National Low-Speed Rail Network Proposal

With all the focus on high-speed rail and urban transit, it’s easy to forget the low-speed rail that forms the backbone of every good national transit network. Switzerland, whose high-speed infrastructure consists of shared passenger and freight rail base tunnels, has a national rail ridership that puts the rest of Europe to shame. Japan may be famous for the Shinkansen, but the enormous low-speed networks surrounding Tokyo and Osaka are the two busiest in the world. Although intercity travel produces disproportionate revenues, most trips are local, even on mainline rail, and government rail planning should make sure to prioritize regional travel.

While the main intercity routes in the US should be eventually upgraded to high-speed rail, rather than rapid legacy rail, the low-speed network should dominate regional traffic as well as intercity gaps in the high-speed network. This means two traffic classes: regional and intercity. The intercity travel in question is for the most part short-distance – for examples, lines in Michigan fanning out of Detroit and not reaching any future high-speed lines to Chicago, lines in Georgia fanning out of Atlanta, and the portions of Amtrak California that won’t be replaced by HSR. The service level on the intercity lines should be a more modern version of the Regional, Keystone, and Empire South services; the service level on the regional lines should be the same as that of regional lines in Continental Europe.

The standards for the low-speed network should be based on the best industry practices. Because those lines are by definition not the highest-volume routes, it’s important to plan them with utmost care to keep costs under control. Federal assistance should aim to do the opposite of what FRA regulations do today. Instead of encouraging outdated practices, the federal government should on the contrary promulgate a set of good practices, based on what is done in Switzerland and other countries with good regional rail.

This is similar to what the various good roads bureaus did in the early 20th century, creating a unified set of standards. That said, the roads movement should only be an inspiration in the vaguest sense, since in reality US road building was much heavier on concrete than necessary and lighter on organization, leading up to the overbuilt Interstate network. This means that, whereas federal-aid highways are required to meet minimum standards for width and speed, federal-aid low-speed rail should be required to meet minimum standards for schedule and fare integration with local transit, signaling, and punctuality. The German motto, organization before electronics before concrete, rings truer here than for other kinds of transit investment, and agencies that ignore it should not receive funding for concrete before they complete the cheaper fixes.

Scandinavia is of especial importance as a rolemodel, because the lower density of its metro areas forces its regional trains to be faster, as they ought to be in the US. Combined with the wider loading gauge, it means that Swedish and Norwegian orders should be one of the sources of early American rolling stock. The lower speeds of Continental Europe (excluding Scandinavia) are not sufficient for more sprawling American urban areas. Instead, regional trains should have a top speed of about 160 km/h or just a little less, except on branch lines. A good example for the service quality to aim for is the Caltrain-HSR Compatibility Blog plans for trains from San Francisco to San Jose: local trains, stopping about once per 3.5 km, average 59 km/h, and express trains average 85 km/h.

While some regional lines in the US already average 60 km/h or even more, the cost is a very sparse station spacing, such that walking to stations is infeasible, even if the station areas are walkable, which they usually aren’t. For example, the Providence Line from Providence to Boston averages 58 km/h, with one daily late-night train with less schedule padding and another that skips stops achieving 65 km/h; however, the average interstation is 6.8 km, and requires skipping or closing down entirely several urban stations (Forest Hills, Ruggles, Readville, Pawtucket).

Instead of current practices, I would recommend a program of federal standardization based on the idea that transit should be able to compete with driving and provide meaningful transportation at all times of day. Federal action means that a few best practices could be violated: most prominently, rolling stock doesn’t have to be completely off-the-shelf if the federal government can induce transit agencies to combine and buy in bulk. However, the most important of the general best practices – perfect schedule and fare integration, allowing seamless intermodal transfers regardless of which agency operates the vehicles – are as important as ever. This leads to the following set of suggestions, in addition to the aforementioned set of best practices:

1. The main lines, both regional and intercity, should be electrified, with 25 kV 60 Hz.

2. Trains’ design speed should generically be 160 km/h, or a little lower on unelectrified branch lines and regional lines with frequent stops, though the track speed could be lower if increasing it is not worth the extra cost. Acceleration should be high, to allow average speed to remain high even with a few more stops. The ideal train should look like an M-7 with bigger doors from the outside and have the performance of a FLIRT. On unelectrified lines, good choices include the diesel Talent, GTW, Desiro, and Coradia. Bilevel trains are useful only in narrow circumstances in which passenger volumes are very high and the higher dwell times coming from the double-deck configurations are not a major problem; with a few exceptions such as the MI 2N used on the RER, this is practically never the case.

3. Subsidies should still be acceptable for regional services, though relative to passenger volumes they should be lower than they are in the US today; they should not be acceptable for the intercity network, though weak lines within a network could be subsidized by stronger lines they connect to.

4. In urban areas, regional service should function as urban transit and not just as peak-period commuter rail from the suburbs to the city center; therefore, there should be frequent stops in the city, replacing the longer-distance functions of American light rail lines. In-city fares should be identical to those of local urban buses and rail.

5. Regional trains should have just one operator, with the fare enforced with random fare inspections; intercity trains, which have lower traffic, can have one operator and one conductor.

6. There shouldn’t be any distinction between regional, intercity, and high-speed rail stations. High-speed rail should be able to seamlessly run through to lower-speed territory when necessary – for example, surplus Northeast Corridor trains that do not need to go to Boston should serve Jamaica at least (with catenary strung over the LIRR Main Line), and possibly even Mineola, Hicksville, and Ronkonkoma.

7. Construction projects should prioritize lines that serve markets that cars can’t, e.g. travel that passes through CBDs or parallels roads that are not freeways.

8. Signaling should be either ERTMS or ATC. Unless the two systems can be made to talk with each other, the federal government should invite delegations from the vendors, pick one, and mandate it. (And unless Hitachi can provide a convincing explanation for why its vendor-locked system is better, the pick should be ERTMS, which has eight vendors.) It can squeeze amazing capacity out of two tracks and, when enabled, provides absolute crash protection.

