The Northern Branch Extension is a Waste

As I mentioned in yesterday’s post, negotiations in New Jersey between Governor Chris Christie and the state legislature have resulted in a significant increase in the state fuel tax. The money will raise $16 billion for funding the eight-year Transportation Trust Fund plan, and be matched with federal funds to bring the amount up to $32 billion. Unfortunately, the money is being wasted. Details of most of the plan remain vague, but it appears most of the money will go to road repair; for all I know, $4 billion a year is a reasonable amount for this. But one component of the plan is extension of the Hudson-Bergen Light Rail system north into Bergen County, along the Northern Branch. This is at best a marginal project, and in the long run would make regional rail modernization in Northern New Jersey more difficult.

Despite its name, the HBLR only operates in Hudson County. Plans for extension into Bergen County along the Northern Branch still play an outsized political role due to the name of the line, but have not been realized yet. Right now, the line is partly the light rail system of Jersey City, and partly a circumferential line linking dense areas west of the Hudson, as somewhat of a circumferential. As such, it is a combination of a radial and circumferential. The Northern Branch would send it 13 km farther north into suburbia, terminating in Englewood, a town center with a fraction of the job density of the Jersey City CBD. Projected weekday ridership is 21,000, a little more than 1,500 per km, weak for an urban light rail line. (The HBLR’s existing ridership is 54,000 per weekday on 55 km of route.)

The original cost estimate of the Northern Branch extension was $150 million, low for the length of the line. While reactivating a closed commuter rail like the Northern Branch should be cheaper, the line is single-track still hosts some freight service, so light rail would have to build new tracks in the same right-of-way, raising the cost range to that of urban light rail. Unfortunately, the cost rapidly escalated: by 2009 it was up to $800-900 million, and in 2015, after the proposal was shortened to its current length from an 18 km proposal going deeper into the Bergen County suburbs, the cost was up to $1 billion. The cost per rider is still much better than that of the Gateway Tunnel, but it makes the project marginal at best.

While the high cost may be surprising, at least to the reader who is unused to the expense of building in or near New York, the limited ridership is not. The original plan, going beyond Englewood, would have terminated the line in Tenafly, a wealthy suburb where my advisor at Columbia used to live. Many people in Tenafly objected to that plan, not so much on the usual NIMBY grounds of traffic and noise as on the grounds that the line would not be of much use to them. They were interested in taking public transit to go to Manhattan, and the HBLR system would not be of any use. Of course, Columbia professors would not be using a rail network that went directly to Midtown or Lower Manhattan, but most of the suburb’s Manhattan-bound residents work in the CBD and not at Columbia.

I would probably not be this adamantly against the Northern Branch project if it were just one more over-budget light rail line at $45,000 per projected rider. The US has no shortage of these. Rather, it’s the long-term effect on regional rail.

The Northern Branch would make a good commuter rail line, going from Pavonia (or possibly Hoboken) north to Nyack, connecting to the HBLR at its present-day northern terminus, with about the same stop spacing as the proposed HBLR extension. Potentially it could even get a loop similar to the proposed Secaucus loop of the Gateway project allowing it to enter Penn Station directly. An even better connection would involve a second tunnel between Pavonia, Lower Manhattan, and Atlantic Terminal on the LIRR, with a new transfer station at the junction of the Northern Branch and the Northeast Corridor. Consult this map, depicting the inner segments of various potential commuter lines: the Northern Branch is the easternmost yellow line, the Northeast Corridor is in red and green.

The importance of the Northern Branch for regional rail is threefold. First, the easternmost line in North Jersey today, the Pascack Valley Line, misses a large swath of territory farther east, which is covered by the Northern Branch and by the West Shore Line. The West Shore Line actually passes through somewhat denser suburbs, with more Manhattan-bound commuters, but is a major freight route, whereas the Northern Branch has little freight traffic, which can be scheduled around passenger trains or even kicked out. Second, again shared with the West Shore Line, the Northern Branch provides a north-south line in Hudson County west of Bergen Hill, where there is suitable land for transit-oriented development. And third, the terminus, Nyack, is a town center with a walkable core.

I wouldn’t really object to making the Northern Branch light rail if it were cheap. At the original cost estimate of $150 million, I would be mildly annoyed by the lack of long-term thinking, but I’d also recognize that the cost per rider was low, and at worst the state would have to redo a $150 million project. At $1 billion, the calculus changes considerably; it’s a significant fraction of what a tunnel under the Hudson should cost (though not what it does cost given the extreme amount of scope creep).

