Suburban Geography and Transit Modes

A post on Let’s Go LA from last year, about different suburban development patterns in different regions of the US, praises Los Angeles’s suburbs for having an arterial grid that allows some density and permits frequent bus service. The Northeast, in contrast, has a hierarchical system, of town centers surrounded by fractured streets and cul-de-sacs, at much lower density. This is how Los Angeles’s urban area has the highest standard density in the US, and one of the highest weighted densities, nearly tying San Francisco for second place after New York. It sounds like a point in favor of Los Angeles, but missing from the post is an analysis of how Rust Belt suburban development patterns reinforce prewar transit. Briefly, Western US grids are ideal for arterial buses, Northeastern town centers are ideal for commuter rail, which used to serve every town.

For a Northeastern example, the post brings up Attleboro as a historic town center. Look at the image and notice the walkable grid and development near the train station, although one quadrant of the station radius is taken up by parking. Attleboro is in fact the town with the oldest development on the Northeast Corridor between Boston and the Providence conurbation, and the only one that, when taking the train between Boston and Providence, I’d be able to see development in from the train. Sharon and Mansfield, both developed decades later, do not have as strong town centers. But conversely, many town centers similar to Attleboro’s exist in the Northeast: Framingham, Norwalk, Tarrytown/Sleepy Hollow, Huntington, Morristown, Paoli.

Now, a careful look at the specific examples of Norwalk and Huntington will show that the most walkable development is not necessarily at the train station. In both suburbs, the old town center is where the original road goes – Northern Boulevard and its eastern extensions in Long Island, the Boston Post Road in Connecticut. Huntington has a second center around the LIRR station; Norwalk has a much smaller second center around the South Norwalk Metro-North station. For the most part, the railroads went close enough to the older roads that the town center is the same, as is the case especially in Attleboro, Tarrytown, and Paoli, and in those cases, commuter rail can at least in principle serve jobs at the suburban town center.

This boils down to the difference between optimal bus and rail networks. Buses love grids: they typically serve the scale of a single city and its inner suburbs, and there it’s feasible to provide everywhere-to-everywhere service, which grids are optimal for. For the suburbs, this breaks down. Buses on uncongested arterial roads are still surface transit; an average speed of 30 km/h is aspirational, and that is for suburbs, not dense urban neighborhoods. On a road where the bus can average 30, cars can average 50, and cars can also use expressways without splitting frequency between different suburban destinations, speeding their journeys up greatly. Meanwhile, commuter rail can, depending on stop spacing, average 50-60 km/h easily, and an aggressive timetable can cross 80 if the stop spacing is relatively express.

There is no such thing as a rapid transit grid. Subway networks almost invariably look like a central mesh, often containing a circumferential line, with spokes radiating out of it in all directions. Mexico City has a larger mesh, approximating a subway grid, but its outer ends again look hub-and-spoke. Counting commuter rail, the hub-and-spoke system is as far as I can tell universal, with the exception of highly polycentric metro areas like the Ruhr. The spokes are rarely clean: they often cross each other (see for example the London Underground to scale). But looking at a city’s rail transit map, you’ll almost always be able to tell where the CBD is, where the inner-urban neighborhoods are, and where the outer-urban and suburban areas are.

At this distance, then, having a bus-friendly grid doesn’t matter much. What matters is having a good network of historical rights-of-way that can be used for regional rail, and a preexisting pattern of development following these lines and their junctions. In the US, the older cities have this, whereas the newer ones do not. In a suburb like Attleboro, good transit means good regional rail, with high all-day frequency, and a network of feeder buses timed to meet the trains. Grids aren’t especially useful for that.

And this is why, despite being so dense, Los Angeles has so little transit usage. Its street network is set up for bare-bones public transit, usable by people who can commute two hours in each direction and will never get cars. Because it was a medium-size city when its car ownership exploded, it doesn’t have as many town centers; its density is uniform. It has a higher weighted density than the Rust Belt outside New York, but its weighted-to-standard density ratio is much lower than those of Philadelphia, Boston, and Chicago. (It barely trails Washington, which has fewer town-center suburbs than the Rust Belt, but made an effort to actually build them around Metro; its Tarrytowns have Metro service rather than infrequent commuter rail.)

The optimal urban geography for urban transit is not the same as that for suburban transit, and the optimal street network for surface transit is not the same as that for rapid transit. Los Angeles could potentially excel at surface urban transit, but there’s only so much surface transit can provide the backbone of public transportation in a city. It has a handful of strong lines for rapid transit, and that’s a serious problem, which a grid won’t really solve.

About these ads
This entry was posted in Development, Regional Rail, Transportation, Urban Transit, Urbanism. Bookmark the permalink.

98 Responses to Suburban Geography and Transit Modes

  1. mike0123 says:

    An ideal bus grid in western North America might be directional in that complete coverage can be provided with only east-west routes or only north-south routes. Connections between coverage routes could ideally be provided with rapid transit stopping only at intersecting bus routes.

    This pattern is approximated by Vancouver’s north-south coverage routes combining with rapid transit on Broadway to generate high ridership. Burnaby’s rapid transit stations are placed well for a similar pattern to take shape, but the bus routes have not been adjusted to extend the pattern of north-south routes eastward. An Expo Line extension down Fraser Highway in Surrey would almost certainly follow this pattern of rapid transit integrated with north-south bus routes deep into Vancouver’s suburbia.

    In some instances, to minimize the number of rapid transit stops, two parallel bus routes can be switched as they cross the rapid transit line, resulting in a typical stop spacing of 1 mile. For example, the lack of a station at Boundary Road might result in a bus on Boundary switching to Gilmore and vice versa so that neither route doubles back.

    • Alon Levy says:

      Just as a nitpick, I wouldn’t call the north-south buses in Vancouver coverage routes. The term coverage service roughly means service that exists to provide coverage on a map, and connotes low ridership and high cost. Of the north-south buses in Vancouver, the only ones that really answer this criterion are the 15 and 17, which basically duplicate the Canada Line. The 3, 8, 16, and 20 conversely are some of the highest-ridership routes in the system.

      More in general, what you say is hard to achieve given that rapid transit trunks tend to radiate in all directions. Vancouver doesn’t just have east-west trains – there’s the Canada Line, plus the diagonal orientation of the Expo Line. So in addition to the north-south buses intersecting current or future SkyTrain lines, there are also east-west ones: the 25, 41, and 49 are all anchored at Expo Line stations at their eastern ends, and of course there’s the why-is-it-not-a-train-already 9/99. In Surrey, they’re still going to have both north-south and east-west buses hitting the Expo Line, in either the current or extended form of it.

      I didn’t deal with Vancouver’s development pattern in this post, but the specific issues raised apply to it the following way:

      1. It is a small, geographically constrained city, bounded by mountains, water and an international border. This means 30 km/h buses are not the end of the world. Honolulu is the obvious US comparison, and it too has very high transit ridership by North American standards.

      2. More than any North American city, even Washington, it could actually build town centers from scratch around SkyTrain, i.e. Metrotown. Burnaby did not have a town center thirty years ago; it does now. Preexisting walkable town centers are useful for producing transit ridership, but being able to slap office towers and shopping centers on top of subway stations is even more useful.

