Thursday, September 11, 2008

Ghettoizing Rail

NOTE: We've moved! Visit us at the California High Speed Rail Blog.

The Progressive Policy Institute released a study earlier this week calling for a major national investment in high speed rail. Their study reads like a greatest hits of the arguments we've cited repeatedly on this blog - the airline crisis, economic stimulus, environmental benefits, and transportation needs.

The study calls for massive investment in five national HSR corridors, including LA-SF. To pay for this the study proposes a dedicated source of funding - a Rail Trust Fund that would provide for HSR as well as other rail service. Its sources would include:

  • A ticket tax on all passenger rail systems, from $5 for Amtrak to $1 for commuter rails

  • Charging freight companies fees to haul along new rails

  • Encouraging state matching funds along the highway model (80 fed/20 state)

  • Proceeds from auction of cap-and-trade (I have also proposed using an outright carbon tax or congestion charge as well)

I would personally just close the Highway Trust Fund, which is obsolete, and turn THAT into the Rail Trust Fund. But the above proposals are a good starting point.

The Progressive Policy Institute, alongside numerous other think tanks, understands the importance of providing passenger rail service at fast speeds to both improve existing rail routes and provide new service where none currently exists. California's high speed rail project will accomplish both goals.

But to hear the HSR deniers tell it, we somehow don't need HSR. That's their new line of attack - HSR is unnecessary and we should not spend $10 billion on it when we could spend the money on BART to Livermore, for example. Seriously, someone proposed that as an alternative to HSR in an email to me, despite the fact that there's about a 100 to 1 difference in the number of riders the two would serve. I'd love BART to Livermore, but come on, it pales in comparison to the service benefits of HSR.

I'll be first in line to agree that we need to improve existing passenger rail service. And hey, guess what? HSR accomplishes precisely that. Those who live along the Caltrain corridor will see major service improvements to Caltrain. Eliminating grade crossings means faster service and shorter trips even on local trains. Passengers could also transfer at Palo Alto or Redwood City (whichever is chosen) to make their journey to SF or San José that much quicker.

The anti-HSR forces are dominated by Northern Californians, which is significant. Southern Californians understand the benefits of HSR, partly because their region is much larger. Anaheim, Burbank, Riverside and Palmdale are already connected to downtown LA via Metrolink and the Pacific Surfliner in some cases, but HSR would make those trips much faster. The Surfliners already connect LA to SD, but HSR will cut that travel time in half if not better - faster than driving. And instead of seeing HSR and other passenger rail as somehow opposed, SoCal rail advocates embrace both systems.

We can look abroad to see evidence of this. In France the TGVs aren't the only form of rail travel. They are well integrated with, for example, the Paris Metro and the RER regional trains. All work together to boost each other's ridership and provide different levels of service that meet most possible travel needs.

What the HSR deniers are trying to do with this "oh we don't need HSR" argument is ghettoize passenger rail and ensure that Californians are never given the chance to use it more widely and frequently than they already do. High speed rail will bring new riders to the rails, rescuing them from a collapsing airline industry and from ever-rising gas prices. The big gap in California's passenger rail network is LA-SF, one of the most heavily traveled corridors of any kind anywhere in the state. HSR would open that corridor to rail, providing a rising tide that lifts all boats and building support around the state for increased investment in rail.

The HSR deniers should be more honest about their motives. It's not that they think HSR is the wrong kind of rail - but that they don't want new rail service period. They believe, against all evidence, that California is just fine relying on planes and cars. They want passenger rail to serve a small niche and not the masses. We reject that narrow, outdated thinking. It's time for California to join the 21st century and build a real high speed train system that can get this state moving again.


Pantograph Trolleypole said...

It's interesting, the worry i think is that it will be successful and people will demand more rail in more places, giving people choices. The worry for these folks for some weird reason is that someone is trying to pry them out of their cars. I never got that, but i guess its the alarmist trying to save their own interests.

