Environmental Assessment of Air and High-Speed Rail Corridors (2013)

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.

9 chapter three MODELING COMPLEXITY RIDERSHIP In investigating the role of air and HSR systems as competi- tors and complementary modes, understanding ridership is a crucial component. Several studies cite the importance of accurate ridership forecasts to understand the environ- mental outcomes of future long-distance transport systems (Burgess 2011; Wang and Sanders 2011; Behrens and Pels 2012; Chester and Horvath 2012). Studies also explore the sensitivity of environmental performance to ridership (Ryerson 2010; Sonnenberg 2010; Burgess 2011; Chester and Horvath 2012). Induced demand is cited as a critical input for understanding future HSR performance (Lynch 1990; Hensher 1997; Cheng 2010; Hsu et al. 2010; Ryerson 2010; Åkerman 2011; Burgess 2011; Carroll and Walton 2011). In circumventing the need for ridership forecasts, ridership has been considered parametrically (Ryerson 2010; Chester and Horvath 2012) or through assumptions. Chester and Horvath (2012) evaluate the regional environmental effects of vary- ing levels of mode shifting to HSR. They start by setting the California High-Speed Rail Authority’s adoption forecast as an upper bound (assuming it is aggressive) and compute the environmental effects at incremental decreases, showing the outcome at a range of lower HSR adoption levels. Jamin et al. (2004) consider 10 U.S. high-speed rail corridors that exhibit redundancy with 220 airport pairs. Under the assumption that rail would capture one-third of the aviation market by 2030, Jamin et al. calculate that overall GHG emissions would decrease by one million tons per year because of the mode shift. However, this assumption may be a large overestima- tion of the potential of mode shift: International Civil Avia- tion Organization found that in Europe there is a maximum of 10% potential passenger shift from air to other modes (Inter- governmental Panel on Climate Change 1999). Furthermore, Jamin et al. (2004) find that the emissions of sulfur oxides (SOx) increased by 10% as the result of a mode shift to rail because of the use of coal for electricity, highlighting the trade-offs between pollutants. Ridership forecasts are not always available to assist with environmental comparisons of future long-distance travel. Those that are available face the challenges of predicting ridership; this is well addressed by Skamris and Flyvbjerg (1997) and Brownstone et al. (2010). The number of riders a new system will attract and how competing modal operators (such as airlines) will respond to a new system are subject to modeling uncertainties. For this reason, environmental comparisons that assess emissions per passenger or trip will be only as good as the ridership models that form the basis for the analysis. Organizations looking to evaluate air and HSR systems in a government review process have taken on this challenge in different ways, with some making coarse estimations of ridership changes (for example, the Chicago HSR Draft Environmental Impact Statement). Morgan et al. (TTI 2009) performed a demand analysis of the possible HSR corridors in Texas and developed a ranking methodology for the potential HSR corridors based on demand, population density, and capacity on existing modes. The California HSR model does consider mode shifting and induced demand from air and auto travel to HSR (Cambridge Systematics 2008); however, this forecast recently has been challenged (Brown- stone et al. 2010). Work is currently under way at the FRA to develop flexible HSR ridership models that can be used by a region to consider HSR operations and costs. Assessing complementarity in comparative environmen- tal models and ridership models is not a simple proposition. For example, HSR was not explicitly modeled as a feeder mode for air in the California HSR ridership model, partially owing to the challenge of developing one modeling tool to capture both competition and complementary (discussion forthcoming in ACRP 3-23). In addition, complementarity is generally not modeled because of a research finding that HSR has little potential as a feeder mode for air. A study by Charles River Associates (2000) found that the potential diversion of connecting air travelers to HSR was less than 1% of ridership and revenue potential. As discussed in ACRP 3-23, this study was based on stated preference data. How- ever, those surveyed would not have been exposed to high- quality transit to airports in California. Since 2000, airport connections by means of transit have increased tremendously in California (e.g., Bay Area Rapid Transit to San Francisco International Airport and the express Fly Away Bus service from downtown Los Angeles to Los Angeles International Airport). These changes and the intensification of interest surrounding comparative environmental studies likely make this an important time to revisit the complementary role of HSR and the environmental impact of this role. The Northeast corridor’s Amtrak Acela service provides valuable information for U.S. HSR planners. There is experi- ence with both competition and complementarity. The clear competition in the mode shares for the Northeast corridor can be seen in Figure 2 (where rail is both Acela and conventional

