The Antiplanner

High-Speed Rail Part 6: The Risks of California HSR

16th September 2008

High-Speed Rail Part 6: The Risks of California HSR

posted in Planning Disasters, Transportation |

Yesterday’s post on the costs of California high-speed rail discussed the likelihood that this megaproject will cost more than projected and the likelihood that taxpayers who pay for construction will have to pay again to rebuild the system every thirty or so years. Such costs are not so much a risk as a certainty. But there are many other risks involved with high-speed rail, some of which could unexpectedly drive up construction costs even more, and others affecting operations.

Some of these risks were identified in the senate oversight report on high-speed rail, including right-of-way, safety, and ridership risks. One important rish that was not brought out by the senate report is the risk that competing technologies will render high-speed rail obsolete.

Right-of-way risks: Rail aficionados often fantasize that high-speed rail can be built in the median strips of existing freeways, thus eliminating any problems with buy rights of way. But freeways have steeper grades and sharper turns than are feasible for high-speed rail, and they often don’t have room in the medians in any case.

The California High-Speed Rail Authority actually planned to use a portion of the right-of-way around Interstate 15 near San Diego for its route. But a recent expansion of highway lanes has eliminated that option. Now the Authority is searching for a new route, which will undoubtedly cost more.

The California project relied much more on existing railroad rights of way for its route. Pretty much most of the line from Sacramento to Anaheim would follow either the Union Pacific (former Southern Pacific) or BNSF (former Santa Fe) routes, while the San Francisco-to-San Jose line would follow the state-owned CalTrains (former Southern Pacific) route.

The senate report questioned whether there was room in these rights of way for high-speed rail. As an example, it pointed to a two-track BNSF route between Fullerton and Commerce that now carries 127 trains a day. Due to the number of trains, BNSF is planning to build a third track, which will leave no space for the double-tracked high-speed rail line.

Even if space is available, it is not clear that the railroads will cooperate. In May, 2008, the Union Pacific Railroad told the Authority that it was not interested in sharing its right-of-way with high-speed trains. “Union Pacific has carefully evaluated CHSAs project,” says a letter from the railroad, and “does not feel it is in Union Pacific’s best interest to have any proposed alignment located on Union Pacific rights-of-way. Therefore, as your project moves forward with its final design, it is our request that you do so in such a way as to not require the use of Union Pacific operating rights-of-way or interfere with Union Pacific operations.”

The freight railroads have good reason to be wary of sharing their right-of-way. Earlier this year, CSX was sued for negligence when one of its freight cars crashed into a commuter train near Boston. A simple way for the railroads to avoid such accidents is to keep passenger trains off their properties. Another way, which is what Union Pacific might be seeking, to for the Authority to sign an agreement accepting sole liability for any problems even if they are nominally the railroad’s fault. In any case, until the Authority reaches an agreement with the railroads, California voters cannot count on cost or route projections that use existing railroad rights of way.

Safety: The senate report points out that rail safety standards follow very different philosophies in the U.S. than in Europe and Japan. Europe/Japan relies on accident avoidance through precise engineering of tracks and vehicle wheelsets. The U.S. relies on accident survivability through careful design of rail cars. In practice, this means that U.S. rail cars are much heavier than those in Europe/Japan.

Obviously, lighter weight vehicles are less expensive to build and require less energy to move. The Authority assumed it would use such vehicles in its 220-mph trains. This allowed it to project, for example, that despite the high speeds its trains would use less than half as much energy, per passenger mile, than Amtrak’s trains.

For the most part, the European/Japanese philosophy has worked very well as there have been few fatal high-speed rail accidents. But the above photos illustrate the differences in safety standards. The photo above is of a high-speed rail accident, caused by a faulty wheel, in Germany. Several of the cars were smashed to pieces and more than 100 people were killed.

