Home Gateway Tunnel Amtrak’s looming Sandy problem

Amtrak’s looming Sandy problem

by Benjamin Kabak

Nearly two years after the storm, it’s easy to consign the floodwaters that consumed New York’s underground infrastructure as Sandy rolled in to memory. Thanks to the perfect storm and tidal conditions, nearly every tunnel into and out of Manhattan suffered from saltwater flooding, and as we’ve seen with the MTA, work to repair the damage has been time-consuming and costly. Even as the G and R train tunnels have reopened, eight other subway tunnels will require some degree or remediation and repair work.

We’ve heard over the years that Amtrak’s tunnels suffered similar fates. Already nearing the end of their useful lives, the saltwater corrosion has sped up the process, and now the rail provider is warning that very disruptive repairs are required to maintain and rebuild the tunnels. In a PDF statement, Amtrak announced that a new engineering report has recommended a phased approach to rebuilding the tunnels that involve taking individual tubes “out of service for extended periods.” The agency had more to say:

Superstorm Sandy created a storm surge that resulted in sea water inundating both tubes of the Hudson River tunnel and two of the four tubes of the East River tunnel. The report found no evidence that the tunnel linings themselves are unsound, but it did find that chlorides and sulfates caused, and are continuing to cause, significant damage to key tunnel components such as the bench walls and track systems as well as the signal, electrical and mechanical systems.

The tunnels are safe for passenger train operations. Amtrak has a robust tunnel inspection program, conducts regular maintenance work and will be performing interim work as needed. However, a permanent fix is required soon so that the tunnels remain available for long- term use by the traveling public. Amtrak engineers are working with expert consultants on designs to rehabilitate the two damaged tubes of the East River tunnel and will coordinate with other agencies to minimize impacts to train service and other projects.

Now, the coverage of this announcement has been rightly dire. The Times, The Journal and Capital New York all ran stories about how problematic service could become. To perform even basic remediation work, which could begin in late 2015, Amtrak needs to close one of the East River Tubes, which could cause a reduction in Amtrak, LIRR and NJ Transit service by around 25 percent. If and when Amtrak has to close one of the Hudson River Tubes, service could fall by as much as 75 percent.

The real problem is that the work that must go on — full saltwater remediation — can’t and won’t happen, Amtrak says, until another Hudson River crossing is built. In a way, this engineering report gives Amtrak another platform upon which to base their argument for the Gateway Tunnel, but as Amtrak officials have noted, it’s likely to be another decade before Gateway is open. That timeline is of course contingent upon funding, and right now, the money isn’t there. One way or another, Amtrak anticipates only approximately 20 years of life left in their Hudson River tunnels.

This news has raised the spectre of the ARC Tunnel, and in a twist of the knife, to The Journal, a spokesman for Chris Christie stated that the New Jersey Governor “has always recognized the need for additional trans-Hudson transit capacity.” For now, Amtrak is moving forward on design and planning while awaiting the money. “Amtrak,” the agency promised, “will ensure the safety of all passengers and balance efforts to minimize service impacts while also advancing as soon as possible the permanent fix needed for the long-term reliability of the tunnels for train service to Penn Station, New York.”

You may also like


Peter October 2, 2014 - 1:04 pm

It’s interesting to contemplate how the transit agencies might cope with a 75% reduction in trans-Hudson capacity. How about re-implementing a ferry connection between Jersey and the West Side for NJ Transit commuters, just like it was before the Hudson tubes were built. With the 7 station at 11th Avenue, the connection between the riverfront and transit actually wouldn’t be THAT bad. It could be called the “Chris Christie Ferry.”

JJJJJJ October 2, 2014 - 1:17 pm

The 75 percent reduction number is fear-mongering based off incredibly poor management practices (also known as standard American train operations).

Yes, if you let one eastbound train in, and then a westbound train in etc, you lose huge amount of capacity.

BUT, if you combine, for example) 5 trains in Newark, and run them in as a single train (that splits up and berths at different platforms in Penn) suddenly your capacity loss isnt that big at all.

That is, the biggest capacity thief is the dead space between trains. Aka, the 3 minute space between each train. Eliminate that wasted space by running the trains together as one and you have yourself less of a disaster.

Patrick O'Hara October 2, 2014 - 1:26 pm

Whatever space and time you save by combining 5 trains together would be lost multiple times over when it comes time to assemble and disassemble the whole mess. Even if you skip all of the required tests and things like that that have to be done when you combine or split train consists, it still takes a decent chunk of time to get everything conjoined together in Newark, and then once you get to New York Penn, you have to unwind that in a very, very cramped interlocking with very little room for error (if it is possible at all). Something like that isn’t very feasible at all, and that doesn’t even begin to consider the difficulties one might encounter with trying to MU and trainline multiple strings of locomotives and cars together.

JJJJ October 2, 2014 - 2:42 pm

And yet countries in Europe manage to do this just fine. Assemble can be done in a minute, disassemble in seconds.

Just another case of American exceptionalism. I guess a 75% reduction in service is preferable to trying to look at global best practices.

sonicboy678 October 2, 2014 - 6:23 pm

Practicality for practicality’s sake. It’s bad enough that this practice may exist to some degree in North America, but it would be even worse here because this also fails to factor in all current and potential problems.

Really, you’re comparing apples to oranges. They may be somewhat similar, but still have differences in nature that render a comparison senseless.

Nathanael October 6, 2014 - 11:57 am

It’s a 75% loss in PEAK capacity. Off-peak will not see nearly such a reduction in service.

And no, there’s no way to do more service than that. The current practice squeezes nearly the maximum possible service out of the North River Tunnels (under the Hudson) at peak. With only one tunnel, capacity will drop drastically; every goofy thing you suggest runs up against some other limitation, such as ability to store trains. At the moment, during peak, trains have to depart westbound in order to make room for others to arrive eastbound… can’t do that with one tunnel. Maybe if LIRR’s West Side Yard was wired with overhead you might be able to rig up something… but it isn’t. Etc.

Alon Levy October 3, 2014 - 12:45 pm

I’m pretty sure nobody splits or recombines five trains. Two trains, sure, but five?

Also, Penn Station’s platforms are capped at 17 cars.

