The RPA Continues to Push for the Flawed Crossboro Plan

As the Regional Plan Association continues to work on its Fourth Regional Plan, expected to be published next year, it’s releasing various components of the upcoming agenda. One, an update from the Third Regional Plan from 1996, is a line variously called Triboro or Crossboro. In the third plan, Triboro RX was meant to be a circumferential subway line, taking over existing abandoned and low-traffic freight rail rights-of-way in Brooklyn, Queens, and the South Bronx, terminating at Yankee Stadium via a short tunnel. It was never seriously proposed by any political actor, but was briefly mentioned positively by then-MTA chair Lee Sander in 2008, and negatively mentioned by Christine Quinn, who called for a bus line along a parallel alignment in her mayoral campaign in 2013. In 2014, Penn Design proposed a variant it calls Crossboro, which differs from the original Triboro proposal in two ways: first, the stop spacing is much wider, and second, instead of the short tunnel to Yankee Stadium, it continues northeast along the Northeast Corridor, making four stops in the Bronx as in the proposed Metro-North Penn Station Access plan. Crossboro is an inferior proposal, and unfortunately, the fourth plan’s Triboro proposal downgrades it from the original alignment to Crossboro.

As I explained a year and a half ago, specifically in the context of Crossboro, it is poor planning to run train service that begins as a radial and then becomes as a circumferential instead of continuing into the center. The route of Crossboro, and now also the Triboro plan, involves going from the North Bronx to the south in the direction of Manhattan, but then turning southeast toward Queens and Brooklyn, rather than continuing to Manhattan. Briefly, in a system with radial and circumferential routes (as opposed to a grid), circumferential service is the most effective when it connects to secondary centers, and has easy transfers to every radial. If a line runs as a radial and then switches to circumferential, its ability to connect to other radials is compromised, making it a weaker circumferential; nor could it ever be even a half-decent radial without service to the CBD. Lines with such service pattern, such as Line 3 in Shanghai and the G train in New York until 2001, tend to underperform.

However, the stop spacing deserves to be treated separately. Under both Crossboro and the RPA’s new version of Triboro, there are too few stops for the line to be useful as an urban rail service. I’m going to ignore the connection between Queens and the Bronx, which as a major water crossing can be expected to have a long nonstop segment, and talk first about the Bronx, and then about Queens and Brooklyn.

In the Bronx, there are four stops in 10 km, starting counting from where the bridge toward Queens begins to rise. This may be reasonable for a commuter rail service with local service extending well past city limits (to New Rochelle or even Stamford), but when it terminates within the city, it’s too far for people to be able to walk to it. The proposed stops also miss the Bronx’s most important bus route, the Bx12 on Fordham Road, which in 2015 became the city’s busiest single bus route. A stop on the Pelham Parkway, the continuation of Fordham in the East Bronx, would be a massive travel time improvement over trying to reroute the Bx12 to meet a train station near Coop City, the proposed northern terminus of both Crossboro and the new Triboro. Conversely, it would delay few other passengers, by very little, since there would only be one further stop north. The result of the proposed stopping pattern is then that most people living near the line would not be able to either walk to it or take a frequent bus.

In Queens and Brooklyn, starting from Astoria and going south, the route is 26 km long, and the new Triboro makes 17 stops. The average interstation, 1.5 km, is noticeably above the international subway average, and is especially high for New York, whose stop spacing is near the low end globally. The original version had 29 stops over the same distance, and one more stop between Astoria and the bridge. Unlike in the Bronx, in Brooklyn all streets hosting major radial routes get subway stops. However, long stretches of the route get no stops. The stop spacing is not uniform – from Northern Boulevard to Grand Avenue there’s a stretch with 4 stops in 2.8 km (counting both ends), but from Astoria-Ditmars to Northern Boulevard there’s a 2.5 km nonstop service, skipping Astoria Boulevard and Steinway, passing through a medium-density neighborhood south of the Grand Central Parkway with mediocre subway access. A stop at Astoria Boulevard and Steinway is obligatory, and probably also one between Astoria and Northern, around 49th Street. To the south of Grand Avenue, the proposal calls for a 2.1 km nonstop segment to the M terminus at Metropolitan Avenue, skipping Middle Village, which is cut off from Grand by the Long Island Expressway and from the M by cemeteries. An additional stop in the middle of this segment, at Eliot Avenue, is required.