High costs, as I said in 2011, should not be an excuse to downgrade transit projects to a cheaper, less useful category (such as from a subway to light rail). In this case we see the opposite happen: high costs are a reason to reject a downgraded project, since the cost per rider is no longer low enough to justify shrugging off the long-term effect on regional rail restoration.


  1. Kenneth Hammer

    I would ask/suggest that with the NYC, Northern New Jersey, and Connecticut transport systems so tied together, that their should be a Transportation Body, that’s maybe in charge of running, but is definitely in charge of long term planning and building of the Transit System. As things stand now, it looks like there are 3 or 4 different groups all in charge of Public Transit in the New York City Area with New York handling New York, New Jersey handling New Jersey, and maybe they sometimes talk to each other over mutual transit projects. The same with Connecticut Transit Systems in the NYC Area.

    • Alon Levy

      In theory Port Authority is supposed to handle capital projects between New York and New Jersey. In practice it’s worse-run than the MTA and New Jersey Transit, because it has a slush fund (namely, the bridges and tunnels) and doesn’t need to justify its spending decisions to anyone.

    • Syd Chan

      There was a reason why the Tri-State Regional Planning Commission failed in the 1980s, and due to the need to coordinate the politics of three states, whose capitals are all on the edges of the region, a single public-sector regional planning and/or transportation commission, is likely to fail again. It’s feasible enough for two states to coordinate, hence why the Port Authority of NY and NJ still survives, but throw in a third state without a strong federal hand to steady the politics, and it’ll just implode again.

  2. Syd Chan

    There’s a lot of negatives than than just the positives that you outlined for regional rail service on the Northern Branch. The Northern Branch corridor is too short to provide regional rail that meets the commuter rail standards of New Jersey Transit, and, although the main portion of the line has little freight traffic, the lower portion of the rail corridor that allows access to Husdon County and Manhattan is a very, very busy freight corridor run by CSX. Even if a schedule is worked out to allow regional rail to run down the CSX corridor, unless an expensive loop is built across the mouth of the North River Tunnels/Gateway, there’s no direct access to Manhattan, which loses that ridership. Regional rail service was considered along the corridor in the past, but was rejected as there was no FRA-compliant train equipment available that would have allowed freight traffic to mix with the passenger traffic of Northern Branch service. Furthermore, the ridership from these northern Bergen suburbs to Manhattan is already more effectively served, time-wise, by buses over the GW Bridge that drop passengers off at the GWB bus station. Also, the politics along the corridor include a socioeconomic factor between the very wealthy residents of Tenafly and points north, versus the more modest denizens of downtown Englewood and points south; this socioeconomic factor encourages Tenafly to oppose ANY public transit improvements, whether light rail or regional rail, from the south of their town, for fear of increased crime. Lastly, the vision of tying in Northern Branch service to a future possibility of connecting Pavonia/Hoboken Terminal to Atlantic Terminal using an RER-based tunnel is so far in the future that it’s not even on the political map.

      • Matthew

        This. NJTransit runs multiple varieties of frequent express buses and a frequent local bus all day to Midtown. During the XBL times these express buses can reach Midtown within 15 minutes from Palisades Park. Those buses are heavily used in the Pal Park – Tenafly – Dumont corridor. (Also there is CoachUSA running some buses express from Pal Park while making some stops in the same towns). Yeah, they go to the PABT, but oddly enough there’s tons of jobs tucked away in the towers behind the touristy spots in Times Square. And better connections than Penn Station has. The NJTransit buses serving those communities via the GWB are infrequent in comparison. The jitneys help a bit, especially for Fort Lee, but still. You have a 20 minute ride on the A train even after you get across the river, if you’re heading to Midtown (but you wouldn’t do that since the buses are much better).

        The extension of HBLR is more about access to Newark, Jersey City and Lower Manhattan. It would be very good for the region in that sense, as a circumferential route relative to NYC, but you’re right about the cost growing out of control. Very sad. I don’t see a commuter rail branch being reactivated here: there’s no advantage over the buses. Especially if the PA ever pulls its head out of its ass and creates a full-time XBL. We’ll see what they do about the bus terminal though, what a mess.

