If you have "no place to go," come here!

Sunday Train: NEC High Speed Rail for Under $20b (from 15Jul2012)

BruceMcF's picture

Sunday Train this week is a re-run from 15 July, 2012

Burning the Midnight Oil for Living Energy Independence

One of the transit bloggers that I enjoy reading is Alon Levy who blogs his observations on a variety of transit topics at Pedestrian Observations . Following the important California HSR funding vote in the California State Senate and the excitement leading up to it, I thought I'd like to take a look at the proposed Express HSR system for the states of the Northeast Corridor.

Of the $53b cost of the proposed San Francisco to Los Angeles Express HSR corridor seems hefty ~ and it seems even heftier when it shows the Year of Expenditure headline value of $68b ~ then the proposed Northeast Corridor states Express HSR will seem massive.

However, Alon claims:

Northeast Corridor HSR, 90% Cheaper
In contrast with this extravaganza, it is possible to achieve comparable travel times for about one tenth the cost. The important thing is to build the projects with the most benefit measured in travel time reduced or reliability gained per unit of cost, and also share tracks heavily with commuter rail, using timed overtakes to reduce the required amount of multi-tracking.

This sounds like an intriguing possibility ... but is it realistic? Or is it wishful thinking? Follow me below the fold, and then let's discuss it.

The Amtrak Plan

Talking Points Memo: Amtrak Unveils $151B Plan For Northeast High Speed Rail By 2040

Amtrak, the government-owned national rail corporation, on Monday released an ambitious $151 billion plan to develop a high-speed rail line along the currently existing Northeast Corridor rail network by 2040.

The proposed high-speed rail line would travel at top speeds of 220 miles-per-hour in some sections and be able to deliver passengers from Washington, D.C. to Boston in a little over 3 hours.

Travel times between other major Northeastern cities would be shortened even more markedly, with travel times between New York and Boston or New York and Washington, D.C. down to 94 minutes, and a little over a half-hour between New York and Philadelphia.

As both Alon and Talking Points Memo note, this $151b plan is a consolidation of two plans from 2010: the Master Plan which focused on improving existing infrastructure, and the High Speed Rail proposal which the Sunday Train covered back on 4 Oct 2010 in 1:36 NYC/Boston, 1:23 NYC/DC, $117b, 30yrs.

So there is not much new to say about this "new" $151b proposal. Indeed, that is one reason why Alon could come up with his "90% off" proposal so quickly.

Why does the Amtrak Express HSR project cost so much?

So its two programs (each consisting of a number of distinct projects) that have been bundled together. The HSR side costs on the order of "only" $120b. So, what accounts for that price tag?

Surely part of this is the seeming inflated price that we pay for rail infrastructure in the United States compared to much of the world ... but that cannot possibly be the major part of it. Alon pinpoints a major part of the construction budget:

... Per kilometer of route length, this means the project has now crossed the $200 million/km mark, a higher cost than 60%-underground Chuo Shinkansen maglev. The primary cause of the high cost of Amtrak’s project is the heavy amount of deep-cavern urban tunneling: nearly a tenth of the cost is the Gateway Tunnel, a rebranded bundling of ARC into the project, and a similar amount is a similar project in Philadelphia.

Now, if Alon is promising Express HSR for "90% off", or in the range of $15b, then he is looking at an Express HSR corridor for less than the cost of those two tunneling projects alone.

Alon proposes some general principles to guide the project:

  • 1. Rolling stock is cheaper than infrastructure.
  • 2. Speed up commuter trains instead of bypassing them.
  • 3. The regulations should be based on service needs.
  • 4. On shared segments that aren’t bypassed, build infrastructure that allows higher speeds
  • 5. Make sure station throats allow full speed
  • 6. Fix curves in higher speed zones.
  • 7. Worry about track capacity when all other capacity factors have been optimized.

And what is the plan? Thing is, Alon had already presented the plan when the question of whether the California HSR system would go ahead was still undecided.

Washington to Boston in 4hrs

This list of projects is taken from Plan B for HSR.

Alon lists four projects which could be pursued in the near term future.

NT1. Constant tension catenary. Much of the NEC has variable tension catenary. In the summer time, when the wires expand, the tension is lower, and the wires sag more, while in the winter time, when the wires contract, the tension is higher, and the wires do not sag as much. Constant tension catenary includes mechanisms that maintain constant tension. The cost of about $2b to $3b for the whole NEC, and would raise the speed limit from 135mph imposed by the catenary to the speeds imposed by the track curves themselves. The faster the top speed of the trains and the more curve easing projects are completed, the greater the benefit.

NT2. New Haven Bridge Line Replacement. Two bridges, the Devon bridge and the Cos Cob bridge, require replacement, one bridge, the Saga bridge in Newport, requires rehabilitation. Given that the Connecticut River bridge replacement cost $500m to cross a wider river, figure $1.5b or less.