In Brooklyn, the route runs express next to the L train, splitting the difference between serving Broadway Junction (with a connection to the A/C) and Atlantic Avenue (with a connection to the LIRR): the RPA’s diagram depicts a station at Atlantic Avenue but calls it Broadway Junction. Farther south, it makes a few stops on an arc going southwest toward southern Brooklyn; the stops are all defensible, and the stop spacing could potentially work, but there are still potential missing locations, and some nonstop segments in the 1.7 km area. For example, it goes nonstop between Utica and Nostrand Avenues, a distance of 1.7 km, with a good location for an interpolating station right in the middle, at Albany Avenue. From Nostrand west, it stops at a transfer to every subway line, except the R. In that segment, one more stop could be added, between the F and the D/N; the reason is that the gap between these two lines is 1.8 km, and moreover the right-of-way slices diagonally through the street grid, so that travel time from the middle to either stop is longer along the street network. However, overall, this is not why I dislike the route. Finally, at the western end, the route is especially egregious. The right-of-way is parallel to the N train, but then awkwardly misses 59th Street, where the N veers north and starts going toward Manhattan. The original proposal had a stop several blocks away from 59th, with a long transfer to the R (and N); this one drops it, so there is no R transfer in Brooklyn – trains express from the D/N transfer at New Utrecht to the terminus at Brooklyn Army Terminal, where there is very little development. There are practically no through-riders who would be inconvenienced by adding the extra two N stops in between. In contrast, due to the low frequency of the N (it comes every 10 minutes off-peak), making passengers originating in those stations who wish to ride Triboro transfer would add considerably to their travel time.

A route like Triboro has an inherent problem in deciding what stop spacing to use, because as a circumferential, it is intended to be used on a large variety of origin-destination pairs. For passengers who intend to connect between two outer radial legs more quickly than they could if they transferred in Manhattan, the wider stop spacing, emphasizing subway connections, is better. However, the mixed radial-circumferential nature of the new Triboro makes this a losing proposition: there’s no connection to any subway line in the Bronx except the 6. Moreover, in Brooklyn, there’s no connection in Brooklyn to the R, and if there’s a connection to the A/C, it involves walking several hundred meters from what on the L is a separate subway stop.

In contrast, for passengers whose origins are along the line, narrower stop spacing works better, because they’re unlikely to cluster around the connection points with the radial subway lines. (The line has no compelling destinations, except maybe Jackson Heights and Brooklyn College; in the Bronx, the two most important destinations, the Hub and Yankee Stadium, are respectively close to and on the old Triboro route, but far from the new one.) The aforementioned Astoria/Steinway, Eliot, and Albany, as well as the skipped stations along the L and N routes, all have reasonable numbers of people within walking distance, who have either poor subway access (the first three) or only radial access (the L and N stations).

What’s more, if trains make more stops, the increase in travel time for passengers connecting between two legs is not large compared with the reduced station access time for passengers originating at an intermediate station. The reason is that passengers who connect between two legs are not traveling all the way. The fastest way to get from the West Bronx to southern Brooklyn is to take the D train all the way, or take the 4 to the D; from the 6 train’s shed, the fastest way is to take the 6 and transfer to the N/Q at Canal or the B/D at Broadway/Lafayette. No circumferential service can change that. The benefit of circumferential service is for people who travel short segments: between the Bronx and Queens, or between the 7 or the Queens Boulevard trains and the lines in Brooklyn that aren’t the F. Given high circumferential bus ridership in Brooklyn – two circumferential routes, the B6 and B35, rank 2nd and 4th borough-wide and 4th and 7th citywide, despite averaging maybe 9 km/h – connections between two Brooklyn legs are also likely. For those passengers, making a few more local stops would add very little to travel time. The subway has a total stop penalty of about 45 seconds per station. Of the ten extra stops I list as required – Astoria/Steinway, Eliot, Albany, 59th, four along the L, and two along the N – three (the two on the N and 59th) are basically end stations, and few passengers have any reason to travel over more than five of the rest. In contrast, adding these ten stops would improve the quality of transfers to the R and A/C and provide crucial service to intermediate neighborhoods, especially Middle Village.

Finally, let me make a remark about comparative costs. The original Triboro plan required a short tunnel, between Melrose Metro-North station and Yankee Stadium; the new one does not. However, a single kilometer of new tunnel in the context of a 34 km line is not a major cost driver. The new proposal is actually likely to be more expensive. It is longer because of the segment in the Bronx along the Northeast Corridor, about 40 km in total, and 10 km would be alongside an active rail line. There are plans for increased mainline passenger rail service on the line: Penn Station Access, plus any improvements that may be made to intercity rail. Far from offering opportunities to share costs, such traffic means that any such plan would require four tracks on the entire line and flying junctions to separate trains going to Penn Station from trains going to Brooklyn. Fare collection would be awkward, too – most passengers would transfer to the subway, so subway faregates would be required, but commuter rail has no need for faregates, so sharing stations with Penn Station Access would require some kludge that wouldn’t work well for any mode. Tunneling is expensive in New York, but so is at-grade construction; a kilometer of tunnel in the Bronx is unlikely to cost more than configuring an active rail mainline for a combination of suburban and high-frequency urban service.

