I’m getting a lot of hits on a post I did awhile back arguing that High Speed Rail is pro-sprawl. If a bunch of people are going to link here off of a HAIF discussion on HSR, I might as well jot down all my views in one place. Starting with Houston…
Use the SP station / Barbara Jordan PO site for the terminal
For one, it’s essentially the same site as the current Amtrak service, so there’s continuity there. For another, it was originally built as a train station, so all of the roadway infrastructure necessary to support a train station is already in place. For a third, it’s within walking distance of a substantial portion of the CBD. The commuter rail plans drawn up by Metro and their consultants tend to look to L.A., which has a station some distance from the CBD with a forced subway/LRT transfer. But it’s much more refreshing to ditch the forced transfer and just walk directly to your destination. And the destinations around the SP / PO site aren’t just dense, they’re 24 hours. Far from just office, there’s already a critical mass of clubs, drinking establishments, and parks. The sort of ecosystem which urban design consultants imagine growing up around train stations already exists, so just dropping a train station into it is the definition of synergy.
290 is the exit route
It doesn’t really matter whether you’re building a “Texas Triangle” or a “Texas T-Bone” or whatever shape you desire, 290 is your corridor. IH-10 is way too built out, the Katy ROW long since subsumed by the freeway. 45/Hardy’s north/south alignment adds needless time for Houston-Austin and Houston-San Antonio trips. And the BNSF corridor from Garden Oaks out to Tomball just isn’t developed enough, not to mention it’s a skinny right-of-way that goes through tons of residential backyards. I’m seeing NIMBY suits.
But the 290/Hempstead corridor isn’t just perfectly-aligned to handle trips to Dallas and San Antonio with minimal delays, it’s also a near-continuous strip of low-rise industrial. That means you’re dealing with CBRE and Transwestern instead of the Smiths and the Joneses. You can get through practically the entire right-of-way acquisition process while speaking only to MBAs. Sure, there might be a suit or two over appraisals, but there’ll be no one trying to derail the project because it destroys the unique nature of their livable community. No one’s heart aches that you might put a bullet train through an industrial park. It works.
Get out of San Antonio on IH-10, but resist the siren song of 130
IH-10 is the natural route out of SA. It’s flat, it’s relatively undeveloped, and it’s a straight shot. But SH 130 isn’t so great. The trouble is, the route between Seguin and Lockhart is a string of almost continuous 1 degree S-curves. That’s nice curvature for an 85mph cruise in a Suburban, but it’s massively inadequate for trains at 220 or even 150mph. There is little getting around the fact that HSR will need a brand new alignment roughly parallel to 130.
The good news is that HSR is skinny by Texan right-of-way standards. On flat ground, you really only need 100′ – same as a standard Houston arterial. And where rails cross active farm or ranchland, you can build overpasses that are little more than culverts. A couple MSE walls and a few beams to span 12 or 16 feet is all you need. Simple.
Build a loop through the Metroplex
The only way to deal with Dallas-Fort Worth is two lines, one running directly into Dallas, one directly into Fort Worth. This wins from a travel time perspective, but gets tricky from a scheduling perspective, since you’re dividing trains into multiple termini. The trick, then, is to “close the loop” with a direct HSR link from Dallas to Fort Worth. Thus any train going to one city is going to the other, and vice versa, it just might take a bit longer. I don’t think the routing matters all that much. You can run it along the TRE for ease of acquisition, you could send it through Arlington or Grapevine, or you could go deep into multi-billion-megaproject territory and tunnel under DFW with a station at the terminal, a la CDG or Narita. As long as the trains can make the loop.
Avoid Downtown Austin
This will be seen as sacrilege by some, but think about it. There’s no conceivable way to run through trackage through Austin. You can’t elevate over 35 because 35 is already elevated over 35. You can’t run it up MoPac because MoPac is ROW-constrained on all sides by well-heeled neighborhoods. And similar neighborhood impacts occur with any other conceivable downtown routing.
But Austin can’t be a stub terminal because it lies between all of the other major cities. You need Austin to be a through-stop for trains from SA to FTW or SA to HOU. Hence the alternatives analysis for Austin should be a couple of routes stopping at ABIA and a couple stations along SH 130 or further out. For my money I’d put the station right at ABIA and then have a driverless metro (a la Skytrain) for frequent service between planes, trains, and 7th Street. The reason you do a driverless Metro and not conventional LRT is because conventional LRT always ends up with poor late-night headways, and Austin is a late-night town. Driverless avoids this, allowing 6-8 minute headways even in the wee hours.
Tying it all together
Once you’ve decided how the trains are going to run in the cities, the next step is just drawing out some 20-minute curves through the cows. For me, I am agnostic here. There remain substantial questions such as how do you serve the substantial ridership base at Temple/Fort Hood, do you run tracks through Waco or College Station, etc. But if you make the right choices about how to run it through the major cities, and if you set clear standards for how the system should run through the rest of the countryside (e.g., minimum curve radii and the like), then any system that results is a good one.
And the technology?
Shinkansen. And here’s why. Most of the off-the-shelf HSR trainsets out there are built to Euro spec, which means they’re in the range of 9’6″ to 9’8″ wide. This is narrower than the traditional US spec of 10’2.” Siemens has a variant of the ICE it sells which roughly parallels the US spec, and is cylindrical in shape, similar to an Amfleet car. But the 700 series Shinkansen is a slab, square-profiled, 11’1″ in width. Which is to say, it’s the biggest.
The 700 is so big, in fact, that in standard-spec it seats five across. Now of course, this will not do. Four-across seating is the order of the day. In Japan, this is the Green Car. Add some legroom and make it the Lone Star Car. In fact Judge Eckels signed up with a company that’s trying to do just that. And while it’s not exactly good policy to preselect a vendor, there are enough JR Group companies that they can each affiliate with a different set of firms and submit a different bid. LSHSR is hitched up with JR Tokai to sell the 700? Some other group can hook up with JR East and sell the E5 Hayabusa. Just write a spec for width or cross-sectional area that excludes all known ICE variants, and voila. Besides, Siemens already has a near-monopoly on US Light Rail Vehicle production. Let someone else sell the high-speed ones.