This photo is of the 2005 commuter-rail accident in Glendale, California. Most of the railcars in this accident are intact and only 11 people were killed. Granted that the Glendale train was not going as fast as the German train, the U.S. rail cars were still more survivable.

Last week’s Chatsworth collision, in which one commuter car “telescoped” into the locomotive and killed at least 25 people, is exactly the kind of accident U.S. safety standards are designed to prevent (and makes me wonder if those double-decker commuter cars are properly designed). Except for the first car, all of the cars on the train were intact. If lightweight European cars had been involved in the accident, it is likely that many more would have been killed. (All accident photos are from Wikipedia.)

The problem for California high-speed rail comes with the Authority’s plans to operate its trains in the same rights-of-way as conventional U.S. trains, as shown in the animation above. The derailment of a conventional train could cause severe damage to a high-speed train. The Federal Railroad Administration, which regulates rail safety standards, might not object to lightweight trains themselves, but might object to operating them adjacent to conventional trains. If so, then the Authority will have to find alternate rights of way even if the railroads were willing to let it use their routes.

Ridership: The California high-speed rail EIS projects (on page 3.2-26) that rail will carry between 32 and 58 million people in 2020. Are these numbers realistic? The Reason Foundation suggests we compare them with Amtrak’s Boston-to-Washington (Northeast Corridor) trains.

The Census Bureau estimated that about 41.0 million people lived in the metropolitan areas of the Northeast Corridor in 2006. By comparison, only 32.5 million people lived in metro areas on California’s high-speed rail corridor. At current rates of population growth (about 1 percent per year in California), this will reach 37.5 million by 2020, or roughly 92 percent of the Northeast Corridor today.

Amtrak reports (on page A-3.3, physical page 21) that, in fiscal year 2007, it carried a record number of passengers: 10.0 million in the Northeast Corridor, just 3.2 million of which rode its high-speed Acela trains. Of course, California’s trains are supposed to go faster, but will that speed be enough to attract ten times as many passengers as Amtrak’s high-speed train or more than three times as many passengers as use the entire Northeast Corridor?

Complicating the question is the fact that, while the California and Northeast corridors are both about the same length, their populations are distributed quite differently. The Northeast Corridor has tens of millions of people living in intermediate metro areas: Providence, Bridgeport, New York, Philadelphia, Baltimore. California’s corridor has most people living near the end points of the corridor. This means California’s trains will attract fewer intermediate riders, who make up the vast majority of Northeast Corridor ridership.

If actual California ridership falls short of projections, then all of the benefits claimed for the project will be overstated. In particular, the revenues that the Authority is counting on to repay the private partner’s share of the costs will not be there. Since the Authority wants to build its network in stages — San Francisco to LA first, with extensions to San Diego, Anaheim, and Sacramento later — a revenue shortfall will make it even more costly for California taxpayers to complete those extensions.

Competing technologies: One of the examples of optimism bias in the California rail plans is a presumption that there will be virtually no improvements in competing technologies — highways, autos, buses, and air — over the live of the high-speed rail network. Autos are expected to travel just as slowly (if not slower due to increased congestion), airport security will continue to delay air travelers, and both autos and airplanes will continue to use just as much energy as they do today.

As I will discuss tomorrow, the energy assumption is certainly wrong. But the other assumptions can easily be wrong as well. There are many improvements in both technology and systems management on the horizon that will keep high-speed rail from being the grand success its planners claim.

Computer-controlled autos: UC Berkeley has developed a bus that steers itself. On city streets, the bus still needs a driver to start and stop, but on freeways any vehicles using this low-cost technology could safely travel at high speeds, thus greatly increasing the capacities of existing highways.

If you can take your own car from, say, Sacramento to Bakersfield or San Francisco to Fresno on an uncongested highway at 90 miles per hour — and let the car do the driving most of the way, freeing you to read or work — then high-speed rail loses most of its advantage. Sure, the train goes faster, but first you have to get to the train station, wait until the train is scheduled to leave, then when you arrive you are dependent on local transport instead of enjoying the convenience of your own car.