9. High punctuality is non-negotiable, especially when timed transfers or overtakes are involved. Trains should be able to stick to their clockface schedule and passengers should be able to rely on transfers even with short connections. Here is a list of ways to maintain punctuality. The ultimate goal is Japan, where, barring suicides and natural disasters, late trains are almost unheard of.

Those requirements are deliberately meant to be as scalable as possible. Although the rolling stock I’m implying is very ambitious for small-scale operations, the advantage of the high top speed is that such operations could piggyback on larger orders by the main established agencies, which could make great use of the extra speed and acceleration and get a more rationalized schedule as a result. The point is to give agencies pricing power coming from pooling together to order multiple thousands of more-or-less identical EMUs.

Although the investment described here is much more intensive than anything done in the US up to now, the true cost is not high. Restoring regional branch lines should be doable for a million dollars per kilometer, bulk electrification of main lines can be done for not much more and has been done on $3 million/km on the NEC, and mainline ETCS installation costs $11.5 million/km. It’s comparable to the per-km cost of the diesel-only, single-track, low-platform, commuter-only Lackawanna Cutoff, and if past results are any guide would lead to a sharp increase in transit ridership, measured in hundreds rather than tens of percent.

The ultimate goal of low-speed rail is to make it convenient to use regional transit. With speeds comparable to those of driving, local fares comparable to those of buses, and a frequent, memorable clockface schedule, transit would be a realistic option for many more people in the US than it is now. Every trip should be serviceable by transit, or else people will find it more convenient to buy a car for their irreplaceable car trips and then drive it for other trips. SBB claims that 32.7% of Swiss travel to work is on mass transit; this is higher than the figure for Greater New York, and about seven times the figure for the US.

Some of this is, to quote James Kunstler, Bill Lind, and other supporters of transit who look backward to the industrial era, merely restoring what the US had in the 1920s and 30s, before cars made all but the most traffic-intensive rail travel unprofitable. But the operating practices I’m proposing are modern, in line with today’s labor and capital costs and with innovations in countries that have kept improving their rail systems. Modern low-speed rail shares many characteristics with old local trains, but it’s fundamentally something that’s never really existed in North America. It’s about time to try it.


  1. political_i

    This post deserves a huge two thumbs up. It is a concept I would like to utilize to add more passenger rail into the country following NARPs vision for passenger rail, it would also open up a new market of higher speed frieght for shippment companies like DHL, FedEx.

  2. Stephen Smith

    Something I’ve often wondered – do you think this plan, or something close to it, would be feasible under current overall transit spending levels, assuming that you have all the power in the world when it comes to things like the FRA, FTA (Buy America), CBAs, etc.?

  3. Aaron M. Renn

    I’ve ridden the Providence Line a few times lately. I think on the whole it is very good price-performance. You can do Providence-Boston in 1:08 for $7.75. Given the distance, this is viable commuting, unlike say Milwaukee-Chicago. The MBTA hasn’t operated as reliably at Metra IMO, and their schedule frequency isn’t as good or easy to remember either. The trains do have free wi-fi though. It shouldn’t be difficult at all to upgrade this line to much higher speeds, since Amtrak is already doing it.

    • Alon Levy

      The Providence Line is decent because of the wi-fi, yeah. But the frequency is horrific; I’m sitting at the Prudential Center right now, having taken the 8:35 train to Boston to make it in time for an event at 12:30. The speed is okay for an average line, but the Providence Line uses high-speed track that Amtrak plows at 150 mph; cutting travel time to 45 minutes is a matter of new (and not particularly ambitious) rolling stock, not of concrete.

      • David Alexander

        Admittedly, that’s a problem that I’ve run into in regards to a transit planner that’s engaged to a friend of mine. We recently discussed the MBTA headways on the weekends, and I argued that the headways are pathetic, encourage auto use, or simply chase away potential visitors to the core from the suburbs. He noted that the schedules were designed to operate with one train set in order to reduce operating costs. Of course, I’ll say that we should simply run more service, but he dismissed that as it required too much operating subsidy to make it work.

        This little scenario that played out between us hints at the fact that many of the transit planners do see their work as simply a matter of keeping costs down for their employers without changing other aspects of the status quo like work rules and salary issues or rolling stock performance. Industry insiders aren’t going to arguably push for greater changes even if they’ve been exposed or made aware of better changes. I suspect that we’re going to need a new generation of transit planners willing to push harder to emulate foreign practices and be aware of the role that transit plays*, but also we need transit agencies willing to push for more concessions from employees in order to reduce employees expenses. Running lightweight UIC-ish stock can help with the scheduling, but it doesn’t solve the “engineer, conductor, and affiliated goons” problem.

        *Especially in older, more established systems with conservative practices.

        • Alon Levy

          I don’t know about the other lines, but the Providence Line operates with two trainsets on Saturdays – and, late at night, those trainsets have multiple cars that are closed off for passengers, in other words complete deadweight.

          The best way to resolve the labor issue is to keep total staffing constant, but run more trains, with shorter consists if necessary. It can’t be done with current rolling stock – as it is, the MBTA already runs late-night trains with half or more of the cars sealed off for passengers, i.e. deadweight, which means shortening the consists is not an option – but with better rolling stock, it’d be easy.

          The Swiss slogan for the issue of setting the schedule around train availability is that trains should run as fast as necessary. In the context of regional lines, it means they should be sped up just to the point that they can reach one connection node from another in an hour, or maybe get from one connection node to an outlying point in an hour, or a suitable multiple, minus the turnaround time. For the Providence Line it means making trips in 55 minutes, so that 4 trainsets can provide half-hourly service; since 160 km/h FLIRTs could do the trip in about 41 minutes including schedule pad, this means the MBTA should fix schedules so that the travel time is nicer: add many stops, send the trains south of Providence to Cranston and the airport, run with enough schedule discipline to turn trains in 4 minutes or higher speed to slightly reduce travel time and then provide half-hourly service with 3 trainsets, or plan this simultaneously with high-speed rail with timed overtakes slowing local trains down at selected stations.