        • Adirondacker12800

          Traffic isn’t going to be any lighter in 2025. Or the bus terminal less crowded.
          The XBL is mostly on state highway. The PA may be paying to set it up and tear it down but it’s the state’s decision. Automobile drivers tolerate it during rush hour because a bus is whizzing by frequently. They may not tolerate it midday. When the bus frequencies are much lower.

    • Alon Levy

      A couple things.

      1. FRA-compliance is going to change very soon; the FRA is supposed to circulate new rules any day now (I think they’re already in the extension period after they blew a deadline last month). What’s more, if Caltrain managed to get a waiver, setting in motion the chain of events leading to the upcoming rule changes, NJ Transit could most likely run whatever it wanted.

      2. For $1 billion, they can build a good fraction of an RER tunnel. Somehow, when the excessive costs come from overruns on small projects, nobody ever says “it’s so far in the future.” (See also the South Coast Rail project – $3.3 billion, 5,000 weekday riders, Charlie Baker’s busy chasing the 50,000-weekday-rider Green Line Extension over cost savings and doesn’t care about cost blowouts on South Coast.)

      3. The West Shore Line is a heavy freight corridor, but the Northern Branch isn’t. They share tracks for a few km, which would have to be quadrupled, but it’s cheaper than doubling the entire Northern Branch (or tripling it through Englewood).

      4. Per this map, the people in the suburbs along the Northern Branch who commute to Manhattan mainly do so by car; along the West Shore Line, it’s a mix of car and bus commuters.

      5. At least this time, the Tenafly NIMBYs complained about traffic impact of being the terminus and about lack of service to Manhattan, and not about crime or drugs or whatever people say when someone proposes school integration.

      • Adirondacker12800

        Well no they never say they are afraid of those people. It’s going to curdle the chicken milk. Make the cow eggs thin shelled. The spaghetti vines will produce mushy pasta. …

        • Alon Levy

          Hey, in Shaughnessy, the response to initial proposals to run the Canada Line down the Arbutus rail corridor included the phrase “creme de la creme” in an unironic reference to the neighborhood (link). The NIMBYs succeeded in getting the line off their backyards and onto Cambie, which happens to be a much better corridor for it anyway.

          • Syd Chan

            And the politics in Vancouver are much different than the politics in New Jersey. A much smaller metro area, for starters.

          • Adirondacker12800

            Someone has to show them how much more they can get for their house if they can put “walk to train” in the ad. ….maybe that’s the underlying problem. They are all Real Canadians ™ that drive everywhere. They don’t want their property to be worth more. With a bigger tax bill.

            …Everybody wants another High Line. It needs Manhattan around it to make it another High Line.

      • Syd Chan

        1. FRA-compliance hasn’t changed soon enough for this project, unless you want to throw away the entire planning process for the past 10 years and start over with the DMU option. The DMU option could work with River Line-style Stadler units, but even the River Line has a FRA waiver with a time sharing agreement. I doubt CSX would agree to a time-sharing agreement on the busy West Shore segment, or even to mixed-running with passenger and freight traffic, so you will still need to spend big bucks to build dedicated tracks. Also, extending Northern Branch service to Gateway or Hoboken, assuming that dedicated tracks on the CSX right of way will be needed, will cost even more.

        2. It’s possible to “build a good fraction of an RER tunnel” with 1b euro, but $1b USD is not the same as 1b euro. Also, a tunnel from Pavonia/Hoboken to Atlantic Terminal in Brooklyn would be much more expensive than tunneling across Paris, due to having to cross two deep and wide rivers that require deep tunneling, electrification, serious waterproofing, and serious ventilation. Also, when excessive costs come in from small projects, the absolute value of the excessive cost is still much smaller than that of a large project (for example, $0.5b in extra cost for a $1b project is still much smaller than the absolute value of a $10b project). Additionally, a tunnel from Pavonia/Hoboken to Brooklyn would also preclude DMU service on the Northern Branch anyway, since DMUs aren’t likely to be run in such a long tunnel.

        3. There’s no plan to triple the Northern Branch through Englewood. While the $150-200m figure was the estimated cost of the DMU option, the $1b figure is the estimated cost today of the EMU option. The high cost of the EMU option is not from building extra track on the upper portion of the corridor, but from electrifying the branch, since new electrical substations and infrastructure are EXPENSIVE – armchair transportation planners always underestimate the cost of electrification by an order of magnitude.