Streamlined airport security: Even at 220 mph, high-speed rail can’t compete with 400-mph turboprops, much less 500-mph regional jets. The Authority’s plans count on airport security delays forcing people to arrive at airports an hour or two in advance of their plane’s departure. But if security is streamlined — perhaps by vetting frequent flyers in advance so they can by-pass airport checkpoints — high-speed rail loses this advantage. Since (according to page 19 of the senate report) the Authority is counting on business travelers for 91 percent of its profits, losing frequent travelers means losing the whole ball game.

End of hub-and-spoke travel: Most airlines today have two things in common: they rely on hub-and-spoke systems and they lose money. The airlines that make money, like Southwest and Jet Blue, are not using hub-and-spoke systems. But people flying from, say, Oakland to Bakersfield or San Diego to Fresno are forced to use the hub-and-spoke system, usually meaning layovers in LAX or SFO. A regional airline that breaks the hub-and-spoke model, flying more planes direct to more California destinations, will offer greater competition to high-speed rail.

Continuing decentralization: To boast rail’s advantage over flying, the Authority is fond of comparing downtown-to-downtown travel times by rail vs. air. But this assumes everybody wants to begin and end their trips downtown. In reality, downtown Los Angeles has only 2.5 percent of the region’s jobs (see page 12), and the downtowns of San Francisco, Oakland, and San Jose together only have about 12 percent of Bay Area jobs. Job decentralization is likely to continue, making high-speed rail attractive to fewer and fewer travelers.

Improved bus service: So-called Chinatown buses are offering increased competition to Amtrak’s Northeast corridor. Aided by Internet ticket sales, these buses offer very low-cost travel and often provide city-to-suburb or suburb-to-suburb service (thus going people actually live). Even confined to highway speeds, they can be competitive because different buses serve different city pairs, thus avoiding the delays of intermediate stops. Such buses can provide transport to today’s decentralized cities in ways that fixed rail cannot.

These are just a few of the ways that changes in technology and other systems can be foreseen to make high-speed rail a risky investment. No doubt there are many unforeseeable changes as well. California voters who support high-speed rail are betting that none of these changes will take place.

This entry was posted on Tuesday, September 16th, 2008 at 12:00 am and is filed under Planning Disasters, Transportation. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

There are currently 17 responses to “High-Speed Rail Part 6: The Risks of California HSR”

Please feel free to submit comments. Constructive debate is welcome. Ad hominem attacks and name-calling will reveal the shallowness of the author. Foul language may be deleted. Your first comment will require moderation; after that, all comments will be approved unless they contain too many links in which case the software may hold them for approval as potential spam.

  1. 1 On September 16th, 2008, Hugh Jardonn said:

    Here’s another interesting take on CAHSR:
    http://www.insidebayarea.com/sanmateocountytimes/localnews/ci_10472331

    posted on September 16th, 2008 at 9:13 am

  2. 2 On September 16th, 2008, PeterSK said:

    As an addenda regarding safety the FRA requires specific track engineering to be in place in order for a train to operate at a given speed. This is referred to as the track class. The physical infrastructure investment required to upgrade a track class, and thereby increase potential train speeds, is substantial: road beds and rights of way need to be improved particularly because more rail ties are needed to keep the rails in alignment, as well as accomodate grades at speed. For rail speeds in excess of 90 mph, this sort of investment would need to occur if existing freight lines were to be used to carry high-speed passenger rail. I would assume that this would be part of the construction costs referred to in the report, but I did not see that this was part of the cost equation. Most assuredly, UP and BNSf will not be picking up the tab for the improvements as they will only provide investment to maintain the existing track class needed to support overall network velocities.

    posted on September 16th, 2008 at 9:29 am

  3. 3 On September 16th, 2008, C. P. Zilliacus said:

    Mixing high-speed rail with freight trains? Not a good idea at all! As an example of what can go wrong, read the Wikipedia article below about the 1987 Chase, Maryland crash of the Amtrak Colonial (running at over 100 MPH) and three Conrail locomotives (the Conrail engineer was found to be under the influence of marihuana and had disregarded a signal prior to the wreck):
    http://en.wikipedia.org/wiki/Chase,_Maryland_rail_wreck

    posted on September 16th, 2008 at 10:25 am

  4. 4 On September 16th, 2008, C. P. Zilliacus said:

    Antiplanner wrote:

    > Rail aficionados often fantasize that high-speed rail can be built in
    > the median strips of existing freeways, thus eliminating any problems
    > with buy rights of way. But freeways have steeper grades and sharper
    > turns than are feasible for high-speed rail, and they often don’t have
    > room in the medians in any case.

    I think the above may have originated in Robert Caro’s badly-flawed biography of Robert Moses, the Power Broker.

    In it, Caro asserts that if Moses had “allowed” the Long Island Railroad to build two tracks for LIRR commuter trains in the median of the Long Island Expressway (I-495) when road was under construction, then there would not be any traffic congestion on Long Island.

    posted on September 16th, 2008 at 10:34 am

  5. 5 On September 16th, 2008, stevepasek said:

    I would argue with your contention that high-speed rail can’t compete with 500 mph planes. In congested urban areas, when you add in the time it takes to get to and from airports to the city center, particularly over short distances, the travel time is relatively the same, and, of course, you can get up and walk around on a train during the time you’re traveling, which is a better quality of travel.

    Also, if you’re seriously arguing that computer-controlled cars on highways are anything close to reality (not to mention that the current time-to-market on autos is something like 6 years), you’ve been smoking hemp. This has been a fantasy since I saw film-strips in grade school in the 1960s. Unless you are inclined to mandate that everyone must purchase these vehicles, this doesn’t seem like it would work. Furthermore, if you’re going to argue that (occasional) tragedies based upon the lightweight engineering of high-speed rail cars is a downside, wouldn’t the (probably less occasional) auto accidents caused by the failure of computer-controlled autos contribute as many or more deaths?

    I do agree that the state’s cost estimates likely are also an optimistic fantasy. However, especially in California, where highway congestion is a never-ending cycle of expansion and increased traffic, high-speed rail would probably be an effective alternative, requiring some subsidy to be built and an effective customer acquisition campaign to succeed.

    I also agree that it’s not very smart to put these trains on “shared” railbed with slower-moving ones, that’s just asking for trouble. It does seem that building high-speed-only lines using existing right-of-way along interstates would significantly lower acquisition costs, but require enhancements to local connecting transit. It would have the benefit of creating “transit plazas” along existing travel corridors and remove political squabbling about routing, which always increases costs and sometimes removes logic from routes.

    posted on September 16th, 2008 at 10:41 am

  6. 6 On September 16th, 2008, The Antiplanner said:

    stevepasek,

    Thanks for commenting. Regarding air vs. rail times: you repeat the mistake of counting times from city center to city center. As I point out, most people don’t work downtown anymore, and many do work near airports. So an HSR system that benefits downtown workers but shuts down air service is harming non-downtown workers who rely on that air service.

    Computer-controlled cars are close to reality. Many cars being sold today can accelerate, brake, and steer themselves. All that is really needed is a software upgrade and a sensor to detect either signals in the road or conventional road signs (stripes, etc.). The main problem is one of transition. As you say, that will take several years, but so will building high-speed rail. Most experts think computer-controlled cars will be a reality by 2020, which is the base year for California’s high-speed rail projections.