Rafael said...

The PPI paper fails to account for the impact of FRA rules on mixing passenger and freight traffic. There are major crash safety, signaling and other compatibility issues with modern, efficient, off-the-shelf rail equipment from overseas vendors.

The Acela Express trainsets had to be retrofitted at great cost to meet FRA's mixed traffic rules, massively increasing electricity consumption and drastically shortening the intervals between overhauls, especially of the delicate tilting mechanism.

The paper also ignores the fact that US freight trains feature very heavy boxcars with doubled-up containers means they cause far more wear and tear on the tracks than the much lighter passenger trains do. For example, it would be extremely expensive to maintain tracks and beds used by a significant number of freight trains to the level required for safe operations above 125mph.

Note that the French mail service (La Poste) owns a number of TGV trainsets converted for hauling light cargo as opposed to heavy freight. The PPI paper does not discuss this third category of rail service, which also includes ground transport of mail order goods, air freight containers, perishable fish and produce, cut flowers etc. It would be possible to connect passenger and cargo trainsets into long single trains, but afaik this is not current practice anywhere in the world.

Finally, the paper does not reflect the scheduling problems inherent in interleaving fast passenger trains with much slower freight trains on the same tracks. European operators get around this by prioritizing passenger trains but that means freight rail have to travel slowly and stop frequently. The result is that freight rail operations in Europe are generally subsidized - in France's case, by TGV operations.

Running steel wheel freight trains through residential neighborhoods at night requires very slow speeds (sometimes also lubrication in bends) to keep the noise down, which limits available capacity. Another reason is that the inherent limitations of human physiology mean train drivers are much less alert in the wee hours of the morning.

Bottom line: it is heartening to see a think tank take such a positive attitude toward investment in high-speed passenger rail service in selected corridors. However, IMHO PPI's arguments would carry a lot more weight if they included some cogent analysis of the gory technical and regulatory details involved in planning, construction and operations at various top speeds.

There's a world of difference between running trains at 79mph, 109mph, 124mph, 150mph and 187-220mph. This last category, which applies to the California project, is called very high speed rail (VHSR) elsewhere in the world for good reason.

crzwdjk said...

Taxing subsidized service? That makes little to no sense as a means of funding. What would make sense is to have some kind of Rail Infrastructure Authority, preferably one representing the interests of the commuter rail and intercity rail agencies as well as the freights, owning all the track in congested urban corridors, and responsible for both operating it and providing new service. Imagine the Alameda Corridor Authority, but with passenger rail participation, and with provisions to allow open access by any qualified company so long as they're willing to pay for track usage and schedule slots.

Oh yeah, and HSR per se won't improve anyone's Caltrain commute, since I doubt they'd even allow people to ride just from San Jose to SF, but the improvements brought about by the project, namely a mostly four track and electrified Caltrain line, would allow for much faster and more frequent Caltrain service, both local and express.

crzwdjk said...

"It would be possible to connect passenger and cargo trainsets into long single trains, but afaik this is not current practice anywhere in the world."
It used to be on Amtrak, though, back in the 90s when Amtrak was trying to attract exactly the kind of light cargo business you describe. Many long-distance Amtrak trains had more freight cars (or road-railers) than passenger cars. It ended because, among other things, it required hour long delays at the terminals to couple/uncouple the freight cars before the passengers could be unloaded.

Rob Dawg said...

I would personally just close the Highway Trust Fund, which is obsolete, and turn THAT into the Rail Trust Fund. But the above proposals are a good starting point.

And how do you propose backfilling the $15b annually siphoned off to subsidize transit?

Robert Cruickshank said...

I don't disagree with rafael's excellent points on the technical issues involved in mixing freight and passenger traffic. But the strength of their study is their reinforcement of the reasons why HSR is badly needed, particularly their focus on economic and environmental needs.

arcady, I don't see anything anywhere that would suggest HSR travelers would be unable to ride just from SF to SJ. It's always been my understanding that the system will run a mixture of trains - some express, some local, some semi-express. If they run trains that originate at the Transbay Terminal and stop at Diridon Station I don't see why they'd disallow people from boarding at the first and detraining at the second.