  4. Alon Levy

    Everyone: yes, there are similarities with the NARP and Steel Interstate plans. The main difference, especially with Steel Interstate, is that my focus is on shorter-distance travel.

    I don’t know what the total cost estimate for a full buildout would be. It’s probably higher than the money currently available for transit, and much higher than the share that goes toward regional rail, though at the local end this could displace some of the longer-distance urban transit projects. But on the currently available money ($15 billion a year nationwide as a first-order estimate) the core suburban networks, e.g. my New York regional rail plan, could be built in 20 years or so. The less traffic-intensive lines probably can’t be done on current spending in a reasonable amount of time, but the idea is that making rail spending more cost-effective would increase the political will to spend money on regional transit.

    • Stephen Smith

      …the idea is that making rail spending more cost-effective would increase the political will to spend money on regional transit.

      This is an important point that I think a lot of “politicals” miss: One reason funding transit in the US is so difficult is because politicians see such a low rate of return on the money that is already spent. Transit activists have big plans that involve lots of money, but they want the (extra) money before they show results with the money they’ve got, and I just don’t think that’s very politically feasible.

      • betamagellan

        In part I think you’re right–even though the Milwaukee-Chicago Hiawatha service is still a diesel-engined, FRA-compliant, not-clockface Amtrak line, it’s still frequent and quick enough to achieve operational self-sufficiency and even garner reasonable support in Wisconsin’s more conservative quarters. Under Scott Walker’s watch as Milwaukee County Exec. local transit deteriorated–one of the first thing he did was to eliminate limited-stop services, and average frequencies went down–but since becoming governor he tried to get federal money for Hiawatha improvements.

        I’m also kind of skeptical, though. In general, politicians tend to be supportive of investments as long as they serve the people who voted them in, thus for example you get conservative support for extending light rail into the

      • betamagellan

        Aargh, was writing on my mobile and knocked the reply button.

        Anyway, you see conservative support for light rail extensions with large park-and-rides in places like Charlotte and Dallas, often at the expense of operating dollars that could be used to improve service in denser areas. And here in Chicago, most of the good advocacy (and occasional semi-good proposal) for improving Metra—improving off-peak and reverse commute frequencies, adding infill stations near secondary urban job centers or in dense neighborhoods, fare coordination, better utilization of ROW and not building Taj Mahals, comes from the city of Chicago, whereas the suburbs tend to be more concerned with extending coverage (there’s even suburban support and advocacy for a circumferential service proposal that would barely-developed parts of the outer suburbs with unwalkable suburban job centers—fortunately the MPO basically shot it down in the most recent regional plan). So, even though it might be possible to eventually garner political support for good intercity rail, the best urban transit investments will likely still suffer from the problem of being clustered in dense neighborhoods that are underrepresented at the state and federal levels.

    • Stephen Smith

      What exactly is $15 billion annually your guess for? Total US regional rail (i.e., commuter and intercity) money, or total US transit money?

  5. ant6n

    One should consider the way Regional operations are contracted out in Europe, to minimize the amount of required subsidies. The infrastructure is provided by the state (one way or another), and it makes a schedule and a ticketing system (i.e. Verkehrsverbund Berlin Brandenburg). The rolling stock is usually provided by the private operators themselves – and if there is one standard for rolling stock, that makes a lot of sense.

    This allows the public to control many planning and ticketing issues, while still operating for cheaper in a competitive environment, without having to buy rolling stock. I think industry best practices might proliferate faster with a little competition, and if operators get some of the fare revenue.

    I don’t know about Switzerland, but in Germany there are still conductors in regional trains – given the ticket prices, it think it sort of makes sense.

    Building a dense regional network to 160km/h is going to take a lot of money and time. While this should be a goal, it should be clear that this could take 30 years, even with very good funding.

  6. Adirondacker12800

    2. Trains’ design speed should generically be 160 km/h,

    Too low for Keystone East and Empire South. Both have sections of track capable of 125 MPH/200KPH. Keystone East should probably just use high speed equipment. Instead of reversing trains at 30 Street just send the train from Ardmore to North Philadelphia and onto New York. Some if not all of the trains to 30th Street would continue at least as far as Wilmington or all the way to DC. I’ve seen claims that electrification to Albany would save 15 minutes, which is easy to believe because Turboliners were faster than today’s trains…. by 15 minutes.

    But then Harrisburg-Philadelphia is along the HSR line to Pittsburgh and beyond and Albany-NYC is on the HSR line to Buffalo and beyond.

    • Alon Levy

      Empire South and Keystone East are in unique positions about rail speeds. The corridors they serve should both be HSR, but until they do, yes, there’s room for higher top speed than 160 km/h. With Keystone East it’s easy – trains already run through to the NEC. With Empire South, to be honest, the best thing that can be done is electrification plus tilting trains, and those tend toward a top speed of 200-250 km/h. Empire South is in the unique position that it has little freight traffic and is straight enough for about 200 km/h on legacy track, with some slow zones, but can’t be straightened further at reasonable cost due to the mountains.

  7. Andre Lot

    The Swiss case is sui generis in the way Switzerland, a country built almost entirely in valleys, dozens of that, surrounded by mountains with variable sizes and a very multi-center distribution of its population. There is a daily, large influx of commuters in/out of Basle, Bern, Zürich to each other city and everything on that triangle. Yet, SBB panders the numbers considering % of population using trains, and not the international used measure for infrastructure utilization, which is km x passenger or mile x passenger. Yet, there is a healthy and expanding network of freeways around there, including a brand-new (opened 2010) Luzern-Zürich highway and Zürich western bypass.

    The built-up environment of US is entirely different. Even other European countries like France or United Kingdom or Italy couldn’t respond well to the Swiss model because their urbanization and dispersion patterns of population and jobs are different.

    To put the other way around, it would be like arguing that the Swiss should adopt an urban freeway program like US, because US perfects the art of moving people by cars. US and Switzerland are entirely different countries, and such comparisons are not good in my humble opinion.

