Friday, July 04, 2008

Doing commuter rail right by 2012

OK, after attending a couple of public meetings, reading the H-GAC report summary, and having an in-depth meeting with Christof (and thinking about his five questions), I think we've come to a consensus on what the right answer is for a "phase 1" commuter rail program (here's my first post on the topic). Those who remember my "Commuter rail is the wrong ride" op-ed from a few years ago - which Barry handed out at the H-GAC public meeting - might wonder why I've changed my mind. I will lay out the reasoning, as well as the narrow set of circumstances where I think it makes sense. I'm not now a blanket supporter of commuter rail. HOV/HOT express buses that travel nonstop at high speed and circulate at their destinations are still the best overall commuter transit solution for the many decentralized job centers of Houston.

The reasons commuter rail can make some sense now for Houston:
  1. I agree with County Judge Ed Emmett that our metro growth is so fast now, and projected to be so large over the next few decades, that we will really need to tap all sources of mobility capacity that we can.
  2. Freight track corridors have been identified that are underutilized, and can support scheduled commuter services. The capacity is there, and we can tap it at marginal cost.
  3. For the most part, the new services will not overlap or eliminate Metro express HOV commuter buses. The goal is to add all-new service and capacity, not replace what we already have.
  4. Commuter rail is more comfortable for very long haul, connecting to places well outside the Beltway. In particular, passenger service stretching from TAMU College Station to Prairie View A&M to NASA JSC Clear Lake to UTMB Galveston is attractive for economic development purposes by tying more academic and technological brainpower to our city. (similar to how Yale and Princeton are tied into NYC)
  5. $4 gas (and increasing) is driving up demand for commuter transit, and Metro cannot buy buses or expand P&Rs fast enough. In fact, some of these rail services may free up more express buses for other in-demand routes.
The trick is doing something helpful, quickly and affordably. The hardest, most expensive, and most controversial part is inside the 610 loop. There are calls for expensive elevated structures and new routes through neighborhoods that will trigger a firestorm of protest. If I heard Judge Emmett right, he wants to do a "quick and dirty" implementation that stops at the loop and transfers to buses. That could be a mess. Cheap, yes, but offering bad service that attracts few riders and undermines public support. We believe he is partially right focusing on Galveston and 290 lines, but there are tweaks that can make for a much better start.

Here are the elements of our proposal:
  • Galveston line that comes all the way in to just east of downtown, with transfers to the new East/SE LRT lines to get into downtown. Christof has worked out how this is possible without using busy freight lines or building elevated structures (which he will detail in his own post soon). The line will have few if any stops inside the Beltway and will not replace Metro HOV services inside the Beltway.
  • Since a 290 line cannot (currently) make it to downtown, would not actually reach the Uptown LRT (it stops at the NW transit center), and is already well served with HOV bus service, it should not be part of a phase 1. Instead, some freight track improvements need to be made first inside the loop (along the Terminal subdivision - and they're needed regardless of commuter rail), which Christof will articulate in detail. Those will pave the way for a future 290 service that does not use the empty RoW through the residential Heights.
  • Instead, the real short-term opportunity is a 249 line to Tomball through the vast 1960-area suburbs, as well as picking up the back side of The Woodlands. This would attract many riders who do not currently have good access to HOV express bus service, and it would take traffic pressure off of both 290 and 45N. In fact, 249 ranked higher than 290 in H-GAC's study (see figure ES-7) This starter line would be used as is, and new track would not be rammed through the neighborhood north of the 290-610 interchange (as shown in the H-GAC plan). Instead riders would ride it all the way in to the North Metro LRT and transfer near Northline mall. A bus transfer stop would be an option down Mangum for those going to Uptown.
  • As part of this phase 1, Metro would experiment with running express LRT trains from the Northline transfer into downtown. The idea is that, after transferring off the 249 commuter train, most of those riders would want an express direct to downtown, with no slowing stops along the way. Also, even if the train stops normally from downtown on, it would make connections from the 249 commuter rail to the med center tolerable (vs. 45+ mins with all local stops). There are different ways these express cars could be run, even without a third track, via simple timing or using crossover tracks for passing (or maybe even via third tracks at selected stations). The expresses might be slightly disruptive to the local trains (making them crossover and/or freeze at certain points for a couple minutes), but it would be worth it for the enhanced commuter service, and would only be used for a handful of inbound trains in the morning and outbound trains in the afternoon rush hours that connect specifically with the infrequent commuter trains. If the expresses can be made to work well on the North line, they could be considered for other LRT lines in the future as needed for other new commuter rail connections.
  • For phase 1, no expensive hub, and only minimal maintenance facilities. Each of the two lines might have separate small maint facilities, or they could be routed at off-hours along the freight tracks to a central facility inside the loop (like Eureka yards). The freight tracks inside the loop are too congested to support scheduled commuter rail, but they still have enough capacity to shuffle around out-of-service commuter trains when needed (as long as they are not in any hurry).
  • The phase 1 target is 2012, to coincide with Metro's LRT plan, since it obviously relies on those connections.
Bottom line: we think both lines might be doable for a total capital cost less than $1 billion, vs. the $3 billion in the full H-GAC starter plan. That might sound like a lot, but it's only about twice what the Main St. LRT cost, and it's about 80 miles of track vs. 7.5 for Main St. It's also way cheaper than any new freeway capacity serving the same places.

