HSR and the Aerotropolis

Frankfurt Airport long-distance rail station - CC image from Heidas on Wiki

Alon Levy has a post up about the potential for high speed rail to fulfill the goals of ‘decongesting’ US airports. Alon looks at origin/destination pairs and compares the flight time to comparable HSR ranges where the technology has a chance to offer a superior travel time.

The takeaway is that the benefits for airport relief are likely bigger in California than they would be in the Northeast Corridor.  Assuming a fully built out rail system, this makes some intuitive sense: California’s big cities are arranged somewhat linearly along the coast/central valley, just as the NEC cities do along the fall line.  The big difference would appear to be the proximity of other cities – much of the California air traffic is intra-California travel:

Anonymouse in comments brings a good point about the distribution of short-haul travel within airport systems: there is often proportionately more of it at the secondary airports…

The five LA-area airports between them have 27.5% of their domestic traffic within 3-hour radius, but this splits as 21% at LAX, 35% at Long Beach, 37% at Santa Ana, 40% at Ontario, and 63% at Burbank. The three Bay Area airports between them have 19% of their domestic traffic going to LA and a total of 35% within 5-hour train radius, but this splits as 14% and 29% at SFO, 27% and 48% at San Jose, and 35% and 57% at Oakland.

In California, this isn’t a surprise, as SFO and LAX are the big international hubs.  In the Bay Area, SFO is next to the proposed line, but LAX is not.  Translating that to the NEC is different, however.  One element is adjacency – the NEC (or a short extension thereof) runs directly next to DCA, BWI, PHL, and EWR – of which only EWR is a major international hub.

This raises a few issues for HSR trips substituting for flights:  will HSR replace an entire trip, or just one leg of a trip?  If HSR is going to replace one leg of a trip, how easy does the HSR-Airport transfer have to be?  What kinds of trips would make sense for this kind of transfer, and are the best airports positioned along the line to handle them?

On the domestic/short haul flights, commenter Anonymouse writes:

In the Bay Area at least, there’s been a slow tendency for flights to get more concentrated at SFO, with OAK and SJC having mostly domestic flights, and most of those short hauls operated by Southwest. With an HSR line to LA, each of OAK and SJC can easily lose a quarter of its passengers overnight, and with good connections to San Diego, Las Vegas, and Reno, that could become half of all their traffic. SFO and LAX would lose some of their existing passengers, obviously, but not as many, thanks to the fact that they’re hubs. And that means that airlines are more likely to cut flights at the secondary airports, rather than SFO, thus filling the SFO capacity right back up with the passengers displaced from the secondary airports.

On the NEC, a similar retrenchment to the big airports would mean more traffic at IAD, EWR, and JFK.  These three are the busiest international airports along the NEC.  Dulles in particular is one of the few with room to grow, and grow substantially.  The one problem: Dulles (like JFK) isn’t adjacent to the NEC.

Maybe the more interesting case is DCA.  With no international facilities (save for border pre-clearance cities), DCA’s traffic is virtually all domestic.  DCA’s old noise-related perimeter rule also limits most of the destinations to a 1,250 mile radius.  The wiki summary of domestic destinations shows several ripe for rail substitutions (in bold):

Busiest Domestic Routes from DCA (May 2011 – April 2012)[26]
Rank Airport Passengers Carriers
1 Atlanta, Georgia 828,000 AirTran, Delta
2 Chicago (O’Hare), Illinois 697,000 American, United
3 Boston, Massachusetts 685,000 Delta, JetBlue, US Airways
4 Dallas/Fort Worth, Texas 489,000 American, US Airways
5 Miami, Florida 449,000 American
6 New York (LaGuardia), New York 362,000 Delta, US Airways
7 Orlando, Florida 350,000 AirTran, Delta, JetBlue, US Airways
8 Fort Lauderdale, Florida 315,000 JetBlue, Spirit, US Airways
9 Charlotte, North Carolina 294,000 US Airways
10 Houston, Texas 264,000 United

The same data for BWI:

Busiest domestic routes from BWI (May 2011 – April 2012)[33]
Rank City Passengers Airline(s)
1 Atlanta, Georgia 720,000 AirTran, Delta, Southwest
2 Boston, Massachusetts 549,000 AirTran, JetBlue, Southwest
3 Charlotte, North Carolina 483,000 AirTran, US Airways
4 Orlando, Florida 463,000 AirTran, Southwest
5 Detroit, Michigan 342,000 Delta, Southwest
6 Tampa, Florida 301,000 AirTran, Southwest
7 Denver, Colorado 294,000 Southwest, United
8 Providence, Rhode Island 293,000 Southwest
9 Fort Lauderdale, Florida 283,000 AirTran, Southwest
10 Manchester, NH 273,000 Southwest

These two airports would seem to be the candidates to see a lot of rail trip substitution, with Dulles remaining the region’s predominant international hub.