Aaron Burger October 3, 2014 - 10:17 am

Hi Patrick–a bit off topic, but do you have plans to bring The LIRR Today back online? I really enjoyed your work.

Michael October 3, 2014 - 12:53 pm

Same Here!

Eric F October 3, 2014 - 2:49 pm

I agree, great resource, very informative!

Patrick O'Hara October 4, 2014 - 1:39 am

Hopefully I will be able to do so sometime soon…but I can’t make any estimate as to when.

Caelestor October 2, 2014 - 1:39 pm

The biggest waste of capacity is terminating trains at Penn, which increases congestion significantly. If only there was a way to through run trains.

johnny johnson October 2, 2014 - 7:46 pm

That’s a great idea, only there is very little demand for off direction trains at peak times (and also very little share capacity in the opposite tunnel. So yes, running a train from trenton to babylon sounds great and will lessen the time in penn station, but how is it going to get to long island, and would anyone care to ride it.

lop October 2, 2014 - 11:35 pm

NJTransit already through runs a lot of their trains. They go through the east river tunnel to sunnyside yard. LIRR does the same to the west side yard. But not all trains. Add catenary to get to near in terminals past Jamaica. Or get trains that can use overhead or third rail. Now you have an NJTransit train going to Babylon. So take the LIRR train that was going to reverse move to Babylon – maybe to make a late peak run back to Penn – and send it somewhere in NJ. The NJ train can make the same late peak run. You might not have that many NJ riders heading to Jamaica and LI riders heading to Newark, but you don’t have to do that much to serve them. They aren’t the only ones on the train too, you still have the reverse peak riders you have today.

Bolwerk October 3, 2014 - 10:52 am

This doesn’t change anything. Trains that run in peak have to return in the off-peak direction at some point anyway. All through running means is taking on some mix of each other’s peak and off-peak runs in exchange for avoiding complicated terminal moves at a super-busy through station.

Nathanael October 6, 2014 - 11:59 am

Through-running is great and would increase capacity in the long run…. but not while there’s a SINGLE TRACK bottleneck.

Currently there are two tunnels under the Hudson. If one needs to close for repairs, that leaves ONE. And there’s not much you can do to maintain service with a single-track bottleneck.

Jerrold October 2, 2014 - 3:35 pm

Let’s just hope that we don’t wind up with a Chris Christie ADMINISTRATION!

Jerrold October 2, 2014 - 3:39 pm

P.S. I intended that to go up right under the statement that ends with “Chris Christie Ferry”.

Eric F October 3, 2014 - 8:54 am

“The One” has been president for 6 years now. How’s the northeast corridor doing? Shouldn’t you be out in the street with a paper maiche puppet protesting endless wars?

Alon Levy October 3, 2014 - 12:47 pm

Obama, for all his many faults, is not as authoritarian as Christie.

And no, I’m not saying this because Obama is supposedly a Democrat and Christie is a Republican. I think most of the people running in the GOP invisible primary so far are less authoritarian than Christie. Ted Cruz has scary ideas about government programs, but he doesn’t deliberately cause traffic jams in cities whose mayors refused to endorse him.

Eric F October 3, 2014 - 1:15 pm

The current administration has brought the IRS down on its adversaries, attacked two countries without Congressional authorization, used recess appointments to staff agencies and then use those agencies as part of its campaign and adversary punishment apparatus… I’d trade a traffic jam for that.

Nathanael October 6, 2014 - 12:02 pm

The IRS accusation is a total fraud by Republicans; it did not happen. In fact, the Obama administration had the IRS preferentially attack left-wing organizations, while allowing blatantly illegal right-wing organizations a free pass.

The recess appointments were necessary because the US Senate (both parties) simply refuses to do its job, which is to confirm or reject appointments — it’s not supposed to leave them in limbo, but it does.

No argument about invading foreign countries; I really wish Obama wouldn’t do that. In that regard Obama is exactly like Bush. Also the drone assassinations.

Christie is all in favor of drone assassinations and all in favor of invading foreign countries without Congressional approval. Just like Bush and Obama.

Bolwerk October 3, 2014 - 1:18 pm

I don’t know about that. Christie is probably fairly live-and-let-live on a lot of issues compared to that mentally ill crop. At least for practical reasons, he dropped things like gay marriage opposition. He just happens to be rather corrupt. Cruz doesn’t deliberately cause traffic jams, but he doesn’t wield executive power either. I don’t know what he’d do with his hand near the proverbial red button, but I’d just as soon not find out.

For some reason, Eric can’t understand that disliking Christie’s boneheaded politics doesn’t make you sycophantic for Obama. But an objective evaluation of Obama? He pretty much saved the world economy. My opinion? He’s still terrible.

Eric F October 3, 2014 - 2:51 pm

I think he’s been involved in the deaths of more people (Libya, Yemen, Afghanistan, Syria, etc.) than any Nobel Peace Prize winner since the Vietnam era, so he does have that going for him.

Bolwerk October 3, 2014 - 6:39 pm


“Political satire became obsolete when Henry Kissinger was awarded the Nobel Peace Prize.” – Tom Lehrer

Eric October 5, 2014 - 4:13 pm

A lot of those you mention were damned if you do, damned if you don’t situations.

Alon Levy October 6, 2014 - 8:01 am

No, actually, they are not. In Libya there was a very clear course of action. Obama took it, at the last moment, after Cameron and Sarkozy dragged him there. In Syria it’s a bit more complicated, but pretty much every single decision the administration has made has helped Assad, who despite being in the news less than ISIS is the biggest mass murderer in the region. Conversely, in Yemen and Afghanistan, there’s no real need for bombing weddings just because metadata analysis suggests there are terrorists there.

Boris October 2, 2014 - 4:35 pm

We can use rail floats, the ferries that carry trains. They probably haven’t been used for passenger service in decades, but I’d imagine they aren’t more dangerous than car-carrying ferries. Since passengers can stay on the train, it can even be considered a one-seat ride.

Eric F October 3, 2014 - 8:51 am

Given that it is taking the MTA over 20 years to build east side access, what help exactly would it be to be 10% of the way through ARC construction?

JJJJ October 2, 2014 - 1:16 pm

The 75 percent reduction number is fear-mongering based off incredibly poor management practices (also known as standard American train operations).

