Tuesday, November 25, 2008

California As Model For the Rust Belt?

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

So argues Ryan Avent:

High-speed rail would, in other words, turn Rust Belt distances into northeast corridor distances, while also shifting the Rust Belt closer to the northeast corridor. It would increase the return to doing business in every city in the region. It would be the Erie Canal and the original railroads on steroids.

And here’s the thing — California is estimating that its 800-mile high-speed rail network will cost it about $45 billion over twenty or so years. The actual cost will probably be higher than that, and a Midwest network would be larger and therefore more expensive, but the total cost is in the same ballpark as the $50 billion in aid automakers are begging for (which wouldn’t even be spread out over a period of years).

Avent's questions come in the context of assessing the proposed auto industry bailout, but it's a worthwhile reminder of the economic stimulus value of high speed rail. It can help a struggling region grow its economy through proximity to regions that are doing better. Affordable and rapid transportation that connects the Bay Area to Southern California, the Central Valley to the rest of the state, Orange County to LA and Silicon Valley to downtown SF, generates its own form of economic stimulus.

HSR's value is that it provides fast trains over long distances. It's not just upgrading existing commuter rail corridors, although that's part of it and is quite valuable. It's also providing new forms of transportation service along routes where it either doesn't exist or is inefficient. (I love you, Coast Starlight, but you're not exactly an efficient passenger rail solution to north-south travel.) That means economic growth that gets generated in one place can have beneficial effects on other parts of the state. In the past the only economic ripple effect has been sprawl - Silicon Valley-fueled growth of housing markets in Modesto, for example, which was not sustainable economically or environmentally. HSR provides a very different kind of economic networking that is more fruitful for the entire state.

Another argument to keep in mind as we prepare to take the HSR fight to Capitol Hill...


Unknown said...


Have other people heard of this project yet? Orange country LA country maglev line?

Robert Cruickshank said...

I was thinking about posting on that today, but I don't know a whole lot about that. It's VERY conceptual at this time. I am doing some research about it and hope to have a post up about it next week.

Anonymous said...

Would it be more likely the Midwest HSR project only starts when CAHSR's spine segment proves the economic viability of HSR in the USA, or within the next four years as part of a stimulus plan by Obama?

Rafael said...

@ Francis -

I think it's safe to say OC maglev is deader than a dodo now that prop 1A has passed. Light rail would make more sense now, the whole point was to improve access to valuable commercial real estate in that area.


Hopefully, someone will talk some sense in to Senate majority leader Reid (D-Nev) so he'll abandon both the maglev-to-Disneyland and the Ivanpah relief airport projects in favor of a spur off the California system at Mojave. Desert Xpress, hitherto a strictly private project to Victorville, would morph into a public-private partnership pursuing this spur.

Basically, the federal government and/or the state of Nevada would pay for electrification with the money saved on the airport and maglev. Jean would still get an HSR station, the strip in Vegas is running out of places to build new casinos.

Palmdale would serve as the relief airport for McCurran, if one is still needed. Fully one-third of its traffic is with California.

A new high-voltage DC trunk line could be strung close to or perhaps even above the the HSR spur. It would deliver solar, perhaps some wind, power from new plants in the Mojave desert and Nevada plus nighttime hydro power from the Hoover dam to California, for the HSR network and other uses.

A fringe benefit of an HSR spur is that the hwy 57 median would be freed up for a spur off the San Diego HSR segment to Anaheim ARTIC.

Metrolink could selectively extend service beyond San Bernardino to Barstow, especially if there's an intermodal with the HSR spur there.

The Riverside UC HSR station should be moved a couple of miles north to Grand Terrace, half-way between Riverside and San Bernardino, to create an intermodal with existing Metrolink service. Two lines that currently terminate should be extended to at least San Bernardino for generate more HSR feeder traffic. SB was very unhappy CHSRA didn't give it a station because running right next to Ontario airport was considered more important.


LA Metro has kicked off a mostly unrelated planning effort about how best to repurpose the old single track freight corridor between LAUS, San Pedro and west Long Beach that it acquired from BNSF in the context of the Alameda corridor project. An extension of the green line (light rail) to LAX is part of measure R. I'm not sure, but perhaps the plan is to steal the inside traffic lane of the access loop for all the terminals. That would also permit circulating trains that remain at LAX, replacing the shuttle buses to compensate for the loss of a traffic lane. Level boarding is essential for trains serving airports.

Light rail might be the best option for the Harbor Subdivision ROW as well, since two light rail tracks can be squeezed into a single track heavy rail ROW. Metro wants to integrate services with the Crenshaw corridor and Long Beach Transit Terminal as well.