Again, the problem is that Dulles is not along the NEC, preventing the kind of air/rail connection you see in Frankfurt or in Paris.  Or, look at the top destination for both DCA and BWI – Atlanta.  Given Atlanta’s massive hub status, what kind of air-rail connection would be required to get passengers to use the train for the first leg of that journey? Making the connection to get an international flight at Frankfurt is one thing, but doing so in Atlanta is another – particularly if the rail and air terminals are not co-located.

Having a great HSR-Airport station at DCA or BWI is relatively easy (compared to, say, IAD).  Alon makes this observation about California:

Note, by the way, how California is planning the Oakland Airport Connector and considering an HSR station at Burbank Airport instead of downtown Burbank. Because if there’s one place Californians would really need to use HSR to get to, it’s an airport 63% of whose traffic competes with HSR.

Now, a DCA rail station has far more potential that just serving the airport (it would essentially replace the current Crystal City VRE station and could easily offer pedestrian connections to both Crystal City and to National Airport), and it would be far more than just the massive parking garage that is the BWI rail station.  Burbank (~2.5 million passengers a year) also isn’t anywhere near National Airport (~18 million passengers a year). So, what might the future look like for DCA and BWI in an age of ubiquitous HSR? Alon offers one possibility:

For New York, the best things that can be done then are to use larger planes on domestic flights, and find relief airports. In Japan, the domestic flights use widebodies, sometimes even 747s, and this has enabled Tokyo-Sapporo to grow to become the world’s highest-capacity air city pair. In the US there are more airlines and the city pairs are less thick, but there is still room for larger planes than 737s and 757s.

Changing DCA’s perimeter restriction could plausibly open the door to such a change, though security and airfield restrictions limit that to some degree. (Boeing did recently bring a 787 into DCA for display, and Delta did operate 767s into DCA to add passenger capacity prior to the Obama inauguration – would’ve been fun to be at Gravelly Point for that.) Larger planes, and potentially different destinations – if HSR changes the distribution of the airline hub model.

Commenter Jonathan English offers a competing hypothesis:

Unfortunately, major American airlines (compared with European and Asian airlines) are obsessed with frequency. They believe that they will only attract business travellers if they offer many flights per day, in many cases hourly or even more. This means that even if high-speed rail really eats into the number of people flying, airlines may just compensate by down-gauging from 737s and 320s to regional jets. They’ll still operate just as many flights and put the same pressure on runways.

Given the importance of frequency to transit, complaining about frequent air service might seem a bit ungrateful, but air travel isn’t really like transit.  Due to limited capacity and security, you must schedule in advance. Security, logistics, and airport location demands an early arrival.

Predicting what the air travel marketplace will look like in this hypothetical scenario is somewhat pointless – what if oil prices spike? What if there is political action on global warming resulting in a carbon tax or something like it? Air travel will most certainly still have a major (and a high value) role in linking these places, but exactly which places are linked could easily be disrupted.

8 comments to HSR and the Aerotropolis

  • An HSR connection to the Burbank airport makes sense to me. BUR would lose most of its short haul flights to rail, but it would make sense to add more flights to the East Coast, now that people in the Central Valley and also downtown LA will have much easier access to BUR. LAX will be relatively less affected on both sides – west side residents that consider it more convenient than BUR for within-California flights will also find it more convenient than Union Station, and it won’t get any new connection synergies for long haul passengers that want to get to San Diego or Bakersfield. (I suppose the former could possibly change if the Sepulveda Pass corridor and the East San Fernando Valley corridor become a single light rail line that runs all the way from LAX to the HSR station in Sylmar, or wherever it will be. But that rail connection would also make LAX more attractive for SF Valley flyers.)