      3. It has high population growth and is relatively tolerant of urban redevelopment, which means that the urban geography can generally change itself based on evolving needs, unlike in the US Rust Belt or California.

      • mike0123 says:

        The spacing between the routes in my so-called coverage network does not depend on ridership. The set of north-south routes in Vancouver (and Surrey and Richmond) provides near complete coverage, while the set of east-west routes does not provide complete coverage (and this is more true in the suburbs). This was even more the case in Vancouver before east-west routes were added on 4th and 16th to relieve Broadway. There are also a few areas where the direction of the coverage network is east-west, like in Kitsilano and the most northern parts of Surrey and Richmond.

        Metro Vancouver is divided up so that contiguous developed areas are no more than about 10 km long in their shorter dimension. In general, the shorter dimension is (or could be) covered by local routes and the longer dimension has rapid transit to connect the local routes together. I agree that this circumstance stems from the city’s unique geography, and that it might not apply in a more radial city.

        In a larger city with more suburbs to its south, the Canada Line might have been built as more of a regional metro, stopping only at intersecting rapid transit lines (e.g. Broadway, 41st, Marine). In such a hypothetical city, the coverage network would consist of local routes spaced 1/2 mile apart, the rapid transit network would consist of perpendicular lines spaced a few miles apart with stations every 1/2 mile or 1 mile, and the regional metro would consist of a few lines with stations at intersections with rapid transit. The problem is that the local routes in such a network would not connect directly with the regional metro.

        • Alon Levy says:

          I can’t think of a single city that would build a greenfield subway with no stops between Broadway and 41st; even BART didn’t do that in the Mission, only in the less dense areas of Glen Park and farther, and BART is one big bundle of incompetence. Even the current Canada Line spacing, Broadway-King Edward-41st, is toward the high end for an urban subway.

          The 4th Avenue buses aren’t really coverage routes, either. When the 4, 44, and 84 merge, they form the second or third busiest corridor in the city, after Broadway and maybe Hastings. The 44s and 84s that are timed to just make the morning classes at UBC fill even before they reach Kits. If your bus route has passups, it’s not a coverage route.

          A 41st Street rapid transit route is intriguing… but probably not worth it. The 41 is very productive, but it’s not remotely close to the 9/99, and it would require more tunneling to get from UBC to Joyce, let alone Metrotown, than to VCC-Clark. There’s some nice stuff at 41st and Arbutus, but less than what there is at any proposed station on the Broadway subway, even Alma and Sasamat, and there’s nothing like Central Broadway. So the 41 and 49 have to stay buses either way, and the 9 is going to stay there while only the 99 will be zapped when the subway opens.

          Things in Burnaby are a lot different, of course. The unique geography there is that there’s no realistic way to get from the Canada Line in Richmond to the Expo Line on the surface, whereas the Expo and Millennium Lines are close enough that at many points it’s better to take a direct bus than to take the Millennium Line the long way around.

          • mike0123 says:

            I think the CP ROW on Arbutus would have been used instead of a subway on Cambie if they had instead built regional metro. The point is that for a regional metro, serving more-distant suburbs than exist in Metro Vancouver, a line connecting just to rapid transit lines might have been built instead of a line connecting to each local bus route. Instead of a network with SkyTrain and local buses, we would have a three-level network of regional trains and light rail and local buses, like the network in the Ruhr (their online maps separate the modes, so even having used it I don’t have a very good picture of the network structure). The geometry problem posed in western North America by regional metro, because of its wide station spacing, is that either the local buses converge on regional stations, breaking from the grid, or there is an additional transfer to a rapid mode like light rail for passengers to get to a regional station.

            Local transit uses the 1/2 mile grid in western North America to cover populated areas, with the routes usually spaced every 1/2 mile in one direction (coverage) and every 1 mile or more in the other direction (connecting). The more closely spaced direction is intended to put every residence within walking distance of a transit stop. A local network consisting of coverage and connecting routes is very common (maybe the most common) in western North America. It is the usually the network in Denver, Las Vegas, and Surrey, and it was the network in 1958 in Vancouver. Rapid transit with 1/2 mile station spacing makes a lot of sense in this context. Regional metro with 2+ mile station spacing (matched to connecting routes) only works with a lot of transfers and a frequent grid of coverage and connecting routes.

            The 4 is a coverage route in the sense that it is close to 1/2 mile from Broadway and has a long distance between intersecting routes (e.g. 1.4 miles from Granville to Macdonald), while the 44 and 84 are adding capacity that would likely be eliminated with a Broadway subway. I agree that a limited-stop bus might be the best rapid transit that can be hoped for on 41st.

          • Alon Levy says:

            Actually, western Vancouver is another good example of how development doesn’t follow old railroads. The CP line on Arbutus serves residential neighborhoods and the retail of eastern Kits and Kerrisdale. Cambie serves Central Broadway. Much as I’d have liked to live within walking distance of the Canada Line, it really wouldn’t have made sense to build it on that alignment. A railroad would have had to deal with grade crossings, and probably could not have run elevated for much of the way because NIMBYs, so the construction costs would have been about the same.

            The stop spacing for regional metro in places that aren’t the Ruhr is actually pretty similar to what we see on SkyTrain. The Ringbahn has 27 stations on 37.5 km, which is a slightly tighter spacing than the Expo Line; since SkyTrain is new and the Ringbahn is old, the Expo Line maintains much higher average speed, 44 km/h vs. 37 km/h. The RER has wider spacing, but that’s because it runs express in Paris proper, with the Metro serving local trips. What S-Bahns and RERs have that SkyTrain doesn’t is the ability to run on legacy single track with grade crossings in the suburbs. But in the specific case of Vancouver, first, many legacy ROWs are heavily used freight mainlines, and second, there isn’t all much suburban development next to the mainlines. BC Electric was comparatively more important, and New West was indeed a preexisting town center served by the Expo Line.

          • Adirondacker12800 says:

            Tax gas like they do in the Ruhr and Canadians might think harder about driving. Make it a bitch to park like it is in the core of cities in the Ruhr that might make them think even harder. According to Wikipedia the Ruhr has four times a many people and twice the population density of metro Vancouver. Things will look different.

          • Alon Levy says:

            I think Canadian fuel prices are closer to German prices than to American prices.

          • mike0123 says:

            The Rhine-Ruhr is an interesting example because it is a dispersed group of large cities that are too far apart for rapid transit to be time-competitive and the cities are often interspersed with farmland. The U-bahn operates like a tramway or light rail depending on context, the S-Bahn operates like faster rapid transit, the Regional Bahn is like regional metro with a few stops every few miles, and the Regional Express stops at only the larger centres.

            Regional bahn

            Of course, the Rhine-Ruhr is on a much larger scale than Metro Vancouver. The Rhine-Ruhr’s centres are much more dense than Metro Vancouver’s less-developed centres, but there is more farmland and less suburb between the centres in the Rhine-Ruhr. The Metro Vancouver density on Wikipedia is useless because much of the land included is mountainous wilderness.

          • Adirondacker12800 says:

            It doesn’t matter much if the regional train is passing through uninhabited mountains or uninhabited farmland.