I do however like your idea of a Rail Infrastructure Authority. And I also believe that there would be use to statewide rail planning, as opposed to the balkanized systems that currently exist.

crzwdjk said...

robert: there's a very simple reason to disallow local SF-San Jose riders, namely that there's quite a lot more of them than there are long distance riders. Imagine getting on a train at the Transbay Terminal only to find that it's standing room only, full of people going to San Jose. Sure, you eventually get a seat after half an hour, but it's still going to be annoying to stand for that first half hour. So either local passengers will be prohibited entirely, or else they'll make tickets expensive enough to discourage most of them (on the order of $30).

As for the Rail Infrastructure Authority, I don't think that needs to be done on a state level. Statewide, between the Caltrans Division of Rail and the HSRA, there's enough planning and operating authority, although it would be nice if those two agencies talked to each other more. The real problem is on the regional level, for example in LA, where Metrolink, UP, and BNSF each have their own lines, where everyone's operations could be made much more efficient by planning and operating everything as a single system. It would also create a reasonable mechanism for funding expansions: on congested lines, the railroads would have to bid for access slots, and the money raised that way should be dedicated exclusively to improving infrastructure.

Rafael said...

@ arcady -

HSR trainsets usually feature Jacobs trucks, i.e. the kind that are located in-between short cars rather than underneath them. Such trainsets are not readily broken up, so the smallest unit for operational purposes is typically 6-8 cars incl. the ones at each end.

In practice, any given trainset would probably be dedicated entirely to cargo or else, entirely to passengers. If a cargo trainset is present, it could be split off at a major station even as a second one was coupled at the other end and, passengers board or alight. The decoupled cargo trainset has its own propulsion system (no locomotives in true VHSR) and is quickly moved to a siding for loading and unloading. When it's ready, it is recombined with another passenger train and the cycle repeats. The whole cargo end of operations would be virtually transparent to passengers and not cause any additional dwell time at stations at all. This isn't the 1950s.

Since cargo and passenger trainsets would feature identical cab layouts, all drivers would be qualified to operate either one or any combination. Only a small number of additional drivers would be have to be hired to shunt the cargo trainsets. The automatic train control (ATC) system would detect the train configuration as part of the handshake protocol during coupling and decoupling and, adjust the train's stopping point at stations down the line accordingly.

Btw, CAHSR absolutely intends to let passengers use the service for relatively short hops like SF to SJ or LA to Anaheim. If demand for local HSR is higher than expected, especially during rush hour, then some HSR trains will ply only those routes. They're not building another BART here, not every train will have to travel the entire length of the spine.

It's not yet clear how the HSR operator(s), Amtrak, Caltrain, Metrolink and ultimately, NCTD will co-ordinate their schedules and fare policies. Operationally, Caltrain may have to discontinue its "baby bullet" service once HSR goes live, reverting instead to local-only trains. The latter will be faster than today thanks to modern EMU equipment. Passengers will also have the option of transferring to semi-express HSR trains in either Redwood City or Palo Alto and, in Millbrae - especially handy if their final destination is south of San Jose.

Note that in France, TGV passengers must always make seat reservations at the counter, over the phone/SMS or the web. This not only avoids overcrowding, it also lets passengers queue up at the right locations on the long platforms. This minimizes dwell times and pedestrian traffic along the aisles inside the cars. If need be, certain doors could even be reserved for exits and others for entrances, creating one-way circulation inside the cars.

Afaik, CAHSR is not currently planning to require seat reservations, fearing that this additional overhead would complicate reservations and put off potential passengers. That might be a mistake: inefficient boarding and alighting procedures are needlessly stressful and reduce system capacity.