        5. If you’re not from New Jersey, you don’t understand how snotty the denizens of Tenafly/upper Bergen County are. Even when they couch their concerns in terms of “traffic impact” and “aesthetics of train yards”, the implication is that they look down on people from lower-income towns. They’re savvy enough to know that, if they directly call out “crime”, “drugs”, or “race”, they’ll get shot down, so they use indirect terms like “traffic”, “aesthetics”, and “no benefit to me/my town”. This is especially true of the relationship between Tenafly and neighboring Englewood; it’s hard for outsiders to understand the socioeconomic antipathy between Tenafly and Englewood. Englewood’s demographics are unusual for a geographically-small North Jersey town; it has a downtown core consisting of low-income residents who are socially segregated into a mediocre public school system, and an east side of town consisting of very high-income residents who send their kids to a neighborhood private school with very prominent alumni. However, when you average out Englewood’s demographics into a town-wide whole, it masks the social issues to outsiders and creates a different impression than what neighboring towns like Tenafly conceive.

        • Alon Levy

          1. Oh, fuck DMUs. Stringing wire over the track isn’t expensive (it’s been done, in the US, for $2-3 million per double-track-km), and semi-reasonable compliant EMUs exist. This is especially important on a line with such short stop spacing – DMUs frankly don’t accelerate fast enough, even the light noncompliant ones. The Silverliner V has an acceleration penalty to 60 mph of about 14 seconds, comparable to the FLIRT. The Talent has an acceleration penalty to 100 km/h of 32 seconds. And as for dedicated tracks on the shared segment with the West Shore, that has to happen under the light rail option anyway.

          2. I wasn’t even thinking of Parisian costs. The original budget for ARC, in 2003, was $3 billion; given difficulties of tunneling under the river, that sounds about reasonable to me. Another local point of comparison: the Cross-Harbor Rail Tunnel, going under about 7 km of harbor, was claimed to be $7.4 billion in one of the studies from last decade. Tunneling under a wide river is expensive, but at the end of the day, it’s around 5 km of tunnel from the portal to Penn Station (on the NEC) or to Fulton Street (via Pavonia).

          3. The light rail option involves new tracks in the corridor! As for electrification costs, California HSR seems to believe it’s $3 million per km, Amtrak electrified New Haven-Boston in the 1990s for around $2.5 million/km, SNCF seems to believe electrifying a single-track line is €1 million per km. Light rail lines are of course more expensive than this, because of track construction, but $70 million/km is very much the high end, especially when it’s in an existing ROW and not on city streets. At the low end, some lines have gotten built for $15-20 million/km (and no, most of it was not electrification).

          5. I think how people phrase “no benefit to my town” is instructive. For example, when the Providence Foundation ran a study of commuter rail from Providence to Woonsocket, with a diagonal timed transfer at Pawtucket between the line and the Providence Line to Boston (so Woonsocket Pawtucket Boston is timed), the people in the high-income suburbs between Providence and Woonsocket were uninterested in the project. They explained that they don’t need transit to Providence because they can just drive there, and don’t need the timed transfer to Boston because they can park at South Attleboro and take a direct train to Boston. This says a lot about how users in those suburbs conceive of commuter rail and of the Providence Line. So the fact that Tenafly NIMBYs chose to use “it won’t get us to Manhattan” as their line says something about what they’d want from the transportation system.

          • Adirondacker12800

            There’s no reason why they couldn’t run Silverliners over it. Or M8s without the third rail capability. Or whatever Bombardier comes up with that looks like an electric multilevel. Or a diesel and some Comets. They run in mixed traffic now. The Princeton shuttle is short, they can be run in short consists.

            It’s very probable that they are class 1 tracks and need to be replaced. A really quick surf through the bus schedule has 31 buses an hour between 7 and 8 AM, to the PABT, through Tenafly. If you get half of that on the rail line it needs to be double tracked. 31 buses an hour is probably a good place to railstitute.
            ……. better than Suffern to Port Chester….