          • Alon Levy says:

            No, but if the farmland is between cities whereas the mountains are all at the edge of the metro area, then it means the built-up area of Vancouver is actually a lot more compact than Metro Vancouver Regional District, whereas the built-up area of the Ruhr is not more compact than the official statistics say.

          • mike0123 says:

            Metro Vancouver’s mountainous wilderness has no population or roads or train lines crossing it. There’s a stark boundary at the edge of often-not-that-low-density suburbs and there’s nothing past it.

            The Rhine-Ruhr has farmland interspersed with populated areas so that one has to cross the farmland to get between the populated areas. There is a similar pattern in outer suburban Vancouver, except in Vancouver the areas consist of low-density suburbs instead of cities. This is the main differences in experienced density between the regions, and I think it is the reason that Metro Vancouver has mainly rapid transit and the Rhine-Ruhr has faster modes with fewer stops.

            I should correct the previous post: the first map is tramways and U-bahn, not U-bahn and S-bahn.

          • mike0123 says:

            Also, the scale is different so regional metro helps less.

          • Adirondacker12800 says:

            Gasbuddy says gasoline is 1.46 a liter in Vancouver. That works out to 5.52 a US gallon. Which works out to 5.02 in US dollars. Assuming I asked Google the right questions. Gasbuddy says gasoline is 3.87 in Washington State.

            They are 8.21 in Germany.


            Last time I checked 3.87 is a lot closer to 5.02 than 5.02 is to 8.21.

            Either Vancouver is Ruhr like or it isn’t. It’s got the same-ish population as Brooklyn. Brooklyn has one regional rail line. It’s got lots of subways. Transportation in Brooklyn is gonna look a lot different than it does in the Ruhr or Vancouver. Pick solutions that make sense and stop pining for ones that don’t fit the problem.

          • Alon Levy says:

            For some reason I was still operating under the assumption that CAD = USD, and then gas is about 1.5 euros per liter in Germany.

            Vancouver is highly unlike the Ruhr; few metro areas are like the Ruhr, least of all Vancouver, which is more monocentric than US metro areas of the same size and age.

    • Alan Robinson says:

      A transit system with parralel lines ending at a trunk is very efficiency for serving the entire city to one destination (when capacity constraints are ignored). They do not serve the anywhere to anywhere travel for which a grid derives its strength.

      In Vancouver, there is a conspicous absence in the grid along 33rd. If you’re trying to get to the Canadian Blood Services clinic on Oak and 32nd from pretty much anywhere, you’re going to have to make an extra transfer onto the 17 to reach your destination.

  2. Joseph E says:

    “There is no such thing as a rapid transit grid”
    That doesn’t mean it is impossible. Paris and Tokyo have extensive rapid transit networks. They are not pure grids, because the streets were not laid out as a grid. But they have the same connectivity. And you agree that the center of Mexico City’s system looks grid-like.
    Tokyo (Only part of the system):
    Los Angeles could have a rapid transit grid, based on a grid of surface rapid transit lines (real BRT, or LRT) to strengthen the current grid of Metro Rapid buses, in addition to the growing network of partially-grade-separated LRT and HRT.
    Los Angeles (Looks a bit like a grid in the center, no?):

    Are you just trying to say that a rapid transit grid is not reasonable to extend to low-density suburbs?

    • Alon Levy says:

      Paris and Tokyo have a tight mesh in the center, but also extend very far out of it. Here is a map of all Tokyo rail services, to scale. I do not know of a similar map for Paris, but here is one just for the RER; it’s not cleanly hub-and-spoke, and branches intersect a lot as they do on the London Underground and New York City Subway, but there’s still an obvious center. And these are both cities with dense suburbs. Paris is planning a circumferential subway connecting the inner suburbs, extending the hub out, but it’s an enormous expense, measured in tens of billions of euros. Tokyo’s railroads do not have an equivalent plan; the two recent circumferential lines, Oedo and Fukutoshin, are firmly within the inner parts of the city, since unlike Paris, Tokyo has retained most jobs within the central city, forming CBDs within the city rather than going out like La Defense.

      Los Angeles is indeed shaping up to look like Mexico City, with the Green Line and the Sepulveda/405 line forming two edges. It works fine within the LA Basin, as long as it doesn’t require too much tunneling. LA is not New York: the per-km construction costs of the Westside Subway are in “higher than I’d like” territory, rather the “what the hell is going on there?” range of New York or San Francisco or Boston, but the per-rider construction costs are fairly high, since LA needs long lines to provide reasonable ridership. The Vermont Subway idea has the same problem, which is why it’s so precious that it can be done above-ground south of Gage. The issue then is that the Green Line is in a freeway median, with all the problems that entails, and more in general it’s expensive to build lines that are grade-separated unless there’s a freeway or a wide enough arterial.

    • Tcmetro says:

      The proposed Bangkok rail system is fairly grid-like, albeit with a number of nuances that arrive when there are three different (government!) agencies building and operating a rapid transit system. There is no real center to the system in that there are only two stations that will serve three lines, and one of those is deep in the suburbs. Many circumferential lines are proposed fairly far out from the city center as well.

      Here are a couple of maps:
      To scale:

      Of course, Bangkok is a radically different paradigm than many cities. It reminds me of LA, except even more decentralized, but with higher densities.

  3. Eric says:

    1. A number of big cities have or are building a gridlike subway network. Obviously, you have to be a very big city for that much subway construction to be justified (multiple lines in each direction). But for cities of that size, it’s rather common. Not just in Mexico City.

    2. Just like the Northeast US, Los Angeles has radial commuter rail – Metrolink. Why did it not merit even a mention in this post? Because it is lightly used and has not led to clustered development around its stations. Why is this? Because downtown LA is less of a business center than Northeast downtowns; because Union Station is a long walk from downtown; and perhaps because zoning around the stations prohibit dense development.

    I think the first factor is most important. I assume people always use cars except when impractical. LA’s downtown is small enough that it is practical to drive there. Manhattan has been extremely dense for a century, and it was never practical to commute to there by car. Everywhere else, in the LA area and in Northeast suburbs, cars are preferred. If traffic continues to grow and LA’s bus grid is given separate lanes, then LA too will have high transit usage. In that case, it will become clear that grids are not incompatible with high transit usage.

    • Alon Levy says:

      No, Beijing and Moscow are both hub and spoke. Moscow has an inner ring and a lot of spokes radiating out of it. Beijing indeed has two rings, but outside the outer ring it’s the same hub and spoke. See maps here and here.

      Re Metrolink, it’s not just much lower-ridership than the Northeastern systems; it’s also smaller in extent, for example there are only two lines in the Valley. There are fewer commuter lines, even deactivated ones available for reactivation (i.e. Harbor Subdivision), than in Boston and Philadelphia. It’s not even that the stations were not developed: Burbank is a secondary CBD, right next to the station. Rather, it’s that LA grew when cars were already available, which meant among other things that people tolerated Union Station’s edge-of-CBD location less than they did Penn Station and North Station’s. The same reason leads to LA’s less developed mainline rail system.

      Finally, I don’t think grids are incompatible with high transit usage at all! I just think that, at larger scale, they aren’t everything, and there needs to also be longer-range transit underlying everything.

