Nebraska, Ohio and Wisconsin) and the FRA. In 2004, the MWRRI finalized a plan for a Midwest Regional Rail System that would incrementally build on existing infrastruc- ture resources. By repairing and upgrading service on current lines, and restoring lines out of use for decades, the system would provide every major city in the Midwest with fast, frequent, and reliable passenger rail service on seven major branches joining in the Chicago hub. The MWRRI plan elements include: • Use of 3,000 miles of existing rail right-of-way to connect rural and urban areas; • Operation of a hub and spoke passenger rail system; • Introduction of modern, high-speed trains operating at speeds up to 110 mph; and • Provision of multimodal connections to improve system access. The goal of the initiative is to develop a passenger rail system that offers business and leisure travelers shorter travel times, additional train frequencies, and connections between urban centers and smaller communities. The proj- ect is funded by the FRA as part of its approved program of HSR corridors. Based on the interviews conducted, the managers of the MWRRS proposal do not plan to divert alignments between Union Station and Milwaukee to serve ORD directly with the 110 mph rail system; this would add more than ten minutes travel time to the city-to-city trip. Alternative Higher Speed Proposals SNCF: Midwest Corridor In December 2008, the FRA announced it would begin accepting Expressions of Interest (EOI) for the development of high-speed lines in the United States. By February 2009,

86 to Indianapolis and Cincinnati (via Gary), and to Detroit and Cleveland (via Gary and Toledo) would all be routed through a new station at McCormick Place, and then to a new, underground Union Station. From there, the 220-mph alignment would proceed to ORD and then north to Milwaukee. Figure 6-8 shows that the alignment would allow trains from all three lines approaching from the south to serve Union Station directly, and proceed on to ORD— if demand and operations considerations determined this was desirable. Functionally, service to ORD before the main terminal would be possible for riders from the north (e.g., Minnesota, Madison, and Milwaukee) with all others having the physical potential of an airport stop after Union To give a sense of the scale of the differences in assump- tions about the system, Figure 6-7 reproduces the summary table from the MHSRA’s recent study of economic impacts (Midwest High-Speed Rail Association 2011). The MWRRI travel time calculations show that the application of 110-mph technology would not provide terminal-to-terminal service in under 3.5 hours in any of the four corridors analyzed. By contrast, adoption of 150 mph technology would provide 3.5-hour service in all four of the corridors; the assumption of 220-mph technology results in 2.5-hour terminal-to- terminal times in all four corridors. In the plan published by MHSRA, new lines providing 220-mph service to corridors to St. Louis (via Champagne), Figure 6-6. Composite map showing several separate Midwest HSR proposals. Source: Midwest High Speed Rail Association.

87 Figure 6-7. System differences, by speed assumed. Source: Reproduced graphic from Midwest High Speed Rail Association. Station, with a transfer for some under most operating assumptions. In interviews, advocates of both the mid-speed systems and the high-speed systems agreed that a diversion of the Union Station to Milwaukee line to serve ORD directly was not practical with lower-speed systems. The same managers agreed that, if the kinds of travel times envisioned in the 220-mph vision could happen, the diversion away from the shortest airport alignment would be logical. The MHSRA plan refers to a 36-minute travel time to Milwaukee, which assumedly refers to the fastest train times (i.e., without a stop at ORD). With corridor travel times this low, a diversion into the airport might be justified without damaging long- distance travel times. HSR advocates in Chicago have some interesting approaches to the question of the possible role of HSR as a feeder mode to airports in the United States. The benefits of the improved travel times occur with longer distances, perhaps well outside of the traditional catchment area of an airport. In Chapter 2, for example, demand for access services was strong in the nat- ural catchment area of Dusseldorf Airport (served by medium speed rail), but poor in the cities served by the longer distance HSR trains. At Newark International Airport, demand for airport access using NJ Transit services is strong but some- what weaker for access using the long-distance Amtrak ser- vices. However, advocates of HSR solutions for the Chicago hub system believe that once new rights-of-way are created for longer distance, high-speed systems, a new form of commuter hybrid service could be created, modeled after the Javelin ser- vice in the London area, which operates in both local and high-speed mode. This is an important concept, as it would create a system in which an air market from the farther points in the metropolitan area of the airport would be served by a newly improved service. In order to create this program, discrepancies between commuter rail schedules (overwhelm- ingly oriented towards commuter movements) and schedule requirements for air services (spread out over several peaks during the day and night) would need to be resolved. Airports Connected in Alternative Proposals In the MHSRA proposal, the new high-speed right-of-way would not serve Midway Airport directly. By contrast, the lines routed through Gary, Indiana, would make possible an air/rail station at that airport. As of the time this report is being writ- ten, debates continue over alignments near smaller airports, such as Madison, Wisconsin in the MWRRS (110-mph) plan. The authors of the SNCF plan for the region were explicit in