    In the Bay Area by contrast, OAK will probably lose a lot of short haul flights (because it’s just as easy to get to downtown SF as to OAK) and not make up for it with rail connections for long haul flights.

  • charlie

    I always thought MAD-BCN was the world’s largest air bridge.

    IN terms of frequency, you’ve got it right and wrong. Yes, frequency is good, but the goodness is bascially for businesss travellers. 20% of the public, but 80% of the profit.

    And of course it that segment which is being destroyed. See Southwest’s new attraction for business travellers, the continuing rejection of RJ service by the major carriers, and the increasing tendancy for consultants to book their own flights.

    The biggest challenge of HSR is the continuing attack on the hub system by removing passenger dollars.

    So, basically, in the future, the only two airlines will be Southwest and Emirates.

  • Alex Block


    I can’t speak to the air bridge part, but I would imagine the new HSR line has cut into that a great deal. And yes, frequency is important to some degree, but not for the same reasons as transit. When push comes to shove, I wonder if the airlines will shift back to less frequent flights on larger planes.


    It’s not that a stop at BUR wouldn’t make sense, but a stop there over downtown Burbank does seem odd. BUR could easily absorb longer distance travel as a reliever for LAX, but I would guess the coordination of where the relief airports should be in LA would matter a great deal, since none of them are truly obvious choices nor do they have the passenger volumes/facilities of the airports along the NEC.

  • I’m not totally sure how much more sense “downtown Burbank” would make. I put it in quotes, because looking on Google maps, I see that that train station is just a park-and-ride lot surrounded by an interstate on one side and a water and power plant on the other side, with several warehouses nearby. There does appear to be a neighborhood with some commercial activity on the other side of I-5, but I don’t know how much of a destination it actually is. I haven’t actually been to downtown Burbank myself, but if it’s anything like downtown Glendale, then it wouldn’t generate too much HSR traffic.

    Of course, the airport location is even worse in those ways – it’s between a cemetary, the airport, and some big box parking lots, though perhaps there is some potential for redeveloping those big box stores.

    But I definitely don’t know what all the relevant issues are for these sorts of plans, and have a naive idea that if there’s an airport on the route that can easily accommodate the rail, then it might as well stop there.

  • Madrid-Barcelona was just the busiest airport pair in Europe. (It may have also been the busiest city pair – it was close to London-Dublin before the AVE opened.) The biggest markets in Asia have twice the traffic Madrid-Barcelona had at its peak.

  • To further clarify my comment quoted above, I don’t necessarily agree with the way the North American “full service” airlines operate, but they have made it pretty clear that they consider an hourly service on major routes to be essential to attract business travelers. I’m personally inclined to agree with you that a flight every two hours should be more than adequate in most cases, but they consider hourly service essential to their business model. Of course, these “full service” airlines have all been through bankruptcy while airlines that fly bigger planes less often (Southwest, Jet Blue) have been much more successful, but that doesn’t seem to change their minds. Look at Air Canada on the Toronto-New York route, for example: when WestJet started operating to LaGuardia, Air Canada didn’t reduce the number of flights. Instead, they switched from operating A320s on the route to Embraers. All I’m saying is that it would be wrong to assume that airlines would automatically reduce the number of flights on a route with new high-speed rail competition, rather than just down-gauging the plane. If high-speed rail is going to provide real congestion benefits at airports, governments can’t just assume that airlines are going to reduce the number of flights on their own as passenger counts drop. The best approach would seem to be encouraging (or even forcing) airlines to completely abandon routes where high-speed rail takes less than about 1.5 hours and connects directly to the airport, and instead to code-share with the railway. There also needs to be some kind of reform of the use-it-or-lose-it slots system at airports, which encourages airlines to operate unnecessary flights to retain the valuable slots. It’s pretty clear that there are just too many slots at airports like LaGuardia to be able to operate reliably in inclement weather, but forcing airlines to give up slots has run into legal obstacles in the States.

  • Further to above, this business model may, of course, change in the future, but it hasn’t shown signs of changing even in this time of relatively high oil prices.

  • John Jamil Brownson

    very interesting blog ,,,, looking forward to further discussion … my interests are networks of intermodal transport hubs … air, road, sea & rail … in particular adjacent port cities & a long term focus on the Arab/Persian Gulf city-states …

Leave a Reply




You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Recent Comments