Yes, if you let one eastbound train in, and then a westbound train in etc, you lose huge amount of capacity.

BUT, if you combine, for example) 5 trains in Newark, and run them in as a single train (that splits up and berths at different platforms in Penn) suddenly your capacity loss isnt that big at all.

That is, the biggest capacity thief is the dead space between trains. Aka, the 3 minute space between each train. Eliminate that wasted space by running the trains together as one and you have yourself less of a disaster.

sonicboy678 October 2, 2014 - 6:26 pm

I hope this was an accidental double post.

Regardless, running trains as such is not practical in terms of safety or otherwise.

Alon Levy October 2, 2014 - 1:34 pm

An agency that wants a $13 billion tunnel is scaremongering about what happens without the tunnel. Not saying there’s no need for maintenance, but outside appraisal is needed.

And no, the GAO and Amtrak OIG don’t count as outside appraisal. The Amtrak OIG’s grasp on best industry practices outside the US lacks opposable thumbs. Or for that matter arms or other manipulatory appendages.

lawhawk October 2, 2014 - 2:12 pm

I’m curious how Gateway would cost $13 billion versus the $8.7+ billion for ARC, even though there wasn’t a deep terminal that is associated with Gateway. The difference simply can’t be solely because of inflation (which is relatively low).

Is it lumping together the Portal Bridge and other projects at either end of the Hudson tubes (like the Penn station move or Harold Interlocking) or other improvements in the NYC metro area along the NEC right of way? It would seem to be the latter by my reckoning.

Portal is $1.8 billion at last check (for a 5-track with flyovers).


Joseph Steindam October 2, 2014 - 7:47 pm

Last time I checked, Gateway also includes the building of Penn South, a southern expansion of Penn Station that would include additional tracks and platforms. This would require taking an additional block from private land owners, demolishing all those buildings, and building out another station. So that’s one component of additional costs vs ARC.

Another large cost is I believe for the Gateway Tunnel to meet the interlockings at the entrance of Penn Station, the tunnel will require a tunnel box that extends out into the Hudson River before the tunnel can go beneath the riverbed, and building the box through the riverbed will require significant environmental remediation and efforts to mitigate the spread of pollutants because the Hudson riverbed is severely contaminated. Supposedly this reason contributed to the rationale for ARC choosing to not connect into Penn Station and selecting the alternative with the cavern beneath 34th Street. Obviously other concerns were probably more critical, but it’s a cost that ARC wouldn’t have faced.

Eric F October 3, 2014 - 8:52 am

The difference is that the ARC figure is a fantasy.

Bolwerk October 3, 2014 - 10:55 am

{{citation needed}}

Eric F October 3, 2014 - 1:16 pm

Cite: east side access, reality, lived experience.

Benjamin Kabak October 3, 2014 - 1:36 pm

Citing ESA ignores the numerous ways in which the projects are different. One involved new build tunnels and an idiotic deep-cavern terminal. The other involves working through the busiest interlocking in the country with limited cooperation from other agencies involved. The feds offered a range of cost overruns for ARC, and Christie latched onto the highest number with no basis in reality for that estimate. The feds have also offered estimates for ESA, and to date, they’ve been right.

Eric F October 3, 2014 - 2:53 pm

Well, agreed that it’s speculation, but these deep cavern things don’t seem to come together well. NJ has done a very good job in keeping its road work on budget the past 5 years, but compare with Washington State’s below ground issues with it’s Alaska Viaduct replacement work. It’s very possible that NJ might have lucked out on the project, but that’s not the way to bet.

Bolwerk October 3, 2014 - 1:58 pm

That basically amounts to “I asserted it so it’s true.” Come on, you aren’t stupid. You must have some evidence!

Alon Levy October 3, 2014 - 12:49 pm

The difference is Amtrak involvement. Even Harold, an overrated and expensive project, was $300 million, which isn’t enough to explain the difference. It’s just Amtrak and its ideas of how much things cost.

Eric F October 3, 2014 - 9:10 am

Additional tunnels are needed anyway. If scaremongering is the way to get the, so be it. I’d rather someone scaremonger for capital construction than over the usual things people scaremonger about.

Nathanael October 6, 2014 - 12:04 pm


We have an outside assessment, anyway; Amtrak hired a company to provide one, and published the report. The current tunnels need to be shut down to replace the damaged concrete and ballast.

Jerrold October 2, 2014 - 1:40 pm

[A bit off-topic, but important enough to put here]

Has everybody here read THIS article?


Benjamin Kabak October 2, 2014 - 1:51 pm

I have to disagree with you that this article is important (and certainly important enough to be so off-topic)! Anyone who’s reading here knows what’s in yet another piece about Upper East Siders whining about subway construction. The video’s neat though.

Kai B October 2, 2014 - 1:54 pm

I like how their graphic assumes there is going to be a T shuttle train running between the three stops on day one.

Jerrold October 2, 2014 - 3:32 pm

I noticed THAT too.
Must have been just a case of sloppy copying off of something they had.

Patrick O'Hara October 2, 2014 - 2:04 pm

It is easy over-exaggerate how ARC would have impacted this situation. At the end of the day, the earliest ARC was planning on being finished was sometime in 2018, which would still be way past when Amtrak would have to start undertaking this work. And even if it was up and running early, such a disruption would still be significantly felt, as the ARC tunnels wouldn’t have tied into the existing station, so you wouldn’t be working with three tunnels like you would on the East River Side, but one tunnel upstairs and one tunnel downstairs. Like it or not, ARC was flawed and we’re probably all better off for Christie killing it. If only we had realized the mistake sooner.

At any rate, this would all be a lot easier if this was 10 years from now and East Side Access and ARC/Son of ARC would have been open for business, but this is a short-term problem that, unfortunately, can’t really be helped by these long-term projects.

The best case scenario is they can consolidate the whole ordeal down to just two years (do one North River Tunnel and one East River Tunnel at the same time, then move to the other two). With NJTransit service curtailed through the North River Tunnels, they can have the fewer trains they do run turn in the station, freeing up space in the three open East River Tunnels. Whether or not they will have the resources to take care of two simultaneously is another question, but we will have to ‘wait and see’.