88 are within 300 miles, representing nearly 5.9 million yearly flights in 2007. At present, ORD has approximately four times as many yearly boardings as Midway. Diverting Air Passengers with Both an Origin and Destination in the Region: Establishing the Range of Diversion It is clear that there is the potential for travel times simi- lar to those in Europe reported in Chapter 2, which have resulted in significant diversions of O-D trips, based on travel times by so-called 150 mph and 220 mph technolo- gies as shown in Figure 6-7. In Europe, HSR between Madrid and Barcelona garnered about half of the air plus rail O-D market—even while the airlines kept an exceptionally high quality of service. The Madrid–Barcelona case study is of great relevance to the United States application, as the Span- ish airlines have made a far more robust defense of their markets than was the case in some early HSR applications in France. the support for HSR (220-mph) alignment stops at airports serving Milwaukee, Gary, Cincinnati, Detroit, and Cleveland. The SNCF plan also includes a Chicago bypass loop that would allow their alignment to serve both MDW and ORD in Chicago, with no HSR track between Union Station and either airport. Part Three: Understanding the Demand to/from Chicago’s Airports Understanding the Market for Air/Rail Competition in Chicago The opening chapters of this report found it important to treat origin-destination (O-D) markets separately from the markets for connecting air passengers. This section of Chapter 6 examines only markets that reflect true O-D travel entirely within the study area. The following section will exam- ine only those currently using feeder air services to begin/end a longer trip within the study area. As shown in Figure 6-9, there are 32 airports with service to ORD and/or Midway that Figure 6-8. Diagram showing HSR alignment between McCormick Place and O’Hare Airport. Source: MHSRA.

89 itinerary information about domestic air travel in the United States. These databases provide the total number of passen- gers traveling from airports within a range in which HSR could reasonably serve as an effective alternative. The range was set as between 50 and 300 miles from Chicago for this analysis. The planned HSR service could certainly serve some trips outside of this range, but the greatest potential would clearly lie within this range. Estimates for the potential size for this market can be determined by comparing the number of passengers that currently travel to these regional airports from ORD or Midway with the future alignment of the HSR network in the Midwest. Focusing now on the markets for ORD, Table 6-1 shows that of the 18 million originating passengers boarding a plane at ORD, about 6% of these O-D passengers are going to an airport within 300 miles of Chicago, or roughly one million passengers. The Research Team has calculated that over 90% of them are in fact going to an airport that is within a 20-mile drive of a planned HSR station. Figure 6-10 shows the loca- tion of those airports and the scale of the volumes carried from ORD. Thus, about 946,000 departing ORD passengers were found to be logical candidates for diversion to rail in its O-D market function. How many of these passengers might be diverted away from their present medium-distance mode (air) into a differ- ent medium-distance mode (rail)? Three diversion rates were The following analysis examines the implications of rail diverting a high range of 50% of the air plus-rail O-D market, with additional sensitivity testing for a midrange at 35% and a low range at 20% of that market. Figure 5-4 in Chapter 5 showed that O-D air volumes between Boston and New York dropped by about 50% between 1993 and 2010. Wardman et al. (2002) reported that when the Paris–Bordeaux corridor attained a travel time of 3 hours, a 20% substitution rate was experienced along that corridor; similarly for the early years of the Paris–London market, they reported a 25% corridor level substitution rate (additional diversion occurred later). In a comparable analysis, Black (2005) reports a variety of substitution rates, from 15% to 53%, along the Northeast Corridor. Thus, the range of possible diversion rates applied seems reasonable for scenario testing purposes. One can posit that, in markets like Chicago–Minneapolis and Chicago–Detroit, airlines will assertively defend their network markets. Thus, feeder flights will remain, and those seats can also easily be marketed to the O-D market. This sup- ports the concept of testing both high and low scenarios for diversion. The O-D Market from ORD—Data Descriptions The DB1B and T100 air passenger data from the Bureau of Transportation Statistics provide origin, destination, and Figure 6-9. Regional airports with service to O’Hare and Midway 2007.