    Now, going to some specifics, en passant, to avoid a very long post:

    Clockface scheduled – vastly overrated feature in an age of smarphones and internet everywhere. There is enough computer capabilities do devise non-symmetrical timetables that still minimize connection times for the whole network. Even if a recurrent scheduled is to be in place, it doesn’t have to be a natural divisor of 60. There is no reason, other than pandering to lazy minds, that make a 15 or 30 minutes frequency-based services better than 17, 23 or 41.

    Fare integration – zonal fare is not the panacea many people think it is. Netherlands, for instance, recently moved ( and related links there) for a small basic fare (free transfers) + km-based add-on national scheme for trains, buses, trams, guided buses etc. In the US context, where “regional” areas of metros like Houston, LA, New York, Chicago are immensely large, you either end with too many zones, or too steep fare boundary zones, rendering certain trips too cheap and others too expensive IF your objective is, as you wrote, to incentive not only express commute traffic from suburbs to downtown areas and back.

    Station placement – lower US densities are to stay (and should stay giving its immense area), which brings, always, the problem of last-mine travel. I think an honest approach would be the combination of TOD around stations plus massive park-and-ride structures that, instead of counting on spotty and unreliable bus service for last-mile connections, rely on cars that are already there. The combination of car-sharing + park-and-ride could vastly increase the utility of regional rail, allowing the use of cars on both ends of the trip. However, that can only work if stations are spaced enough so that the average commercial speed is decent. A good idea might be placing additional stations in areas prone to redevelopment (former industrial sites, brownfields, even open space) so that, instead of dealing with recalcitrant local communities, rail development can be furthered by real estate profits of developing around stations. So many times the compromises required to build a new station limit its potential.

    HSR on legacy lines – if you are not willing to install the same signaling all over your network, you need to start balancing the costs of fitting HSR stock to run on regional lines, which are not trivial. Good connections might work better in certain cases. At least, having a small network leaves US in a comfortable position to design a nation-wide set of in-cab signaling that could be deployed from the construction. It would be like if Europe were to build its network from scratch and adopted ERTMS protocols level 3 from start. There is an advantage of starting after others, in this case meaning avoiding the patchwork of incompatible signaling, PTI, safety, electric power systems that plague countries like Italy, France and Germany.

    Financing a system – this is the major hurdle. Costs and absence of market price equilibrium aside (meaning: many regional rail systems in US don’t have any possible equilibrium that would allow them to run with a fare enough to cover direct operating costs – fare too high: people flee to cars, fare too low: even with packed trains they don’t break even), the kind of integration you envisioned is, in a larger scale, only manageable with a more central planning authority in the regional level. As ant6n pointed out, it doesn’t matter much whether the operator is private or public or a hybrid entity, but that the planning is integrated – and, here, the operations planning is more important, and difficult, than infrastructure building planning. For a variety of reasons, I don’t think there is scope for regional authorities take over local and state functions in a way that would be politically acceptable. In a realistic fashion, there should be mandated, federally, some sort of uniform electronic fare collection protocol (like there should be one for electronic toll collecting, but that is another discussion). Then, each region has to work out its integration schemes as they seem pleased. Bear in mind, though, that monthly passes are usually a money drain in any agency, as they encourage further use of the system once the installment is already paid. For a financially sustainable strategy, monthly passes need to be couple with very high one-ride tickets, like it happens in Netherlands, Germany, regional London transportation etc.

    • Alon Levy

      The headline figure of passenger-km is used only in very specific contexts, comparing national rail networks. It’s not too useful if you want to know how many people are being served by rail, especially regional rail. Compare and contrast France and Germany: France has 50% more passenger-km per capita, on the strength of the TGV network, which has higher ridership than DB Fernverkehr; but Germany has much better regional rail, leading to higher total rail ridership. And for comparisons of transit vs. driving, passenger-km masks the way the voluntarily transit-dependent arrange their life so that they can walk to their minor errands instead of drive.

      Anyway, with regards to your other points:

      Scheduling: smartphones make the schedule more easily available; they’re a replacement for pocket timetables, not for a good schedule that one can memorize and compare with other options. If smartphones obviated the need for an easily memorable schedule, we’d see the gap in ridership between systems with takts and systems without shrink in the last few years. Instead, the opposite has happened. The MBTA’s commuter rail ridership has barely budged in the last 10 years. In contrast, SBB’s ridership has surged 50% since 2001, and the passenger-km figure has gone up even more. In addition, French TER services are seeing ridership surges where they adopt a clockface schedule (“horaire cadencé), for example PACA.

      Fare integration: I’m willing to sign off on every word you say. However, as the example of the Netherlands shows, fare integration doesn’t mean zone fares. The main need for zones comes from usable season passes; Japanese distance-based passes are station-to-station, limiting flexibility.

      Station placement: park-and-rides are extremely costly. A replacement structure they’re building near Hicksville is coming in at $25,000 per spot, which means that just the parking is about 5 times per rider as many French light rail lines, and about on a par with Second Avenue Subway. They also raise operating costs, because they concentrate service in the peak: reverse-peak service is impossible since the area is unwalkable, and off-peak service is impossible since the car-train transfer is invariably annoying enough that it’s only used to bypass rush hour city traffic. Best practice for a low-density city is Calgary’s light rail system, which has few park-and-rides for the reasons mentioned here; Calgary preferred to upzone near train stations to generate reverse-peak service. More on this later, but in the US stations are often dreary even in relatively dense outlying city; Providence is a prime example of urban renewal hell.

      HSR on legacy lines: precisely… one of the reasons for deploying ERTMS in the US is that it’s cheap on an individual line, and there’s no legacy PTC to interfere with, except to some extent ACSES and that’s only installed on about half of the NEC.

      Financing: yes! Current state politics is such that it’s very difficult to create an interstate transport association. This is where federal action is required. The precedent for this is Port Authority, which was created by federal courts in response to fights over port jurisdiction between New York and New Jersey. There already exist in-state metropolitan planning organizations (MPOs), but those tend to be toothless when it comes to operations, leading to continued agency turf wars.