With just LIRR trains and the handful of Amtrak Long Distance and Northeast Regional trains in the East River Tunnels, the LIRR could probably get on with minimal impact. Lines 1 and 2 of the East River Tunnels (the ones NJT and Amtrak primarily use) are the ones that will need to go down, not Lines 3 and 4 (the ones LIRR primarily uses), and with fewer NJTransit trains scurrying back and forth to and from Sunnyside Yard, the reduction in capacity would not be felt as hard. Running full service might not be likely, but I would be surprised to see anything more than a rather small handful of combined trains at rush hour.

Amtrak, the landlord, will also probably not cancel too many trains. Perhaps they’ll chip in and have the Keystones only run between Harrisburg and Philidelphia or Newark, but I would imagine they’ll be able to secure space in their tunnels for the Long Distance/Acela trains.

NJTransit won’t have it as good, as they’re sure to get the brunt of the headache. Boot the Midclowns to Hoboken, drop some Secaucus stops from the Main/BC/PV/ST Line trains, start trains out of Newark, have PATH step up their game, perhaps adding a NWK-33rd Street service, a ferry service from Midtown/West 39th to Hoboken Terminal, and hopefully they’ll get through the mess without too many riders flocking to the buses or their cars.

But they ought to get the show on the road quickly…let’s just cross our fingers and hope that while tunnel A is undergoing the repairs part of the ceiling in tunnel B collapses. Then we’ll all be in, to quote a MTA Board Member, “deep doo-dads.”

JJJJ October 2, 2014 - 2:44 pm

I can also see more long distance ferries from places like Perth Amboy.

Eric F October 3, 2014 - 9:48 am

Presumably, NJT would send its midtown direct trains back to Hoboken, and it would close Secaucus Junction for the duration of the project.

eo October 3, 2014 - 10:21 am

No, it won’t. There is no spare capacity in Hoboken during peak hour. Off peak hour, yes, peak hour there is no free platforms. Have you noticed what happens when there is a temporary problem in the tunnels and they reroute the Midtown directs to Hoboken? All lines starting and ending in Hoboken and the M&E trains in there start experiencing delays exceeding 2 hours. Hoboken is an option now because it is temporary disruption that works itself out in a few hours, you cannot run the M&E lines continuously there on top of everything else.

At this point the ridership is such that you cannot close Secaucus either. There are people who live in those developments next to the station. I am sure they have state representatives and a congressman — the station will never close as long as there is a single NJTransit train going into Penn — they are not going to let the “rich” from Princeton or Millburn have their expresses through Secaucus while they schlep backwards to Newark to get on the train into Penn. Remember, when they restored the first Hudson tunnel after Sandy and some NJTransit trains started going into Penn they did have service at Secaucus.

Patrick O'Hara October 3, 2014 - 10:43 am

Did NJTransit demolish a bunch of platforms at Hoboken Terminal when the Kearney Connection opened? No, they didn’t, so there’s no reason to assume that they couldn’t fit those trains, especially considering the amount of trains NJT runs now on the M&E and the Monty-Booty is not all that more than the number they did when the Kearney Connection opened in 1996. When they up and send all MidTOWN Direct trains to Hoboken things get crazy since the sudden influx of trains can cause lots of congestion, but considering trains dwell at platforms in Hoboken during rush hour for very long periods of time (upwards of an hour for many trains), there’s no reason to presume that NJT couldn’t handle all of MidTOWN Direct in Hoboken with a little bit of planning.

lawhawk October 3, 2014 - 12:43 pm

That doesn’t quite make sense since those Midtown Direct trains used to terminate in Hoboken before they were able to run direct to NY Penn. The delays are the result of NJ Transit apparently forgetting how to shift trains around in the yard to accommodate the additional traffic due to problems on the NEC.

They have the capacity to get the combined traffic into Hoboken. There’s sufficient space in the yard, especially since they reconfigured it in the past decade. At the moment, NJ Transit is currently rebuilding one of the interlockings just outside the yard, limiting through traffic to three tracks (though all four tracks in the Bergen tunnels are available).

Eric F October 3, 2014 - 1:17 pm

You could keep Secaucus open as a station, but eliminate Bergen country trains’ stooping there.

eo October 3, 2014 - 4:42 pm

Try telling that to the people people in Rutherford and Fair Lawn. Or prevent the people from Ridgewood going to the Main Line trains. Politically it will not happen. Also you will lose at least half the ridership to cars — the rich do not like the idea of changing trains or ferries that much, probably 30% of them will start driving.

If I was to predict what happens it is the following: For another 10-15 years lots of studies and talking and no much action. Then one day Amtrak discovers an issue that is deemed safety critical and they close one of the tunnels for say 2-4 weeks while urgent repairs are made. Politicians start drawing plans for new tunnels and whatever else. Then the closure ends and people forget. The project keeps moving at a snail’s pace. Then another 15 years later another issue crops up. This time they cannot correct it quickly and one of the tunnels stays closed for 2-2.5 years. By then the politicos have finally figured out how to finance the $100 billion needed for the new tunnels and the tunnels are finally complete in 2075 long after I have been buried somewhere …

It is the American way of building public infrastructure …

Larry Littlefield October 2, 2014 - 2:36 pm

If the situation is that dire, the right thing to do is build a third tunnel under the river, hooking into the existing Penn Station trackage, pronto. Rather than waiting for a massive $15 billion upgrade to be funded in its entirety.

That would allow service no worse than today with one track closed for rehabilitation.

Once rehabilitation was completed, it would allow New Jersey Transit to use one Hudson River track and one East River track in the peak direction for a series of one-trip (per rush hour) trains to/from Sunnyside Yard.

The LIRR would use one East River track for one-trip trains to and from the West Side yards.

And the LIRR, NJT and Amtrak would use two tracks under each river for through trains and those going back and forth throughout the day.

lawhawk October 2, 2014 - 3:19 pm

That’s entirely sensible, which is why they wont do it.

Then again, why build one when you can build two at twice the price.

Bolwerk October 2, 2014 - 8:00 pm

Yes to a tunnel, but if you do one there is little reason not to bore a wider tunnel with room for two tracks.