The RPA proposes the London Overground as a model, treating the new Triboro as a commuter line offering subway service levels. Everywhere else I’d support this idea. But here, it fails. First, as I explained in a previous post, the routing is an awkward mix of radial and circumferential. But second, the stop spacing only works in the context of a long suburban line feeding city center, and not an urban circumferential line. In the context of an urban line, more stops are needed, to let people walk from more neighborhoods to the train, or take a connecting bus. For the most part, the original Triboro plan, designed around interstations of about 900 meters not counting the water crossing, would work well. Crossboro, and its near-clone the new Triboro, is inferior to it in every respect, and the RPA should jettison it from the Fourth Regional Plan in favor of the old proposal.


  1. Union Tpke

    Great pos! I really wonder why this has happened to the RPA. It used to propose innovative and useful ideas, but now they are proposing garbage proposals. I don’t know why they changed their old proposal.

    • electricangel

      Also, the tunnel under St. Mary’s Park has been filled in to create a parking garage at a new “green” development in the South Bronx. That means unless they want to take that parking back, they cannot go to Melrose station.

        • threestationsquare

          The obstruction is south of 156th St (and is a high-rise mixed-use building, not just a parking structure). I suspect that building (which opened 2012) is the reason recent iterations of the TriboroRX proposal have used the inferior routing towards Parkchester/Co-op City. It looks possible to get around it via adjacent lots, but the difficulty of doing so might tip the balance in favor of just turning under 149th and terminating the line at the Hub (149th and 3rd Av).

  2. Adirondacker12800

    30 pounds of train in a 20 pound of bag.
    Metro North wants to send 6 trains an hour over the bridge, Amtrak will want to send 4. The Port Authority and New England want to use it for freight. Sumpthin is gonna have to give. And the route in Brooklyn and Queens is the same ROW that the Port Authority and New England want to use.

    • Alon Levy

      The route in Brooklyn and Queens carries not much freight. Just kick the freight out – Triboro (the original, not this version) is a higher and better use of the ROW, and the freight volumes don’t justify four-tracking the few two-track narrows just to let the few daily freight trains back. Hell Gate has four tracks (reduced to three, but it’s wide enough for four), and it’s possible to give away two for Triboro and have intercity and regional trains share the other two. It’s not as if there’s much of a speed difference on a curvy bridge without stations. The freight trains can go in commuter rail slots in the evening off-peak.

      The problem is that between the bridge and Coop City, there are only two tracks to begin with, and there is a substantial speed difference between commuter and intercity rail, so that scheduling everything to avoid overtakes is a pain. If there’s money for four tracks on the line, it should be used for untethering commuter and intercity trains from the clusterfuck that is the New Rochelle-Stamford segment (in a perfect world it would have six tracks).

      • Adirondacker12800

        You have a choice in 2030. A longer subway ride between Queens and the Bronx or eating. The road bridge is grid locked with drywall going to Connecticut the truck going to the supermarket in Bayside or Bensonhurst gets stuck in it.
        The New Haven acquired ROW on the assumption that automobiles were playthings for rich men. The the ROW in the Bronx is 8 tracks wide.

      • kevd

        The Bronx section can also be four tracked just like the hell gate bridge.

        If freight needs to use the bay ridge line, they could be run at night. A triboro Rx could shut down 1-5weekdays and be replaced by night bus service as happens in many metro systems around the world

        • Adirondacker12800

          They aren’t going to build a tunnel to Jersey City for a few trains. The 20 million people on Long Island and in New England will generate a lot of traffic. That won’t be on I-95.

          • threestationsquare

            Based on p51 of this pdf from the Cross-Harbor Freight Program DEIS, they apparently expect 16-21 freight trains per day, which by the standards of passenger rail is “few”. (Significantly less than the North London Line.) Though if you just mean “they aren’t going to build a (freight) tunnel”, then I agree, they probably aren’t.

          • Adirondacker12800

            None of the pages reference how much New England wants to ship through it. Outside of the scope. If I picked the right page 51, the report seems to go up to page 5-47, page 53 says 48-51 trains a day.

          • threestationsquare

            PDF p51 of that document depicts the selected “preferred alternative”; PDF p53 shows an alternative that was eliminated from consideration (see here). The projection is 7-12 daily freight trains (out of those 16-21 total through the tunnel) running over Hell Gate to/from New England, up from about 4 today.

      • Union Tpke

        Isn’t there room for six until Port Chester. The New York, Westchester and Boston Railway used to share the right-of-way. I am not sure after Port Chester

        • Nick Gorski

          Room for 6 in the ROW? Maybe. But you’d be looking at replacing all the catenary equipment (the masts are, aside from a short stretch north of New Rochelle, only four tracks apart) and rebuilding all the stations along the line.