Long-term potentials include a 290 line that connects on to Prairie View and College Station while continuing through the core on to Galveston (parents: easy-ship your teens to the beach and theme parks during the summer!), a downtown connection at Amtrak/UHD that would be more effective than Metro's northern intermodal terminal, and maybe lines to Alvin/Pearland and Ft. Bend. That's all speculative at this point, and I'm not prepared to pass judgment on their cost effectiveness. But this 2-line, "lite phase 1" proposal seems like it could be a cost effective way to "test the waters" by going after the low-hanging-fruit commuter rail opportunities while taking advantage of the Metro Solutions core rail network to save money and neighborhood disruption.

I will update this post with a link to Christof's excellent new maps when they become available (and here they are). If you know anyone attending Judge Emmett's commuter rail summit this coming week, please pass this along for their consideration. Much appreciated.

UPDATE: Christof's much more detailed explanation of this plan, with great maps.

Labels: , , , ,


At 7:13 PM, July 04, 2008, Anonymous Anonymous said...

Sounds like a good plan. However, I don't remember reading if Park and Ride will be expanding along 290, 45, and 59. Hopefully, those areas transportation options will improve as well.

At 7:42 PM, July 04, 2008, Blogger Tory Gattis said...

Absolutely. Metro will need to be aggressive to keep up with P&R demand.

At 12:05 PM, July 05, 2008, Blogger engineering said...

Thank you Tory for the summary. For over 10 years I been advocating the 290 commuter rail corridor.
A map that triggers my curiosity is one that combines METRO's service are and that of commuter rail. Have you seen this map? If not maybe you and Christof can put one together.
Have not read the commuter rail report but does it suggest what type of entity will operate this commuter rail system? The fun part will be funding for infrastructure and O&M.
Any lessons learned from the DFW commuter rail that Houston can benefit from?

At 12:29 PM, July 05, 2008, Blogger Tory Gattis said...

The map in my post, if you click on it, will take you to a bigger version with both the commuter lines and the LRT lines. Christof is working on a better one, which I will link to when it's up (keep an eye on his Intermodality blog).

There are lots of entity options: Metro (if they expand), HCTRA, TXDoT, GCFRD.

I don't know too much on DFW, other than it's a mistake to try to use one system for both long distance and local trips. You end up with either too few stops, or too many and it's too slow. We're doing the right thing with frequent stops in the core system, but very infrequent stops on the commuter rail to keep up speeds.

At 3:04 PM, July 05, 2008, Anonymous Anonymous said...

I'm not too familiar with the Northwest 1960 area, is that an area already served by Metro (buses)?

At 4:36 PM, July 05, 2008, Blogger Tory Gattis said...

Not really. See here for P&R locations.

They're only on 249 inside the Beltway, where it's not a freeway and there are no HOV lanes. If you look at a Google satellite map, you can see vast residential areas out there, and they're all forced over to 45 or 290 to get into the city.