90 at ORD. Table 6-2 presents the results from three diversion scenarios ranging between 20% and 50%. The number of boarding passengers with both origin and destination in the study area that may be diverted at ORD in the scenarios range between 190,000 and 473,000, depending on how competi- tive the HSR service is in the future. If 473,000 of these origi- nating passengers—who formerly made regional flights from ORD—now traveled via HSR, the number of regional flights from ORD could be reduced by about 5,750 flights per year, or about 16 flight departures per day, as shown on Table 6-2. However, building directly on the analyses of air/rail com- petition presented earlier in this report, the question—again— turns to the speculation over how the airlines would react to a 50% reduction in the O-D component in any given city pair market. Thus, with the focus of this section of Chapter 6 only on the O-D passengers in the HSR study area, one simply can- not predict the change in number of flights that would occur. Rather, if the airlines wanted to do it, the number of departing applied to that candidate market; however, for brevity’s sake, this text will focus on the high end of the diversion estimates— with about 50% of the O-D passengers choosing rail instead of the base case assumption of air. Assuming that the new rail could reach out and attract the higher diversion number, this means that 473,000 of these annual O-D passengers would become rail users. Remembering that there are 34 million annual air passen- gers boarding a flight at ORD, this diversion would represent a decrease of approximately 1.4% in ORD passenger volume. From this point, further sensitivity analysis can be applied; if the reader posits that HSR will capture 75% of that logical mar- ket, then the diverted passenger volume would jump to 2.1% of ORD’s departing passengers. What is important in this scenario testing exercise is to establish the scale of change without imply- ing the performance of actual forecasts, which were not done. Figures 6-11 and 6-12 show the number of airplanes and number of seats in planes used in flights under 300 miles Total Boardings With Destinations within the region % of Total Originating Boardings 18,154,235 1,019,710 6% Boardings from Transfer Flights 15,932,074 3,843,273 24% Table 6-1. Trips from ORD to study area airports. Figure 6-10. Study area airports to be served by HSR.

91 Figure 6-11. Annual flights in study area to MDW and ORD (2007). Figure 6-12. Average number of passengers and seats on study area flights (2010).

