Regional Rail to Lower Manhattan

Staten Islanders’ desire for a subway connection got me thinking again about my previous proposal for a tunnel from Staten Island to Manhattan, possibly with a cross-platform connection to Brooklyn and New Jersey. From all points of view, it is desirable to build a regional rail hub near Fulton Street, and connect it to nearby commuter lines, creating a second pole for an RER- or S-Bahn-like system in New York, in addition to a Midtown pole centered around Penn Station. My intention in this post is to discuss tradeoffs in choosing how to build it.

As the source of the ARC and ESA cost overruns is the station caverns in Manhattan, the Fulton Street station should be minimalistic: as close as possible to the surface (closer to 20 meters underground than to 55), without a full-length mezzanine, and with only four tracks, one to each of Midtown, Brooklyn, New Jersey, and Staten Island. It should under no circumstances look like the ESA extravaganza. This economizing means it’s difficult to build this as a terminal: ideally the station box would be built and then tunnels would be built to all four destinations, making two two-track lines.

The length of this station should be 300 meters, to accommodate 12-car trains, but if it’s too hard to build within the available footprint, then a shortening to 10- or even 8-car trains is feasible. Regional trains in both Paris and Tokyo are usually only 200 meters long. In addition, unlike its Parisian inspiration, Chatelet-Les Halles, Fulton Street need not have very wide platforms. When I took the RER A and changed trains at Chatelet, at 8:40 in the morning, the platform was about as crowded as that of a normal subway station, and needed much less than its full width of 17 meters. Indeed, the Chuo Line, with far higher peak load than anywhere in the West, terminates on a single 10-meter-wide island platform. A single-level four-track station with a cross-platform transfer should be about 30-35 meters wide; since it is too wide for most Lower Manhattan Streets, another option is a bilevel station with two tracks per level, with useful cross-platform transfers, and about 15-18 meters wall to wall per level.

The best time to have built this project was right after 9/11. Because of the connection to World Trade Center, it could have been funded out of 9/11 recovery funds, which instead went to the Calatrava PATH terminal. In addition, the rebuilding of WTC and the PATH terminal could have been done in tandem with the new train station, making its placement far easier. Alas, this did not happen, and all available space east of the WTC site is gone.

The difficult part in building this is the train station, not the access tunnels. Access tunnels can be built very deep using tunnel-boring machines and can even go directly underneath existing subway tunnels. For example, Paris Métro Line 14, whose construction cost was low relative to its depth and Paris’s underground complexity, goes parallel to and under Line 7 for a short segment, and a few kilometers of the RER D were constructed alongside the preexisting RER A tunnel. While I’m not aware of similar examples outside Paris, this shows that it is doable. In contrast, I don’t know of new stations built right underneath preexisting stations in parallel except for some high-cost US projects like the BART Market Street Tunnel Sixth Avenue Subway (thanks to Adirondacker for the correction). Stations-under-stations in oblique configurations exist in Paris, but the biggest station, namely Chatelet-Les Halles, avoids this.

The upshot is that many streets in Lower Manhattan are suitable for new tunnels, and it’s even possible to have access tunnel go deep under building foundations. However, since nearly all suitable north-south streets already host subways, a new station is more difficult. The option for a station under an existing subway station exists, but would be expensive. Let us then scout alternative locations.

Although it’s desirable to have cross-platform transfers, this configuration is more difficult than a cross configuration, in which the Brooklyn-New Jersey line goes east-west in Manhattan. This is because under a north-south configuration it may be difficult to have the Brooklyn-New Jersey tracks dive and turn west fast enough to connect to the desired New Jersey end. An east-west configuration also permits narrower single-level station caverns.

Finally, we should consider the fact that peak employment in Lower Manhattan is around Wall Street, judging by where skyscrapers are located, whereas the main station to connect to is at Fulton Street. In both cases, the key location is well east of the WTC site. Thus a Cortlandt Street location is suboptimal, and a West Street location, where there’s enough space for everything, is too far away. We will repeatedly look at the location relative to existing subway stations in Lower Manhattan: close is good, intersection is not so good.