    To clarify about shared roadbeds, the California project does not propose to mix high-speed trains with conventional trains on the same tracks. But it does propose to put them on adjacent tracks. The Chase collision involved a freight engine that wandered onto tracks used for a passenger train. The danger in California would be if freight cars derailed or accidentally unloaded onto the high-speed tracks.

    posted on September 16th, 2008 at 10:54 am

  7. 7 On September 16th, 2008, C. P. Zilliacus said:

    Antiplanner said:

    > To clarify about shared roadbeds, the California project does not propose to mix high-speed
    > trains with conventional trains on the same tracks.

    Good!

    > But it does propose to put them on adjacent tracks.

    Not good!

    Long sections of Washington’s Metro Red Line are between the two CSX Metropolitan Sub tracks
    (in particular from just north of Silver Spring to just south of Brookland.

    Click this link for a Google Maps view of the Silver Spring station: http://tinyurl.com/58load

    > The Chase collision involved a freight engine that wandered onto tracks used for a passenger
    > train. The danger in California would be if freight cars derailed or accidentally unloaded
    > onto the high-speed tracks.

    Agreed. Getting back to my example above, there have been several CSX derailments on this
    segment of its Metropolitan Sub, doing severe damage to the Metrorail tracks. But thanks to some
    very good luck, the Metro tracks were not occupied by a train at the time of the CSX derailment – if
    there had been a Metro train on those tracks (and in the way of the derailed CSX freight cars),
    there would have been many fatalities.

    posted on September 16th, 2008 at 11:08 am

  8. 8 On September 16th, 2008, Francis King said:

    One thing that has not been said so far is this:-

    High speed track is very different from regular track. Because of the speeds of the trains, the trains cannot stop very quickly. Therefore it is necessary to grade separate the trains from the roads. It is not possible to have at-grade crossings, signalised, barriered, or otherwise.

    As far as competing technologies go, this is a real threat for cars as well as high-speed trains. There is an assumption that cars are a superior transport technology to everything else out there, for the foreseeable future. This is only an assumption. Many technologies, including bicycles and the internet, have not been developed to anything like their true potential.

    Electronically guided cars, however, are mythological. There is no technical reason why they can’t be built right now. Legally, however, there is a problem. If a car driver crashes a normal car, it is the car driver’s fault. If an electronically guided car crashes, it is the fault of the manufacturer. It’s a big legal risk.

    As for airplane security, it can be streamlined. A lot of the new security rules can be ditched, and access to the airplanes much speeded up. It is as important to actually enforce the rules properly. Some of the 9/11 terrorists, when at the check-in desk, came up as suspects on the computer screens, but they were allowed to board, without any further security checks.

    posted on September 16th, 2008 at 12:05 pm

  9. 9 On September 16th, 2008, the highwayman said:

    These are just cop outs O’Toole, the streets that you drive on also have tractor-trailer.

    I know you’re anti-commerce, though you also come off as being more paranoid than Howard Hughes.

    posted on September 16th, 2008 at 12:18 pm

  10. 10 On September 16th, 2008, the highwayman said:

    CPZ, regarding the Conrail case, if any one in any mode is stoned or drunk and driving they are looking for trouble.

    A few years ago America West came under fire for having pilots that had been flying drunk. No one was hurt, though there could have been a real mess if the pilots in question were not grounded.

    posted on September 16th, 2008 at 1:58 pm

  11. 11 On September 16th, 2008, the highwayman said:

    Mr. King, your comments about grade seperation strange in that this is also what makes a motorway a motorway. This is why I think that most of this HSR project should be built in the centre I-5, in that it is already grade seperated.

    posted on September 16th, 2008 at 7:14 pm

  12. 12 On September 17th, 2008, Joseph P. Thompson said:

    Dear Mr. O’Toole,
    1. Thanks for allowing me to comment.
    2. Is this an appropriate place for me to state my position from the Gilroy, CA., Chamber of Commerce Government Review Committee debate with our former Supervisor, Hon. Rod Diridon?
    3. While head-on and rear-end collission train wrecks are specially engaging transport disasters, there is another train accident that must be considered when planning for safe rail transport. “Sideswipes” is the word we used to describe a kind of accident where one railcar, or the freight on it, hit another car, or cars, while passing on adjacent tracks.