          That said, rebuilding New Rochelle to 6 tracks (with two island platforms) is certainly necessary and probably do-able (though they’ll have to rebuild the North Av Overpass, again, and possibly demolish the multimodal transit center / parking garage). At that point you can grade separate the junction between Metro North and Amtrak.

  3. threestationsquare

    I think it makes more sense to build along the TriboroRX route as a series of extensions/branches of the L and M trains, something like this. Benefits:
    1. Saves about four miles of construction and operating costs (between Metropolitan Ave M and New Lots Ave L) while still serving all of Alon’s recommended stations.
    2. One-seat service from new stations to Williamsburg and Manhattan is likely more useful than one-seat Brooklyn-Queens service.
    3. Smaller minimum-operable-segments can still be useful, so the line can open incrementally as funding and political will become available. (First to the Brighton Line and to Queens Boulevard, tackling the more expensive and politically difficult sections later.)
    4. Ensures that the line is operated as a subway with compatible equipment and fare integration, not some special-snowflake light rail line calling for its own shops/car order/other contractor handouts, and not some infrequent labor-intensive LIRR shuttle.

    Regarding freight: Along the northern section of the route, freight trains (which run at off-peak hours) can share tracks with Amtrak trains across the Hell Gate bridge and to Sunnyside yard, then use the chord from there along 25th St to reach the Lower Montauk Branch and the LIRR freight network, so the two-track New York Connecting Railroad between Metropolitan Ave and the bridge could be used exclusively by the subway. Along the southern section, the right-of-way is wide enough for four tracks everywhere except for the less than 1 mile segment between the Brighton and Culver Lines; the political decision whether to spend the money to widen that segment to three tracks or eliminate freight service to Brooklyn Army Terminal can be made after the earlier phases have opened. (A branch of the L that only reached the Brighton Line would still be very useful.)

    • Adirondacker12800

      The Port Authority wants to spend billions of dollars building a tunnel from Brooklyn to Jersey City. They don’t want to do that for a few trains a day. There’s 8 million people on Long Island and 14 million in New England. They use a lot of stuff. Which mostly moves by truck. Moving some of it to rail postpones the day the region gridlocks.

      • threestationsquare

        If the Cross-Harbor Freight Tunnel actually happens, the cost of widening a one-mile trench in Brooklyn is chump change by comparison, and that’s the only place where there’s any conflict.

        If you’re concerned about track sharing between freight and Amtrak/MNR, you should keep in mind the two-track North London Line carries about 50 freight trains each way per day slotted in between a 6tph all-day intensive passenger service. The Hell Gate Amtrak tracks aren’t getting close to that much off-peak use any time soon.

        • Adirondacker12800

          They don’t have to widen anything to move freight. Brooklyn real estate isn’t as pricey as Manhattan real estate bu tit ain’t cheap. It would probably be cheaper to elevate the passenger trains over the trench.

      • Joey

        Adirondacker: All the lines that freight could possibly connect to are passenger primary anyway. Go north along the Hell Gate line and you hit the NEC at New Rochelle. Go east and you hit Jamaica.

        • Adirondacker12800

          You find the track space for the freight trains or you do your food shopping in Pennsylvania. Take your pick.

          • Adirondacker12800

            If everything inside I-287 is gridlocked the trucks can’t get through. Stuff might trickle in via the Oak Point Link or by wandering through Connecticut. But then the trucks can’t get to Hunts Point. Or the population and consumption has to freeze. Or you do your grocery shopping in Pennsylvania.

          • threestationsquare

            If it’s gridlocked, raise the Hudson tolls until it isn’t gridlocked anymore. We’ll pay a bit more for food in exchange for better rail service.

          • Adirondacker12800

            Why should people in New Jersey pay higher tolls when people crossing the East River can avoid them?

          • Adirondacker12800

            SIlly me. Taking “raise Hudson tolls” to have something to do with the Hudson River.
            The crossings teeter on the edge of gridlock now. If there’s another million people in the city by 2030 how high do the tolls have to be to keep the food moving in and the garbage moving out? And beer and drywall and paper goods and….

          • Joey

            Adirondacker: Point is that nearly all the possible routings for freight trains are already saturated with passenger trains at peak. They can run as many freight trains as they want at night.

        • threestationsquare

          Negative progress actually; Conrail abandoned the Poughkeepsie Bridge because there wasn’t enough freight demand to justify spending a few million to repair it after a fire in 1974.

    • Alon Levy

      Two disadvantages:

      1. Service between Queens and southern Brooklyn gets awkward. Service to Manhattan opens up but is not that useful, since the L and M would be taking circuitous paths; regardless of what the alignment is, the fastest way to Manhattan from Middle Village would be to take the train up to Jackson Heights and switch to the E/F.