With the above in mind, here are the options:

East-West Lines

Liberty Street

Advantages: wide enough for everything for a long stretch, from Cortlandt to beyond William. A possible 18*300 box exists from Nassau to Cortlandt, requiring demolition of at most a few low-rise buildings west of Church Street; another box with a slight curve exists from Church to William. There’s a 200-meter stretch that’s 25-meter wide from building edge to building edge. Zucotti Park permits a main entrance with enough space for pedestrians that commuters wouldn’t saturate the neighborhood’s narrow streets. Liberty is close to quite close to the center of Lower Manhattan. Brushes off against the existing Cortlandt Station on the R, but stays away from the other subway stations.

Disadvantages: far from Fulton Street (200 meters); closer to the Wall Street subway station than to the Fulton Street station on the 2/3 and J/Z. Also far from one of the easiest if not the most convenient locations for a north-south line. (But see at the end of the post for update.)

Vesey Street

Advantages: one block from Fulton. There’s an easy location for a north-south line (even a four-track one) next to City Hall, right north of Vesey. St. Paul’s Churchyard can act as the equivalent of Zucotti Park for the Liberty option. Intersects just one subway line, the E. There’s a very easy 200-meter box beginning at Broadway and continuing east.

Disadvantages: far from where most Lower Manhattan workers want to go to. A 300-meter box requires going under Vesey north of WTC, where building foundations may be a problem, or through narrow right-of-way under Ann. Continuing west requires threading the narrows between 1 WTC and 7 WTC, and continuing east requiring threading the Ann Street narrows.

John/Dey Streets

Advantages: one block from Fulton. Almost as close to where most people work as Liberty. Close to the Fulton Street stationhouse, which may be developed into a major retail destination because of the subway station.

Disadvantages: even a 200*15 box would require a few low-rise demolitions. To the east it could thread the John Street narrows, but to the west it would dip under WTC foundations and the memorial. A 300-meter box is impossible without going under skyscraper foundations (fortunately, nothing supertall). Crosses right under the J/Z and 4/5, though the active constraint for depth is probably the WTC foundations to the west.

North-South Lines

City Hall Park

Advantages: enormous space for everything, including a 35*300 box that misses all building foundations, all subway stations, and possibly all subway tunnels. Can bend to the south under Broadway and go under a mid-rise building to the north. A lot of space for pedestrian circulation both in the park and on Broadway.

Disadvantages: closer to the City Hall subway stations than to Fulton – in other words, located about half a subway stop farther away from the CBD than Fulton, which is already at the margin of the main cluster of skyscrapers. If it were picked, then there would probably be a need for a South Ferry train station serving the southern end of Lower Manhattan, which is about a kilometer from the southern end of this location – and South Ferry is so close to the water the station would have to be deep-level.


Advantages: close to everything. The street offers about 21 meters of width until well south of Zucotti Park, which is more than enough for a box, and the only stretch narrower than about 24 is at Fulton and John, where the buildings are lower-rise.

Disadvantages: right under and parallel to the 4/5. If this is a four-track option, then this means putting the station at levels -2 and -3. It’s no big deal by the standards of Chatelet-Les Halles or especially Auber, but the RER A was expensive for its time. And even the Chatelet train box was built between two Métro stations (Chatelet and Les Halles, hence the name) rather than under a station.

Church Street or Nassau Street, taking over a subway line

Advantages: could leverage some existing structures, no interference from the subway for rather obvious reasons; under both options the J/Z would terminate at Chambers, where there are existing tracks for it, and under the Church Street option the R would be rerouted along a short new tunnel (cheap by the standards of what we’re talking about here) and thence the Nassau Street tracks. Nassau is literally in the center of the Fulton Street complex.