            It does squat for the stereotypical Columbia professor who lives in Leonia. It does a lot for the people who are taking a bus to the PABT to get on the subway and go to Wall Street. Or Tribeca or Soho or Greenwich Village… CSX is gonna want a lot for the flyovers. The flyovers are cheaper than building more Lincoln Tunnel so the buses can go to a ten billion dollar bus terminal. … the ARC project was cheaper than more Lincoln Tunnel and a ten billion dollar bus terminal…

          • Syd Chan

            1 & 3. It’s not just stringing wire over rail; if a new electrical substation needs to be built, then that costs a lot of money. This is usually financed by the electric utility, hence why other agencies have lower costs per km, but it doesn’t seem to be the case here – perhaps the local electric grid is already near capacity? I believe that the currently HBLR electric grid is nearing capacity. Also, the new electrical substation needed in this case is going in on a wetland due to land acquisition issues, and nearby wetlands were flooded by Hurricane Sandy, so there will be additional resiliency costs. Also, costs will be high for safety and overtime when track workers are building new track next to the active tracks of the narrow West Shore right-of-way that host trains 24/7.

            2. The costs you cite are from at least decade ago. Gateway is projected to cost several billion more than ARC; Cross-Harbor, undoubtedly, if built today, will cost much more. Time is no friend to project costs, which is a criticism of Christie’s when he canceled ARC. Also, even if a new tunnel is built from Hoboken to, say, Fulton St., a new terminal in New York will need to be built, which will cost a lot of money. (One way to avoid building a new terminal downtown is to utilize the infrastructure already there, which gives me an idea: it might be possible to extend PATH from Hoboken up the Northern Branch. But then you’ll be starting the planning process anew, and dealing with questions of demand, NIMBYism, etc.)

            5. On the ground, comparing a project to case studies around the world is as useful as the technical ability to utilize lessons learned from the case study to the local condition of the project in question. The lesson learned from Providence is that super-wealthy people don’t need public transit since they can afford to drive and park anywhere. The lesson learned from Vancouver is that it’s hard to reactivate disused surface rail corridors due to neighborhood opposition. However, the technical lessons of RER in Paris is that it costs less to build a tunnel under land where you can drive construction and ventilation shafts most places as needed, than it costs to build under a wide river, where construction and ventilation shafts can’t be constructed just anywhere mid-river for technical and navigational reasons.

          • Alon Levy

            1, 3. Amtrak got to tap into the electric grid, yes. But new high-speed lines don’t seem to do that, since they’re often built in relatively out-of-the-way areas (and if anything, they cost more when they’re in denser areas, because of the extra viaducts and grade separations required).

            2. Gateway has at least one obvious example of unnecessary scope creep (Penn Station South); I’ve read, I forget where, that the cost of the bare tunnel is around $7 billion, which is still a factor of 2-2.5 too high, but is still less than what ARC was at cancellation.

            The PATH option is interesting! I’ve thought about the possibility of extending PATH north from Hoboken, along the waterfront and then west or northwest under the Hill. Especially in any context where the Erie lines get moved back to Pavonia, which independently of any Lower Manhattan tunnel is a more central place than Hoboken, it’s possible to just move the underground Hoboken station.

            5. In Vancouver, I think the plan would have been to build the Canada Line under the Arbutus corridor, not on it, in similar vein to how a short segment of the Red Line in Boston is built under the Somerville Community Path and not on it. And in Paris, it’s not really true that they can drive shafts anywhere – the RER has to go under the entire legacy Metro network, and so does Metro Line 14, with a lot of deep-level stations and really expensive (i.e. $450 million/km for the RER A) segments built with TBMs. In general, lines built under large legacy systems, e.g. in Paris and London, tend to be very expensive by local standards; what annoys me about New York is that Second Avenue Subway is a much simpler project but still manages to cost way more per km.

          • Henry Chin

            As far as extending PATH goes, PATH is already basically at capacity. You’d need a third PATH tube to absorb any more traffic; but then again, you might need a third PATH tube anyways, maybe under 57 St to the East Side.

            Extending from Hoboken adds another wye to the PATH system and makes optimizing service patterns hard. Maybe have a line from Journal Square under Bergenline, a line from Secaucus, and the Northern Branch all funnel into 57 St.

          • Syd Chan

            The Hoboken PATH station is already on it’s own wye. The wye to Hoboken is why the late-night 33rd-JSQ train enters the Hoboken station, stops, and then reverses out of the station to continue its route. Why it was designed that way, I don’t know.

          • Adirondacker12800

            Because ships would bang into the tunnel if they made it any shallower.