      2. The M and half the L wouldn’t be frequent. A non-branching line could be run more frequently – if there’s no demand for 5-minute off-peak service with 10-car trains, then run 5-car trains.

      • threestationsquare

        1. Most travel from Queens to Southern Brooklyn is likely to be faster via Manhattan no matter what.
        2. Given MTA loading guidelines (and more generally the fact that labor is the main operating cost) I expect an isolated line would run even less frequently than the M. The G only runs every 10 minutes even with short trains.

      • Eric

        I agree with threestationsquare. The Triboro ROW serves residential suburbs, and the demand for travel between any two suburbs is very low. Many of Triboro’s riders would transfer from it to a Manhattan-bound subway, and running Triboro trains to Manhattan in the form of the L/M makes this much more attractive. And Queens-south Brooklyn trips would become a bit less convenient under this plan (1km more distance and a transfer, but trains would likely be more frequent, so less waiting), trips to Williamsburg and Bushwick would be more convenient, so the overall effect on outer-borough trips would be somewhat of a wash.

  4. Pingback: Weekend work affecting 13 subway lines - Second Ave. Sagas :: Second Ave. Sagas
  5. mister

    This is an excellent post Alon. I have been looking at the new alignment in the Bronx for a long time and thinking that it’s inferior in many ways. Connecting to the B/D/4 at Yankee Stadium, and the 2/5 at a new station where the abandoned Port Morris Branch crosses the 2/5 lines would make it a much better circumferential line. Stopping at Broadway Junction is a must. I’m less concerned about the “express” runs that the line makes, because people who live along other stops on the line will have to transfer too. And whatever happens along the southern end of the line needs to allow for a future expansion to Staten Island.

    I also think that freight rail doesn’t need to be kicked off the line. Right now, the plan is to 4 track the Hell Gate bridge anyway for Penn Access, so MNRR and Amtrak can share 2 tracks. There is very little freight on the line. Let what few freights there are between Fresh Pond and the Bronx share tracks with TRX trains and operate them during less busy periods. We need to start looking at the possibility of allowing more flexibility with passenger rail by allowing light rail and rapid transit vehicles to share tracks with Commuter and freight services.

    • Alon Levy

      The ROW is for the most part four-tracked, including on Hell Gate, as you note, but there’s a segment in southern Brooklyn with a two-track trench. Squeezing three (two passenger, one freight) might possibly king of work; four would require widening the trench (=$$$). The New York Connecting Railroad is also a two-track ROW with nontrivial widening costs between Hell Gate and the Lower Montauk Line, but freight trains can get around that by detouring via Sunnyside Yard.

      • mister

        Well I think right now, freight is infrequent enough that there isn’t much need to widen the right of way. Construction of a few passing sidings and full double tracking of the line means that you wouldn’t need to have that third track.

        Whenever the PA gets around to building the Cross Harbor Freight tunnel, and freight demand goes way up, then you can think about adding more tracks for freight. Without widening the existing trench, it might be possible to get 4 tracks, even in the 2 track segment. The total ROW through the 2 track section (including embankment) is wider than the 4 track Sea Beach ROW (Google maps measurement shows the Sea Beach ROW at non-station areas at around 52ish feet (with buildings right up against the retaining walls in some cases). A conservative measurement of the width of the 2 track segments of the Bay Ridge is approx 70′. The tricky part will be places where the ROW has additional constraints (for example, between Nostrand and Flatbush, where a new mall was built over the ROW).

        The Chokepoint really would be the NYCRR, because the kind of volume one would expect once the line is connected to New Jersey and becomes a major freight artery would be too great to send everything through Sunnyside/Harold. At that point, it might make sense to double stack the NYCRR (getting past the cemetery remains difficult), and add two outer tracks on the Hell Gate Bridge (it certainly seems to have space for that).

  6. Bronx Resident

    Bronx route shouldn’t go into the East Bronx. It should go west utilizing the existing diagonal ROW which runs under St. Mary’s Park. A short tunnel under Via Verde will be necessary.

  7. Sydneysider

    I think the plan is still not the best but the changes are a step in the right direction. This is at the, end of the day, still a commuter line and 900 metre spacing is probably a bit too close for my liking given the natural of this line. I think of the proposed spacing and plan as less London Overground and more London Crossrail (without City Centre) or even Milton Keynes to East Croydon Line. Let’s not forget the Crossboro line passes through areas well outside of Manhattan, the London Overground orbital lines doesn’t go well outside of Zone 2 (which is a commonly accepted definition of Central London).