I've argued for commuter rail based on bilevel cars between LAX and LAUS before, but perhaps light rail delivers more utility and capacity. Connecting to HSR isn't its only or even the most important purpose.

Rafael said...

Cities in the Midwest are are arranged in a kind of hub-and-spoke pattern centered on Chicago. The area also has a very large - if weak - auto industry. If GM and/or Chrysler go under, then Toyota, VW and Renault/Nissan will all be interested in their physical assets, their established supply chain and their skilled workforce. The big three still have a combined share of almost 50% of the US light duty market. New owners wouldn't offer the current union labor rates, health benefits or retirement plans, though.

My point is, the Midwest is not going to stop building cars. Their suppliers will still deliver high value parts just-in-time to assembly plants. That means it would make sense to upgrade freight rail lines to "rapid rail", essentially creating a European-style regional network use for both passenger and priority freight trains at 110-125mph, the limit FRA has set for grade crossings.

Freight rail companies would contribute their existing ROWs and tracks to a public-private partnership responsible for upgrading to dual track, modern signaling, quiet zone grade crossings and maintenance. Heavy freight trains would still run in rapid rail corridors, but there would be minimum speeds and a timetable organizing all traffic. The states would subsidize those regional rail operations deemed essential (e.g. to avoid freeway expansion or exposure to high fuel prices) but with fare box returns below 100%.

Where appropriate, new shortcut tracks would serve as bypasses that would be off-limits to heavy rail. FRA-compliant and non-compliant rolling stock would be allowed to share track in rapid rail corridors. That would require a major relaxation of the mixed traffic rule, made possible by signaling and grade crossing upgrades, positive train control (mandated by HR 2095), and Caltrain's proof of safety in grade crossing accidents.

Electrification (25kV AC @ 60Hz) and grade separation would be implemented selectively as and when needed. By default, rapid rail would be based on diesel locomotives and self-propelled rolling stock.

On selected routes, service at 220mph on dedicated, electrified all-new passenger-and-light-cargo-only tracks may be justified. With rapid rail as the foundation to support just-in-time delivery of auto parts, true HSR in the Midwest could follow the French model of laying fancy new tracks only in the countryside. The trains would generally revert to rapid rail operation to reach downtown stations.

In some cases, capacity constraints may dictate construction of dedicated passenger service tracks inside built-up areas, e.g. via grade separated aerial alignments immediately above the legacy freight tracks. Those could continue to rely on quiet zone grade crossings, which would be separated with full underpasses as and when needed.

This three-class model of traditional heavy freight, European-style regional rapid rail and true HSR corridors, with mixed traffic permitted in rapid rail corridors and trains transferring between corridor classes is very flexible. It allows for incremental improvements to the grid in decades to come and may be well suited to other areas under consideration for true HSR. For example, it would greatly simplify implementation along the I-95 median in the NEC while preserving access to existing downtown stations.

The cornerstone of this concept is getting FRA to relax its mixed traffic rule without compromising operational safety.

Andrew said...


Concerning LAX rail access: I think the current plan is to simply run a Green Line spur up Aviation Blvd. to a station at Century Blvd., and count on LAX to build a tram connecting it to the terminals. It's a rather less-than-ideal arrangement in my view.

Ideally, there would be an underground station located directly between the three terminals, underneath the parking structures. It would be stub-end in layout with the end pointed towards the Tom Bradley International Terminal, and feature four tracks and platforms, two for the Green Line and two for a line running to Union Station. Leaving the station, tracks would run under Century Blvd. and emerge at Aviation. South-turning tracks would connect with the existing Green Line, and north-turning tracks would run on an elevated structure built along the Harbor Subdivision to Downtown. Intermediate stations along this LAX-Downtown line would feature sidings, allowing for both express and local services.

Anonymous said...

Building rail travel in the Midwest should be cheaper than California because of the geography of the Midwest. Most lines needing to be built would be built in similar terrain to the Central Valley — flat and farmed. There would be few, if any, tunnels and no major seismic concerns.

The Midwest also has plenty of railroad right-of-ways which can be used for HSR. Consider that between the Twin Cities and Chicago, for example, there were three competing "high speed" lines (operating over 100 mph) in the 1950s and 1960s (The Milwaukee's Hiawathas, the Chicago and Northwestern's 400s and the Burlington Route's Zephyrs). Much of these corridors are now abandoned or barely used, but might be perfect for high speed conversion.