92 the focus is now on those connecting on to a later departing flight at ORD, usually from a long-distance flight. The viability of HSR as an alternative to regional flights is related to the number of flights from each of the regional airports and the capacity of those flights. Not surprisingly, the largest number of flights to ORD and Midway come from the other large airports in the Midwest, including Detroit Metro politan Wayne County Airport (DTW), Lambert- St. Louis International Airport (STL), and Port Columbus Inter national Airport (CMH). These airports also happen to be near the border of the 300-mile radius of Chicago. Fig- ure 6-12 includes the total number of flights from all of the regional airports to ORD and Midway; Figure 6-13 shows the average number of seats and passengers on those flights to ORD and Midway. Average load factors on these flights to ORD vary between 58% for Fort Wayne International (FWA) and 81% for Louisville International Airport (SDF). The majority of the flights to ORD are between 50% and 70% full. The distribution of load factors for flights traveling to Midway is more clustered, with a minimum of 62% for DTW and maximum of 72% for SDF. A significant justification for connecting a HSR service directly to an airport is the concept that connecting air pas- sengers would abandon the feeder flights to the extent that the number of flights could be lowered. The question is raised about what impact such diversion might have on the need for continued feeder flights to ORD. Getting the holders of a long- distance airline ticket to voluntarily choose rail for the first segment can be done, but only through a major institutional commitment by many parties. The data from Chapter 2 on consumer choice between an access flight and an access train ride is varied: from Stuttgart to FRA, approximately one inte- grated ticket holder in seven chooses the train as the feeder mode over air as the feeder mode. In a geographic area where both trains and planes are available to Paris CDG, it is esti- mated that 45% of air travelers select the train as feeder while 55% choose the plane as feeder, as reported by the French Civil Aviation Authority (DGAC). That study concluded that there was great variation in the market behavior: in essence, where the quality of competing air service was weak, rail mode share was strong; where the quality of competing air service was strong, TGV share was weak. In the base case scenario, about 3.8 million persons from the study area are gaining access to ORD by airplane to con- nect onto a departing long-distance flight. Based on lessons flights could decrease by 16 departures per day, assuming the size of the planes remains constant. Who is on Those Flights? It is important to review the basic scale of the two markets during the transition to the analysis of the feeder flight func- tion in the study area. Table 6-1 showed that 14% of airport passengers boarding at ORD were destined to disembark at airports within the 300-mile radius study area. Of the people on those flight segments, almost 80% were not O-D passen- gers, but were connecting passengers. Thus, if the total volume of O-D passengers on these within- study area flights were to decrease by half, the total volumes remaining on the flights would fall by only 10%. All of this is to set some scale for the amount by which the airlines will want to decrease the frequency of the feeder flight system. Phrased differently, it suggests that in theory the market for diversion of connecting air passengers from ORD is four times the size of the market for diversion of those traveling completely within the region. Understanding the Market for Air/Rail Complementarity in Chicago: Rail Providing Feeder Services to the Airport The following two sections of Chapter 6 will review the use of HSR to carry air passengers from ORD by diverting them from their present use of feeder aircraft to rail service designed to mimic the trip quality of the feeder airplane. These sections will also examine the scale of the market for HSR to carry air passengers from Chicago airports by diverting them from their present use of highway—using ground access modes, including auto and airport bus. The analysis in these sections assumes that the number of feeder flights and passengers into a major airport (ORD or MDW) is mirrored by an equal number out of that airport. Thus, the terminology “to” an air- port is used to be more brief than the more accurate phrase “to or from” said airport. Rail Complementarity: Diversion to HSR from Feeder Aircraft at ORD Next examined are passengers flying in flight segments between ORD and other airports within the HSR study area— Boarding Passengers Diverted, by scenario Potenal Reducon in Number of Deparng Planes 20% 35% 50% 20% 35% 50% 189,256 331,198 473,140 2,300 4,024 5,749O’Hare International Airport Table 6-2. Passengers diverted by competing rail, and market for reduction of planes.