    I first became aware of this while working for SP at the old piggyback ramp in San Jose (1964-1970). I think it was in 66 or 67, our graveyard shift was given instructions to unload aluminum bins of ammonium perchlorate (solid rocket motor fuel) and load them on flatbed trailers for early AM delivery to UTC’s rocket motor test site at Coyote (south of SJ), CA. After the cables were undone on the flatcar, forklifts were used to transfer the bins to nearby flatbed trailers.

    While doing this we noticed that one of the bins had a horizontal slash all the way through to the yellow “cake” like material, some of which was scattered on the deck of the flatcar. An old hand said “sideswipe” to explain. I won’t tell you what he did next because you’d think I was making it up.

    While tracing and expediting and diverting cars when I worked for UP’s Traffic Department in San Jose (1970-1980), I learned about other such “sideswipe” accidents. More recently, a client’s driver parked a flatbed set of doubles carrying lumber but didn’t not pull forward enough when he did, and left the rear trailer sticking out onto the Caltrain tracks in Redwood City. The ensuing “sideswipe” accident did not involve any injuries, but it did cause one driver termination, and some crushing-swiping damage to the Caltrain locomotive.

    More recently on 7/26/08 while coming south on Hwy. 85, on the overcross over the Coast Main Line near the junction with US 101 in South San Jose, I noticed that a lumber car on the UP train passing underneath had some protruding lenghts of lumber (in bundles) sticking out in the direction of the adjacent track (double track zone between SJ and Morgan Hill). It was flailing in the breeze, and based on what little I know about loads shifting, etc., I thought I ought to let UP know about it before something bad happened. So, I called UP’s Harriman Dispatch Center in Omaha and described what I saw.

    Later, I stopped in Morgan Hill and waited for the freight train to pass. Turns out he stopped just north of Morgan Hill, I assume so that one of the crew could walk along to see it for himself. Later I learned that they held trains while this freight (from your neck of the woods) pulled into the yard at Watsonville. And guess what? I now have something to go with the ten year pin UP gave me when I “retired” to practice transport law. I think you’d like this engraved hunting knife, and since we owe you more than we can ever put into words, I think I’ll leave it to you in my will. Caveat viator!

    posted on September 17th, 2008 at 9:51 am

  13. 13 On September 19th, 2008, the highwayman said:

    I’ve dealt with load shifting aspects too, though in my case it was truck related.

    Pallet loads that have broken apart in side dry vans or containers, with LTL(less than a truck load) cases this can be more of a problem if some one else’s pallet broke stacked on top of yours.

    Trucks have even tipped over, because of load shifting too.

    posted on September 19th, 2008 at 10:02 pm

  14. 14 On February 15th, 2009, Scott said:

    Highman,
    Now I see your angle & ignorance. You are a truck driver.
    Nothing wrong with that or even inherently uneducated.
    People trying to make comments about topics in which they have have limited knowledge usually don’t have much substance. And if you add in a person’s inability to communicate a concrete point with logic & reason…

    posted on February 15th, 2009 at 9:44 pm

  15. 15 On March 19th, 2009, the highwayman said:

    There are plenty of dumb people in this world with PhD’s too.

    Scott you have one hell of a big ego to maintain!

    posted on March 19th, 2009 at 11:51 pm

  16. 16 On June 10th, 2009, Scott said:

    And you have plenty of ignorance which repeatedly show & try to cover up by addressing unrelated items & typing nonsense.

    posted on June 10th, 2009 at 6:36 pm

  17. 17 On July 28th, 2009, the highwayman said:

    Scott, you’re an ignorant clown.

    posted on July 28th, 2009 at 5:59 pm

Leave a Reply

You must be logged in to post a comment.