      • grendelkhan

        FRA-compliance is going to change very soon; the FRA is supposed to circulate new rules any day now

        What exactly happened here? You and some other folks were banging the drum about the issue back in 2011 and earlier, and nothing happened. It was sort of common knowledge among wonks that US safety regulations made for expensive, not particularly safer trains. This was the Way Things Were, as it’s hard to get people excited about loosening safety standards for passenger rail. But now this is actually going to change? Why? What are we expecting? And when’s the actual deadline?

        • Alon Levy

          In 2012, the FRA announced it was going to change the rules, and listed a lot of good vendors as reviewers for the new rules it was writing. It was supposed to circulate the new rules in 2015, then it changed to August of this year, and then it got delayed to September.

          The origin of this is the combination of the Acela debacle and the Caltrain waiver, apparently. The Acela is unreliable because of these regulations, and once Caltrain got the waiver, the FRA (slowly) started thinking about how to change the rules, even while it maintained that its rules had a good reason and shouldn’t be questioned by outsiders.

          • Adirondacker12800

            The FRA knows about crash energy management.

            Somebody has to make a proposal. Apparently they have. once the vendors get involved a lot of proprietary information gets discussed. There isn’t going to be an update about it twice a week.

            The railfan Acela meme is that the FRA “changed it’s mind”. about the trains. Bureaucracies don’t work that way. Much more reasonable scenario is that the signal companies, who had never done this before. went to Amtrak and promised that ACSES would be cheap and easy to install. And that with all this train protection they’d be able to run lighter trains. It wasn’t cheap or easy. They had to start over. The signal schedule kept slipping. The FRA said the light trains can’t run on tracks signaled with 1920s technology. Which is the same thing they have been saying for decades. Amtrak made commitments the signals weren’t able to meet. So they went with tanks on steel wheels.

            All of that takes synthesizing a few thoughts. It’s much easier to say the evil FRA changed it’s mind.

          • grendelkhan

            Would it be possible, maybe when the new regulations come out, to get an explainer on this? I’m seeing a lot of reasonable-sounding bits both from you and from Adirondacker12800, but you can’t both be exactly right. I’m trying to get some good information on this, but I’m no expert, there’s over a century of history here and people grinding axes on every side.

            Okay, here’s what I’ve got. In 2011, there was a lot of argy-bargy over Amtrak buying very expensive heavy trains because the FRA required that for crash management. (According to this post, one of the effects is more than an order-of-magnitude worsening of mean-time-to-repair.) The wait has been for installation of PTC on the Northeast Corridor, expected to have been done in 2015 (seems to be active now, but not on every line), after which near-off-the-shelf European and Japanese trains could be imported to run here, or at . This is the culmination of something that’s been brewing since the Obama administration started pushing for high-speed rail around 2008. Bad FRA regulations don’t stop with over-heavy trains, but it’s the biggest factor.

            (Confusingly, the video that Adirondacker12800 posted above actually shows an FRA-compliant high-buff-strength train at top experiencing override, and a European-style crash-energy-managing train at bottom performing much more safely, according to that guy.)

            So, in short: FRA regulations required either heavy trains or PTC; PTC was difficult, but is now rolled out or rolling out everywhere, so cheaper, more maintainable trains will follow. Was the requirement for PTC a reasonable blocker? Could American passenger trains have used crash-energy management instead of pure buff strength in the meantime? Do I have all this right?

      • Nathanael

        “3. The West Shore Line is a heavy freight corridor, but the Northern Branch isn’t. They share tracks for a few km, which would have to be quadrupled, but it’s cheaper than doubling the entire Northern Branch (or tripling it through Englewood).

        See my comment below. The shared section is a big pain because it has active industrial customers and CSX wants to expand the rail yard. So I say, reuse the Edgewater tunnel, bypass the entire shared section, and gain a bunch of useful riverfront stops.

        • johndmuller

          The Edgewater tunnel is in a reasonable position for a new tunnel under the Hudson to connect to 125th Street in Manhattan and the SAS. The SAS would then run on the Northern Branch (and/or run along the CSX / Susquehanna lines to Paterson if sufficient ROW width is available). I don’t know if there is sufficient horizontal/vertical space to fit an Edgewater station and the appropriate lead-ins to both of the tunnels.