      • Eric says:

        1. Besides the two rings, Beijing has lines 13/14/15 which also run tangentially not radially. Overall, most of the system is grid-like.

        Similarly in Moscow, the construction of four separate new ring-like lines (11, 13, Little Ring, and Alekseyevskaya) will make the whole system a lot more grid-line, although the spokes outside the rings will still be quite long.

        2. Grids lend themselves to uniform density and hub-and-spoke systems to spiky density. I guess what you are saying is that LA does not have high enough uniform density for high transit use. But in the spiky parts of the Northeast (and not elsewhere), the density is high enough.

  4. Aaron M. Renn says:

    I think the real differentiator is not the suburban development pattern, but the presence of a large employment base in the CBD. None of these suburban stations would exist without that. The development patterns of the entire region are linked.

    • letsgola says:

      Modern CA suburbs are still denser than their East Coast counterparts, though. San Francisco can probably be seen as something of a hybrid – it has a big CBD like an East Coast city, but its suburbs (at least, where not constrained by urban growth boundaries) are quite similar to LA’s. Are there any modern East Coast suburbs that would comparable to, say, Antioch or Eastvale?

      • Adirondacker12800 says:

        Modern CA suburbs are still denser than their East Coast counterparts

        No they aren’t

        • letsgola says:

          Eastvale is 11 square miles of pure suburb and it’s 4,700 people per square mile… and it’s still getting denser. Where is a suburb like that being built on the East Coast?

          • Neil says:

            That’s not exactly a fair comparison. There are few suburbs being built from scratch in the Northeast. Eastvale densities aren’t hard to find NYC auto-era suburbs (such as Nassau County as mentioned that are not legacy commuter rail burbs. But there aren’t much of any greenfield NYC suburbs (or most places in the Northeast) being built of any kind, so it’s hard to make a comparison.

        • letsgola says:

          In fact, Eastvale is denser than many legacy East Coast commuter rail suburbs, like say Attleboro or Paoli.

          • Adirondacker12800 says:

            It’s got the same density as Nassau County NY then?

          • letsgola says:

            Yes, the difference being that Nassau County’s population hasn’t changed in 45 years.

          • letsgola says:

            To clarify, the original post was intended to be about differences in suburbs currently being built around the country. Auto-oriented suburbs on the East Coast don’t look like LA, they look like Atlanta.

          • Adirondacker12800 says:

            well yeah, it’s kinda hard to build Levittown unless you have a potato field to plow under. Building a dozen McMansions 1000 feet farther away from the station here and two dozen there 1500 farther away from the station won’t look like Levittown. 60 years ago they were busy building Levittowns on the East Coast.
            Get back to us when they start building Great Neck Plaza within walking distance of the station. Though you need a station to be able to do that.

  5. I disagree that buses can’t achieve subway-like speeds, but only if they also use the expressway grid. Golden Gate Transit generally hits speeds of 40km/h using freeway bus pads at exits and can get up to 70km/h for some skip-stop runs. The problem, of course, is that freeway BRT or freeway rapid bus service is on a freeway and, therefore, encounters all the problems of freeway-running rail.

  6. Pingback: The Plan to Transform Houston Transit Without Spending a Dollar More |

  7. letsgola says:

    In defense of the original post, the intent was more to compare recent patterns of suburbanization, since I think the differences are underappreciated. Downtown Attleboro is great, but Massachusetts hasn’t built anything like that in what… 70 years? New development in Attleboro is unfortunately not being built near the center. Instead you get mini-estates like this. In the entire city, the densest residential zone allows duplexes on quarter-acre lots by right.

    The same is true of, say, Paoli. The zoning ordinance isn’t free online, but the township containing the station added all of “270 people between 2000 and 2010, and is about as dense as Lancaster, which includes enormous swaths of undeveloped Mojave Desert in its city limits.

    That’s not to say that development in Fontana is ideal for transit but at least it has a chance. I don’t see the same potential in suburbs currently being built towards the edges of Northeast metros, like southern New Hampshire or southeastern Massachusetts. Most California cities are happy to be thoroughly suburban, while New England towns worry about maintaining a village or rural feel.

    Returning to the larger question of your post, if existing development in Los Angeles will work well for transit, I’d say the jury is still out. We don’t know because no one’s really tried. While much of the LA Basin and the older parts of the Valley developed along streetcar or interurban lines, there’s obviously been an enormous amount of development since the last one of those cars rolled. For example, will commuting from Reseda to UCLA or Santa Monica by transit be practical? We don’t know, because we’ve never had a transit service between the two worth riding if you could afford a car.

    The other side of low transit ridership in LA is the promise of free (or very cheap parking) at both ends of your trip. Transit in LA is often stuck in the same traffic as cars, and if parking is free, there’s very little incentive to take transit. My current carpool ride to downtown LA is paying $90/month to park at 9th and Figueroa, right in the booming part of downtown. I have to think that if you could park for that price at Grand Central or Penn Station, it would put a dent in commuter rail ridership.

    We’ll probably find out over the coming decades if you can make transit really work in a place like LA. It’s a different sort of city that’s going to require a different sort of transit network. Personally, I think that a rapid transit grid on LA’s arterials will work better than many expect. With the exception of parts of the Westside and Central LA, the rights-of-way are wide enough to provide dedicated lanes. Again, it just hasn’t been tried.

    Transit services planned like traditional commuter rail will undeniably always be a challenge in a place like LA. We just don’t have the huge employment center that works well with commuter rail. I’m working on (or claiming to work on, I keep ending up writing other posts first) a post about options to convert some Metrolink lines to services that would be more appropriate for LA. As you note, Burbank is right on a line to downtown; we ought to be able to do better than we are at the moment.

    On the other hand, I’m not sure I can be as optimistic about feeder services to commuter rail stations in the Northeast. I’d be happy to be proven wrong but it would seem to be difficult to lay out cost effective feeder bus routes to Paoli or Hanson or Sharon. And it would seem to be very difficult to lay out cost effective transit that doesn’t serve the city center, such as to commercial nodes like King of Prussia or Burlington. Many of these legacy commuter rail stations serve places that were specifically designed to be semi-rural bedroom communities and that history is hard to shake. The one exception is the Washington DC area, which really has made a concerted effort to channel development to WMATA stations.

    The really funny thing is that if you could drop a place like Eastvale on a place like Sharon, you probably would be able to design some good feeder bus routes, but no one seems to really have that option (well, maybe the older industrial cities of MA – they should try that if they’re not already doing it).

    • Adirondacker12800 says:

      if existing development in Los Angeles will work well for transit, I’d say the jury is still out. We don’t know because no one’s really tried.

      All the parts of it that are trolley car suburbs will probably work. Wide swaths of it are trolley car suburb.

      I have to think that if you could park for that price at Grand Central or Penn Station, it would put a dent in commuter rail ridership

      Barely if at all. Sitting in traffic for an hour at the crossings wouldn’t go away.

      • letsgola says:

        What do you think about the vast swaths of LA that weren’t streetcar suburbs but are overlaid on that pre-existing pattern?