Eric F October 3, 2014 - 9:00 am

At least two, three makes more sense, just to have redundancy for routine issues. It’s insane that the federal government, which has been running inter-city passenger rail for longer than most of the commenters here have been alive, has not fixed this by now. When did Amtrak start up, 1971?

Joey K October 3, 2014 - 9:25 am

Yes, 1971. You are talking about a fundamental problem with the way that our national rail infrastructure is financed. Unlike infrastructure improvements to the highway system, that (while currently screwed because of mismanagement) is funded by a replenishing trust fund from gas taxes, improvements to Amtrak’s infrastructure come from a balancing of internal profits and losses, federal subsidies, and a hodgepodge of federal grant and loan programs. This results in an unsustainable system of funding large-scale projects, like the Gateway Program and what not. It’s an unsustainable system too because, unlike the FHA, Amtrak diverts profits and capital spending allocations to its losing routes, which are most of them. They average over $1 billion in losses every year, and they have to because Congress has mandated that they continue running long-distance routes that are incapable of making money.

In short, overhaul Amtrak, you’ll see the opportunities for infrastructure improvement to open up. Of course the cities and states could step it up, but these tracks and Penn Station are owned by Amtrak. You don’t see these problems in Europe because, in part, there are laws against having a single entity control both the infrastructure and the operations. Here, it’s how we do business.

Nathanael October 6, 2014 - 12:13 pm

FWIW, the vast majority of Amtrak’s routes are quite capable of making money… except that they are at the mercy of hostile “freight railroads” (NS, CSX, CN, CP, UP, BNSF) which are trying to sabotage them. πŸ˜›

Yeah, the routes across the Mountain Time Zone are hopeless. But NY-Chicago and NY-Miami *should* make money… it’s just being sabotaged by the refusal/inability of the “freight railroads” to run the trains on time.

“Private-public partnership” — the private company takes the profits, and the public gets screwed. πŸ˜›

Larry Littlefield October 3, 2014 - 9:41 am

“If you do one there is little reason not to bore a wider tunnel with room for two tracks.”

I can think of one reason. Let’s say you do a one track tunnel, and then plan to extend the Flushing Line to Secaucus as well.

You can tell the union and the contractor that if they don’t rape us so much, and finish the damn thing as fast as the original tunnels were built 100 years ago by hand, they might get the contract for the next job, which might go ahead.

If the two track tunnel is the only plan, the SOBs will never allow it to finish (like East Side Access).

eo October 3, 2014 - 10:41 am

I think that the 7 to Secaucus is a pipe dream. Here is why. The MTA, NY state and NYC have relatively little to benefit from such an extension. Why would they want to put up significant money into it? Yes, NY state gets all the income taxes of people commuting from NJ, but they get them anyway — there is no reason to spend $5 or $8 billion to get them. So even though the MTA, NY state and NYC might give a token amount on such a project there is no real reason to push for it — the people in Jersey do not vote in NY. So the Feds and NJ will need to pay most of this 7 line extension. At that point comes the Environmental Review with its alternatives and guess what happens there? The cost-benefit analysis of sending one of the narrowest rolling stock in existence the same distance under the Hudson and the Palisades as the Gateway portion between the existing Penn and Secaucus indicates that you can move a lot more people and have much greater benefits to Gateway (the part equivalent to that 7 extension to Secaucus) than the subway to Secaucus. So the extension dies in the environmental review. The only way the 7 extension can come out ahead as the preferred alternative in an environmental review is if it serves large markets such as Hoboken and Edgewater with several stations before going to Secaucus, but then you have killed its viability as a desired way to get train commuters from Secaucus to Manhattan because it will be quite a bit slower.

It is not that the idea of sending the 7 to Secaucus is bad — it is that when compared to others in a world of limited resources it will lose out.

Larry Littlefield October 3, 2014 - 10:48 am

“The MTA, NY state and NYC have relatively little to benefit from such an extension. Why would they want to put up significant money into it?”

They wouldn’t. New Jersey benefits, so New Jersey pays. Despite its problems, it could do so if it wanted to.

An income tax surcharge could be imposed within X distance of Northern New Jersey rail stations, for example. Rational people might be up for the deal, given the effect of transit on property values and the rising hell of bus transit through the PABT.

I do think the service could come close to covering its operating costs. New York could contribute by operating the service, and perhaps adding a mezzanine level over the Flushing Line platform at GCT to prevent an overcrowding crush there.

Phantom October 3, 2014 - 12:31 pm

NJ is not the only state that benefits.

I live in Brooklyn take NJT – SEPTA to the Philly area a few times a year. If you could catch the subway at Secaucus, I’d certainly be willing to do that.

Lots of NYC residents go to NJ for football games, other social / recreational reasons, or because they reverse commute – they’d be willing to catch the subway at Secaucus.

A major connection like that would be a big improvement for mass transit in the entire region. It is in the interest of everyone in NYC / NYS / CT as well as NJ to have proper, free flowing mass transit -and- road traffic between NYC and NJ.

There are problem

eo October 3, 2014 - 4:51 pm

Yes, there are benefits — they are not worth even $3 billion to NY though, so the politician who rails against NY paying for any of it is likely to win. You are an exception, say to you it is worth $20 and you use it 5 times a year. That is $100 per year. We need therefore 30 million person years to get to that $3 billion. Even over 100 years that is still 300,000 thousand people per year. I do not see that many people clamoring to go from NY to Secaucus. The demand just is not there from the NY side.

BruceNY October 3, 2014 - 2:04 pm

“…so New Jersey pays.” If NJ is going to pay for anything, I can imagine them wanting to pay for a 4th tube of the Lincoln Tunnel so that their constituents can drive their cars into the city more easily. But for rail? Not so much.

AG October 3, 2014 - 5:01 pm

simply because NYC needs the workforce of NJ. They are suburbs of NYC which happen to be in NJ. There is a reason NJT goes into Manhattan. PATH is shared… as is one of the lines of Metro North.
Both places benefit. The provincialism has to stop… that’s why the regional transportation is so disjointed.

Nathanael October 6, 2014 - 12:11 pm

NYC technically doesn’t actually need the workforce of NJ… but changing that would require MASSIVE changes in policy, involving intense promotion of Westchester County development and giant expansions of Metro-North. Not seeing it.