          • Alon Levy

            The problem with this: demand along SAS and 125th Street is far higher than along the Northern Branch. This means that any subway-based solution needs to either terminate in a dense place (so, maybe Fort Lee), or branch in the suburbs. Two branches are probably still too few: the tunnel needs to be carrying a 180-meter-long train every 2-3 minutes at rush hour and every 5-6 minutes off-peak, which means that each branch needs to be able to support a train every 5 minutes at the peak and every 10 off-peak. With commuter rail it’s possible to vary train length (since less of the demand comes from inner-urban areas) and also cut branch frequency to every 15 or 20 minutes off-peak.

          • Syd Chan

            Because the eastern portal to the Edgewater tunnel is obviously above the level of the river, it’s most likely not technically feasible to reuse the Edgewater tunnel for an under-river tunnel to Manhattan that would be built to a much deeper level.

  3. Adirondacker12800

    Gateway, when it was announced was going to cost twice as much as ARC. And rising. On the “wrong” side of Penn Station. With one level instead of two. And without all the entrances/exits that were going to be on the north side of 34th Street. That LIRR and Metro North passengers could have used.

    The bumper blocks at the PATH station in Hoboken are on the river side of tracks going east/west. The same orientation as the suburban tracks above them. That are already packed during the peak.

    Someday, I think it’s going to be as soon as NJTransit can accept delivery on more cars, Penn Station is going to full again. It’s going to outrageously expensive to carve a deep cavern station under Fulton Street. Connect all of the suburban lines in New Jersey to all of the suburban lines on Long Island and it shifts 20, 25 trains loads out of Penn Station. And off of PATH and subway. And buses out of the PABT. Aim it northwest to southeast there can be another level for Metro North trains that go to Tottenville. Well some of them to New Dorp.

  4. Nathanael

    Problem is that that the Northern Branch is too far west, with an urban void to the west of it.

    I actually think that HBLR should be extended north on a new branch to Edgewater and then through the Edgewater tunnel. This bypasses the obnoxious, busy, active part of the freight line where it will be expensive to build LRT tracks, and runs the trains through a much denser area along the way.

    • Nathanael

      Anyway, this would not preclude future regional rail service, as there’s plenty of room on the ROW north of Fairview where the Edgewater tunnel comes out. The real problem with future regional rail service is figuring out how to get the trains to NY Penn.

  5. Nathanael

    So, the more I look at it, the more I see the core problem as a lack of rail tunnels under the Hudson. Suppose we build:
    (1) The Cross Harbor Freight Tunnel
    (2) The two new tunnels to Penn Station
    (3) #7 to Secaucus
    (4) Conversion of Lincoln Tunnel bus lane to HBLR and extension across 42nd St in Manhattan
    (5) extension of Second Avenue Subway along 125th St not just to Columbia but right under the Hudson to Edgewater and under the Palisades to Palisades Park

    I think this takes a lot of the pressure off, right?

  6. Adirondacker12800

    Cross Harbor moves the date where everything east of the Delaware gridlocks from 2030 to 2050-ish. It gives everybody another decade or so to upgrade other options. ARC was projected to reach capacity again in 2040. I suspect Gateway will too, even though it’s going open a decade later. 7 to Secaucus gets a few thousand people to Hudson Yards a few minutes faster, maybe, than walking from Penn Station. It’s not worth billions to build a tunnel for that. If the HBLR is going to share the tunnel with buses, the buses need a ten billion dollar terminal. So that people can walk down to their suburban train station and get on a bus to the PABT. It make much more sense to spend the billions so they can walk down to their suburban train station and get on a train.

    • Nathanael

      The point of HBLR through the Lincoln tunnel is that *right now* a huge line of buses crawl through the Lincoln Tunnel, take the circular ramps up to PABT, and people then get out and overcrowd the subway trying to get to the other side of town.
      Instead, these buses can stop in New Jersey and a *much higher capacity* train line can whisk them to both the west side and the east side of Manhattan. That’s the idea.

      It only makes sense because apparently PABT is falling down and has to be replaced anyway. It’s a hell of a lot cheaper to replace it in NJ than in NYC.

      For reference, the buses through the Lincoln Tunnel are mostly fairly local buses from the densely populated areas around the Palisades *which do not have commuter train service and never will*. Best they’ll ever get is trams, aka HBLR, so that’s their best chance for a long-term one-seat ride.