        • Nathanael says:

          Honestly, it’s going to take a while before we even get to them. Just restoring decent levels of service to the streetcar suburbs of LA is going to take a loooong time before we have to look at anything else.

          • Adirondacker12800 says:

            ..well back when the trolley cars in Los Angeles covered 25 % more miles than the NYC subway you could, if you were a masochist, take the trolley from Chicago to New York. The legend has it that there was a mile and half walk somewhere in Upstate New York where the trolley lines didn’t connect one way or another.

          • John Hupp says:

            I think the moral of the linked story is that LA’s Red Cars failed for the same reason as most streetcars: it was cobbled together as a real estate development subsidy with little regional planning and no high-speed trunk line capacity. Thank god somebody else paid for it, though, so we aren’t still stuck paying off bonds.

          • @Adirondacker12800: There were two gaps, one of about 80 miles (as the crow flies) from Tarrytown, NY to Hudson, NY and another of about 25 miles from Fonda, NY to Little Falls, NY; I suspect the NYCRR had something to do with both of these. From Little Falls there were continuous trolley routes west to Chicago (and beyond, past Milwaukee). North from NYC trolleys reached as far as Fairfield, ME; south from Jersey City they reached to Wilmington and Harrisburg. is a book from 1915 about all the places you could go taking only trolleys (there were, of course, transit geeks then as now); has maps from around that period. Of course, many of the lines involved were not particularly cheap or frequent, but I think the same goes for some of Pacific Electric’s longer lines (I believe LA-San Bernardino ran 10 trips each way per day).

    • Alon Levy says:

      Well, Sharon isn’t Attleboro – Attleboro has a bigger center. When I’m on a Providence Line train at Attleboro, I see development from the station; at Sharon (and Mansfield, and Canton Junction) I see only parking and maybe the odd single-family house.

      Anyway, if the town center is located at an old crossroads, you can run a bus on the road perpendicular to the railroad. If there is a pre-railroad town center and a rail station area, as in Huntington and Norwalk, you can also connect the two centers. I can see a bus running on New York Avenue that’s timed to meet at least the westbound LIRR, with the same frequency on the trunk between the old town center and Huntington Station as the LIRR off-peak takt (which doesn’t currently exist, of course), with lower frequency on various branches of the bus line. At Attleboro I can see a bus running on Main Street.

      The issue I have with LA is that surface transit will always be surface transit. Even with dedicated lanes, there’s a limit to how fast it can be without absolute priority at every intersection. In Vancouver, the 4th Avenue express buses average 30 km/h between Vine Street and UBC: they do 7.3 km in 15 minutes. This is with only 7 stops (of which they’ll typically skip about 2 because no riders will want to get off or on). 4th Avenue isn’t congested except at rush hour, and all traffic on it has permanent signal priority over intersecting traffic, complete with pedestrian beg buttons. West of Alma the street has a suburban feel, and for about a kilometer and a half on Chancellor Boulevard the buses run through a park. The park and the low density west of Alma ensure very high housing prices for less than premium housing, pushing people farther out, so people who don’t get paid a postdoc salary have hour-long commutes (in Vancouver, that’s considered a lot). You can improve effective access by raising densities – i.e. picking a fight with the Point Grey NIMBYs about allowing redevelopment – but then your bus has to fight more traffic and no longer averages 30 km/h, even with dedicated lanes (Parisian tramways average about 18).

      That’s the advantage of rapid transit: even the slow legacy subways that average 30 km/h can do so in very dense environments, so the access is high even if the mobility is superficially the same as your suburban bus.

      And that’s the problem with Eastvale. Nobody there has any reason to take a bus – cars are faster. It’s possible to run connecting buses to East Ontario Metrolink, but there’s nothing at East Ontario worth running buses to other than the station. Corona is more of a town center, but it’s on the 91 Line, the most freight-primary and hard to modernize of the Metrolink lines.

      What do you think LA could do with Metrolink, anyway? To me, several lines seem like fairly straightforward modernization cases – many are passenger-primary at least. Metrolink is already working on run-through tracks; add electrification and you have a couple of diagonal-shaped through-services from the Valley to Orange County. The current HSR plans may be implicitly counting on such a program in the first place, since the official Initial Operating Segment plan takes HSR to the Valley but not to Union Station in order to avoid explicitly saying “we will blend service with Metrolink.”

      It’s not the best, because the development in the Valley doesn’t really track the Metrolink lines, except in Burbank. That’s the problem of LA’s relatively uniform, relatively new development – the rail backbone doesn’t serve the existing secondary centers that well (again, Burbank is an exception and LA should be paying more attention to it than it is currently doing). I guess the Harbor Subdivision serves existing density better because of Inglewood, but that requires actually thinking hard how to serve LAX, and may require reconfiguring the entire airport to be serviceable by a train station without an additional people mover. The Inland Empire is also hard – not in the same way the Valley is, since San Bernardino actually has a center, but the development elsewhere doesn’t really follow the tracks, as in Eastvale.

      • Adirondacker12800 says:

        reconfiguring the entire airport to be serviceable by a train station without an additional people mover.

        There’s too much terminal at LAX ( or O’hare or Newark or JFK or Hartsfield or .. ) to be served by one train station. The people mover keeps the shuttle buses to the parking lots and car rental out of the terminals which is useful. The cell phone lot and the kiss-n-fly can be remote from the terminals. So can the hotel shuttles if it gets really crowded.

        • Nathanael says:

          Unfortunately, LAWA seems intent on designing the worst and stupidest possible people-mover. I could sketch out a fairly decent pair of peoplemover routes: one loop which stopped at all the terminals, and a longer route which stops within a block or two of all the hotels, at most of the remote lots, and at the Metrorail station, but LAWA is instead coming up with looney routes which won’t replace the buses.

          • John Hupp says:

            If only LA Metro and LAWA were the same agency, like the Port Authority in New York, then we’d get obvious solutions like the AirTrain (It stops at like every single terminal! And it connects to a major intermodal hub! But it still doesn’t provide a single-seat ride into Manhattan! *wah-wah*), along with a nice, multibillion-dollar real estate boondoggle. Or perhaps a PATH train to Newark Liberty?

            More ideally we’d probably get something like the KLIA Ekspres in Kuala Lumpur, where you can through-check your bags at the Malaysia Airlines desk at KL Sentral before taking the 90mph express train to the airport. (Incidentally the RapidKL LRT lines use the same technology as the AirTrain JFK, though the KLIA Ekspres is different.)

            Like, duh, we should have through-checking airline counters at Union Station, with a crazy-ass maglev to LAX, or something.

          • Alon Levy says:

            Just as a small comment: KL is basically the LA of the Asia-Pacific region, and Malaysia is the US. It has a network of expressways and not much transit ridership for its size, due to decades of government policy of promoting road construction to create an internal market for the state-owned car industry. It also has puzzling things like splitting its not very large rail network among four different operators. So I wouldn’t trust its transit construction priorities.

            I’m more interested in a solution similar to that of Zurich, Frankfurt, and Paris: there’s a train station at the airport, which serves both regional and intercity trains. People may be able to check their luggage at the train station they board at, or they may not; it’s less important than having a train station that allows people to check in for the flight and go through security without passing through additional hurdles like a people mover transfer. People mover transfers are a major slowdown, especially over large distances. A service disruption made me discover that the fastest way to get from Manhattan to JFK is to take the E to Forest Hills and a taxi to the airport, which does the Forest Hills-Terminal trip in 13 minutes.