    You also want to make the Crossboro line as attractive as possible so that passengers don’t have to travel on the subway through Manhattan. Reducing subway crowding should a goal too, not just local access. Reducing travel times, no matter how trivial, by reducing stops can help achieve this. It’s a balancing act.

    Given the commuter rail characteristics, I also think that LIRR should be the legal operator of this line, if it comes to fruition. With Metrocard fares fully integrated and different rolling stock of course. While I’m talking about commuter rail, I also think that Metrocard express fares should be accepted on MNR and LIRR within NYC. But what do I know about Crossboro, fares and LIRR, I’m not from the New York metro area.

    • Alon Levy

      Some of the neighborhoods Triboro passes through – the South Bronx and Astoria, for two – are denser than any London borough. It may seem like an outer-urban line because it misses Manhattan and most of the important secondary business districts (and the new proposal even misses one that the original proposal hit, the Bronx Hub), but when it intersects the subway it’s in the middle of most lines.

      The original ridership estimate for Triboro was 76,000 commuters each traveling a roundtrip per weekday, of whom 32,000 would be new subway riders. Generally, people in New York who work outside Manhattan drive, especially if they don’t work in Downtown Brooklyn or maybe Long Island City. A few years ago, I read a report that I can no longer find that put Downtown Brooklyn’s transit mode share as a job center at 50%, and Jamaica and Flushing’s at 30%. I forget where LIC was, probably in the middle. A very large fraction of Triboro’s impetus is to attract new riders to the subway who are currently driving, or maybe taking a ghastly crosstown bus, or just not taking jobs in parts of the city that aren’t on their subway line. It’s unrealistic to expect such a line to reduce subway crowding much.

      The other issue with crowding is, if you’re traveling to work through Manhattan, you’re probably not traveling at the same times as people who work in Manhattan. It can easily be 30 minutes just to get from where you enter the Manhattan core to your destination, if you’re doing something horrible like living in Brooklyn and working at Columbia; in the reverse direction it can be 45 minutes, because of some destinations that are really far from the subway, like Queens College. If you work the same hours as everyone else, you’re taking the subway in before the morning rush and back after the evening rush.

      The big difference with Crossrail is that Crossrail is suburban rail. The line goes as far west as Reading. In contrast, Triboro stays within the city in all proposals. Yes, there could be transfers from commuter rail… but who would use them? Long-distance commuters into the city are disproportionately likely to work in a major business district – the Manhattan core primarily, but if not, then probably Downtown Brooklyn, LIC, and other secondary nodes that the Triboro alignment completely misses. The reason is that Manhattan, and to a lesser extent the other major secondary nodes (both in New York and in the suburbs, e.g. Newark, Exchange Place, White Plains), is where the regionally and internationally important employment clusters. Off-core job markets tend to be more local. Someone from the Bronx might still end up working in Queens, especially if transportation links are improved, but someone from the Metro-North suburbs would probably only do it if they worked for a specialized employer, which would most likely be in a secondary business district (JetBlue’s headquartered in LIC).

    • threestationsquare

      This is not a “commuter rail line” in the US sense; it reaches neither outer suburbs nor major destinations. It’s an urban subway feeder and should be operated accordingly. Even orbital journeys will mostly be transferring to/from the subway. (The LIRR with its antiquated labour-intensive operating practices should ideally not be operating /any/ lines, but that’s another matter.)

  8. Sydneysider

    Thanks for your response Alon. What I was trying to compare was Central London and Manhattan, they both have a population of around 1.5 million. The London Overground’s circumferential/orbital lines do not venture well outside of Central London (or in NYC, Manhattan). The Triboro Line ventures well outside of Manhattan, the central part of NYC.

    For this reason, despite the fact that Triboro travels near Manhattan at some places, I believe that the Triboro Line is better suited as a suburban line operated by LIRR with subway like frequencies and subway fares. More stops in big activity centers will be nice but not the 10 extra stops you’re proposing.

    You’re right that London’s most density populated borough (at 18000 people per km2) is nothing compared a lot of NYC neighbourhoods though.

    As you would have noticed from my name, I’m from Sydney, the government here is building a new 15km metro/subway line with only 9 stops so I don’t see wide station spacing as big deal for a metro style service. The project is called Sydney Metro Chatswood to Sydenham. It was originally planned as commuter rail line and I still prefer the commuter rail option. But clearly, wide station spacing should not disqualify Triboro from offering a subway style service.

    Continuing on with my comparison with Sydney, the mode share to work the other centers you mentioned sounds quite low. North Sydney, just across the harbour from the city center, has a public transport share of around 65% compared with 50% for Downtown Brooklyn. Parramatta, over 20km west of Sydney city center, has a public transport share of 45% compared with 30% you mention for Jamaica and Flushing. However, 30% is on par with Macquarie Park, 15km from city centre, which I consider pretty car orientated. This is all despite Sydney metro having a lower public transport mode share overall, quite surprising. Triboro will definitely help with this for Jamaica and Flushing but I still think it’s a stretch to suggest that wider station spacing will hurt this.