Rob Dawg said...

Btw, CAHSR absolutely intends to let passengers use the service for relatively short hops like SF to SJ or LA to Anaheim. If demand for local HSR is higher than expected, especially during rush hour, then some HSR trains will ply only those routes. They're not building another BART here, not every train will have to travel the entire length of the spine.
Not physically possible. Short turn HSR would exceed the maximum circuit density of the line. Please don't suggest a third track, there isn't enough row to operate safely as it is.

Anonymous said...

I would personally just close the Highway Trust Fund, which is obsolete, and turn THAT into the Rail Trust Fund.

Either you don't know what you are talking about, or you're so off the map you might as well be advocating building a HSR system on the moon.

Most realistic proponents of rail acknowledge that it will not replace the interstate highway system any time soon.

Anonymous said...

The Highway Trust fund can no longer cover highway construction in this country. I think what Robert is implying is that another revenue source needs to be found other than the Trust Fund to fund the highway system. (The general fund would work just fine).

Then the remaining HTF funds can be used solely for rail. I say, why not just hike gasoline taxes, that way you reduce carbon emissions, reduce oil imports, fund much needed infrastructure, and encourage people to use mass transit. Then you can fund both rails and roads. Kill many more than two birds with one stone.

Anonymous said...

SoCal train wreck death toll rises to 17

How prophetic are Rafael's remarks regarding safety when we read today about the terrible train accident down south.

Safety is of course a major concern; The UP refuses to allow HSR on its corridors. Can there now be any doubt that they are serious.

Rafael said...

@ rob dawg -

HSR technology permits the safe operation of trains with headways of as little as 3 minutes. In real-world operations, the minimum headway will likely be 5 minutes. However, the experience in other countries has been that initial passenger volume does not justify running long-distance trains at anything like that frequency.

CAHSR itself is predicting around 34 million passenger trips per year in 2020, growing to 90+ million in 2030 (an ambitious number). Network capacity using full-length double-decker trains with 1000+ seats each will be around 120 million annual passenger trips. SNCF and JR already operate such trains on selected routes, with very high seat utilization during rush hour.

Technically, it would therefore be quite possible to utilize a greater share of the California system's available capacity on certain sections, especially early on. Later on, I think this is actually a problem CAHSR would quite like to have.

@ anon @ 6:56am -

my sincere condolences to all those who lost loved ones in this major accident.

Note that this was a head-on collision of two trains on the same track at a relative speed of 60mph. Most likely, the signaling failed or else one or both of the drivers ran a red light. We'll have to wait for the NTSB report for the official cause.

This specific crash scenario could not happen with HSR because it would not ever share track with freight or non-HSR passenger trains. HSR also features advanced - and proven - signaling and positive train control technologies that very effectively minimize the risk of a pile-up of two HSR trains. Head-on collisions don't happen on dual track alignments. IMHO, it is not appropriate to compare the accident risk of antiquated legacy systems in the US with the state of the art in HSR technology.

The crash scenario UPRR had advanced entailed the highly unlikely but theoretically possible combination of a severe derailment, fouling of adjacent track and subsequent collision of a second train with the debris of the first. This is not what happened in Chatsworth today.

For reference, note that 41 people died on California roads over the Labor Day weekend. Annually, around 4000 do.

I'm well aware that anyone who has ever lost a loved one on any vehicle couldn't care less about statistics. However, they are relevant for planning purposes, to keep the relative level of risk in perspective. In spite of today's tragedy, trains are still far safer per passenger-mile than cars ever will.

Spokker said...

"Anonymous said...

SoCal train wreck death toll rises to 17"

Your comments have already been refuted in this thread and the other, but I just wanted to take this opportunity to call you a piece of shit.

Thank you.

Rafael said...

@ spokker -

language, Timothy. Please.

Spokker said...

"language, Timothy. Please."

I stand by my remark. :)

Great reference by the way.