          • John Hupp says:

            I usually stay with my cousin in Forest Hills. I’ll keep in mind that I should take a taxi, rather than the AirTrain, next time I visit her. As you may know from Chicago, the Blue Line to O’Hare is quite direct (even if it’s slow), and sometimes the trains will make an exception for you and continue on past the last station…×468.jpg

            KL’s similarity to LA was pretty obvious to me when I was studying abroad there from LA. Blog comments are weird in that replies tend to tell people what they already know. Also, I have a very dry sense of humor… How about we build a maglev to nowhere, like Shanghai? Perhaps I should qualify that with “lol”, because I’m being facetious.

            Mainly I mentioned KL because I like it despite its constant frustrations,—I like LA, too!—and because of the factoid that RapidKL uses driverless Bombardier ART people-mover technology for a long-distance line, which is weird, and is the same technology as AirTrain JFK.

            KL also has a frustratingly opaque planning process. All official government documents are in English; they just don’t deign to tell you much of anything. Like the new MRT there is sort of kind of just happening, whether anyone likes it or not. Some things get built and never used, like the monorail suspension bridge in Putrajaya. And, hey, let’s build another flyover because fuck pedestrians, and we can plaster it with 1Malaysia logos!

          • Adirondacker12800 says:

            Or perhaps a PATH train to Newark Liberty?

            Nobody calls it Newark Liberty unless their employer tells them they have to. It’s just Nork Airport.
            They have been talking about PATH trains to the airport since the 70s. The Port Authority might actually do it sometime soon. They are running out of space to store cars and the next available space big enough is out at the airport. Since they have to drag the tracks out there anyway they might as well build a station. Space was allocated for PATH when they built the airport train station.

            Forest Hills and a taxi to the airport, which does the Forest Hills-Terminal trip in 13 minutes.

            The fastest way to any of the airports is to hail a cab or call for one. Faster than driving because you don’t have to park the car. Don’t schedule your flight for rush hour. But you know not to schedule your flight for rush hour because that’s when airport delays happen. All of the airlines schedule their flights to leave at 9:00 or 5:00…

          • John Hupp says:

            I call it Newark Liberty because United calls it that, and sometimes it’s cheaper to fly there instead of JFK. Being from Chicago and living in LA, I think it’s annoying how United doesn’t really fly to JFK…

          • Alon Levy says:

            I fly long-distance, usually internationally; my flights leave whenever is convenient based on time zones, which is when it’s convenient for everyone else based on time zones. Even for a 9 pm flight, subway or LIRR to AirTrain will beat a taxi from most Manhattan destinations.

          • John Hupp says:

            Wouldn’t it be nice if you could check you bags for United flights at the Amtrak counter at Penn Station? I mean, they do code-share, don’t they?

          • Adirondacker12800 says:

            There are other airlines besides United. If you are going to Forest HIlls LaGuardia is an option.

          • John Hupp says:

            “Being from Chicago”, no, United is the only airline in the world! And I like riding trains…

          • Adirondacker12800 says:

            You can ride the train if you go through LaGuardia. It’s just that you have to use a bus to get to the train station, Apparently United doesn’t fly to JFK much because they are busy flying to LaGuardia, More or less once an hour from 6 AM to 8 PM.

          • Adirondacker12800 says:

            the people checking bags at Penn Station are the land cruise passengers. People who aren’t on a cruise have to carry their own onto the train.

          • John Hupp says:

            Yes, that is the present-day situation, but through-checking from Amtrak to Newark and/or JFK would be awfully convenient if you were traveling from, say, Stamford to Stuttgart.

          • Alon Levy says:

            Checked baggage means longer dwells at stations. It’s fine for multi-day Amtrak land cruises; it is very much not fine for what’s supposed to be high-speed rail.

          • John Hupp says:

            I’ll point out as an occasional land cruiser myself (not the Toyota variety), you probably don’t want to check your bag for a long rail journey because you’ll want to access the contents of your bag during the course of the trip. So I guess checked bags are overrated for everyone?

          • Adirondacker12800 says:

            It would be much more convenient to check them in Stamford. Wrestling them off the train in New York so that they could be checked there instead of in Stamford doesn’t sound like much fun. Or rolling them off the train – from Stamford – in Newark Airport – that was designed for people with luggage. No to mention that wrestling them off the train and getting back on the train might not happen.

          • John Hupp says:

            I do not believe we are in disagreement!

          • Nathanael says:

            Just for reference, LAWA is proposing a people-mover which will not take you from one terminal to another. I really don’t understand LAWA.

      • letsgola says:

        For sure, the freight is a big challenge on a level not really faced by any other US commuter rail operator.

        There’s definitely potential to do some good transit with the rest of the Harbor Sub, but I don’t think it’s worth the money to rebuild the entire airport to get rail there. Rail to the airport might only make sense if there’s a big CBD at the other end of the line.

        The 30 second version of the “improve Metrolink” post (since I’m apparently never going to actually write it ;) is to connect the Red/Purple Line at Union Station to the passenger-oriented Metrolink lines and run service like Seoul Line 1. Service pattern would be something like Purple Line – SB Line, Antelope Valley Line – Orange County Line, Red Line – new line funded by Measure R2 south then southeast through Gateway Cities (South Gate, Paramount, Bellflower or South Gate, Downey, Norwalk).

        Any preemptive suggestions for that idea would be welcome!

        • Alon Levy says:

          Are the loading gauges at all compatible? I’ve always thought Wilshire-SB made a very good retro-fantasy, but I don’t know whether there’s any compatibility between 25 kV 60 Hz 3.2-meter wide mainline trains and the Red Line. Dual-voltage EMUs exist of course, but the train dimensions might be a problem.

          As for airport rebuilding, about the most comprehensive rebuild I can think of is actually not much more than what seems to be on the table with the ITF and Terminal 0. The idea is to have a single groundside terminal, roughly where the Terminal 0 plan is, at Sepulveda, and drag mainline rail there, on a viaduct over Century if necessary. Maybe have additional shuttle trains provide extra frequency from the Crenshaw Line. The point here is that Terminal 0 should be the only groundside terminal, with the current terminals turned into airside terminals, in similar vein to the terminals of Atlanta.

          • letsgola says:

            Loading gauges is the big question. Red Line vehicle half-width is given as 5′ 2″ and platforms are specified as being 5′ 4.5″ off center & 3′ 8″ above top of rail. So for a 3.2m wide vehicle to sneak by would be tight. Of course, Metrolink equipment doesn’t use the same boarding height. Hope would be that width issue could be overcome w platform extenders if necessary, a la old school South Ferry. (CPUC’s clearance requirements notwithstanding.) Red Line cars are 75′ long & Metrolink coaches are 85′.

            Regarding LAX, any Terminal 0 or ITF at LAWA’s location (Airport & 98th) would probably require some sort of APM. At Sepulveda/Century might not need it for air side (leaving ConRAC out of picture). Rather than bring Crenshaw Line to Sepulveda/Century, I’d rather see an APM connect to both Crenshaw Line & a future Sepulveda Line. Thoughts?