Disadvantages: beyond the immense subway disruption this would cause, the tracks are old and have insufficient loading gauge. In addition, Nassau, the less disruptive option and the one that is closer to the peak employment center, is a narrow street and even its two subway tracks are on separate levels (see track map). Existing track geometry may impose unreasonable curves, causing squeal.

Greenwich Street

Advantages: serves WTC. Already has a grand terminal because of PATH, reducing the likelihood that politicians will spend billions on starchitecture. I believe that when the street is remapped after WTC construction is complete, it will be 18 meters wide from building to building, permitting an ample train box, even a four-track bilevel one. There’s an easy way to continue both north and south, allowing shallow construction, right underneath the 1 and PATH.

Disadvantages: too far west of where most Lower Manhattan employment is. The PATH terminal may impose unreasonable constraints. Taking over the 1 is a possibility, but presents the same difficulties as taking over the R or J/Z, and would also require an additional South Ferry station, at great depth. The proximity to WTC means the national security agencies may shit bricks about such good transit designs as shallow platforms, maximally free pedestrian circulation, and an open station. Of course the TSA and associated paranoid agencies should be (and is) fought whenever it is required, but ideally agencies should avoid situations that invite a fight they may lose.

Pearl/Water Street

Advantages: Pearl is wide enough for everything north of Fulton, and Water is wide enough south to Pine but with a 22-meter narrows between Fulton and John. Free of obstructions, as the only intersecting subway line, the A/C, is already deep. Can continue south fairly easily as well as north, curving west under the Brooklyn Bridge ramps.

Disadvantages: far east of the existing train station – 300 meters just to the 2/3 at Williams. On a similar note, east of most development; the block east of Pearl and north of Beekman is a parking lot. Conflicts with Second Avenue Subway Phase 4, but this is a small problem as the box could be built with provisions for a two-track subway underneath at little additional cost.

Gold Street

Advantages: close to both the subway station and most Lower Manhattan development. Crosses the A/C line but no station. Can continue north very easily because of the Brooklyn Bridge ramps.

Disadvantages: some room to the north of Fulton, away from the development, but very little south of Fulton; even 250*15 would brush up against the southern end of the street, which is flanked by skyscrapers. In a similar vein, continuing south requires bending west under William or maybe east under Pearl, and both require tunneling under buildings with three-figure height.

Dutch Street, or between Broadway and Nassau

Advantages: both options offer the ability to thread through some gaps in the skyscrapers, going under lower-rise buildings or lower-rise sections of the skyscrapers and limiting the amount of demolition required. Literally at the same site as the subway station. Quite close to most commercial buildings.

Disadvantages: even a 200*15 box would probably require deep-level construction to limit demolitions. Has some leeway to the south and north (assuming shallow construction) but much less than options using continuous streets. In other words, very expensive.

Update: another option is to route the Staten Island-Manhattan tunnel through Brooklyn, in which case it comes into Manhattan from the east, allowing a four-track east-west option. Both Vesey and Liberty present the option of the line to Grand Central branching north under West, then diagonally once deep enough that it can freely go under private property. Vesey presents the additional option of going north under Greenwich, which conflicts with nothing. Conversely, Liberty has the advantage of being south of the 9/11 memorial rather than south of skyscrapers, allowing the turn to West to have much wider curve radius (about 200 meters) without going under any building or the footprint of the Twin Towers.


    • Alon Levy

      I keep vacillating on this. Yes, it’s possible with the SI-Manhattan straight shot alignment. There may not be enough on the island to justify the expense and the slowdown, though.

        • Alon Levy

          Well, if there’s a concerted plan to build some TOD there in tandem with the rail link, then it’s something different of course. It could be additional NYU space, but also other things: Lower Manhattan overflow, a public square for major events (in case Union Square is insufficient), extra residential space, whatever.

          • Adirondacker12800

            Roosevelt Island South? Roosevelt Island has road access which means you can stock the bodega and haul out the garbage. Bring the fire trucks in and take the ambulances out. Before you get all TOD-y you have to figure out a way to supply all that TOD. Building it and operating it.