At 5:06 PM, July 05, 2008, Anonymous Anonymous said...

I guess what I should have asked is if the 249 rail route would also be served by local bus. I was just wondering if in the area outside of Houston, will be basically "park and ride" commuter service or do they have Metro NW Harris county?

At 5:47 PM, July 05, 2008, Blogger Tory Gattis said...

NW Harris County is part of Metro, but has pretty limited local bus service. See overall system map here:

At 5:48 PM, July 05, 2008, Blogger Tory Gattis said...

But I expect most of a 249 commuter rail would be about Park and Ride, with very little connecting local bus service. Although that might change once the line is in place.

At 7:39 PM, July 05, 2008, Anonymous Anonymous said...

That's what I figured, a park and ride commuter rail line. I don't believe there's anything wrong with that since it's suburban and a starter line.

How is this going to be funded? I haven't heard anything thing about whether Metro was paying for it or will the feds be helping.

At 9:30 PM, July 05, 2008, Blogger Tory Gattis said...

That's the big question. Metro's service area only includes western Harris County. And they don't really have the money anyway, at least in the short term (all going to LRT). All the surrounding counties have maxed out their sales tax under state law. They will certainly go after federal money, but there will have to be a big chunk of local money too, and it's very unclear right now how that would happen. It might involve a sales tax increase by the surrounding counties, authorized by new legislation in the legislature next session (early 2009).

At 10:10 PM, July 05, 2008, Anonymous Anonymous said...

The Dallas commuter rail line, Trinity Railway Express, (not to be confused with the DARt light rail system) is doing pretty well for a commuter rail line, with 9,000 average weekday boardings. The lessons are pretty much those I've pointed to for most successful commuter rail lines ( activity center on both ends (Downtown Dallas and Downtown Fort Worth), connections to local rail and local bus, off-peak as well as rush hour service, and service that is faster and more convenient than other transit in the same corridor.

At 1:25 PM, July 06, 2008, Anonymous Anonymous said...

Aren't you forgetting the fact that 290 is about to be torn up.

At 2:42 PM, July 06, 2008, Blogger engineering said...

Tory, the map in mind is one that includes areas with the one cent sales tax "Communities that are part of the METRO area include the cities of Houston, Bellaire, Bunker Hill Village, El Lago, Hedwig Village, Hilshire Village, Humble, Hunters Creek, Katy, Missouri City, Piney Point, Southside Place, Spring Valley, Taylor Lake Village and West University Place. Major portions of unincorporated Harris County are also included." found at

The Trinity Express used existing tracks and used equipment vs. new tracks and new equipment, meaning it might be low cost. As Christof indicates it has few stops and links downtown Fort Worth and Dallas and some of the mid cities plus current bus connection to the DFW airport. Can take the train in Fort Worth, transfer to light rail at Reunion Tower in Dallas and go all the way to Plano, etc.

Also, if not mistaken, there is current plans to link DFW airport to downtown Fort Worth and further to the southwest with commuter rail - this currently under implementation.

By the way, some existing commuter rail service has two types of service using the same tracks, express services that has few stops vs. local service with more stops. I think some of the NY subways do the same, i.e. reduce stops to increase speed.

At 3:31 PM, July 06, 2008, Anonymous Anonymous said...

To be clear, TRE used an existing rail line, but the tracks were extensively rebuilt, and a new connection was built at the Fort Worth end. The whole thing still cost $300 million, not counting the cost of the original rail line, which had been bought by Dallas and Fort Worth a decade earlier. To out that into perspective, that's roughly the same cost as as the Main Street light rail line for 1/5th the ridership. But it's cheap compared to urban freeway expansion.

At 4:10 PM, July 06, 2008, Blogger Tory Gattis said...

Christof has a new map up at the end of this post combining our proposed commuter rail starter plan with the LRT lines and HOV bus service:

Even if the 290 HOT lanes are built, space is being reserved for the freight/commuter tracks.

Most lines that offer express vs. local have 3 tracks to create passing opportunities. Metro LRT does not have that, so some creativity will have be used.