NT3. B&P tunnel replacement immediate west of Baltimore. These tunnels were designed with excessively sharp curves, and

NT4. Pelham Bay Bridge Replacement. This is a bridge like the Cos Cob bridge with sharp turns in the bridge approaches that impose speed limits on trains using the bridge. This is $500m in the Amtrak Master Plan, and this project alone would convert a speed limited section to a 125mph section for a few miles around the bridge.

Building on this are seven medium term projects. Since these were outside of the time frame if California had played games with its HSR funding and lost it, Alon does not have indicative costings for these. However, some of these are included in a set of projects in the Amtrak Master Plan at a cost of $4b.

MT1. New rolling stock.

The Acelas are heavy, low-capacity, low-performance, and high-maintenance. New trains can’t be FRA-compliant, and in practice some time (measured in years, not decades) can pass before the best rolling stock is legal on US track. But Amtrak and all involved in HSR on existing track should be at the forefront of asking for an overhaul. High-acceleration trains, capable of about the same cant deficiency as the Acela (for example the E5 Series Shinkansen and the high-speed Talgos), can achieve much faster trip times than possible today, with trivial changes to right-of-way geometry. Of course the tracks would have to be maintained to higher standards, but that’s much cheaper than moving a viaduct or carving a new right-of-way through a residential suburb.

This is the key to much of the extra cost of the Amtrak $120b Express HSR plan. The Amtrak plan treats existing FRA regulations as sacrosanct, as equivalent to laws of natures. But FRA safety regulations do not obtain any greater safety for high speed passenger rail operations than the Europeans and the Japanese are able to obtain with regulations that permit much more capable passenger trains.

If we would rather spend $10b-$20b than $120b-$150b to bring the travel time between DC and Boston down to four hours or less, we need a regulatory regime for a primarily passenger rail corridor that fits the realities of the 21st century, rather than a regulatory regime focused on the needs of mid-20th century freight hauling.

MT2. Elizabeth S-Curve Modification. This is in northern New Jersey. The current speed limit is 55mph, imposed by the transition between the two curves in the S-curve. If rebuilt, with the track provided with aggressive banking and using modern rolling stock, the speed limit could be raised to 135mph, pushing the northern end of the high speed section in New Jersey closer to Newark. This requires taking a newly built $48m building and part of a parking garage. This is included in the pool of improvements that Amtrak lists at $4b.

MT3. Metuchen S-Curve Modification. This is an S-curve which is, I'm guessing, further south in New Jersey. The Metuchen S-curve does not impose as severe a speed limit as the Elizabeth S-curve. However, it is in the middle of what is otherwise full Express HSR speed zone, so the impact that it would have on 220mph capable rolling stock would be substantially greater. This is included in the pool of improvements that Amtrak lists at $4b.

How fast the curve can be made to go depends on how much neighboring properties are taken by the project. The takings involve some residential properties on one part of the S-curve, and a strip mall and industrial site on the other part. Alon believes that with relatively modest takings, the curve could be built for 150mph, with more aggressive takings from the strip mall and industrial site, the curve could be built for 180mph, and taking the whole strip mall and industrial site would allow for a full speed 220mph curve.

MT4. Port Chester-Greenwich Bypass.

Most of the slowness of the segment between the NY/CT state line and Stamford comes from Cos Cob, but part of it comes from a sharp curve in Port Chester that can’t be modified without too many takings. As an alternative, trains should leave the existing line just south of Rye, travel along I-95 and its gentler curves, bypass Port Chester and Greenwich, and rejoin in the vicinity of the newly-raised Cos Cob Bridge. Curve radius without significant residential takings would be about 1,300 meters through the I-95 S-curve in Rye and Port Chester, and 2,000 meters elsewhere.

Fixing the Port Chester curve and the Cob Cos bridge, the NEC between the Connecticut Border and Stamford, Connecticut could be upgraded into what I have been calling a "Rapid Rail" corridor for local commuter trains. The current permitted top speed is 75mph in MetroNorth rail territory in Connecticut. With these two projects and with improved banking of the tracks, the track geometry would allow 105mph from New Rochelle CT to Harrison CT, 125mph from Harrison CT to Stamford, and 100mph through Stamford.

And if the local commuter rail runs faster, that is a benefit to the HSR on the corridor as well, since the faster the commuter rail can run, the fewer times a given HSR train must pass a commuter rail train.

MT5. New Rochelle Interlocking Grade-Separation. This is the flat junction with a speed limit of 30mph. It also creates capacity constraints during rush hour, when heavier use by the local commuter rail implies only limited slots are available for the intercity trains running through. The project would run a rail overpass for the intercity trains over the top of the interlocking, allowing them to run with speed limits imposed by slow speed switches and allowing them to be scheduled without concern for bottlenecks caused by peak demand commuter rail services.