      • David Alexander

        A replacement structure they’re building near Hicksville is coming in at $25,000 per spot

        Said replacement structure was recently opened, and it appears to be smaller than the original structure. Admittedly, I have no evidence to prove it, but I suspect that some of the riders that were using Hicksville as a park and ride may have switched to my home station’s park and ride lots since parking has become more scarce. Of course, due to overcrowding there, I end up going to Mineola and paying to use the garage there.

        BTW, I’ll quickly note that the LIRR does not pay for maintenance for parking lots at my home station which is owned by the Town of Hempstead, or at Hicksville which is owned by the Town of Oyster Bay. In other words, the cost for the structure wasn’t borne by the LIRR, but by the town governments, and despite being heavily dependent on property taxes for revenue, there’s little political incentive for TOD.

    • Alai

      In regard to the station placement paragraph:

      It seems very unlikely to me that many people would bother with driving into a station, parking, waiting for the train, riding, finding a carshare car and driving to their destination when the the alternative is simply driving the whole way. Gas prices or traffic would have to be incredibly bad for people to see that as a reasonable option.

      As far as park-and-rides go, that’s the standard approach here in the SF Bay Area. It seems to me that once you build the park-and-rides right next to the station, plus all the access roads that are necessary to deliver the cars to them, there’s not much of an incentive to build any sort of TOD. The areas are unpleasant for pedestrians, and who would pay a premium to live or shop in the shadow of a parking structure, when they can live in a low-density house a mile or two away, and drive in to use the station?

      From personal experience, I would get off at MacArthur Station. It’s in a fairly dense area, but immediately adjacent to it is a large parking lot. Despite the fact that there are thousands of commuters who use the station, there is little evidence of prosperity for the local businesses, and it seems to me that few passengers are interested in anything but getting to their cars and leaving. And if there’s no benefit to being located near the station, there’s no profit to be made from TOD, and so financing it that way doesn’t seem likely to work.

      There is a plan to replace the lot with TOD, which I am cautiously optimistic about. However, the first step is apparently building a parking garage at over $100K/space, which is less encouraging.

      • David Alexander

        finding a carshare car and driving to their destination when the the alternative is simply driving the whole way

        If the middle portion of my trip is some type of long-distance train ride or plane ride, then of course, that may make some sense. If I’m headed to some low density area with ample parking, then it makes little sense unless my car is hideously unreliable. So the ideal option is somebody heading from the low density periphery to the high density core, but the problem is that in general, what’s the incentive to the average suburbanite to do so even if there’s limited parking at the core? Even if we build the transit infrastructure with ample headways, we need to build reasons for them to go to core as well.

        when they can live in a low-density house a mile or two away, and drive in to use the station

        Is the market for low density housing the same as the market for TOD? If you’re the type to demand low density housing, in most cases, you’d complain about the TOD because it’s “too small” or doesn’t offer a private backyard or whatever reason for supporting the choice for traditional suburban housing.

      • Wad

        It seems very unlikely to me that many people would bother with driving into a station, parking, waiting for the train, riding, finding a carshare car and driving to their destination when the the alternative is simply driving the whole way.

        With a mentality like that, it’s no wonder Americans can’t fly on no aeroplanes. 😉

        • David Alexander

          I’ve actually met an American who wondered why people fly when they can drive and bring their own car with them, so those types of people do exist…

        • Alon Levy

          There’s a very big difference between the average speed of a plane and that of a commuter train. It’s no different from how, before the Interstates and jet travel, people would drive to and from the stations to take the California Zephyr, the Super Chief, or another premium long-distance train.

    • Wad

      Snipping Andre’s points down to the main bullet points as copying them would be too long.

      Clockface scheduled – vastly overrated feature in an age of smarphones and internet everywhere.

      Clockface is tacitly demanded by consumers. It wasn’t really tyranny of the printed timetable that compelled timetables. Technology won’t change this, or “overrated” could be a justification for the provider to backslide on service.

      Think about this. A timetable with a rhythm to it calls less attention to itself than one with an irregularly divisible pattern. This is just me, but if I see a schedule with a 17, 23 or 41 minute frequency, I would think the service provider is cash-poor or sloppy at its job.

      Fare integration – zonal fare is not the panacea many people think it is.

      There are several other fare regimes besides geographic boundaries. You could have time-based fares, as San Francisco Muni has (you’re given a transfer that can be used in any direction for a set period of time), or point-to-point fares like on BART.

      Also, an integrated fare could be so due to the acceptance of a common card-based medium (say, every California transit system having a reader that accepted the HSR system’s adopted reader). Or, the rail carrier’s ticket is good for passage on a connecting system, and the agency bills the rail system for reimbursement.

      Station placement – lower US densities are to stay (and should stay giving its immense area) … . I think an honest approach would be the combination of TOD around stations plus massive park-and-ride structures …

      I agree in principle. On the details, I would first say: 1) No free parking. Land, especially near a transportation portal, is too valuable to piss away on mobility machines to do nothing. Also, free parking is a symptom of very low-value land and possibly a broader economic weakness.

      2) Remember that Transit Oriented Development is three separate concepts that must operate as one feedback mechanism. TOD does not mean you want dense housing directly atop the station, unless you’re running a peak-hour unidirectional train to downtown. Development must match the transit. If service runs at a regular interval, or perhaps even better-than-clockface, you want high-turnover traffic.

      For lower-frequency service, a sports venue or convention center would be a good TOD. For higher-frequency service, it ought to be a Class A office, hotel, college/university or museum. If you cross the clockface threshold, the above plus intense retail is a must.

  8. Richard Mlynarik

    I strongly question the notion of a “national” anything in the USA. It’s an immense, continent-spanning nation — let a thousand flowers bloom! (or, more likely, wilt.)

    Very good arguments can be made that it is rather the intrusion of federal politics and agencies (politicized, cost-is-no-object New Starts capital giveaways, FRA regulation, Buy America, APTA neanderthal culture) into all local and regional projects that encourages the wretched outcomes which we all live with and expect.