93 the airport. This might include creating a legally binding sys- tem wherein the air passenger is guaranteed a flight, even if the feeder mode fails to perform. This might be the case with airlines with very long-distance (presumably inter national) tickets to sell. On the other hand, an airline with a hub in Milwaukee might strongly oppose the Chicago airport’s investment in a program to attract more Milwaukee residents to another state’s airport, and offer litigation against such an investment. The Rail Operator The rail operator must be convinced that there was a strong business case for veering off the straight shot align- ment between Chicago and Milwaukee, adding time for the long-distance corridor riders (thus lowering the number of said riders)—all for providing access to the smaller number of riders destined for a long-distance flight, as analyzed in Chapter 2. Under the rules established by Fakiner in Chap- ter 2, that rail company could be providing at least hourly service to every corridor under consideration for the rail-as- feeder model, including hours in which the airport volumes peak and metropolitan volumes do not peak. The Public As noted in Chapter 3, the method for soliciting passenger feedback is unclear. In a multi-state consortium, is it realis- tic that a State DOT representative from Nebraska will lobby heavily for more expenditure in Illinois? In a fragmented decision making process, it is hard to locate any given leader that has the power to impose institutionally complex solu- tions which, when accomplished, might benefit everyone. At the time that the O’Hare Modernization litigation was being settled, the only organization lobbying for a better pub- lic understanding of Terminal 7 was the Midwest High-Speed Rail Association, which it did through its website. Lowering the Number of Feeder Flights? This analysis has not included a quantitative prediction of the diversion of connecting passengers away from feeder flights to feeder rail. Given the fact the largest airports in the study area are distant from Chicago (e.g., 300 miles), the most logical markets from such diversion seem to be those closer to the city, which now tend to be served by smaller aircraft. It is simply unclear how many of the flights from smaller markets would be eliminated, if any. It is unclear that the airlines would choose to reduce their network presence much lower than the present frequen- cies, where flights are timed to meet major activity periods at the hub airport. The airline’s business case arguments against a complete abandonment of air feeder services for learned from the previous chapters of this report, at the far- ther destinations of the study area, the rail travel times may not provide compelling alternatives to the flights. Thus, as feeder to ORD, it will be difficult for rail to be competitive from Detroit, St. Louis, Columbus, Cincinnati, or Minneapolis, based on dis- tance alone. In addition, there is no compelling argument for the airlines to abandon air services to/from those markets, as discussed in Chapter 2. This market assessment leaves open the question of mar- kets for rail-as-feeder to longer distance flights at ORD from HSR stations like Springfield, Decatur, Champagne, Lafayette, Fort Wayne, Gary, Milwaukee, Madison, and La Crosse. With some exceptions, this list of logical potential markets is con- sistent with the conclusion of the French DGAC study that rail was chosen over air-as-feeder primarily in places where the air service was not very good. Of course, not all HSR corridors would have equal access to ORD. Geography dictates that rail users in the best position to get the travel times required for this concept live along the align- ment from Milwaukee, Madison, La Crosse, and Minneapolis. For every other corridor, some passengers on trains will have to transfer at Union Station with some, but probably not all, rail services continuing on to ORD. In short, some trains will get to serve the airport directly, while others will not. What is Needed to Make Feeder Rail Work at ORD? Consistent with the conclusions of Chapters 1 and 2, all parties would have to work together to produce ORD services that are comparable to the services produced in Frankfurt by the legal agreement between Fraport, Lufthansa, and Deutsche Bahn. To recap, the roles needed from the players are: The Airport The airport must buy into the concept by investing in a passenger and baggage movement system similar to the high- quality, but expensive, people-mover system envisioned in the Terminal 7 concept developed for the O’Hare Modernization Program. The standard set by all parties at Frankfurt Airport was that the entire connecting time—as published in the CRS—would be under 45 minutes, a challenging logistical task in an airport of the size and layout of ORD. That airport would have to be in a legal position to defend such an investment in case of litigation against it by an opposing airline. The Airlines If some airline found a market niche in which this combi- nation could improve their product’s competitiveness over other airlines, they might want to invest the energy into joint ticketing, and some kind of specialized baggage handling at