MT6. Eastern CT I-95 Bypass. This is the big bang part of the project. Upgrading the transit for a fast train on the NEC New York through Western Connecticut to an hour. The NEC from Kingston, Rhode Island through to Boston is already the fastest section on the NEC. The I-95 Bypass bridges the gap between the two:

This project would start right at New Haven Union Station, cross the Quinnipiac River at a new bridge near US 1 and the new I-95 bridge, follow I-95 to the state line, and then cut across barely-populated territory to the Shore Line at Kingston, where it straightens.

Cost: this is 121 km of tunnel-free route, and based on similar costs in Europe, it should be $2.5 billion. Carefully tracing through the unit costs implied by the Penn Design group, following California HSR costs, produces a figure of $2 billion. But this assumes much lower costs for the bridges over the rivers than Amtrak has produced so far; Amtrak costs are likely much higher, though not by orders of magnitude.

Benefit: New Haven-Providence in about 40 minutes, New Haven-Boston in about an hour. Current travel time can be improved using better rolling stock, but there’s a point of diminishing returns, and reliability with present-day movable bridges, especially over the Connecticut River, is low, requiring extra schedule padding.

MT7. Commuter Rail-HSR Compatibility. This involves four track overtakes for the High Speed Rail so that it can overtake local and regional rail service without interference. It also involves upgraded passenger trains for the Maryland local train system, MARC, and the Greater Boston train system, the MBTA.

So, does it all add up?

Does this all add up? Well, the its reasonable that this would upgrade DC to NYC to about 2 hours and Boston to NYC to about 2 hours.

As to whether it would be a 90% discount off of $151b ~ that, I don't know, as a price costing requires more detailed engineering. But it certainly ought to cost well under $20b. And it can be done in incremental segment ~ indeed, even the East Connecticut I-95 bypass can be done in incremental segments ~ with travel time benefits for each incremental segment.

What it does require is tackling institutional obstacles with institutional reforms, rather than with concrete and tunneling.

This is not precisely the same performance characteristics as the Amtrak Vision, which proposes Boston to Washington DC in 3hrs. However, it proposes Boston to Washington DC in 3hrs in 2040. Alon's plan could give Boston to Washington DC 4hrs in 2025.

Neither is it precisely the same capacity as the Amtrak Vision. However, as Alon points out, a system that is running eight car long passenger trains can't be considered to be at capacity, even if there is no opportunity to run any more trains on that system. First extend the trains to 16 passenger cars, and then talk about having reached the limit of capacity.

Which is where the upgrade gets self-sustaining: if Amtrak was filling up 16 passenger cars up to four times an hour in peak demand periods, then the operating surpluses would be in a position to finance further upgrades. So building the four hours DC to Boston system would itself be how you finance the construction of the upgrades required for a three hours DC to Boston system.

So, for full disclosure, put it at 80% of the Amtrak HSR system, over 60% sooner at under 20% of the price.

And at the same time, since it deals with the "obstacles" of local rail systems sharing the NEC corridor in large part by speeding those local rail systems up, rather than by tunneling underneath them, its an approach with substantial side benefits to the urban areas along the corridor even for those who will not be taking the HSR.


Well, as y'all know, the Sunday Train does not conclude when the opening act essay draws to a close, it just opens the floor for discussion. As always, any issue in sustainable transport is fair game for discussion, whether or not it was touched in in this week's essays ...

... but the question that Alon's proposal raises is, how does a progressive populist movement organize to push through the kind of regulatory changes requires to allow us to get sustainable transport between DC, NYC and Boston this fast, this soon, for this much lower cost than a system designed around the current regulatory system? Is it even possible to organize around such a technical issue as that, no matter how appealing the end result might be?

No votes yet


Jay's picture
Submitted by Jay on

I don't see any point to a Washington-Boston HSR project that skips Stamford, Bridgeport, New Haven, and Providence. White Plains to Hartford looks like a fairly straight shot on a map but it goes over the Litchfield Hills, and omits probably at least 65 percent of the population of Connecticut. Danbury and Waterbury are just spur lines off the main New York-New Haven line. Plus, if you're going to Hartford you may as well go to Springfield, the capital of Mass, and Worcester. It's important to get people to and from DC, Baltimore, Philly, Newark, New York, and Boston quickly, but it's also important for people in second-tier cities already on existing lines to get to those cities, otherwise they'll just use automobiles. The point should be to move people along a heavily populated line, not elites from megalopolis to megalopolis.

BruceMcF's picture
Submitted by BruceMcF on

I'm not sure where you got that it skips New Haven and Providence: "Benefit: New Haven-Providence in about 40 minutes, New Haven-Boston in about an hour."