  1. Eric

    Um… any chance of using the PATH tunnel as the western approach to the regional rail station, and continuing the PATH tracks eastward to Brooklyn? Hoboken-bound RR trains would be rerouted to the PATH tunnel west of Journal Square, and all eastbound PATH trains would go to Midtown.

    • Alon Levy

      Maybe if the loading gauge were compatible with mainline trains, which I believe it isn’t (it’s compatible with the IRT instead). On top of that, the WTC station would have to be rebuilt from scratch to convert the loop into a straight two-track station; this of course applies equally well to the proposal to connect PATH to the 6.

    • Adirondacker12800

      No. What Alon said and the PATH tunnels are at capacity. You’d lose a lot of riders to the through trains but through trains would induce demand. It’s the same problem they ran into trying to bring trains to downtown from JFK. There’s lots of tunnel but they are running at or close to capacity.

  2. Steve S.

    What about Water as an N/S station? The major strengths would be that it would be on a wide street (~ 70 ft. IIRC) for the area, and remains so literally all the way down to South Ferry, and that a station box–especially one with narrower platforms, or bilevel–may be built directly intersecting Wall and wherever the E/W line is located; and its south approach would offer an optimal angle for a straight shot tunnel to Staten Island; the weakness is that it is much closer to the water line than all the others, being at the edge of where the natural topography ends and fill begins, and thus (possibly) much more expensive to construct and maintain.

    • Alon Levy

      That’s the Pearl/Water option I brought up – the wide street is Pearl north of Fulton and Water south of Fulton, and a station could even straddle both.

      You’re right, the closeness to the water line could make this less constructible than I thought. I was thinking of this as one of the easy but far-from-everything options, even more so than City Hall Park. But the water table could pose problems, yes.

  3. jim

    In Madrid, they used large bore tunnels (I believe about 40′ diameter) for one of their Metro extensions. The tunnel could accommodate two tracks side by side (which eliminated the need for blasting caverns for crossovers) or two tracks vertically stacked with room next to each for its platform (which eliminated the need for blasting station caverns). To create a station, there only needed to be an access shaft which touched the tunnel next to the platforms and which contained enough vertical access (elevators, since these were deep tunnels) to lift passengers from the tunnel to ground level. The size of the access shaft was to some extent independent of the length of the platform.

    It strikes me that two such tunnels at right angles with a common access shaft might be a useful concept for a lower Manhattan station. Ground or near ground level facilities would be designed solely around paxflow. Since ground or near ground space is hard to come by in lower Manhattan, this would help.

    Large bore tunnels have other advantages. If the Staten Island-Grand Central tunnel did, in fact, run under Governor’s Island, it would be relatively easy to create a station there. The platforms could be built into the tunnel when the tunnel is being constructed and an access shaft added later, if required. It may well be there are other locations where that sort of pre-planning might be helpful.

    • Alon Levy

      Oh, there’s that. You’re right, it could make a deep-level option much easier, although at the cost of cross-platform transfers. It might need to be modified somewhat to improve passenger circulation, incorporating escalators and not just elevators to avoid the crunches seen at 168th, but that shouldn’t be a big problem. On the other hand, looking at this diagram of Barcelona’s L9 shows a pretty hefty station cavern next to the platforms – though, on the third hand, despite the immense cost overrun, L9 is still ungodly cheap by American standards.

      But yes, if this can be done deep-level on the cheap, then it’d be ideal. Lower Manhattan has enough public plazas like Zucotti that could be used for easy pedestrian circulation.

      • anonymouse

        The two-track tunnel works great when your trains are 7.5 feet wide and you have overhead power supply. The M-series cars, on the other hand, are close to optimal for a mix of single track tube tunnel and cut and cover box.

      • jim

        You’re looking at slide 60, right? It’s not that large a cavern. There’s a bunch of elevators surrounding a circular shaft (inner diameter around 10-15m, so there’s somewhere between 32m and 48m for elevator faces: figure 2-3m per elevator, provides for 20 or so elevators; assume the elevators are on the order of 5m deep, then the outer diameter of the shaft is 20-25m) and a fairly narrow passage from the shaft to the tunnels taking the place of a couple of the elevators. Look at the figures, especially in the upper right corridor, to get a sense of the scale.