94 which attracts over 30% of its users from greater distances, as reported in Chapter 2. Of the Chicago air passengers who travel this distance, roughly two thirds come from areas to be served by the new HSR system. Twenty-five of the airports with service from ORD are within 25 miles of a proposed HSR station. These 25 airports represent roughly 90% of the regional air service from ORD. Similarly, the 24 airports that are within 25 miles of an exist- ing Amtrak service account for 83% of regional travel from ORD. Among the 29 total airports that are near future HSR or existing Amtrak, 20 airports will be serviced by both future HSR and Amtrak, four will only be serviced by Amtrak, and five will only be serviced by future HSR. Data Sources and Analysis In 2003, Chicago DOT undertook research for a ridership study of a proposed Rail Express service from downtown Chicago to ORD and Midway airports (City of Chicago, 2003). RSG conducted an O-D and stated preference survey to pro- vide insight into these air traveler markets. The surveys were administered to passengers at the airports and collected infor- mation about origin, destination, itinerary, and mode choice to the airport. The survey data collected were expanded to rep- resent the total number of trips made in 2003, as applied here. An estimate of the divertible air passenger ground access market can be determined by comparing the geocoded home ZIP code information from the 2003 study to the proposed future HSR network, as shown in Figure 6-6. For the pur- poses of this analysis, it was assumed that passengers who lived between 50 and 300 miles were eligible to be diverted, see Table 6-4. Passengers who live outside this range were assumed to be unlikely to change their ground access behav- ior. According to the 2003 data, approximately 970,000 of the 22.1 million originating boardings at ORD and MDW came from passengers who traveled from a location between 50 and any given city pair were noted in Chapter 2. The question of further reduction in feeder flight volumes because of diver- sion to rail in this mid-distance market must be placed in the category of an unresolved issue, needing better tools and better data. Rail Complementarity: Diversion to HSR from Roadway-based Modes As discussed in Chapter 3, most ground access issues at airports in the United States are primarily metropolitan in scale, not supra-metropolitan in scale. This tends to explain the high ratio of NJ Transit riders over Amtrak riders at the Newark Liberty International Rail Station, for example. How- ever, the focus of this report is the examination of the possible role of longer distance rail services in solving airport-based problems—including airport ground access. The goal in this section is to explore the scale of the market for a possible major shift in the access patterns to Chicago airports if a radi- cally improved rail system were built in the region. Such a multi-billion dollar investment might fundamentally change access patterns, and even choice of airports. For example, assume that HSR supplied fast travel times to ORD but not to Midway (MDW) from Champagne, Illinois. Under this scenario, the Champagne air traveler might switch to ORD for the trip. This section reviews the scale of the market of Chicago region air travelers starting the trip from more than 50 miles away from both Chicago airports. Although major international airports, such as ORD, often have large geographic catchment areas, the research has estab- lished that less than 5% of Chicago airport passengers access- ing the airport by ground do so from a distance of more than 50 miles (Table 6-3). From the vantage point of the study of long-distance rail, this represents a much smaller catch- ment area than that of FRA, which pulls about half of its pas- sengers from distances greater than 61 miles, or Paris CDG, Boardings from 2003 Chicago Survey Data Total Boardings % Total % of ground origins between 50 and 300 miles Total Boardings 22,113,959 NA NA Ground origins between 50 and 300 miles 967,459 4.4% NA Ground origins between 50 and 300 miles within 25 miles of future HSR 622,737 2.8% 64.4% Ground origins between 50 and 300 miles within 25 miles of existing Amtrak 850,704 3.8% 87.9% Table 6-3. Scale of the long-distance ground access markets to two Chicago airports.

95 Rail Diversions from the Highway System This exercise examines the number of air passengers who could be diverted from their present “rubber tired” mode to HSR in the future. For a variety of reasons, moving passen- gers from the congested highway system to the rail system is generally seen as beneficial. But it cannot be assumed that all those coming to O’Hare from a distance of more than 50 miles are coming by automobile. According to the sur- vey results, about half of these air passengers were gaining access to O’Hare by bus. Exactly how the rail would compete with specialty bus services is beyond this report’s scope. The present ground access mode shares for air passengers from between 50 and 300 miles from the Chicago airports is pre- sented in Table 6-4. Conclusion: Demand in the Midwest, Market Categories Together This section of the Chapter 6 analysis has explored the potential size of the future Midwest HSR network by examin- ing two potential sources of ridership: HSR as a replacement for regional “feeder” flights and HSR as a replacement for long- distance ground access to ORD and Midway. The concept of rail as a feeder mode has been explored, but the results were not 300 miles of the airport. The majority of these airport access trips would be served by a future HSR line (65%) and nearly 90% are currently served by an existing Amtrak line. Fig- ure 6-13 illustrates the geographic distribution of the home ZIP codes for passengers accessing ORD or MDW who are located near the proposed HSR lines. The analysis reveals that the potential diversion of these ground access trips from highway using modes to HSR is considerably lower than for air trips. Considering present demand to both airports, these passengers could provide a market for an airport HSR station service of between approximately 120,000 and 300,000 trips per year, based on the assumption of 20% to 50% diversion as summarized in Table 6-6. O’Hare Midway Both Auto 34.8% 71.4% 47.2% Rental car 13.5% 13.4% 13.5% Public Modes 51.7% 15.2% 39.4% Total 100.0% 100.0% 100.0% Table 6-4. Mode share of air passenger from 50 to 300 miles from Chicago airports. Figure 6-13. Markets between 50 and 300 miles located along HSR lines.