The best thing is, yes, to get rid of the provincialism.

Or, you could embrace the provincialism, and shut down all cross-border traffic, and things would adjust.

But what we have now is government which *depends* on the cross-border traffic and *assumes* it’s happening, but *pretends* that it doesn’t depend on it, which is head-in-the-sane behavior.

You can’t depend on Mexican immigrants for your farm labor *and* seal the border — and you can’t depend on NJ commuters for your work *and* ignore the transport links to NJ.

Bolwerk October 3, 2014 - 10:49 am

I really can’t see why this would be an issue with a wider diameter and not a narrower one. An objective evaluation of our needs is probably: we need four more tracks under the river no matter what.

Alon Levy October 3, 2014 - 12:54 pm

All the plans call for twin bores, rather than one large-diameter bore. This is perfectly fine. Single-bore tunnels do not have any advantage over twin bores in an environment in which the station tracks preexist (canceling large-diameter TBMs’ advantage in allowing platforms to be built within the bore) and speeds are not so high as to require a large amount of free air around the trains (allowing twin-bore tunnels to be built with small diameter). Twin bores have more redundancy because it’s easier to shut down one track for maintenance, and are favored for difficult fixed crossings of mountains and water.

Bolwerk October 3, 2014 - 1:23 pm

That doesn’t matter much here, where there are already two tracks and you only need to shut down one at a time. Off-peak maintenance of the double-tracked tunnel still leaves two tracks available, and that can be done at off-peak times.

But, even if some environmental issue precludes double-track bore, the calculus is not that different: may as well do two single-track tunnels and have a complete system. I think you’ve said that yourself.

Just do whatever’s cheaper.

Alon Levy October 3, 2014 - 12:57 pm

A twin-bore tunnel costs much less than twice as much as a single-track single-bore tunnel. The geologic and engineering work is reused; only the physical construction needs to be doubled.

For the same reason, it’s cheaper to build a new tunnel next to a recently built tunnel, even if this other tunnel served a different purpose. In California, a water tunnel under the San Francisco Bay in the vicinity of the Dumbarton Bridge required extensive seismic and environmental engineering, which high-speed rail could use to build its own tunnels along the same alignment at relatively low cost.

Alon Levy October 3, 2014 - 1:02 pm

Actually, a single-track tunnel would have limited usefulness. What tracks of Penn Station would it connect to? Unless it connected to the same tracks as the existing tunnel pair, it would not be able to substitute for these tunnels if they were closed for repairs. But then it would force the entire construction to follow the existing station tracks, limiting its ability to provide additional connections to the southern tracks (the ones that stub-end to the east and should be connected to Grand Central) or to the northern tunnel pair used by the LIRR.

Nathanael October 6, 2014 - 12:07 pm

“If the situation is that dire, the right thing to do is build a third tunnel under the river, hooking into the existing Penn Station trackage, pronto. ”

Well, Larry, Amtrak is in fact asking for exactly that. Might as well build 2 while you’re building 1.

Kevin S October 2, 2014 - 3:15 pm

In all of these discussion I never see a mention of a plan to prevent flooding next time. Sure this was storm of the century, blah blah blah, but that doesn’t mean it will be another 100 years before the next time the water rises that high.

Rob October 2, 2014 - 3:34 pm

I would think with east side access, the east river tunnel traffic could easily be reduced enough so that 3 tubes would be fine.

Patrick O'Hara October 2, 2014 - 5:17 pm

East Side Access is at least seven years away from completion…this is something that Amtrak is going to begin undertaking towards the end of next year. Way before ESA, ARC, or Son of ARC will or would have been open for business.

Eric F October 3, 2014 - 9:01 am

Yup. If that project was anywhere near on time, this would not be a big issue for LIRR. They could just lean on GCT a bit more heavily for a year or two. What a fiasco.

Patrick O'Hara October 3, 2014 - 10:46 am

This is not a big issue for the LIRR anyways. If you look at the numbers of the amount of trains going through each of the tunnels at the various times of the day, you’d see that there’s actually a considerable amount of free space in the East River Tunnels. Couple that with fewer NJT Trains during such a disruption and the LIRR could get on with a very, very close to normal schedule.

JebO October 3, 2014 - 4:56 pm

You are assuming they’d do one Hudson tube and one East River tube at the same time, when they explicitly said they are not planning on touching the Hudson tubes until they get Gateway. So, sorry, I don’t think the LIRR will be lucky enough to be able to bank on a huge reduction in NJT trains using the tubes.

Patrick O'Hara October 3, 2014 - 7:33 pm

On the average weekday, NJTransit runs 96 trains through the East River Tunnels, most of them in the midday period. The LIRR runs roughly 568. I did the math earlier, the three railroads can get along very close to as-is with the three tunnels. The only occasion when things become remotely tight is during the 9am hour and one or two hours in the PM rush (depending on which line is closed). But even then, the overflow is a single-digit number of trains in each of those hours, which could more than be accommodated in the reverse direction in one of the other two open tubes. All you’d need is to have a few more NJT trains turn in the station, which is certainly feasible.

Do you have any numbers to contribute or sources to cite to claim the contrary?

Joey K October 3, 2014 - 9:33 am

This is true, as well as the improvements to Harold Interlocking. These two things will reduce capacity constrains more than enough to facilitate the temporary closure of one East River Tunnel at a time. Amtrak/NJ Transit average 270 inbound trains per weekday for one inbound tunnel across the Hudson River. Amtrak/LIRR on the other hand average around 350 inbound trains per weekday for two inbound tunnels across the East River. You can argue that Amtrak/LIRR could already accommodate shutting down one of the four East River Tunnels at a time today, but if not, certainly by the opening of East Side Access.

The bigger problem is cross-Hudson capacity constraints, which is also where ridership growth is occurring to a much larger extent.

Ed Unneland October 2, 2014 - 3:40 pm

Um, how much are we talking for a new set of tunnels on the East (North) River from the Long Island City LIRR station, going across 23rd Street, and under the Hudson River to Secaucus?

Larry Littlefield October 2, 2014 - 4:04 pm

With the NIMBY’s we’ve got, the construction costs we’ve got, and the politician’s we’ve got, we can’t get bus rail transit across Manhattan let alone a rail tunnel.