      • Anon256

        Couldn’t you do cross-platform transfers with two large bore tunnels (with two vertically-stacked tracks each) next to each other, and a large number of short passages punched between them? Something like many deep London Underground stations, but with bigger tubes.

  4. ant6n

    One question is how long RER/S-Bahn trains should be. Generally the service should be very frequent in downtown; and there should be more than one station to distribute the load of passengers, so trains/stations can’t be/don’t need to be too long.

    Most of Germany is built for 200m trains (during rush hour, 3x67m); Berlin S-Bahn trains are only 160m long. Paris’ RER appears to be longer (although i couldn’t find numbers).

    To what extend do longer trains increase costs for train stations (need for bigger caverns, need to accommodate more passengers every time a train arrives)? To what extend do longer trains decrease the maximum possible frequency?

    • Alon Levy

      Paris’s RER trains are a bit longer. Each trainset is 112 meters, but they couple 2 sets, so the box is approximately 224*infinity. Tokyo’s JR East network is mostly 200 meters, with some newer lines allowing 300. The reason I’m proposing 200/250/300 in New York is that the trains used by the LIRR and Metro-North are normally 200 meters long, with some 300-meter trains on busier lines or at busier times of day, and the cars are 25-meter long and normally go in married pairs.

      Longer trains increase station cavern size linearly, so the variable costs go up linearly. But some fixed costs, such as the train box, are not dependent on train length. For subways at least there’s too much noise in the numbers to get a good formula for the dependence of cost on station size.

      Longer trains also increase the minimum headway since the distance between a train’s end and the following train’s front has to be kept constant, but the effect isn’t large. The time needed to clear the extra 100 meters, at an average speed of 45 km/h (reasonable for entering and leaving stations), is 8 seconds. For what it’s worth, the 224-meter trains of the RER A achieve 120-second headways.

  5. MNF

    Intriguing. What’s old is new again. This has been kicking around for about a century. I’d actually suggest (and it’d be easier to create) an extension of the ESA lower level to GCT further south to Fulton St. Then you get access to the East Side as well as potentially a one-seat ride east to geological and geographical Long Island.

    • Alon Levy

      For access to Long Island, you can just go on the LIRR Atlantic Branch. The only thing that connecting to ESA gives you that connecting to the existing Grand Central tracks won’t is access to the Port Washington Branch, the Main Line west of Jamaica, and the NEC between Sunnyside and New Rochelle. Your mileage may vary, but I think these are less important than connecting to the Harlem Line.

      And on top of that, the ESA TBM is being disassembled and left south of the tunnel, so that it’s going to be difficult to extend.

  6. jim

    Following my usual scope reducing tendency (sorry; old habits die hard), I ask whether the links to Staten Island and Brooklyn are worth the cost. Brooklyn is already well connected to Manhattan and there’s not that much in Staten Island to connect to. A line from New Jersey (somewhere around Hoboken Terminal) into lower Manhattan that then turned to the north and followed the Bowery into 4th Ave, thence up Park Ave. into Grand Central (yes, breaking through the south wall of the lower level into tracks 105-112, as in ARC Alt. G), though, would be a useful “son of ARC”. It would achieve for NJT most of what ARC would have and allow Metro-North to directly serve lower Manhattan and run into Hoboken, Newark, Meadowlands …

    That would suggest an E-W station under Vesey, perhaps between Church and Broadway. Good connections to Fulton. Close to the developed portion.

    • Alon Levy

      The Brooklyn-Manhattan connection by itself isn’t useful, you’re right. The point is to connect the LIRR to Lower Manhattan, without requiring a transfer at Atlantic. Not only is that transfer complex, but also commuters tend to dislike transferring at the downtown portion of the trip. Brooklyn at the same time gets connected to Hudson County in an easier manner; the Jersey-to-Brooklyn commute market isn’t huge, but it exists.