    In a fantasy world, one might imagine some urban area/region/state coming up with some idea (well, adopting well-tested Not Invented Here ideas), implementing them at lower cost and with far greater public benefit than the US National Standards for Uniform Transit Failure, and becoming an exemplar of “it can be done” and “why can’t we here in XXX have a transit system as nice as YYY?” (where “YYY” = “Portland” or “Philadelphia” or “San Francisco” or “Cleveland” or whatever) in the same way as Dutch regular-interval timetables or Munich’s 1970s S-Bahn remaking or Japan’s Shinkansen inspired others and spread over time.

    Problems with this: finding even one transit agency in the USA from which to start, one that isn’t irrecoverably sunk under Olde Tyme US Transit Dude “culture”. Since nobody in the industry even wants success (they simply want larger budgets, without exception), it’s hard to know where to begin, let alone where a good start might spread.

    But it’s vastly easier to imagine this than to imagine the FTA issuing world-cass uniform national standards for transit operations, funding only world-class projects, and funding such projects nation-wide — even if there were reason to have, say, standard S-Bahn/RER trains and standard articulated buses nation-wide, which I strongly doubt.

    Swiss transportation planning and operating culture doesn’t have much of a foothold right next door in Italy, after all.

    In short, I’d prefer one half-way-decent small Verkehrsverbund anywhere on the continent over a federal transit moon shot.

    • Alon Levy

      The problem is that small agencies simply don’t have the in-house expertise to modernize. Large ones could potentially hire the expertise, but there’s a small number of those, and they’re no more capable of reform right now than Amtrak and the FRA. In some ways they’re worse – the sausage factories in Albany and Springfield make the workings of the current Congress look downright reasonable. Congress and USDOT don’t know how destructive FRA regulations are; with Albany, even if it did know, it wouldn’t want to dispense with them unless the major power brokers got their favorite pork projects through first.

      The reason why I think federal coordination is necessary is that in some regions, there would be too much overlap between Verkehrsverbünde. It’s not really important in California, but it’s crucial in the Northeast, and useful in the South and Midwest. Caltrain can build a decent system under present FRA regulations by using time-share and kicking out UP as it’s entitled to; the MBTA can’t. And once federal action is required, it might as well also standardize rolling stock, so that if one regional operator loses 75% of the rolling stock to a sudden recall, it can raid nearby agencies’ spares without having problems with incompatible electrification. It’s not an issue in California, where even Metrolink-Caltrain compatibility is a nice-to-have, but it’s important in the eastern two-thirds of the US.

      • Richard Mlynarik

        I’m starting to sound like a right-wing strangle-the-public-sector nutcase, but it’s hard not to believe that the problem is generally too much “expertise” and too much funding.

        Here in the San Francisco Bay Area, the mafiosi have taken $400 million or so — so far — to gift the populance with a (sole-vendor, vendor-capitive, naturally) “smart card” ticketing system, for which absolutely zero justification (no cost/benefit analysis of any type) was ever provided, and of which no post-implementation analysis (“was a single extra rider gained anywhere from this investment?”) will ever be undertaken.

        That’s what happens when just a regional “coordinating” agency gets its hands on the till. (This “public” agency even issued edicts to all its “coordinated” transit agencies that they’d lose regional funding if they dared to issue paper tickets for anything but single trips and day passes — got to keep the vendor’s cash flow strong!) The problems only get worse scaling up to a state or continental level.

        The stuff that most needs to be done often comes with a negative price tag. It isn’t a lack graph algorithm experts that keeps transit lines from scheduling transfers, or a lack of bus drivers that allows 8mph average speeds, or a lack of “smart card” technology that prohibits multi-modal trips, or a shortage of rail transportation consultants that results in recommendations for high-floor FRA DMUs. They get away with this harmful nonsense because we collectively encourage and reward repeated failure in the public sector, not because we don’t have a policy-packed Federal Transit Administration.

        • anonymouse

          That’s a perfect example. The lead agency in charge of implementing things is the one agency with the least experience in the relevant fields. At least any of the local transit agency has the expertise to collect money from the public in some fashion, even if it’s not the most efficient. At worst, something chosen by one of them would die a quiet death after a pile of money was spent on development and trial deployment, and it wouldn’t reach mandatory status if it were an epic fail of Clipper’s magnitude. It’s telling that BART is keeping their magstripe tickets and VTA is actually getting its own smartcard system, which happens to be interorable with MST and SCMTD, something which Clipper and the MTC refuse to think about because it’s outside their sphere of influence.

  9. anonymouse

    Alon, I don’t get your obsession with ERTMS. By itself, it’s not going to make the signal system considerably more efficient, especially not compared to US best practice. In fact, ERTMS depends on the existing interlockings and train detection mechanisms, and in most cases in the US, that’s the limiting factor for train capacity. Having a fancy computer on the train getting instructions by GSM radio won’t change the fact that there’s just one really long block covering Ruggles and Back Bay.

    • Richard Mlynarik

      You have to pick something.

      And given that, there’s no good (in any engineering or public policy sense) reason not to pick something that exists and works adequately. (“Adequately” because the perfect is the enemy of the good in pretty much every field of engineering. If you think you’re Special and your needs are unique … you’re either wrong or you’re solving the wrong problem, nearly every time.)

      Developing life safety critical systems is eye-wateringly expensive and prone to schedule blow-outs and cost escalations and scope redefinitions that make even the most egregious civil engineering fiascos look like cake walks. So why borrow trouble and attempt to do so all over again, reinventing a wheel in a square shape, all for marginal, if any, gain? The fiasco of development of ETCS/ERTMS itself is a great example of these problems. Better to have let others undergo the pain (SBB were the ones who finally beat the vendors into functional submission) and stands on their shoulders, not on their feet.

      Don’t borrow trouble. Don’t re-invent the wheel. Good enough is good enough.

      This is a pretty good overview of everything possible being done badly:

      (It’s bizarre that you imagine or imply that other people in other parts of the world might not also have encountered the issue of train throughput limited by block length. Bizarre.)