    The other thing you mention is crowding through Manhattan, in Sydney at least, my experience is that these people do unfortunately travel at the same time as city center workers. I should know, I was one of them, especially having to commute every day from Parramatta to St Leonards through the city center but I understand that this is different in NYC, as you explained.

    Also about freight, freight trains share track with suburban trains in Sydney all the time outside of peak hour. Some tracks operate at 6 passenger tph off peak and there is still plenty of freight even during midday off peak. Kind of makes it fun when waiting for the passenger train to see an 1800 metre train pass.

  9. Philip Marshall

    The frequency of freight trains on the line in Brooklyn and Queens may seem low in comparison with passenger traffic elsewhere in the region on the LIRR, MNR, the NEC, etc., but that doesn’t mean it isn’t critically important. The Hell Gate Bridge/NY Connecting RR at one end and the carfloats at the 65th St./Bay Ridge Yard at the other end in Brooklyn represent the only freight rail connections that Long Island has with the rest of the country, so I think the existing freight use needs to be prioritized. In practical terms there is no other way for freight trains to access the LIRR system, especially now that the former LIRR Yard A at Sunnyside has been removed and the MTA has petitioned for abandonment of the Montauk Cutoff loop that connected Yard A to the Lower Montauk Branch. (Without the Montauk Cutoff, the only way for a freight train to get from Sunnyside to the Lower Montauk would be to make a reverse move at Long Island City, and there isn’t enough room to do that except with very short trains.)

  10. kevd

    Ummmm. What is so horrible about eventually tunneling to upper Manhattan across the east river? I realize it’s then the quadboro Rx, but it makes it more of a true circumferential line of it hits the a,c & 1, right?

    • Alon Levy

      It adds 2 km of tunnel, including an underwater crossing. It should be looked at, and you’re right that it offers more connections. I just doubt the extra 2 km of tunnel are cost-effective. Of course the question is, cost-effective at what construction cost…

      • Union Tpke

        I don’t think it should be a priority, but it could run via 125th Street via Randall’s Island, or it could curve into Manhattan at 145th Street after being in the South Bronx.

  11. Union Tpke

    One proposal that I have scans from that is from 1997 suggests that it be a light rail line. Then, it argues that it could therefore do street-running in some segments, such as getting closer to the 2 and 5 at Flatbush Avenue. I will have to look at the scans again. This proposal had the line ending in Jackson Heights. Because it was light-rail in this proposal, it was suggested that provisions be made to allow it to run over the Verrazano because of the lighter weight.

  12. Juan

    What I would suggest as a first “phase” of the Triboro RX would be to extend the M train via the freight tracks up to about Queens Blvd, then build a small connection with the unused platforms at the Jackson Heights/Roosevelt Ave Station with at least 3 stations in between (one by Juniper Blvd N/Eliot Ave, another by Grand Ave, and another at Queens Blvd before terminating at Roosevelt Ave).

    I would like to see a full-build of the Triboro RX, but only if it connects to the 59th Street station in Brooklyn, and perhaps of running to 161st St/Yankee Stadium, it can run under 149th St in the Bronx up to 145th St in Manhattan as a crosstown, making it easier for people moving around to Upper Manhattan or other parts of the Bronx without making a transfer at 125th St or even 42nd St. If it can connect to a 125th St crosstown, maybe that could work but I like the former because it could help to connect people from point A to B by avoiding Midtown if they don’t need to go there, all together.

    The current plans miss the mark. I don’t think 161st St/ Yankee Stadium would work due to the closeness of the Stadium to the station and because I think it wouldn’t serve as much as a 149/145 St crosstown would. Just my 2¢.

    • mister

      Why would relative closeness of the stadium have any bearing on whether or not this is workable? If anything, extend it even further to connect with the Husdon line at E. 153rd street. Then it will open up connections for Brooklyn and Queens residents to upstate jobs, and vice-versa.

      • Alon Levy

        But the biggest Westchester job center is White Plains, on the Harlem Line… the Hudson Line isn’t great at serving suburban jobs. There’s a secondary downtown around Yonkers, and some suburban employment in Tarrytown, but neither is near the waterfront.

  13. johndmuller

    Just like mainline railroad lines rarely return to railroad use after being “Rail to Trail”ed, so also I see mainline routes which have been “subway”ed rarely returning to full service railroad use. Thus, I have a jaundiced eye when it comes to appreciating propositions such as the Triboro, of whatever variation, which feature wholesale “subway”ifiation of numerous segments of still functioning railroads, especially when one of the most iconic railroad structures around, the HellGate Bridge, is potentially being tossed to the wolves.