Urban areas might be easier to access as well. In San Francisco, there is precisely one rail corridor in to the city for forty miles, and it currently has dozens of trains daily. It will need major reconstruction and a capacity increase in order to keep current services and add high speed services, which won't come cheap. Many Midwestern cities, especially Chicago, have literally dozens of rail routes accessing the city center. While most of Chicago's see some commuter services, the corridors are often three tracks wide (or wider) and don't generally see the same level of service as Caltrain. In most other cities, there are two or three railbeds accessing the city, some of which may only see infrequent freight use if they have not been abandoned completely. Since these freight corridors often parallel each other, one could be converted to high-speed use without major changes to freight service.

Now, the main issue is that the Midwest is not a single state. Thus, we'd need some federal support to bring the states together (and give them some money. Stimulus, anyone?

Rafael said...

@ Andrew -

if Metro's plan is as you describe, then it is asinine. Passengers with bags hate transfers.

Here's what Caltrain customers currently have to do to to get to check-in at SFO:

- get to a Caltrain station
- buy a Caltrain ticket
- board a Caltrain (upgrade to level boarding is part of Caltrain 2025 master plan)
- alight at Millbrae
- go up to concourse level
- buy a BART ticket (work like debit cards)
- go down to BART platform
- board a BART train (level boarding)
- timed transfer to another BART train on other side of island platform at San Bruno
- transfer to AirTrain (level boarding) at SFO BART
- alight at the right terminal
- walk to check-in

It's hardly surprising that hardly anyone uses Caltrain to get to SFO. Once HSR trains run, BART will re-instate direct shuttle service to its SFO station to eliminate a transfer and increase service frequency.

No word on integrated ticketing yet, airline passengers using HSR for feeder service would presumably hate having to wait in line to buy a $10 BART debit card they'll only use a fraction of. It would be nice if they could get a globally unique single-use barcode ticket just for the BART shuttle and print it out as part of their e-ticket.

As for LAX, a dual track light rail alignment should run on an aerial across Aviation, that's how the BNSF tracks runs. However, a wye would provide access to two tracks in the median of Aviation, initially also an aerial. Those tracks would then be used for a one-way loop with a stop at each terminal.

A single-track connector at the east end of the terminal loop road would permit a few light rail shuttle trains to run in loops all day long, never leaving LAX. No unmanned people mover needed, just let Metro provide the courtesy shuttle service. Shuttle bus service would be canceled.

Ergo, multiple light rail lines would share tracks to and around LAX and between Cranshaw and Redondo Beach. BART leverages its tracks in SF and Oakland for multiple lines as well. Headways are short and passengers need to pay attention which train they board but it works just fine.

The only question is how the loop would be implemented. If permanently sacrificing a traffic lane is unacceptable, perhaps doing so temporarily would be ok. The single track loop would run in a covered trench and cars on that cover. That leaves pedestrian traffic, platforms should be on the outside of the loop, with underground passages into the terminals. These components of the system would be constructed by tunneling.

Rafael said...

@ Ari -

I'm aware that the legacy ROWs and the terrain in the region will make implementing HSR easier in the Midwest than between SF and Anaheim.

That doesn't invalidate my point that the Midwest would benefit from a three-class approach. Laying separate tracks everywhere for passenger rail would do nothing to get diesel-guzzling trucks off the region's freeways. Rapid freight rail would.

You want to minimize the amount of new track you need to construct and maximize the throughput on that track, so it's worth spending money on maintenance after it's built.

Right now, the freight companies like to defer that because they can. They also have to pay property tax on their land and infrastructure, which becomes more valuable by adding tracks, grade separations, fancy signaling etc. That is why hiving ownership of freight ROWs and tracks to public-private partnerships makes sense. The freight operator is liable for just a fraction of the higher property tax but enjoys the speed and safety benefits of upgraded and properly maintained track. The public side of the partnership uses tax proceeds to pay for infrastructure maintenance.

You also want to minimize the number of ROWs that you press back into service, because of grade separation overheads and the desire to create intermodal stations for multiple classes of passenger rail service.

Jeffery Atik said...

I suspect the initial LAX Green Line configuration will involve a single station with a transfer to a bus circulating among the terminals. Were the line to run under the center axis to a stub terminal (off Aviation and down Century, then down the center axis under parking and the Themes Building), at least one intermediate station and a terminus in the restored Bradley international terminal would be needed. The chief problem with Green Line access to LAX remains: a very unpleasant transfer to the Blue Line at Rosa Parks involving hauling heavy bags up and down a long series of stairs/escalators.

Anonymous said...