Ed Unneland October 2, 2014 - 4:42 pm

Yeah, guess I may as well suggest powering a spacecraft to Alpha Centauri with unobtainium …

Joey K October 3, 2014 - 9:42 am

While tunneling cost would be high, it would not be the reason why this idea is not practical. Where would the stations be? How many tracks/platforms would they accommodate? This is the real problem. People want to terminate in Manhattan, and getting passengers on and off a commuter train takes longer time than just shuffling them through like a subway. So we need track and platform capacity, and enough of it. If Penn Station unboarding and boardings took half the time they do today, we would not have the current capacity constraints with just two cross-Hudson tunnels. Unfortunately, building a large scale transit hub today somewhere in midtown Manhattan is way too expensive. Just look at East Side Access. That’s not even a very large station, 8 tracks I believe. And its going to cost somewhere around $20 billion??

There needs to be a third transit hub, I agree with you on that. But midtown is way too expensive. Have you read about a proposal to build a waterfront transit hub on the west side called the the Hudson Terminal Plan. You may be interested in giving it a read. It suggests some interesting stuff.

JebO October 3, 2014 - 4:58 pm

East Side Access is $10 billion (at least, as of now), not $20 billion.

Joey October 3, 2014 - 12:29 am

Is the NJT/Amtrak set of East River Tubes even fully utilized? Amtrak uses them for a handful of NEC runs and NJT uses them for deadheads to Sunnyside. I’m having trouble believing that this uses all of the capacity, particularly in the peak direction where there’s no need for deadheading.

Also, could Gateway be expedited somewhat by eliminating the new “Penn South” station box? I would argue that the additional platforms are unnecessary anyway, but if there need to be tunnel closures then it certainly wouldn’t be needed immediately anyway.

Joey K October 3, 2014 - 9:16 am

Eliminating the Penn South station box eliminates the bulk of the capacity savings and benefits from the two proposed tunnels. The design of the Gateway Program proposes two cross-Hudson tunnels, but only one of which would link to the existing Penn Station track infrastructure. The southern-most tunnel would only link to the new Penn Station South complex, which would consist of seven tracks and four platforms. The second benefit of the Penn Station South complex would be to provide another terminus for Amtrak’s Empire Corridor trains, which now only have essentially Tracks 5-8 to terminate at. This would open up more tracks for NJ Transit, who really need the space within Penn Station.

That all being said, at this point, the most sensible solution is to scrap Penn Station South and the southern-most tunnel of the Gateway Program and just build the northern most tunnel to connect back to Penn Station. If we wait to implement the entirety of the Gateway Program, it will be decades, when really we just need a third tube to connect to existing Penn infrastructure and allow Amtrak to shut down one cross-Hudson tunnel at a time for repairs. Not to mention the cost savings… In the interim, we should be looking to build a third regional train station in Manhattan that is completely independent of the mess in Penn Station.

Eric F October 3, 2014 - 9:30 am

They could build the tubes and make the station a second order priority.

Joey K October 3, 2014 - 9:46 am

They could, but why spend the money on the second tube when the benefits of that tube are decades away, if ever. I think more effort should go into looking for alternative sites to build out a third commuter station in Manhattan. If you look more closely at the Gateway Program proposal, there is a second phase to it that proposes building high-speed rail through a second pair of tracks traveling under 30th Street, through a lower level Penn Station south, and connecting with East Side Access tracks. So, you could defer the construction of the second Gateway tube until funding appeared (magically) for this other phase of the Gateway Program.

Really though, the whole Penn Station site is a mess and we should be focusing on planning for a new transit hub elsewhere in Manhattan. WTC would have been ideal, but they squandered that opportunity and invested in materials for an overly ornate PATH station. But I recently read about a plan to build a waterfront train station on Manhattan’s west side called the Hudson Terminal Plan. Ironically, the original Hudson Terminal was torn down to make the current WTC complex.

Joey October 3, 2014 - 4:37 pm

The station throat can be reconfigured easily. And anyway, what’s your source for only one of the tunnels linking into the existing station? All of the conceptual track maps I have found show both new tunnel tracks linking into the existing platform tracks.

Nathanael October 6, 2014 - 12:15 pm

Yes, both tunnels would link into the existing platform tracks.

Patrick O'Hara October 3, 2014 - 10:51 am

No, the East River Tunnels are not at full capacity, unlike their North River counterparts. (Though there is a significant need for equipment moves, as the amount of yard space is finite and can accommodate only a fraction of each railroad’s rush hour service).

The biggest capacity problem at New York Penn Station is the station itself not necessarily the tunnels. Not building the extra Penn South platforms as part of Gateway would allow you to run a grand total of 0 additional trains through the station itself.

Larry Littlefield October 3, 2014 - 12:59 pm

They just need to cut dwell time on those one-trip (per rush hour) trains.

Joey October 3, 2014 - 4:35 pm

The North River Tunnels already run 24 tph, which is very close to if not at capacity. You could maybe push it to 30 with better signaling and OTP, but not much further. The station itself can be improved with shorter dwell times and through-running.

In any case, my point was that if the existing tunnels need to be taken out of service for repairs, then it’s fine to build the new tunnels without the station box because the additional platform tracks wouldn’t be necessary until the existing tunnels were back in service.

Nathanael October 6, 2014 - 12:15 pm

What Joey said.

JebO October 3, 2014 - 5:03 pm

Is the NJT/Amtrak set of East River Tubes even fully utilized? Amtrak uses them for a handful of NEC runs and NJT uses them for deadheads to Sunnyside.

Yes. There’s this little thing called the Long Island Rail Road. You may have heard of it? It’s actually the heaviest user of the East River Tubes, and from what I gather it has a fairly substantial rush-hour operation.

Patrick O'Hara October 3, 2014 - 7:46 pm

Not quite. Despite the Amtrak, NJT, and the LIRR’s “fairly substantial rush hour operation”, the three agencies don’t come close to bringing the four lines of the East River Tunnels to capacity. The only occasion where they do actually meet capacity is during the 9am hour in Line 3. Everything else misses the mark by at least 4 to 6 tph. In fact, there’s enough room that the three railroads could run as many as 77 more trains, in either direction, during the AM rush hour period, and as many as 87 more trains, in either direction, during the PM rush hour period. There would just be no space for them in the station.