At 6:52 PM, July 06, 2008, Anonymous Anonymous said...

Tory, Sorry my post might not have been clear. My point was one of the reasons to get commuter rail in faster on 290 is to have it availabe prior to the havoc 290 reconstruction will cause.

At 12:13 AM, July 07, 2008, Anonymous Anonymous said...

Rail ROCKS! I am in love with Houston all over again.

At 8:42 AM, July 07, 2008, Anonymous Anonymous said...

249, YES... that is the right answer. For the first time in years, I am agreeing 100% with Tory!

Make sure this line connects with major employers along the 249 corridor, or those employers run shuttle service and the line can be a true reverse commuters dream come true! HP, are you reading this?

At 2:39 PM, July 07, 2008, Anonymous Anonymous said...

Good plan, but the it does not do enough to help curb green house emmissions.

At 4:26 PM, July 08, 2008, Anonymous Anonymous said...


Since you've blogged about cost of gas and driving versus household inocome, I though you might like this link.

At 5:28 PM, July 08, 2008, Blogger Tory Gattis said...

That is cool. Thanks for the heads up. It illustrates why fuel increasing a few more % of income does lead to some behavior changes on the margins, but is not going to create a wholesale change in society (like the much overhyped "death of the suburbs").

At 1:39 PM, July 11, 2008, Blogger Dane McKitrick said...

With respect to the Gulf Freeway rail service (Actually it is along Highway 3), there is one very significant issue.

There are three main ingress/egress routes from the pretty heavily populated area once known as Clear Lake City. This is the area north of the lake, east of Highway 3, south of Ellington Field and west of Armand Bayou.

While there are other routes out of the area (Space Center Blvd to the Beltway and a couple of routes to Highway 146), the predominant traffic flow is toward the Gulf Freeway. The routes available are NASA Parkway, Bay Area Blvd. and Clear Lake City Blvd. There are a few lesser routes like Medical Center Blvd. and El Dorado but these both have their limitations.

The thing all these roads share is an at grade crossing of the railroad tracks. When a relatively short and quick freight train happens through this area at rush hour, it creates a traffic back up that extends as much as two miles in both directions and takes fifteen to twenty minutes to clear.

I am sure that a commuter train will not close an intersection for as long a period of time, but I am also sure it will upset the timing, such as it is, of the traffic signals. Running a commuter train through that area every thirty minutes or so will guarantee a constant traffic snarl that will make it nearly impossible to get into and out of Clear Lake

If there is any thought about using the rail as a means of evacuation, take of the foregoing situation and increase it by about an order of magnitude. Clear Lake would become un-evacuable.

If this plan has any hope of actually improving conditions, there would have to be grade separation at the major ingress/egress points. I do not believe that has been taken into consideration in the cost of this plan. Currently NASA Parkway by-pass is under construction and that would become the only reliable way out of the area.

There are other areas that would suffer nearly as badly as Clear Lake. League City, for example, also has few egress routes and a population growing exponentially.

I am not as familiar with the 290 route, but assume that it uses the existing track along Hempstead Highway. This too has a large number of grade crossings that stay very busy throughout the rush hour. I am sure that regular gate closings along this route will wreak havoc on these already busy crossings.

Having said all of that, I agree in principal that this plan has significant merit. It simply requires more evaluation before receiving any endorsement.

At 2:37 PM, July 11, 2008, Blogger Tory Gattis said...

Dane: good analysis and important concerns. I went ahead and forwarded it to the official public comment email address for the commuter rail study. In fact, all my readers and commentors should consider sending in their feedback, ideally before the end of the July:

PublicComments (at)

At 12:24 PM, July 18, 2008, Anonymous Anonymous said...

Why doesn't the plan include commuter rail (instead of HOV/HOT lanes) up Hwy 59 North?

Has there ever been any talk of underground mass-transit, like the subways in other large cities, or even the "tube" in London?

At 12:34 PM, July 18, 2008, Anonymous Anonymous said...

Another question...

I understand the benefit of less cars on the highways from having expanded and more options in mass transit...but will there be a benefit for the commuter in less time spent commuting, when using the rail system?