          • Adirondacker12800 says:

            so instead of taking the people mover from baggage claim to the train station you take the people mover from the gate concourse to the far end of baggage claim and hike to the train station? and the taxis. Or the hotel shuttle bus. or another trip on the people mover to the car rental

          • Joey says:

            letsgola: For the purposes of these types of discussions, it’s probably worth completely ignoring the current Metrolink fleet and platforms (most of which could not be level boarding for any type of rolling stock). The issues I see are these:

            – Horizontal clearance is probably fine, but vertical clearance is much more of an issue. I haven’t done a detailed survey, but my understanding is that, like most American subways built in the last 50 years, there’s very little clearance between the tops of the subway cars and the ceilings of many tunnel structures. This means that you probably couldn’t accommodate a pantograph, even if it was folded down, and you’d have to lay third rail everywhere you wanted through-running.
            – IIRC the Red/Purple line platform height is in the vicinity of 3 feet above the rail. This is never going to compatible with HSR, and honestly, compatibility with HSR is probably more important for Metrolink than compatibility with the Red/Purple lines.
            – It doesn’t really make sense to send these trains all the way to the ends of most Metrolink lines. On the inner parts of these lines where it would make sense, there’s enough demand for additional service patterns that you’d need additional tracks either way. In some cases it would make sense to deviate from the rail ROW entirely – for instance if you were sending trains to Burbank it would probably make sense to route them via downtown Glendale.

            If the Red Line was built to RER standards to begin with this would make sense, but that isn’t the case. As a possible alternative might be a sort of “Metrolink DTX” tunnel to at least route Metrolink trains directly into downtown LA (via 9th and Grand to Union Station or something like that), but that in itself is questionable because (a) LA isn’t very centralized and it wouldn’t provide direct access to jobs on the west side (b) It would more or less parallel both the current Red/Purple lines and the Regional Connector.

          • letsgola says:

            Joey: current Metro criteria is for 14′ vertical clearance above top of rail in bored tunnels for heavy rail, based on a 12.5′ tall vehicle dynamic envelope. I’m not sure if the existing Red Line tunnels were built to that standard. If it proves to be impossible to design a vehicle that could use the existing tunnels w/ pantograph locked down, that probably kills the plan. I don’t think anyone wants to start throwing down 3rd rail on surface running lines w/ at-grade auto & pedestrian crossings.

            Platform height is 3′-8″.

          • Adirondacker12800 says:

            The suburbanites would object when the people already using the tunnel try to sit on their laps.

      • Neil says:

        The problem with many of the town centers mentioned is that the residential areas near the town center are low density, there’s not many in walking distance of the town center and the population is too spread out for anything resembling a high-frequency network of local transit (which would be buses) to work. I’d guess a large majority of visitors to Huntington Village drive, though once visitors park they walk. The radial train stations are used mainly by those going to the city center. Western Nassau has the density to support usable surface transit, though still most drive. Even more so with Eastern Queens, though driving is still faster than surface transit, but bus ridership is still relatively high.

        Attleboro looks rather compact and is great from a pedestrian perspective for those who can live there (or drive and then park) But the usefulness of commuter rail is more from having hard to drive Boston at one end rather than compact Attleboro at the other. As for its zoning being a minimum of 1/4 acre, most of the housing within walking distance of Attleboro looks much denser than that.

      • letsgola says:

        Does anyone in Norwalk or Attleboro have a reason to take a bus to the commuter rail station? I don’t think traffic on the local roads is any worse than Eastvale (though if memory serves MNR has wait lists for parking spots, right?). And traffic on the 91 between Corona and Anaheim is horrendous. I think the problem for serving Eastvale with transit isn’t so much Eastvale itself as it is the dispersed nature of employment and poor transit service offered in Orange County.

        • nei says:

          Norwalk, maybe. There’s a lower income population and is somewhat dense in the center, over a larger area than Attleboro. Except many of the densest parts are already within walking distance to a Metro North station [Norwalk has three train stations] or the downtown business area, so it removes some of the need for bus service. Traffic on I-95 is horrendous, but local bus service wouldn’t help people avoid I-95 traffic; MetroNorth would. Norwalk’s bus system average 6000 daily riders, with a city population of 87,190, seems low but I don’t know how Los Angeles exurbs compare. As for the MNR wait lists for parking spots, don’t know if Norwalk stations have a parking wait list. A few circulator buses connect to Scarsdale station with a limited frequency, they probably exist much of a parking wait list.

          Attleboro is a bit similar to Norwalk (except the proportions between the two types are different) in that it’s either compact [near the train station] or very spread out. Anyone in the latter area would always have car access and probably wouldn’t have any reason to use a bus.

          • Adirondacker12800 says:

            Norwalk isn’t exurban. The population density is 3,744 per square mile. Almost as dense as Eastvale will be someday. Or Nassau County like.

  8. Pingback: Lumpiness: in cities’ property values, and in metro structure | west north

  9. John Hupp says:

    Skimming the comments to make sure someone hasn’t mentioned this… There is a big difference between the street-running light rail lines and the ex-freight light rail lines in LA. Even with its street-running sections, the Blue Line in LA runs roughly twice the average speed of the Purple Line Express in Chicago. By contrast, the Gold Line East Side Extension and the Expo Line run hella slow because of their far greater interaction with cars.

    If the Harbor Subdivision were developed as a high-speed, high-capacity trunk line like Metrolink, it would work quite well with a grid of street-running light-rail feeders. That level of rail density is still decades out, however.

    Likewise the Blue Line could be upgraded to be even faster, and competitive with driving, if ridership were high enough to demand additional track capacity. At the moment, yes, Blue Line trains are running at capacity, but that can still be resolved by buying more rolling stock and reducing headways (which is precisely what they’re doing).

    Even then, it takes a long time to shift commuting patterns in wealthier, more peripheral suburbs. LACMTA is being very aggressive in its infrastructure expansion, but demand at the fringes can still be tamped considerably by inaction from neighboring county transit authorities like OCTA, San Bernardino or Riverside. In that case, we may see density in the Southland become more centralized around LA County, or at least the LA Basin, though not necessarily downtown Los Angeles itself.

    • John Hupp says:

      It’s worth noting that the LA basin has a much higher concentration of pre-war streetcar suburbs than neighboring counties and valleys. At the very least, the mountains segregate the air pollution, so LA would not have to deal so much with its neighbors’ smog if car usage rates continue to diverge.

      Though a good chunk of LA’s air pollution comes from freight operations within and going in and out of the ports of Los Angeles and Long Beach, so perhaps we should be focusing on increasing port-side ship electrification and increasing freight rail mode share.

      (If your main concern is global warming… we’re fucked regardless. But at least most of Southern California’s coastal cities are up on bluffs, rather than level with the beach. And there is an extremely low frequency of hurricanes to begin with, so we don’t have to worry so much about being ravaged by superstorms.)