In reality, the East River Tunnels are far from operating at full capacity.

Joey October 5, 2014 - 3:03 pm

There are two sets of East River Tubes. One is used by the LIRR and one is used by Amtrak/NJT. The LIRR’s peak direction tube is saturated at rush hour of course. The reverse-peak-direction tube less so since the LIRR stores trains in the West Side Yard at midday rather than turning them all around.

The other set of tubes is used by Amtrak and by NJT for deadheads to and from the Sunnyside Yard. I don’t have numbers but I have trouble believing that this set of tubes is anywhere near capacity.

Alon Levy October 6, 2014 - 8:04 am

This is not true. There are indeed two pairs of East River tunnels, but the LIRR uses both, as it runs far too many trains at the peak (about 37 in the peak hour) to use just one tunnel pair.

Joey October 6, 2014 - 2:35 pm

Hmm … I guess you’re right, though I only count about 33 in the busiest hour. I suppose I assumed that because the platforms and western approaches were segregated that the tunnels were too but that’s evidently not the case.

Eric F October 3, 2014 - 8:56 am

One thing they could do is rush through the going on decade long plan to expand PATH to accommodate 10 car trains from 8 car trains. The modifications to allow that are not all too involved. Hardly a panacea, but at least you;d have an immediate 25% capacity increase in the Newark-WTC line.

Eric F October 3, 2014 - 9:02 am

By the way, why is this a train tunnel issue and not an auto tunnel issue? Were the Lincoln/Holland/Midtown unscathed? I know they had problems in the battery Tunnel.

Joey K October 3, 2014 - 9:27 am

They weren’t unscathed, but there is less of a dire situation in terms of maintenance work. Lincoln Tunnel has six total lanes, Holland has four. And tunnel traffic across the metro region has been steadily decreasing over the past two decades. People want their trains!

Eric F October 3, 2014 - 9:29 am

Tunnel traffic is very high and decreases are probably a function of the massive toll increases. If the PA charged a PATH fare of $15 each way, PATH traffic would decrease as well. The Holland Tunnel has two narrow lanes in each direction, no shoulders. Are there plans to take tubes out of commission? I hadn’t heard any.

Joey K October 3, 2014 - 9:49 am

No plans, no word. But it’s not just tolls. People are choosing to live in areas where commuting by rail is convenient because they don’t even want to own a car let alone travel into Manhattan by car. The largest population growth and commuter growth has come from counties that have a diversity of rail infrastructure. PATH growth has been tremendous over the past decade, as has Hudson-Bergen Light Rail. I guess the point is that you can shut down one cross-Hudson auto tube, whether Holland or Lincoln, and disaster won’t ensue. But cross-Hudson regional rail only has one tube in and one tube out. You see what happens when there’s an emergency shutdown. It’s chaos.

Eric F October 3, 2014 - 9:59 am

Will they also bring in freight on the PATH? Perhaps each passenger can bring in one box each day to supply the stores and restaurants catering to the smug urbanists.

“People are choosing to live in areas where commuting by rail is convenient because they don’t even want to own a car let alone travel into Manhattan by car.”

In the NY region, there is no ‘convenient’ commute. All commutes are just different circles of Dante’s inferno. And the death of the demand for a free standing house is greatly exaggerated.

Bolwerk October 3, 2014 - 10:33 am

Who or what is a “smug urbanist”? Unless you think people driving their sedans in are all carrying trunkloads of freight, the tunnels aren’t exactly critical freight routes. The GWB is probably the elephant in the room with regard to NYC freight from the west and south.

The most critical facet of either NJ/Manhattan tunnel is probably the Lincoln Tunnel’s insanely high bus traffic.

Joey K October 3, 2014 - 10:59 am

C’mon, freight? There are already significant restrictions on Lincoln and Holland “freight” traffic. Holland cannot accommodate any tractor trailers (3-axle max) and the center tube of the Lincoln Tunnel doesn’t allow any trucks. Both tunnels don’t allow trucks carrying compressed gases or combustible materials, and both tunnels have height restrictions. The chief “freight” crossing in between NJ and NY is the George Washington Bridge.

The Hudson River crossings carried a million less trucks last year than they did ten years ago. The GWB alone carried 5 million less vehicles last year than it did ten years ago, and the combined traffic from Holland/Lincoln is just 70% of what the GWB’s annual traffic was last year. You could divert half of all Lincoln or Holland tunnel trips to GWB and still not get to peak GWB traffic numbers. In terms of trucks, there were only around 1.5 million truck trips across the Lincoln and Holland Tunnels (combined) compared with 3.6 million truck trips across the GWB. The ability to divert traffic is not a problem and disaster would not ensue (unlike shutting down one of the cross-Hudson train tubes). Like I said before, automobile and truck traffic has been steadily decreasing over the past two decades.

Bolwerk October 3, 2014 - 10:38 am

lop pointed this out a while ago, maybe on Streetsblog. Traffic congestion only permits about 3200 Holland Tunnel fleshy human body trips in the peak direction and not even 4000 in the Lincoln Tunnel. See HubBound 2012 (APPENDIX II- Summary Tables – 2012, page 29 or so).

It’s very possible a strategic toll increase could increase tunnel throughput.

eo October 3, 2014 - 10:51 am

Possibly, but I see no other way to increase movement of people through the tunnels other than booting more cars and replacing them with buses. The issue is that the PABT also has no spare capacity during rush hour, so one cannot really replace the regular NJTransit trains with buses.

Bolwerk October 3, 2014 - 10:55 am

Booting some cars and replacing with buses might literally get more cars through. Right now, traffic congestion is making it so they can’t get to the tunnel.

And many buses probably don’t need to go to PABT. Let people out on streets and loop back to the tunnel, sort of like a trolley operation.

Nathanael October 6, 2014 - 12:18 pm

Some buses already do that; there’s not really much street capacity for doing that either.

Fact is, the Lincoln Tunnel corridor traffic has exceeded the capacity of buses. What do you do when you’ve exceeded the capacity of buses? Yeah, you all know this… trains!

Leave a Comment