The midwest project as someone posted is many states and I know some of thoses states away from Chicago..Ohio for one are very anti
tax when it comes to transit. So unless they get it from DC with no match needed our true HST will probally on line first!
I have also looked at that Orange County Maglev and it shows that system running almost on the same route as HSR ,how can that be?

BruceMcF said...

If you look at the Kerry bill, there is definitely funding that the existing semi-HSR projects would be able to pursue.

The existing Ohio Hub project has two options, one for 79mph and one for 110mph (passive tilt diesels). Heading out of Ohio into the Midwest, I am not sure whether the Midwest Hub has a 79mph option, or whether it is primarily focusing on 110mph, but of course the improvements under way at the moment as part of the project are for existing Amtrak services. The Ohio Hub fully built out for 110mph, circa '04 or '05 I think, that was around $3.3b all up.

AFAIU, the Empire Corridor extension and and Keystone Corridor extension are both along similar lines ... so those four projects would be a system of 110mph semi-HSR services between the NEC and across the Mississipi ... AFAIR, the western ends of the Midwest Hub extends as far as KC and Minneapolis.

Rafael's discussion of a three class model fits in here very well. Most relevant Ohio and Midwestern rights are under-built, and at the speeds and frequencies laid out in the Ohio Hub, sections of high speed single track with double tracked passing zones for passenger, and where required for improved freight capacity, high speed switches to allow FRA compliant high speed freight with PTC to shunt between the heavy freight line and the express line.

The focus on freight rail electrification should be long distance freight routes, mostly on STRACNET, as that is where the electric rail will have a substantial competitive advantage for general time sensitive ground freight when the global economy picks back up and crude oil kicks back over $100/barrel.

Looking at true-HSR, you could well see a true HSR corridor running from somewhere east of Gary to Northeast Ohio, substantially extending the reach of the Chicago three hour zone toward Buffalo and Pittsburgh. The real kicker would of course be a true HSR corridor through northern PA from the Ohio Hub to NYC, but that would seem to be a quite expensive proposition, with a lot of tunneling and viaducts.

Anonymous said...

The BART Millbrae-SFO line could be reconfigured to be free. Both Millbrae and SFO are overbuilt for the service they handle. BART only uses one of three platforms at Millbrae and SFO (for revenue service, for you nitpickers).

Dedicating a platform to a shuttle train at each station would allow the removal of faregates. Reinstating the "Purple" line would also require a labor negotiation. Currently, BART operators get a break at the end of the line- Logical. The SFO-Millbrae service was defined as an entire line- Illogical. This meant operators were getting a break longer than their time operating the train on each run. Labor costs for the shuttle were obviously insane.

This can be fixed through a logical exception for the Purple shuttle in the next BART labor negotiations. It could also be fixed by operating the trains without a driver. The service is such that it doesn't have to interact with any other BART operation, so it could be safely isolated.

In the long term, an extension of SFO's AirTrain (peoplemover) to Millbrae would be the best fix, as it serves all terminals.

Rafael said...

@ anon @ 10:19am -

running the AirTrain down to Millbrae would obviously be the most convenient for customers, and it looks like it could handle the gradient.

However, I recommend you go to Google Maps, find the BART/AirTrain intermodal station at SFO and switch to Satellite mode. You'll se right away that BART's broad gauge tracks run a level below the AirTrain's concrete guideways.

Anyhow, it is what it is now. One low-cost option for getting the BART shuttle working again is to slightly redefine the two lines.

The Pittsburg line would run south to San Bruno, then SFO. Driver #1 is tired and takes a break. Driver #2 is refreshed and hops in the same cab, which is no fasting east. At the same time, driver #3 hops in the cab at the other end, facing west. His job is to drive that train to Millbrae while driver #2 twiddles his thumbs. At Millbrae, driver #3 dismbarks and instructs driver #2 to take the controls. He then drives the train back to Pittsburg.

Meanwhile, driver #3 walks down the platform and waits.

Driver #4 drives the train from Richmond (or Dublin/Pleasanton, as the case may be) to San Bruno and Millbrae. He goes to take a break, his buddy driver #5 gets in the same cab, which now faces south. At the same time, driver #3 enters the cab at the other end and drives the train to SFO while #5 twiddles his thumbs. #3 hops out at SFO, #5 takes over and drives the train to San Bruno and beyond.

Dwell times always remain predictable, so operators can keep them on a strict timetable - essential on the busy core alignments in downtown SF and Oakland.