      The Staten Island-Manhattan connection is about connecting Manhattan to, well, Staten Island. It’s expensive, but it would replace the much slower ferry, and also provide a one-seat ride for people living near the Staten Island Railway (and the North Shore Branch, eventually).

      The Jersey-Grand Central connection you propose could work, but not through Hoboken, which is too far north. If they send Erie trains into Jersey City, either into the old Erie Pavonia terminal (the right of way already exists) or via some new construction along the PRR and PATH, then it could work, yes. But otherwise, trains would curve sharply to the south and thence sharply to the north.

      • Eric

        Wouldn’t taking two lanes off the Narrows bridge for a subway connection be a much cheaper (order-of-magnitude) way of connecting SI?

        • Alon Levy

          It would be cheaper, but not by an order of magnitude – the bridge rises too fast for a connection to the R at Bay Ridge, which means that it would require extra tunneling from another point, or intensive tunnel realignment. Besides, I’m still not sure whether the bridge can accommodate trains at all. The grade is doable for EMUs, but I’ve also heard in comments that it’s very light and couldn’t take the weight of a train.

          For what it’s worth, a subway connection under the Narrows is $1.3 billion on my spite map, under the dubious assumption that per-km cost is the same underground and underwater; under the same assumption, St. George-Fulton direct is about $2.5 billion. (For the record, in my regional rail posts, the implicit assumption is that SI-Fulton is the same as the budget for Bay Ridge-Jersey City, $7.4 billion.) But the cheaper option would come together with much lower benefits. A Narrows connection would serve Grasmere rather than St. George, and would take a long time to reach Manhattan, saving some time for people south of Grasmere but none for people on the North Shore.

  7. Adirondacker12800

    Went back and reread because things are brewing in my addled brain like “how many passenger at peak of the rush hour?” I’m coming up with 30-40 thousand an hour. In rereading, this caught my eye
    In contrast, I don’t know of new stations built right underneath preexisting stations in parallel except for some high-cost US projects like the BART Market Street Tunnel. Stations-under-stations in oblique configurations exist in Paris, but the biggest station, namely Chatelet-Les Halles, avoids this.

    The Muni stations under Market Street were built at the same time as the BART stations and the concourses. Before that the Muni streetcars… ran on the street east of Castro St.
    There’s oblique stations in New York City. Stations along Broadway. Grand Central on the Lex is at an angle…. which is why the “new” platforms were called the diagonal platforms.
    The H&M built their tunnels under the 6th Ave. El. The IND built the 6th Ave tracks around the H&M.. but only local tracks. In the 60s the NYCTA comes back and builds the express tracks between W4th and 34th. I’m sure it was very interesting threading the BMT around the IRT downtown and then coming back 10 years later and threading the IND through both. 59th street on the Lex was a local stop until the 50s.
    Built all at once over/under? The Lexington Ave line and 8th Ave line along Central Park West. There’s all sorts of interesting things going on under Flatbush Ave. All sorts of interesting things went on in East New York as the Els came through, the subway followed and the LIRR went underground and Conduit Ave gets added to the mix. In Chicago, try to figure out Upper Wacker Drive, Middle Wacker Drive and Lower Wacker Drive and the streets between there and Millenium Station.

    • Alon Levy

      Yeah, Sixth Avenue Subway is a better example than BART – it was built first around and then under an active subway. It also was really expensive and was one of the reasons the IND ran over budget so much (no link, sorry).

      For some reason, I keep forgetting that Muni Metro was only put underground in the 1960s and not in 1912. Clever’s paper bashing the Bay Area for lacking the foresight to make the same-direction BART-Muni transfers cross-platform should have clued me in.

