202_Cyclist
Aug 5, 2013, 9:32 PM
GAO published an interesting study last Friday on airport-passenger rail intercity connections (http://gao.gov/products/GAO-13-691). Newark (EWR) appears to have the most successful intermodal intercity rail station, with an estimated 126,000 passengers in FY2012.
Air-rail intercity connections should meet the following three conditions to be a worthwhile transportation investment:
1) Airport should be a large-hub airport with sufficient passenger volumes.
2) There should be multiple Amtrak stations in a metropolitan region before locating a station at an airport, and
3) The feasibility of an air-rail intercity connection increases if the rail station is served by multiple modes (intercity rail-- Amtrak, light rail, commuter rail, etc....).
Regarding point #1, Newark is a large-hub airport served by Amtrak service on the Northeast corridor, the busiest rail corridor in the US. Amtrak had approximately 350 daily boardings at this station (http://www.amtrak.com/pdf/factsheets/NEWJERSEY12.pdf). This is a large capital investment and it is difficult to justify this expense at anything but the largest airports.
With point #2, intermodal rail stations are potentially huge economic development engines. A Washington Metropolitan Area Transit Authority (WMATA) report notes that there is $235B of property value within 1/2 mile of metro-rail stations, create an estimated $3.1B annual property tax revenues (http://www.wmata.com/about_metro/makingthecase.cfm).
Intercity passenger rail stations (i.e. Amtrak) should only only be located at an airport if the region has multiple Amtrak stations. Airports are usually located away from both the central business district and major population centers of a region. With transit, on the other hand, ridership increases if it is within close proximity of residential and employment centers.
Another related point, locating a intercity passenger rail station that is situated either next to surface or garage parking or a terminal with the related runways and airport infrastructure, limits the potential for economic development within 1/4 to 1/2 mile of the rail station. Much of the potential for the the rail station to be a significant economic generator is lost (especially if the airport operator just builds parking next to the intermodal station rather than pursuing revenue-generating uses).
With point 3, if other transit modes (commuter rail, light or heavy rail--subways) also serve the airport, some of the capital costs of the station, right-of-way acquisition, and annual operating costs can be shared with these other modes. With the example of Newark, there are 126,000 annual Amtrak boardings at the Newark Liberty Airport station. There are approximately 1.7M - 1.8M annual NJ Transit local rail boardings (http://www.ci.newark.nj.us/userimages/downloads/2013_04_05_FINAL_Newark_MobReport.pdf) at the airport.
One of the most curious parts of the GAO report is that it totally fails to look at the cost of avoided vehicle-related infrastructure costs by having good rail connections. Newark has nearly 4600 daily rail boardings at the airport (http://www.rpa.org/pdf/RPA-Upgrading-to-World-Class.pdf). The Regional Plan Association notes that nearly 50% of the Newark passengers drive to the airport in a private vehicle. In the absence of rail at Newark, assume there would be 2300 more daily private vehicle trips to the airport. Also assume that one-third, 759, of these vehicles would be parked at the airport. Structured parking is very expensive because of the materials and extensive engineering required. Estimated cost per space for structured parking ranges from $30,000 - $40,000. The cost of the parking for the 759 additional vehicles would be $22.7M. Additionally, there is also the opportunity cost that the land required for this parking would take away from other revenue-generating uses at the airport (hotel, air freight facilities, etc...).
Similarly, the GAO report fails to look at the cost-savings for taking the vehicles off the highways that are accommodated by passenger rail. At Newark, in the absence of rail, the 4600 daily rail passengers (assuming the continued mode-split of 50% private vehicles, 30% taxis) would be an additional 3680 vehicles (rough estimate assuming one passenger per vehicle). Spread from 6 AM - 10 PM, this is 230 vehicles per hour. Some peak hours, however, would likely have twice that number of vehicles. I am not a traffic engineer but at some point, 500 additional cars on the local access roads per hour might require road widening, with the related construction and land-acquisition costs.