      • John Hupp says:

        It sounds like the Northeast is destined for an indefinite future of station parking and two-hour train rides, whereas LA County is destined for a future of pedestrianization and slower but shorter-distance commutes. Yes, you could still take two hours to commute in LA, but you can already do that while driving! Taking the train from Palmdale to Orange County takes more like four hours, and that’s with express commuter rail! Most people are not masochistic enough to commute that far—the cost savings are negligible—so it’s kind of a moot point.

        • John Hupp says:

          Job centers here are more dispersed than you’d imagine.

        • Adirondacker12800 says:

          Port Jervis to Montauk would take 5 and that’s if you are lucky enough to make good connections. New Haven to Trenton would take around four. But then if you really wanted to be a masochist you could start out in New London and use commuter trains all the way to Wilmington. Newark Delaware, not Newark New Jersey if you time it right. Google says the road mileage is 256. Someday SEPTA and MARC will meet somewhere and the MBTA and SLE, assuming Rhode Island doesn’t decide to stop contracting with the MBTA and Delaware doesn’t decide to stop contracting with SEPTA. You’ll be able to go from Virginia to tantalizingly close to New Hampshire. Maybe New Hampshire will get over it’s fear of trains. Fredricksburg VA to Haverhill MA is 512 miles by road. I didn’t bother to drag the route to the places where trains run.

          • Alon Levy says:

            Through-running isn’t for Montauk or Port Jervis; it’s for Rockville Centre and Paterson.

          • Adirondacker12800 says:

            Gotta electrify to Paterson first. To Ridgewood might make more sense. Paterson is a destination and the suburbanites can get on the local to get to Paterson. Off peak the express from Suffern to Hoboken and the local from Ridgewood to Rockville Center can have a cross platform transfer on the lower level in Secaucus….

          • Alon Levy says:

            I’ve always thought of Suffern as the natural terminus of the line (or else Harriman); once you’ve electrified 75% of the line, electrifying the last 25% is almost always highly cost-effective. It’s just that the Port Jervis Line is a basket case and doesn’t serve anything interesting. The only density in that area is in the Hasidic suburbs, which aren’t built next to the railroad, and which run their own bus lines direct to South Williamsburg.

          • Adirondacker12800 says:

            Suffern as the terminus for local service between Ridgewood and Suffern. The train can then express to Hoboken or Penn Station on the Bergen or Main and meet the local that ran on the other branch at Secaucus. Where everybody does cross platform transfers for Hoboken or Penn Station. The next set of trains can do the opposite. Paterson was there before the railroad came so it’s kinda tight. Ridgewood was farms when the railroad came so there’s plenty o’ space to put a pocket track in where the trains can do whatever it is trains do when they turn around.

            Local Port Jervis to Suffern, express to one of the terminals. Meet the express that will run local to Ridgwood in Suffern. Fairly infrequent. Since the infrastructure is already there and the NIMBY’s have forgotten about it, it’s better that they get on a train someplace vaguely local instead of driving to Ramsey-17…. Orange County communities decided they didn’t like having those icky trains on what used to be the passenger main and had them move to the freight bypass out in the middle of nowhere. Park-n-rides out in the middle of nowhere are better than park-n-rides in the middle of downtown. Or using the park-n-ride in Ramsey. Pity that the people who could walk to where the train station used to be, can’t anymore.

            Some of the Hasidim want to get to their jobs on 47th Street, a bus to the train station might work like it works along the Harlem or the Morris and Essex. Though from the looks of it, they can get buses to Manhattan at the same bus stop that takes them to Williamsburg. Unclear if it’s the same bus…. and the bus means they don’t have to mix.

          • Nathanael says:

            “It’s just that the Port Jervis Line is a basket case and doesn’t serve anything interesting.”

            Ah, the Erie Railroad, with its mainline which avoided all population centers except Binghamton. I have always wondered how that got built. It did go bankrupt repeatedly.

          • Alon Levy says:

            The old mainline is a bit more populated (Middletown). But the Erie was a longer-distance railroad anyway – it was actually the bigger partner in the Erie Lackawanna merger, for some reason.

          • Adirondacker12800 says:

            The Erie was chartered for Piermont to Dunkirk. The DL&W was cobbled together from other railroads. Where people and freight was in 1850 is usually a bit different from where they are now.

          • Nathanael says:

            Well, that explains it. The cobbled-together railroads usually hit more population centers than the “single long charter” railroads.

    • Alon Levy says:

      Yes, the light rail lines that are operationally tram-trains (slow at the inner end, fast farther out) are often quite fast, but there’s still a finite number of ROWs for them, and those ROWs tend to be ex-Pacific Electric lines, which don’t serve the newer suburbs. PE had one line connecting LA with the Inland Empire.

      • John Hupp says:

        Inland Empire would be Metrolink because it crosses county lines. There are plenty of freight rights-of-way there that Metrolink could mooch off of. Granted, quality of service could be an issue, though the Orange County Line brought double-tracking and ongoing investment in grade separations. With regard to street-running light rail as a feeder system, I was referring to the main LA Basin and perhaps the San Fernando Valley, which are the LACMTA’s home territory.

    • Jonah says:

      The average speeds on the blue line isn’t that much faster than the gold. Sure the blue line flies for most of its trip, but it then crawls in DTLA and DTLB.

      If you calculate out each route’s distance divided by its scheduled run time you get Blue: 22.75 mph, Gold: 22.3 mph, Expo: 18 mph. Oh and the Green flies at 35.3 mph.

      • John Hupp says:

        Using the driving distance instead of as-the-bird-flies, for an apples-to-apples comparison, 25 miles in 52 minutes (from Pacific Station to 7th St / Metro Center) is 28.8mph. The Purple Line Express in Chicago is covers half that distance in the same time.

        To use Jarrett Walker’s classification, the Blue Line is a light rail line with streetcar segments at the ends. Streetcar segments without signal priority. I’m sure they could make a huge improvement with signal priority.

        • John Hupp says:

          I’m cheating slightly by using Pacific Station instead of Transit Mall, but driving to 7th Street / Metro Center at this exact moment (7:45am on a Friday morning) would take 45-53 minutes, depending on which freeway you take. So there’s that.

  10. Jonah says:

    Also, is it really fair to say LA has “so little transit usage”? I’d characterize it as “medium” as otherwise what would you call cities where transit usage is actually “low”?

    • Alon Levy says:

      All of them are low. When the metro area is below 10%, there’s no other way to classify it, really. To me, medium is a range of cities starting at the low end with Chicago and San Francisco and ending at the high end with Sydney – the 15-25% mode share area. To be honest, New York, at 30%, should probably also be classified as medium, considering that there are many first-world cities in the 40%+ region.

      • BruceMcF says:

        At least with transit as the dominant commuter mode, if NYC were classed as medium transit usage across the board, it would still qualify as high commute transit usage.

        • Alon Levy says:

          Not really – it’s medium all the way. I’m only using trip-to-work mode share to begin with; the reason is that if you look at all trips, then you’ll see fewer transit trips and more walking trips, and either more or fewer car trips. Suburbanites might do 4 car trips a day, whereas urban transit users take 2 transit trips per work day and 1-10 walking trips, which could skew the numbers. So when I say that there are metro areas in the 40%+ region, I’m already talking about work trips, same as New York’s 30%-ish figure.

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s