96 Table 6-5. Passengers diverted to rail for study area O-D market, by airport. Passengers Diverted, by Scenario Potential Reduction in Number of Planes 20% 35% 50% 20% 35% 50% O’Hare International Airport 189,256 331,198 473,140 2,300 4,024 5,749 Midway International Airport 109,320 191,310 273,300 831 1,454 2,076 quantified. This analysis used readily available data sources and the experience from other corridors to determine the assump- tions used to estimate the size of the potential HSR market. While the true size of the HSR market will depend heavily on the service characteristics of the future service and the integra- tion with other transportation infrastructure, this analysis pro- vides an initial outline for what may be possible in the future. Table 6-5 presents three estimates for diversion of pas- sengers from air to rail with both origin and destination in the study area—in total and broken out by airport. Table 6-6 summarizes the work on the scale of markets for each cate- gory, for both airports together. It shows the scale of the markets for diversion under three scenarios and indicates that between 400,000 and just over 1,000,000 trips are can- didates for diversion to HSR in the two markets quantified. The analysis presented earlier suggests that the market of air- as-feeder passengers who could be converted to rail-as-feeder passengers would focus on the closer-in airports, whose pas- senger volumes are lower than those from the further out airports of Detroit, Minneapolis, Columbus, and Cincinnati; given that the response of the airlines in eliminating shorter flights cannot be estimated, the rail-as-feeder mode remains unquantified. Part Four: Importance of Planning Tools and Data Availability For most of the managers and planners interviewed, the lack of tools concerning the multimodal aspects of their jobs were not seen as a major problem to support current decisions. The Airport Viewpoint For the planners at Chicago ORD, the primary need is flexi bility to deal with long-term options that might emerge at a much later date—and to avoid actions now that would preclude later solutions. During the development of the full O’Hare Modernization Program plan, airport planners devel- oped the concept of the Western Terminal to deal with long- term issues of additional gates, additional highway access, additional parking—all above and beyond the desire to plan for possible rail investments, which might or might not be made in the future. Simply stated, there would be a role for the Western Terminal whether or not data about air/rail transfer demand were correct. Their primary job as planners was to keep options open should there be an interest by others, later, to provide high- quality rail services to the airport. They did not feel a strong need to procure highly refined data on num- bers of passengers transferring from rail, as it was too early in the implementation process. With the litigation-based decision to delete the Western Terminal from the airport plan, airport planners still had a strategy to allow their airport to be served by improved rail service, should that option be proposed and financed by others. Despite the loss of Western Terminal from the airport plan, a back-up location for an improved rail connection was already proceeding forward in the planning process for the re-use of the land at Economy Parking F. As for questions about a long-term role for ORD in a sophisticated world of 220-mph rail services, airport plan- ners felt there was no immediate urgency to deal with these issues at this time; they felt that others would be examining the issue at an appropriate level. At the same time, all parties concurred that having the rules spelled out, with a commonly accepted set of planning procedures, would be desirable in principle, and helpful when these issues require solutions. Rail Planner Viewpoint The responsibility for planning rail improvements over a six-state area is somewhat dispersed and not centralized in any one place. Those responsible for implementation of the present FRA HSR corridor program felt that issues—like how to serve airports—could be dealt with at some higher level, but they were Table 6-6. Categories of diversion to HSR for ORD and MDW together. Low Diversion (20%) Medium Diversion (35%) High Diversion (50%) Trips within the study area (O-D Market) 298,576 522,508 746,440 Ground Access Market 124,547 217,958 311,369 Rail-as-Feeder Mode (not quantified) Total 423,123 740,466 1,057,809