-----------------------------------------------------------------
Intermodal Transportation:
A Variety of Factors Influence Airport-Intercity Passenger Rail ConnectivityGAO-13-691, Aug 2, 2013
http://gao.gov/assets/660/656553.pdf
Air-rail intercity connections should meet the following three conditions to be a worthwhile transportation investment:
1) Airport should be a large-hub airport with sufficient passenger volumes.
2) There should be multiple Amtrak stations in a metropolitan region before locating a station at an airport, and
3) The feasibility of an air-rail intercity connection increases if the rail station is served by multiple modes (intercity rail-- Amtrak, light rail, commuter rail, etc....).
Regarding point #1, Newark is a large-hub airport served by Amtrak service on the Northeast corridor, the busiest rail corridor in the US. Amtrak had approximately 350 daily boardings at this station (http://www.amtrak.com/pdf/factsheets/NEWJERSEY12.pdf). This is a large capital investment and it is difficult to justify this expense at anything but the largest airports.
With point #2, intermodal rail stations are potentially huge economic development engines. A Washington Metropolitan Area Transit Authority (WMATA) report notes that there is $235B of property value within 1/2 mile of metro-rail stations, create an estimated $3.1B annual property tax revenues (http://www.wmata.com/about_metro/makingthecase.cfm).
Intercity passenger rail stations (i.e. Amtrak) should only only be located at an airport if the region has multiple Amtrak stations. Airports are usually located away from both the central business district and major population centers of a region. With transit, on the other hand, ridership increases if it is within close proximity of residential and employment centers.
Another related point, locating a intercity passenger rail station that is situated either next to surface or garage parking or a terminal with the related runways and airport infrastructure, limits the potential for economic development within 1/4 to 1/2 mile of the rail station. Much of the potential for the the rail station to be a significant economic generator is lost (especially if the airport operator just builds parking next to the intermodal station rather than pursuing revenue-generating uses).
With point 3, if other transit modes (commuter rail, light or heavy rail--subways) also serve the airport, some of the capital costs of the station, right-of-way acquisition, and annual operating costs can be shared with these other modes. With the example of Newark, there are 126,000 annual Amtrak boardings at the Newark Liberty Airport station. There are approximately 1.7M - 1.8M annual NJ Transit local rail boardings (http://www.ci.newark.nj.us/userimages/downloads/2013_04_05_FINAL_Newark_MobReport.pdf) at the airport.
One of the most curious parts of the GAO report is that it totally fails to look at the cost of avoided vehicle-related infrastructure costs by having good rail connections. Newark has nearly 4600 daily rail boardings at the airport (http://www.rpa.org/pdf/RPA-Upgrading-to-World-Class.pdf). The Regional Plan Association notes that nearly 50% of the Newark passengers drive to the airport in a private vehicle. In the absence of rail at Newark, assume there would be 2300 more daily private vehicle trips to the airport. Also assume that one-third, 759, of these vehicles would be parked at the airport. Structured parking is very expensive because of the materials and extensive engineering required. Estimated cost per space for structured parking ranges from $30,000 - $40,000. The cost of the parking for the 759 additional vehicles would be $22.7M. Additionally, there is also the opportunity cost that the land required for this parking would take away from other revenue-generating uses at the airport (hotel, air freight facilities, etc...).
Similarly, the GAO report fails to look at the cost-savings for taking the vehicles off the highways that are accommodated by passenger rail. At Newark, in the absence of rail, the 4600 daily rail passengers (assuming the continued mode-split of 50% private vehicles, 30% taxis) would be an additional 3680 vehicles (rough estimate assuming one passenger per vehicle). Spread from 6 AM - 10 PM, this is 230 vehicles per hour. Some peak hours, however, would likely have twice that number of vehicles. I am not a traffic engineer but at some point, 500 additional cars on the local access roads per hour might require road widening, with the related construction and land-acquisition costs.
-----------------------------------------------------------------
Intermodal Transportation:
A Variety of Factors Influence Airport-Intercity Passenger Rail ConnectivityGAO-13-691, Aug 2, 2013
http://gao.gov/assets/660/656553.pdf