10 rail operators, classic airlines and low cost airlines on each route.” Although it is not an environmental impact study, it does address issues of market share and ridership using historic data reported by air and rail operators. As shown in Figure 3, the travel time on HSR is a strong determinant of rail market share when compared with air transport. For intercity transport travel times of less than 3 to 4 hours, the rail market share is consistently higher than 50%. This high- lights the strong potential for intermodal competition and complementarity in short- to medium-haul intercity trans- portation corridors. The Steer Davies Gleave study was then used in Eurocontrol’s Challenges of Growth report in 2008 (Eurocontrol 2008). The report demonstrates that there is a recognized contribution from HSR to alleviate airspace needs, further underscoring the role of complementarity. Furthermore, Patterson and Perl (1999) found that the ini- rail) (ACRP 2009). Figure 2 shows the rail and air mode shares on the Northeast Corridor but only as a percentage of the total air and rail market share (auto excluded). Rail dominates in the corridors of less than 300 miles, and air dominates for the corridors of more than 300 miles. Embedded in these results is some experience with complementarity. United Airlines has a code share relationship with Amtrak at Newark Liberty Inter- national Airport. According to Negroni (2012), 24,000 people a year use this service, with the overwhelming majority coming from Philadelphia (a 79-mile journey). Similar results were found in a study for the European Union by Steer Davies Gleave (2006) comparing HSR and aviation ridership. In the study, eight routes with both air and HSR travel options were considered with the objective “to understand the main factors driving the market share of FIGURE 2 Mode share (and miles by driving) for certain Northeast corridor city pair markets. FIGURE 3 Rail market share (compared with air) against rail travel time for select European intercity transportation corridors. (Source: Steer Davies Gleave 2006.)

11 tions. For understanding the impacts of air or HSR systems on regional emissions inventories, energy, emissions, and noise results are often presented per vehicle-mile traveled. Many studies focus on understanding the long-run per passenger-mile traveled footprint of passengers to understand the environmental intensity of service. Per-trip measures are common and valuable for eliminating the differences in trip distances to reach the same origin-destination pairs. Regard- less, current air and HSR transportation systems that offer lower emissions might appear attractive from an environ- mental standpoint, but if such a system is unable to attract passengers, it will produce negative environmental benefits: a train or an aircraft with no payload does all harm and no good. Without ridership forecasts for corridors, environmen- tal assessments will focus on vehicle-mile, passenger-mile, or trip comparisons of air and HSR travel, likely by assuming some ridership range or average ridership. FORECASTING TECHNICAL AND BEHAVIORAL CHANGES The analysis of future air and HSR travel sometimes includes projections of energy use, vehicle technologies, and mode shifting, and forecasting is constrained by limited information for emerging vehicle technologies and ridership outcomes for corridor alignments. The environmental impacts of air and HSR systems will be influenced by future electricity mixes (Jamin et al. 2004; Åkerman 2011; Chester and Horvath 2010, 2012), emerging vehicle technologies (Janic 2003; Jamin et al. 2004; Givoni 2007; Scott 2011; Chester and Horvath 2012), and mode shifting (van Wee et al. 2003; Chester and Horvath 2012). Advanced vehicle technologies coupled with cleaner electricity inputs have the potential to reduce both future air and HSR footprints (Chester and Horvath 2012). Cleaner electricity mixes will also improve local air quality (Jørgensen and Sorenson 1997). For HSR, optimizing opera- tional characteristics such as acceleration, braking intensi- ties, maximum speed, and distance between stations has been shown to reduce the energy footprint of trips (van Wee et al. 2003). New engine technologies are expected to significantly reduce aircraft fuel consumption (and corresponding GHG emissions) and NOx emissions (Jamin et al. 2004). Optimal trip substitution distances have been computed to evaluate the GHG break-even points for air and HSR. For the Span- ish AVE HSR lines, medium distance trips (i.e., fewer than 600 miles) were shown to produce a lower GHG footprint for HSR (Rus 2011; Tucker 2012). Emerging air and HSR systems have been contrasted with emerging automobiles (Kageson 2009; Chester and Horvath 2012). Higher economy gasoline and diesel automobiles, hybrid electric vehicles, and plug-in hybrid electric vehicles can significantly reduce the GHG footprints of on-road, long-distance passenger travel (Kageson 2009). The environmental trade-offs of future long- distance travel will be largely affected by mode shifts. For new systems such as HSR, deployment to short and dense corri- dors is expected to lead to the greatest environmental benefits tial operation of the French TGV in 1981 produced drops in air passenger traffic at several major airports in France. This was dubbed the “TGV effect,” and Patterson and Perl (1999) discuss how at a journey time of less than 3 hours, significant shifts in the market will occur toward HSR, which is reflected in the French experience. Academic literature tends to develop comparative analy- ses at a commensurate spatial scale because it ignores the role of local and regional decision makers to evaluate air and HSR travel at a macro level. Academic literature often com- pares the modes at corridor, regional, or even country scale, ignoring the differences in system operators; that is, at macro scales air travel includes airports, airlines, and air traffic con- trol, whereas HSR travel often consists of a single operator. These differing decision-making layers make government environmental review that spans multiple stakeholder inter- ests challenging. For example, efforts have been made to estimate environmental effects when air and HSR systems are configured as complementary services. Janic (2003) pro- poses three configurations: (1) HSR partially replacing air on spokes; (2) HSR completely replacing air on spokes provid- ing feeder services; and (3) air used exclusively for spokes with HSR connecting airports. For each of the three cases, regions in Europe that have been configured for comple- mentary service are identified [i.e., (1) Frankfurt, (2) Paris to Rome, and (3) Paris Charles De Gaulle airport to Lyons Satolas airport]. However, an environmental assessment of these three cases is not performed. Hsu et al. (2010) discuss the possibility of increasing corridor air and HSR travel when complementary service is offered, yet no study identi- fied contrasts the environmental outcomes when competing versus complementary service is offered. California corri- dor HSR market share was investigated, based on data from Eurostar, for the conditions in which air is best substituted by an HSR system (Behrens and Pels 2012). The authors found that travel time and frequency are the major decision crite- ria for business trips, and for leisure trips, price is the most significant factor and travel time is not a major contributor. Zanin et al. (2012) find that in Spain, corridors with HSR have lower GHG footprints than do corridors without HSR, and understanding the regional interactions of transportation modes is critical for understanding countrywide effects of transportation configurations. METRICS Energy and environmental measures for air and HSR tend to be normalized per trip, vehicle-mile, or passenger-mile; how- ever, comparing across studies remains a challenge because of the different operating characteristics and goals of worldwide long-distance transportation systems. Although temporal and geographic (e.g., track alignment, electricity mixes, market demand, etc.) differences are often masked in normalized results, additional challenges remain when comparing the footprint of different train technologies and operating condi-