      • anonymouse

        I’m just saying that ETCS alone does not fix anything. If ETCS plus a complete resignalling will increase capacity, and just a complete resignalling with no ETCS will increase capacity, so it stands to reason that ETCS is not the primary factor in increasing capacity. Furthermore, much of the US already uses coded track circuits for signalling, and using them for ATC is a pretty trivial matter of swapping some equipment for other, similar equipment and mounting pickups on trains. ETCS is still very much under development as it undergoes the painful transition process from a requirements document made by EU bureaucrats to a set of not-quite-compatible systems designed by different vendors to a single stable and unified platform in actual use by railroads. It’s certainly further away from feasible implementation on, say, Caltrain than a coded track circuit cab signal system, which could probably be installed in a couple of months just by swapping out parts from the existing Electrocode transmitters and receivers.

        • Alon Levy

          ETCS 2 is a lot more advanced than you’re making it look like. There are plenty of implementations worldwide, some with capacity that beats anything else available on mainline rail. The capacity pains of the North River Tunnels could be postponed 10 years if the headway could be shortened to the 110 seconds achieved in Switzerland with ETCS 2.

          • anonymouse

            There really aren’t a whole lot of ETCS 2 implementations worldwide, though the number is growing, and interfaces between different implementations are only now starting to appear and reveal the bugs in the specification. As someone who works in the software industry, I take “advanced” to mean “complicated and prone to subtle bugs that lead to catastrophic failure”, which has certainly been the case with ETCS.

            Anyway, I’m not sure how much of a fundamental difference in efficiency there is between the ETCS model, where a train shunts a track circuit, which reports occupancy to the central computer, which updates a limit of movement authority and sends a telegram to a train, which then computes a braking curve, and the CSS/ACSES model where a train shunts a track circuit, which causes a specific code to be emitted into the next circuit, which gets picked up by the train as a speed code and compared against the current speed. The total throughput still depends on block lengths and arrangements as well as latency of updates, much more so than on how the speed code is communicated to the train. Finally, you should be careful to distinguish between minimum design headway (the closest that trains could run at full speed) and actual operable headway, which needs to be maintained for extended periods and be at least a bit robust in the face of delays. I suspect that regardless of signal system, if you tried to schedule 32 trains in one hour through the North River Tunnels, you’d end up with major congestion anytime a train from DC was more than 10 seconds late.

          • Alon Levy

            There are ways to deal with schedule slack on those 32 tph lines. In the case of New York-Newark, it’s actually very easy: there are four tracks at Secaucus and even more at Penn Station, Newark Penn, and Newark Broad; a train that misses its slot (an occurrence that should be rare with good schedule discipline, which Amtrak doesn’t have and the operators in Switzerland do) can sit on the tracks there and wait. The Zurich S-Bahn, and perhaps also the rest of the Swiss rail network, inherently pads all schedules by 7% over the best technical time, which means that if a train falls behind schedule or misses a tight slot, there’s enough room to recover.

          • anonymouse

            Alon, or Richard for that matter: could you link me to a timetable on a mainline network with more than 24 tph, sustained for at least an hour? I can easily imagine some Swiss S-Bahn having 2 or 3 or 4 trains arrive at 110-second headways every half hour for connections, but that’s not the problem that NJT needs to solve here. Keep in mind, also, that the population of New York City alone is larger than that of Switzerland, and the population of metropolitan region is at least double or triple that of Switzerland, and so the traffic densities are much higher.

          • Alon Levy

            I don’t have a Swiss peak-hour timetable – I have the general national takt, which peaks at 16 tph into Zurich – but I have Japanese timetables. If you go to Hyperdia and look for trains from Tachikawa to Tokyo, you’ll see 14 trains entering Tokyo between 8 and 8:30 am, including one express train. This has two extra limits to train throughput that do not exist in the North River Tunnels, or the Swiss line in question: first, the trains have to terminate at Tokyo, on just two tracks, whereas the other segments in question have no stops at all in two-track territory; and second, the dwell times in Tokyo are very high, due to the famous overcrowding on the city’s trains, which reduces capacity. New York City Transit claims to only be able to run 24 tph on the Lex express line, compared with a nominal 30 tph capacity, and the crowding on the Chuo Line makes the 4 train look roomy.

          • Alon Levy

            I’ve just read Railway Gazette on the Copenhagen S-bane signaling project. Denmark’s existing signaling is functionally obsolete, so it decided to go with ETCS 2 on a nationwide scale. The contract with Siemens for resignaling the Copenhagen regional network is €252 million for 170 route-km, and features moving block signaling to allow 90-second headways in the central area, beating the 120 seconds achieved in rush hour with older moving block systems on the Berlin S-Bahn and RER A.

          • Richard Mlynarik

            (Not sure if this reply is going to end up threaded correctly.)

            Re Anonymouse

            could you link me to a timetable on a mainline network with more than 24 tph, sustained for at least an hour?

            I’m not sure what you’re trying to get at here.

            There are scores, indeed hundreds, of trunk routes around the world that sustain 24tph. If you’re after (as I suspect) 24 trains per track per direction per hour, you’re not going to find many examples, because that is bleeding edge, with near-zero reserve capacity for any sort of recovery from any sort of schedule upset.

            Planned London Thameslink (a neat little Piers Connor capsule summary of the hard limits here) and Crossrail (also 24tph scheduled with theoretical 30tph 2min signalling) and today’s München S-Bahn (25tph, with no alternative but to construct a new parallel relief tunnel) are examples of pushing to the edge. And then of course there’s Paris’ RER Line A, which at 30tph is well over capacity (indeed, it’s has the heaviest non-metro passengers per track per hour) despite heroic measures.

            I suspect you’re trying to make a point that NYC is Special because is is American. I’ll grant you the subway system is something else, but on the whole the regional services (MNCRR, LIRR) are nothing to crow about.

            Keep in mind, also, that the population of New York City alone is larger than that of Switzerland, and the population of metropolitan region is at least double or triple that of Switzerland, and so the traffic densities are much higher.

            Teeny Switzerland’s national network has the highest train density (trains per track per day) of any in the world.

            And for your amusement, here’s a diagram showing the scheduled passenger train arrivals and departures at the dwarf city Zürich’s main station over one day this year. NY Penn is pathetically, embarrasssingly, scandalously inefficient, NY metro population and MYMTA subway scale notwithstanding.