Drivers #1, #2, #4 and #5 all get to take the regular long break at the end of each long stint at the wheel. Driver #3 has a different contract: he gets many short breaks and has to do a fair amount of walking, but his stints at the wheel are always very short.

This arrangement would mean hiring some #3 drivers but on any given day, they would support two lines.

Alon Levy said...

The Empire Corridor needs way more than 110 mph. Either it or the PRR Main Line needs to be made capable of speeds higher than 220-250 mph, so that as the technology becomes feasible, it'll form the bulk of the New York-to-Chicago high-speed line. This is still not ideal at current speeds, which on the best estimates will make it a 5-5.5-hour journey; however, in the future, it can become competitive with air.

Andrew said...
This comment has been removed by the author.
Andrew said...


I've been to SFO and am well aware of the situation there. I think running BART straight to the airport was a sound idea since you can walk right off it into the international terminal, but trying to use it as a shuttle to Caltrain at Millbrae was misguided. An AirTrain spur to Caltrain at Millbrae or San Bruno would've been a better idea. But for the time being, perhaps BART could dedicate one train to running between Millbrae and SFO that is either free or somehow included in Caltrain fare.

As for LAX, running the Green Line around the terminals is a novel idea, but even with such an extension the Green Line is still pretty useless. If you want to go Downtown you have to transfer to the Blue Line, and again to the Red Line if you want to go to Union Station. It's eastern terminus is two miles, only two freaking miles, shy of meeting Amtrak and Metrolink at Norwalk/Santa Fe Springs Station.

In my opinion, anything other than a mostly (if not fully) grade-separated line using heavy rail EMU's from Union Station directly to LAX would be half-assed. No light rail, no BRT, no bus or people mover transfers. A city the size of LA really deserves airport rail service on the Level of London's Heathrow Express, Tokyo's N'EX, or Hong Kong's MTR Airport Express. Seoul's incomplete A'REX particularly interests me, because when it's finished it will provide both express and local services like I suggested previously.

BruceMcF said...

@ Alon Levy

"The Empire Corridor needs way more than 110 mph. ... it'll form the bulk of the New York-to-Chicago high-speed line.

Why would NYC north to Albany, west to Buffalo, then southwest to Cleveland form the bulk of an HSR focusing on NYC to Chicago? Heading halfway from NYC to Boston to get to Chicago is a long way around (remember from your American history and the "Mohicans and Mohawks" that Boston is just about a straight shot east of Albany).

The Empire Corridor will be on a semi-HSR that will run from NYC to Chicago, but that's to increase the number of intermediate trip-pairs without a break in service ... the NYC/Chicago patronage could be important for that route, but it won't be a big share of the NYC/Chicago travel market. What it will do is to capture a big share of Buffalo and points east / NYC, and Buffalo and points west / Chicago.

A true HSR line focusing on NYC to Chicago would, in a world without budget constraints, run through northern PA, northern Ohio, and northern Indiana. In a world with budget constraints, it might run from NYC to Harrisburg and then share a straightened Keystone Corridor to Pittsburgh, then run through the rural gap between metro Cleveland and metro Columbus toward a connection with the Midwest Hub somewhere southeast of Gary, Indiana.

James said...


The airport shuttle would have been a better solution, but why cannot BART do the same job since we already paid for it? Let the scheduled trains arrive and depart from Millbrae/SFO to points east and keep a single Type C (or a pair of Type C back-to-back) and have it shuttling back and forth between Millbrae and SFO. It would have to be timed to be on a different track or on break at the end of the spur when the trains come and go and otherwise it can be zipping back and forth. Maybe it could alternatively make a triangle run between Millbrae, San Bruno and SFO.

What of the concept of strictly local BART operations, and besides SFO/Millbrae/San Bruno is there any other use for a BART shuttle locally? Berkeley to Oakland airport maybe?

Rafael said...

@ james -

you probably mean Oakland Coliseum, there are no BART tracks all the way to the airport. BART has plans for an unmanned people mover to replace the current bus service, but the project remains unfunded.

As for Berkeley to Coliseum or any other short route, you need to consider that BART is essentially a subway/R-Bahn system with extremely long lines. There are no bypasses for express trains, though BART is now working toward a crossover facility in CC county. There is also currently no way for a train to reverse direction at some intermediate point in its line - BART would have to install wyes, hire extra drivers to minimize delays related to reversing train direction etc.

Millbrae to SFO is a special case because both stations already serve as the endpoints of existing lines.

Alon Levy said...

Bruce, you could use either the NYC or PRR route for NY-Chicago service. I'm just guessing the NYC route will connect to more things (namely, Toronto and Montreal)...