For it going to take me any less time taking the commuter rail from out around Jersey Village to UHD, then it does driving?

At 5:14 PM, July 18, 2008, Blogger Tory Gattis said...

Melinda: They looked at 59N, but it wasn't very promising: not that much population out there, plus a very big relatively new freeway with good HOV lanes.

Underground = 3x the cost, and these lines can already be extremely expensive. It's generally not an option.

In most cases, commuter rail will be slower (the biggest commuter rail cities, like NYC, DC, and Chicago, all have longer average commutes than the rest of the country). It usually nets out about 35mph with stops. Then there are the waits at each end for the transfers, or the walks. Definitely slower than HOV express buses at 60mph. Speed vs. traffic depends on congestion.

The good news is that the line might actually terminate right at UHD, in which case it could be a very attractive option for you personally. Even if it's not the fastest, it's comfortable and you can do other things while riding (read, phone, laptop, etc.)

At 9:06 AM, July 19, 2008, Anonymous Anonymous said...

Thanks for the response, Tory!

At 9:41 PM, July 29, 2008, Anonymous Anonymous said...


How soon could a signficant part of this bi-line plan be started?

Then implemented?

We should go for it now!!

This would also help Emmett's election, because he'de be the first local politician to really do anything positive for traffic in the metro.

And yes I haven't thought about safe tow :)

Keep it up Tory, I'm glad you had a change of mind and heart on this one.

Mike M.

At 8:20 PM, October 03, 2008, Anonymous Anonymous said...

This is a brilliant plan, apart from the fact that 290 is about to get torn up and will no longer exist..


At 2:24 AM, October 22, 2008, Anonymous Anonymous said...

Running a line to Tomball is all well and good, but the "express" LRT simply wouldn't work without additional capital expenditure - namely, constructing "off-line" stations where local trains can pull out to stop while express trains skip past.

Which isn't actually a bad idea, when you think about it, but the funding would still have to come from somewhere.

At 9:45 AM, October 22, 2008, Blogger Tory Gattis said...

Because this segment is so short, Christof thinks it's possible to run express trains without bypasses by timing them property (the local gets enough of a headstart before the express leaves). Alternately, there are various crossover points along the tracks, and a train can switch over to the other side temporarily to allow a pass. Of course, this also requires stopping local trains going the opposite, non-demand direction for a short time.

At 8:46 AM, October 13, 2009, Anonymous Anonymous said...

I can't believe the author ever even questioned the rationale of having a commuter rail. He obviously have never personally experienced any of the Metro Commuter buses for any length of time, or seen the obvious lack of forethought in the design of the current commuter systems. As a person who rode both the light rail and the 290 #214 Metro commuter bus for several years, I have witnessed both routes reach the point of overflow in a very short period of time. The 214 park&ride has been overflowing for years, the only reason more people dont ride the bus is becuase there is no space to park. As a solution, 100s of drivers come through and pick up "slugs" or ride commuter vans, only to be able to drive on the HOV, clogging it to a near stand-still during rush hour. These commuter bus HOV systems are Not "high-speed" as the author says. Also, Metro is losing millions in fares for the sole fact that they cannot accomodate the load of riders. Imagine a commuter rail that only has to stop at the park-and-ride stops and at Downtown, with no traffic crossings in between. All of the current bus riders, and those who "slug" or ride Vans on the HOV would switch immediately knowing they don't have to compete with car traffic. It so obvious it is mind boggling. People want this, they are not as attached to their cars as you think when it comes to daily commute.

At 11:05 AM, October 13, 2009, Blogger Tory Gattis said...

I think you just made a good argument for better control of the HOV lanes (to keep them high-speed), more park-and-ride lots to spread the demand load (lease existing giant private lots that are underutilized on weekdays, like churches and malls), and more buses to accommodate demand. With more buses, they could offer more destinations and more frequency.

I'm not saying people are attached to their cars. I'm saying express bus can be superior service to commuter rail when done right. Faster, cheaper, more frequent, and more directly to your final destination building (as opposed to long transfers and/or walks from train stations).


Post a Comment

<< Home