But if we were aiming for a one-size-fit-all service, what we'd end up with is a service that doesn't fit a large number at all, and doesn't fit very many people very well. What sustainable transport requires are services that fit their purpose well.

The purpose of an Express HSR service is for people to take trips like Boston/NYC and NYC/DC ... not just elites, but a large share of the people making trips like that. And while an Express HSR pays a much smaller penalty for a stop than a airplane does, it still pays a penalty, and above a certain stop density, it no longer makes sense to provide a service with a 160mph train.

So, none of this implies that the Northeast Regional services stop running. To the contrary, all of these imply that the Northeast Regional will offer substantially faster trips. The NEC Regional trainsets, after all, have a top speed of 125mph and will also benefit from many of these same investments. And by taking the Express HSR off of the low bridges, it will be possible to increase the number of NEC Regionals so, for instance, Bridgeport, which is not at present an Acela station, will receive more frequent and faster direct intercity service as a result of this plan. The Amtrak $150b Express HSR plan allows more NEC Regionals, but does not allow them to run faster.

As far as "elite" service ... as the French have shown, that depends upon how you price the tickets. With the low capacity of the present Acela service, there is no opportunity to gain additional revenue by pushing the price down, since the seats are mostly booked out at the current prices (within the limits set down by Congress, of course, which by law forbids Amtrak from using discounts in excess of 50% to fill what would be empty seats to optimize revenue). With more than double the capacity per train and with the ability to run substantially more trains per day, due to not running them over the low bridges in Connecticut, that would allow far more reasonable ticket prices to be offered than for the current Acela.


Submitted by lambert on

I think the seats in the NER trains are more roomy, and in my experience the WiFi is better. (This may have changed. And of course, I like the lower price, even if I don't get to eavesdrop on Senators and so forth.)

And you make a very good point on the parallel lines being able to increase NERs.

Submitted by lambert on


The Amtrak plan treats existing FRA regulations as sacrosanct, as equivalent to laws of natures. But FRA safety regulations do not obtain any greater safety for high speed passenger rail operations than the Europeans and the Japanese are able to obtain with regulations that permit much more capable passenger trains.

If we would rather spend $10b-$20b than $120b-$150b to bring the travel time between DC and Boston down to four hours or less, we need a regulatory regime for a primarily passenger rail corridor that fits the realities of the 21st century, rather than a regulatory regime focused on the needs of mid-20th century freight hauling.

I never thought of the problem that way before.

As for the Acela, I don't remember the exact word that the Quebecois workers at Bombardier used for the train sets, but IIRC it translated to "pig."

BruceMcF's picture
Submitted by BruceMcF on

Surely "boar" would be more fair ... it is a strong train, but then, given its weight, it has to be.

BruceMcF's picture
Submitted by BruceMcF on

This confirms my point! A boar when charging may be a fearsome thing, but graceful like a lion it most certainly is not.

And, like a boar, the Acela takes more time than some other beasts to get up to full speed, or to slow down once it gets to full speed.

Jay's picture
Submitted by Jay on

As far as I know, no rail yet exists between White Plains and Hartford. The NEC as it currently exists is dotted with many large population centers with established rights of way, train stations, and the expectation of HSR to NYC, Boston, and DC. The first map above bypasses all that existing infrastructure and those populations in favor of large stretches of nowhere beyond Hartford. If it were implemented, the separate-but-equal New York, New Haven, Boston service would not be augmented. They'd cut it. We all know that. It's like drawing a straight line from New York to DC: the people in Baltimore and Philly would be hopping mad, and as a result it would be politically unfeasible.

Another thing: Maine needs rail. No one ever talks about them.

BruceMcF's picture
Submitted by BruceMcF on

"Next Gen" plan for a dedicated, mostly all new, Express HSR corridors.

So, yes, it selects alignments that are mostly distinct from the NEC since it is based on a premise that the NEC has hit capacity ... as Amtrak understands it.

Alon Levy's approach is to upgrade the NEC where that is compatible with operating a 160mph Express HSR service, with 16 car trains for improved capacity as well. In sections where the track would allow effective transit speeds but the corridor is shared with local transit (Metro Boston and Maryland's MARC), install overtakes to allow timed overtakes. And in one section where upgrading the existing corridor to that speed would be particularly expensive, due to the number of low bridges that would have to be rebuilt, look for an alternative alignment to connect Hartford to Providence.

Way back when the $8b for HSR was provided in the Stimulus II bill, MA DOT set out a vision for New England Intercity Rail (pdf) that included improvements to the Downeaster with extensions to Brunswick and Auburn, and the Auburn extension as the foundation for a possible northern route to Montreal.

Funny how available money sparks creativity in state departments of transportation.

And absolutely insane that the didn't include rail in the road tunnel from the Boston South to North tunnel ... that is why the NEC terminates at Boston rather than extending, as the NEC extends south from DC to Richmond.