      • Nathanael

        Oh. And I didn’t mean for HBLR to share the tunnel with buses. I meant, take one of the tubes of the tunnel, and make it 100% HBLR, kick out all the buses and all the motor vehicles. Bus-to-rail transfer terminal at the entrance of the Lincoln tunnel before the tollgates.

        • threestationsquare

          HBLR is much, MUCH lower capacity than the Lincoln Tunnel buses. During the 8:00-9:00 peak hour, more than 900 buses enter Manhattan through the Lincoln Tunnel, carrying more than 32000 people. That’s more than any two-track subway or rail line brings into the Manhattan core during that hour (the busiest are the E/M with about 27500 people, followed by the 6 with a bit over 27000; the midtown and WTC PATH tunnels carry 15000 and 13000 respectively). You’re proposing taking a passenger load that a subway line would have difficulty handling and cramming it onto the short (40m), low-capacity (300 passengers per vehicle) HBLR trams. It sounded like you were also suggesting then running them on the street in Manhattan, where stoplights would further severely limit their capacity? 6000 inbound passengers in the peak hour is the absolute maximum you could hope to carry with such a system, where are the other 26000 bus passengers going to go?

          If you want to make a dent in Lincoln Tunnel bus ridership with a rail transfer station, it needs to be a fully grade separated subway or regional rail line with 150m+ trains and two full tracks of capacity minimum, and even that would struggle to carry as many people as the buses.

          Data from the 2014 Hub Bound Travel survey, mostly p31.

          • Adirondacker12800

            Make the train in their suburb faster than the bus and they will abandon the bus. Like what happened when Midtown Direct opened. You don’t need a transfer station, you don’t need more road to get the bus to the transfer station.

          • sonamib

            In a fully automated subway, it’s possible to run 40+ tph in each track, as Paris does in their line 14. With 800 people per train, you get 32k people transported per hour per direction.

            But wow, 900 buses driving into Manhattan through a single express bus lane in one hour? That’s one bus every 4 seconds. How is that even possible?

          • threestationsquare

            @sonamib: A new fully automated subway could have that capacity yes, though not one that tied into the non-automated rail on either side (unless you spent the money to resignal and automate all of that). But that’s a world away from Nathanael’s proposal shoe-horning trams into a retrofitted Lincoln Tunnel tube with street running at either end. As for how they manage a bus every 4 seconds… bumper to bumper, with separation regulators would never ever allow if they were trains? Though also I assume some of them use the non-XBL lanes so those aren’t all in a single lane.

            @Adirondacker12800: I agree that it would be best for new Hudson capacity to focus primarily on getting more regional rail trains into Manhattan, rather than on serving a bus transfer station. The points I wanted to make are 1) there are still going to be quite a lot of buses using the Lincoln Tunnel in any plausible scenario and 2) attempting to repurpose Lincoln Tunnel tubes for rail, even if feasible, would lower their capacity.

          • Adirondacker12800

            There will still be buses from places that will never have rail. Many many of the suburban buses stop at the train station, that has train service. People use the bus because the train is too slow. It makes more sense to spend the money on making the existing suburban trains faster. If for any other reason trains don’t get stuck in automobile traffic beyond the XBL. And trains are cheaper in the long run.

          • Alon Levy

            And yet the bulk of bus riders is in north Hudson, Bergen, and Passaic Counties, where the rail is shit.

          • Adirondacker12800

            Yes the rail is shit. Which is why they go down to the train station. Where there are trains. And get on a bus. They used to do that in Morris and Essex counties but don’t much anymore. They get on the train because Midtown Direct made the train faster than the bus. so many of them got on the train the rail tunnels were at capacity 15 years before they thought they would be. We can spend a gazillion dollars to clog the roads with more buses that go to a new 10 billion dollar bus terminal. So that people can go to the train station where there are trains. That could, by spending less money, become attractive enough that few people still want to get on the bus. Or confect even more bizarre solutions like making them transfer to a trolley car.

      • Adirondacker12800

        Last figures I saw were that most of them walk from the PABT. A lot of them get to the East Side on the E train. Or the N,Q or R. Lots of them get off the bus and go places on West Side or Wall Street. Instead of getting on the bus and walking or changing to the subway, most of them will have to get on the bus, change to trolley and then walk or get on the subway. Trading their one seat ride for a two seat ride, expect for the really lucky people who live along the trolley line or their two seat ride for a three seat.

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