    Given a few reasonable destinations in Queens, such as an Astoria station connecting with the subway (whatever the extravagant cost), and direct lines to LGA/CitiField and Jamaica/JFK, I could get behind a commuter rail link using the HellGate Bridge connecting Queens to at least one South Bronx stop like the Hub, on its way to the Hudson line (possibly with a Harlem line option) on to Yankee Stadium and however far up the west Bronx or beyond was convenient logistically.

    That isn’t really the Triboro you are talking about, however; you are looking more for a mass transit option connecting the Hub, arguably one of the most connected parts of the city, to areas in Brooklyn and Queens with value mostly as transfer stations to other subway lines. Insofar as people wanted to go to those locations or further out from midtown, this would be a decent enough route, but for those wanting closer in locations in Brooklyn or Queens or the more westerly parts of Brooklyn, it would likely be faster for someone at or around the Hub to just bite the bullet and take the IRT or the IND thru midtown to connect with whatever radial worked for their trip (if you are starting in the Bronx, you can maybe even get a seat). If the SAS gets up to the Bronx, it could bake it even better that way (or there could be a continuation of the 125th St line into Astoria and beyond accomplishing much the same thing.

    As to an infill station at Pelham Pkwy, it would be good enough for a bus connection, but otherwise there is a lot of nothing right there – although to look on the bright side there is a very large amount of potential for TOD. Extending the 6 line a short, uncomplicated distance to the Coop City station (and beyond into Coop City itself) might be a good idea. There is also a reasonably large gap between the Parkchester and Hunts Point stations which might merit another stop if there is a (not easy to find along this route) decent location. Pelham Manor, in Westchester, which formerly had a station, could possibly be both a reasonably popular origin and destination on this line were it added to the plan.

    • Eric

      Subways are used by many more people than any other mode of transportation. Why would you convert from a more useful mode to a less useful mode? You wouldn’t.

      Rail trails are used by relatively few people, so they should be considered for conversion to rail service which will serve many more people. Not the reverse.

    • Alon Levy

      Not all of Hell Gate Bridge, obviously – it’s a key intercity link. Only two tracks out of four.

      Extending the 6 to Coop City (both potential station and the actual neighborhood) is useful. That said, for Penn Station Access, a 6 connection, while very useful for Coop City, might not necessarily be as useful as a Bx12 connection. The analogy is to York Mills in Toronto: there’s not much of anything next to the subway station, but thanks to the bus-subway transfer, the subway station gets high ridership, 14,600 daily boardings (link), more than North York Centre, which is flanked by high-rises but is not on a grid arterial and has no bus transfer.

  14. Pingback: Changing from a foamer to more of a pragmatist – newyorksfuturetransit
  15. Pingback: RPA Fourth Regional Plan: LaGuardia Airport and the Astoria Line | Pedestrian Observations
  16. Pingback: The RPA Construction Cost Report | Pedestrian Observations
  17. Pingback: Missed Connections | Pedestrian Observations
  18. RapidTransient

    What if the Triboro were to be road, not a railway? This would solve the issue of transit/freight conflicts, as trucks and buses can and do run on the same roads at the same time.
    It would also provide more flexibility for making connections. Instead of heading into the Bronx, bus routes could use new ramps to access 125th St in Harlem and to connect to every major north-south Subway + all 3 Metro North lines. Buses could also operate much more frequently than commuter rail trains, and an express/local model could speed up journeys. The route would be just two lanes wide in between bus stops, with pull-over areas at the stations to allow other traffic to pass. Additional bus routes could also break off onto roads and highways. For example, ramps to the BQE East could allow for a spur to LGA, or ramps to the Gowanus Expressway could allow for an express bus service to Manhattan from the southern portion of the Triboro.
    Dynamic tolling could keep traffic on the Triboro moving at free-flow speeds, so buses could move just as fast or faster than trains. (It’d be a ‘virtual exclusive busway’). The toll, applied to all cars and trucks, could also generate revenue to fund the new bus services and other city priorities.
    This would be good for freight too. Trucks could a lot more freight than trains do today, and they’d still carry more relative to projected freight train traffic with the cross-harbor tunnel. Connections to the Gowanus, Long Island, and Bruckner Expressways would make the Triboro a viable alternative to city streets and the BQE.
    Two key measures could minimize negative local impacts of the road:
    The dynamic tolling enforcement system could also be used to impose low-or-zero emissions requirements for all vehicles using the road to mitigate or eliminate local pollution and reduce carbon emissions.
    Decking over segments of the route near residential areas would mitigate noise concerns, and the city could sell plots on top of the deck to help defray construction costs.

  19. Pingback: The New Triboro/Interboro Plan | Pedestrian Observations

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.