12 the long-distance air and HSR vehicle or passenger trip and exclude door-to-door access/egress. The overwhelming majority of the literature falls in the line-haul category, with very few studies drawing the system boundary around the door-to-door category. LIFE-CYCLE ASSESSMENT Life-cycle assessment (LCA), a framework for assessing cradle-to-grave effects, is used for assessing the comprehen- sive footprints of air and HSR travel beyond vehicle propul- sion. The discussion of new HSR systems and existing air systems created a demand for comprehensive environmental assessment frameworks to understand how upfront construc- tion or sunk environmental costs could be included in the long- run benefits and costs of different modes. LCA is needed for determining the time until payback. LCA studies include all or a subset of vehicle, infrastructure, and energy production com- ponents, in addition to propulsion effects. LCAs are expected to use an analytical system boundary that is larger than those required by governmental environmental review processes. for the region (Burgess 2011), and new analytical methods continue to be developed to determine the conditions under which this is true. ENVIRONMENTAL INDICATORS Environmental indicators are the measures by which human health, ecosystem services, climate change, resource deple- tion, and other impacts are assessed. They include those involved in a government environmental review process (see chapter four). A study may focus on a single indicator or a group of indicators. Studies that evaluate a broad suite of indicators often show that a reduction in one pollutant can lead to an increase in another. The indicators are fundamen- tally different in their significance. Indicators such as noise and criteria air pollutants have significance thresholds such that if an infrastructure project will result in pollutant levels above this threshold, the project might be altered. In contrast, GHG emissions do not have an established upper limit, and studies are generally accounting for the overall level of emissions without a threshold by which to compare. DOOR-TO-DOOR ASSESSMENTS Few studies consider door-to-door trips in the comparison of air and HSR systems but instead focus on the line haul sec- tion of trips, leaving a gap in the understanding of how the first and last mile contribute to environmental effects. The few HSR studies that exist have evaluated European con- ditions (Givoni 2007). Given the challenges of evaluating emerging HSR technologies deployed in the United States, most academic and government-driven studies identified consider only the competing legs of air and HSR travel. However, the process by which passengers access HSR sta- tions, in particular, remains unclear. Although airport access behavior has been evaluated (Airport Cooperative Research Program 2009), no equivalent studies were identified for U.S. HSR systems. The environmental effects of access to and egress from airports and HSR stations may have non- negligible effects in the total footprint of the long-distance transportation system. The two common analytical system boundaries that are used in the literature are door-to-door and line haul. The synthesis of door-to-door environmental assessments will include research that compares the long-distance air and HSR vehicle or passenger trip and information about how passengers access/egress airports or train stations. Research that falls under the line-haul category will compare only ADVOCACY DOCUMENTS The studies considering GHG and energy in an air and HSR corridor comparison are overwhelmingly from the academic and advocacy communities. Regarding the advocacy community, HSR is largely characterized and marketed by rail advocates as an environmen- tal improvement when compared with alternative modes such as personal vehicles and air transport. Rail advocacy groups maintain emphasis on CO2 emissions reduction (American Public Transportation Association 2012). However, maintaining an emphasis on CO2 reductions may ignore increases in other pol- lutants. For example, Chester and Horvath (2012) and Givoni (2007) show how decreases in one pollutant may lead to increases in another. Air advocates main- tain that the majority of GHG emissions occur from very long-haul air trips, for which there is no viable alternative to aircraft; as such, they call for air traffic management and airframe improvements (European Regions Airline Association 2011). Such a statement disregards the fact that short-haul flights tend to be the most inefficient from the perspective of fuel consumption per passenger and that HSR has the potential to alleviate delay, which is a significant contributor to fuel inefficiency (Ryerson and Hansen 2010; Ryerson et al. 2011). In the literature synthesis, advocacy documents are not synthesized because of concerns about modal bias.