            Re Alon

            I’ve just read Railway Gazette on the Copenhagen S-bane signaling project.

            I think you may have taken the wrong conclusion from this.
            The S-Bane (1500v overhead metro-ish, operated as a stand-alone alongside but completely separate from the (inter)national rail system) is getting a non-ETCS proprietary Siemens moving-block CBTC system.

            The rest of the mainline Danish network is being 100% ETCS-2 resignalled in a hurry, partly for technical and end-of-life replacement reasons, partly for familiar political panic reasons.

          • Alon Levy

            Ugh… upon rereading, Railway Gazette doesn’t say ETCS anywhere in the article I quoted – it just says it’s part of a national effort and I blithely assumed it used the same technology. Never mind. My bad.

          • anonymouse

            NY Penn is pathetically, embarrasssingly, scandalously inefficient, NY metro population and MYMTA subway scale notwithstanding.
            Well, yes. And if Switzerland can be so much more efficient with their automatic block signals and intermittent inductive train stop system, then maybe ETCS is not necessary to improve capacity. And in the one real bottleneck in the NY area (and there really is just the one) through the North River Tunnels, you can’t fit more trains through the tube regardless of signalling system if you want things like any slack for schedule recovery, then ETCS is not sufficient either. Sure, it might be a good idea for other reasons, but improving capacity is not one of them.

          • Alon Levy

            You’re missing the part about how Switzerland is moving 32 tph, today, on the ETCS-signaled 200 km/h line that required such capacity.

          • Richard Mlynarik

            You’re missing the part about how Switzerland is moving 32 tph, today, on the ETCS-signaled 200 km/h line that required such capacity.

            Alon, it seems you’re making the mistake of deriving throughput by dividing 60 minutes by the minimum achievable headway.

            In fact the maximum scheduled throughput on the Neubaustrecke Rothrist-Mattstetten (between Olten and Bern, at the very heart of the entire national precision-timed system) is about 11tph, as far as I can see here

            They could put more trains through this pipe, but they don’t need to. (One of my favourite Swiss transportation planning slogans is “Trains run as fast as necessary, not as fast as possible.” Likewise for number of trains.)

            The close headways, which require excellent operating discipline at all points leading up to the junctions entering the NBS, are required by the national Takt, not because SBB enjoys flighting 4 trains in 7 minutes (which happens twice an hour, every hour) or because 32 trains are being shoved into it every hour.

            PS For those interested (and everybody should be!) in the Swiss national rail network, all graphical timetables can be found at

            PPS For my money, the 7tph that run through the network-critical single track tunnels between Zug and Thalwil are far more impressive than 11tph@200kmh on the NBS (or the 16tph on the triple track Lausanne-Renens or even the 20tph on the double track narrows (complicated by a non-flat junction for extra fun!) in and out of Luzern. But no 32tph here or anywhere else that I know of — reliability and network stability prohibit that level of over-subscription.

        • dejv

          ETCS does fix two crucial things:

          1) it safely reliably grants movement authority over several non-uniform blocks (length, slope) allowing multi-block MA in places where you could have traditionally MA only for single block (that hampers efficient operation of mixed operations with the same average speed on the same track)

          2) unlike systems based purely on coded track circuits (or any other continuous information carrier), it forces stopping at signal, so it allows granular time separation.

          Note that 2) can be accomlished by L1 LS that needs just onboard unit and trackside balise that “reads” aspect of nearby signal, making it very cheap retrofit if trains are to have OBU anyway and you don’t need real resignalling on particular line. In addition, the existing coded track circuits can be used as infill information carrier, preventing capacity loss by purely point ATC’s.

          Most of ETCS pains are coming from it implemented under oversight of some 20+ authorities with 20+ different operating rulesets – whereever was ETCS implemented in hore homogenous environment, it was successful. Because North Americas railroad environment is order of magnitude more homogenous than that of Europe and because it is regulated by a single authority, I dare say ETCS implementation would be order of magnitude smoother there.

        • anonymouse

          “The contract covers the supply and installation of Trainguard MT communications-based train control”
          Sounds like they’re using a custom signal system and not ETCS at all, which makes sense, given that the S-tog is a mostly isolated system with almost no connections to the mainline network and an incompatible electrification with the standard used on other lines in Denmark.

        • Richard Mlynarik

          … ETCS is not the primary factor in increasing capacity.

          Exemplary work on erecting and demolishing a strawman!

          Furthermore, much of the US already uses coded track circuits for signalling, and using them for ATC is a pretty trivial matter of swapping some equipment for other, similar equipment and mounting pickups on trains.

          That’s an interesting theory you have there. Especially interesting since nobody even in the US is pursuing it.

          ETCS is still very much under development as it undergoes the painful transition process from a requirements document made by EU bureaucrats …

          Cue the blue-helmeted UN storm troopers!

          … It’s certainly further away from feasible implementation on, say, Caltrain than a coded track circuit cab signal system, which could probably be installed in a couple of months just by swapping out parts from the existing Electrocode transmitters and receivers.

          You ought to getting into the lucrative Brooklyn Bridge sales market, as well as bidding on the CBOSS contract and saving us all a couple hundred million.

          • anonymouse

            Cue the blue-helmeted UN storm troopers!
            And I assume you’re posting this using ISO-standard internetworking protocols like X.25 and TP4 and so on? It’s not about politics, it’s about what works and what doesn’t. Anyway, at least we know that Amtrak is on the right track with its dedicated 25 Hz network, since so many European countries (none of them English-speaking) have similar things.

          • anonymouse

            Also, in an earlier post, Alon said that “The FRA should mandate a modern system, preferably ETCS, which permits a throughput of up to 37 trains per hour at standard speeds”, which is not something that ETCS can do, and there is just about nothing that ETCS can do in terms of pushing more trains through the North River Tunnels, which is realistically just about the only place in the US rail network that is seriously capacity-limited like that. An FRA mandate for ETCS is neither necessary nor sufficient to solve that sort of problem.

  10. Pingback: News & Notes — Greater City: Providence

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.