      • Adirondacker12800

        They would have needed fare control in the middle of the platform or had Muni riders swipe out…. I’m sure part of it is that it’s slightly cheaper to dig a deep hole two tracks wide than it is to dig a shallow hole 4 tracks wide. I’m sure the over/under configuration on Central Park West and Lexington saved money. Makes sense for other reasons on CPW, residents don’t have to cross CPW or you don’t have to build a mezzanine across CPW. Twice as many token booths unless you build the mezzanine. With a mezzanine a round trip takes four flights of stairs, two of them in the up direction. Over/under means three flights of stairs and only one is in the up direction for trips to the south.

        • Alon Levy

          Don’t be so sure about the cost thing – extra mezzanines cost more, and the deeper the tunnel, the more expensive it is to cross later (they had to have planned on some subway crossing Market in the future… right?). Also, the correct thing for Muni and BART to have done was integrated fares, Swiss-style, or at least cross-platform transfers NJT-PATH style.

          • Adirondacker12800

            You couldn’t integrate fares Swiss style in 1969, the filing cabinet sized micro computer to do it and it’s air conditioning unit isn’t the kind of thing you want to put on every bus in the Bay Area.

          • Alon Levy

            You don’t need a micro computer for integrated fares. Just print a ticket on paper. Works for the Swiss today. Smartcards are a nifty innovation, but they’re not strictly necessary.

          • Adirondacker12800

            And what did Swiss fare integration look like in 1969? It’s relatively cheap and easy to print out custom tickets in 2011. Not so easy in 1969.

          • Alon Levy

            If I’m not mistaken, the same as today. It’s really not difficult. Even in the US, buses gave transfers, and previously disparate subway systems were within the same fare control. It’s purely a matter of organization.

          • Adirondacker12800

            So the fare from Van Ness to the Ferry Building is going to be the same fare as from Van Ness to Fremont – the municipality south of Oakland not the street?

          • Alon Levy

            No, why would it be? The fare from Columbus Circle to Times Square was lower than the fare from Far Rockaway to Times Square in the first few years after the subway took over the Rockaway branches.

            For an example without any fare barriers, travel that crosses from outside Berlin to within Berlin’s Ringbahn costs more than travel that stays within city limits or outside the Ringbahn. In fact usually a coherent zonal system is what comes out of fare integration – instead of paying based on how many transfers you make (the two-fare zone in New York, etc.), you pay based on distance. New York was an exception, but this is exactly what was done in Singapore recently.

  8. Joseph E

    I’d forgotten about the bad transfer between BART and Muni along Market Street. It’s really a shame – even if they could not have built it all on one level (or two, with a concourse above), they could at least have put both east-bound tracks on one level, and the west-bound tracks on the other level, with the same construction format. But this would have required coordinating fares and enforcement between Muni and BART, so of course it did not happen.

    Currently, you have to go up two level from the BART station to the concourse, out of the faregates, buy another ticket, go thru another faregate, and then back down another escalator to the Muni platform. It would cost a fortune to fix this, after-the-fact.

  9. Pingback: Staten Island Rapid Transit | Pedestrian Observations
  10. Pingback: Transit Alternatives to the Tappan Zee Widening | Pedestrian Observations
  11. Pingback: Quick Note: How Much Tunnels Really Cost | Pedestrian Observations
  12. Pingback: Troll Rail Projects | Pedestrian Observations
  13. Pingback: Large-Diameter TBMs | Pedestrian Observations
  14. Henry

    I feel like the amount of cross-platform transfers is underestimated – a more apt comparison is the chronically crowded Admiralty station in Hong Kong, since everyone wants to transfer to the northbound Tsuen Wan Line during the PM peak. Similarly, the Midtown bound platform will see very heavy inflow during the AM peak, since that’s where nearly all non-Downtown bound passengers are headed. This becomes even more true if the fares for the regional rail systems are rationalized (or at least made less mind-bogglingly expensive) within the city limits, since nearly everyone in Brooklyn wants cheap, fast Midtown access.

  15. Pingback: New York Regional Rail: the Central Segments | 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 )

Twitter picture

You are commenting using your Twitter 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.