97 With all of the needs for early project development, and the attainment of support for long-term investments, high levels of quantification for these patterns did not emerge in inter- views as a high-priority activity for the rail advocates. Several rail managers interviewed did suggest that having the planning method rules established early in the process would result in a more stable planning process. And, of course, if better data were to become available, the planners would benefit from using it. Bibliography Black, R. C. 2005 (March). “The Acela Express.” Japan Railway & Trans- port Review. http://www.jrtr.net/jrtr40/pdf/f18_bla.pdf. (As of Spring 2011). City of Chicago. 2003. Survey of Air Passengers, RSG Inc. for Wilbur Smith Associates. Freemark, Y. 2009 (September 19). “Breaking: SNCF Proposes Devel- opment of High-Speed Corridors in Midwest, Texas, Florida and California Corridors.” Transport Politic. http://www.thetransportpolitic.com/2009/09/19/ breaking-sncf-proposes-development-of-high-speed-rail-in- midwest-texas-florida-and-california-corridors/. (As of March 19, 2011). Midwest High Speed Rail Association. 2011 (May). “The Economic Impacts of High Speed Rail: Transforming the Midwest.” Highlights from the research project prepared by AECOM and the Economic Development Research Group, Inc. and sponsored by Siemens. O’Hare Airport. 2012. (http://www.cityofchicago.org/city/en/depts/doa/ provdrs/omp/svcs/about_the_omp.html). As of Summer 2012. Wardman, M., A. Bristow, J. Toner, and G. Tweddle. 2002 (May). “Review of Research Relevant to Rail Competition for Short Haul Air Routes.” European Organisation for the Safety of Air Navigation EUROCONTROL, Brussels, Belgium. SNCF. 2009. Expression of Interest and Qualifications, for the Mid- west High Speed Rail Corridor, submission to the Federal Railroad Administration. Accessed at http://www.thetransportpolitic.com/ sncf/Midwest.pdf. not united about what form such oversight would take. One interviewee commented that FRA itself made all the decisions about the content of the multi-state project, as exemplified by decisions on how to serve the airport at Madison, Wisconsin. At the early, conceptual phase of a nearly $80 billion invest- ment in a new program, there is a tendency to make the assump- tion that “others” understand how to forecast complex tasks, like the demand for longer distance HSR services as a feeder mode to air, at a quality level that could result in the cessation of feeder flights. On face value, the need to better understand the tools and methods needed to assess and understand the impacts of rail investment on air travel would seem self-evident, as the scale of impacts in the Midwest is large; the SNCF proposal to the FRA predicts that HSR rail investment would result in the diversion of 3 billion miles of air travel, with air travel delay improved by a value of 2.2 billion dollars (2009) “by taking over some regional air traffic” (SNCF, page 27). They forecast that 3.8 million trips would be diverted away from connect- ing flights at ORD, with an additional .8 million at MDW. Of the users of the high-speed train, 15% were expected to be diverted from local, point-to-point trips, and 11% from connecting air trips. Together, they forecast that over 3 bil- lion trips would be diverted to Midwest rail, from air. (Time period not specified.) Diverted trips made by users who previously used a differ- ent mode accounted for about 86% of the total system demand (58% from auto, 1% percent from conventional rail, 15% from local, point-to-point air trips and 11% from connecting air trips). Transfers of air-connect traffic to high-speed trains have been considered at O’Hare Airport (3.8 million trips in 2028) and Midway Airport (0.8 million trips in 2028). The planning documentation supplied by the advocacy groups is full of references to how this is a success in other areas (Paris and Frankfurt) and would be a success in Chicago.