Thursday, April 14, 2011

No More Commuter Rail Starts

If there is one thing we've learned over the last few decades, suburban political forces are a drain on cities. For everyone wants to be connected into the downtown and its vibrancy but at arms reach. So the wall was put up many years ago, you must have a car to get there. Now many are wondering if there is an easier way to get back downtown, wherever downtown is. And then they say, well it's too congested to drive, how can we get downtown to pay and appease the folks that want transit but aren't quite sure what it is they need. Then the answer comes, commuter rail.

It's a perfectly acceptable form of transit and has its place in the hierarchy, but for some reason regions get stuck on building rail and they look at what it will cost to do the first part right and they balk. How can we appease our overlords in the suburbs so they will give us something we in the city want in the future? Why do I say overlords? Because Metropolitan Planning Organizations and transportation providers as well as the congress is stacked with people who want to suck money out of cities and into their suburban and rural districts. The safe bet is to appease them right? Wrong.

What we've seen over the last ten years is the monumental failure of commuter rail to do any regional work of value as a first line. The millions of dollars for a couple thousand riders at best is disheartening to those of us who want to see regional transit systems, not just a one and done. I've started to think about this with more clarity as the research comes in and I believe that the places who really are in it to win it will build destination based regional transit that connects a major employment corridor in the region. The headways need to be frequent and the line must run up the gut, not on the perimeter.

Houston's LRT Line

Here is the political reality facing regions today that don't have transit, especially in conservative or timid parts of the country. There seems to be this weird wishfulness that somehow commuter rail taking 2,000 people a day is exactly the way to cure congestion or spruce up economic development. However its basically a ticket to political backlash. Sure the line might have met ridership expectations but who cares? That's only 2,000 voters a day. How much induced voting demand did you create by freeing up room for 2,000 others cars on that freeway carrying 100,000 a day? Zero. It just means 2,000 more freeway voters can move into that district.

Here's what you can do. Put a light rail line down a major arterial between major destinations and all those haters that work downtown have to see the train pass them full at rush hour every day. When I was little my dad liked to play a joke on me that there were no boxcars in Bakersfield California. He still to this day will not acknowledge their existence because he knows it gets me really worked up. But the reason it got me really worked up is because I saw them in the yard downtown next to the high school ever single day. I saw them every day we would go pick up my sisters at school. If you saw a light rail train full of people at rush hour every day wouldn't you start to believe too? Once entrenched as something that works, no one pushes back, rather they want it in their part of town too. That is how systems get built.

But let's look again at why commuter rail is not the start.

1. It's got low ridership. These lines don't have that many people on them period. So people don't see the effects and they don't want more because they don't feel like they are missing anything. Lines like the Music City Star, Capital Metrorail, and Northstar are all carrying minuscule amounts of voters.

2. It's got low ridership because the schedules are bad and the schedules are bad because you're second fiddle to freight lines. If you're not giving commuters priority, why should they give you priority?

3. It was too easy. If a region builds a line because it was cheap to do, don't you think people are going to see through that and understand that you're not really putting a full effort in? I know I do. Indianapolis wants to build a cheap line because its politically feasible now. What about in 5 years. The harder the fight and the more work you put in, the more likely you'll be in good shape down the road. In running, you get out of your training what you put into it. I think the same applies here.

4. You're enabling the enemy. Same as the last point, but if you're not putting voters and supporters on the trains, you don't have a constituency for extensions or stopping service cuts.

Look at these lines according to the Q4 ridership numbers, you can quibble with these a little bit as the agencies have different numbers in the news recently but 500 +/- riders isn't going to make a huge difference.

Recently Opened Commuter Lines

1. Northstar Twin Cities - 2,000
2. Capital Metrorail Austin - 800
3. Rail Runner New Mexico - 3,800
4. Music City Star Nashville - 800
5. Frontrunner Salt Lake - 5,400
6. Portland WES - 1,400
7. Oceanside CA - 4,100

Some of these places like Portland and Salt Lake City already have regional light rail systems so a Commuter Line connecting in isn't as bad of a decision for later when you have the internal network.

Single Destination Connecting Lines Opened in Last 10 years. Again the ridership differs due to gas prices but these are in the rough area of current reality

Houston - 34,600
Phoenix - 40,300
Minneapolis - 30,000
Charlotte - 14,000
Seattle - 24,700

Now the difference between people packed into trains running downtown as well as the number of carried voters is huge. It doesn't take a rocket scientist to tell you which ones are going to be more palatable for expansion. So instead of looking at the "cheapest" alternative, let's find the two major destinations in a region that need more capacity and need to be connected. This is what we should be thinking of when we're starting a system. No more commuter lines as regional rail starters.


Sam S. said...

I think that Sounder Commuter Rail in Seattle, Dallas, and Metrolink in L.A. have been rather successful for newer systems.

Allie Cat said...

To be sure, the SPRINTER rail line in Oceanside (which I'm assuming you're referring to, since COASTER commuter rail ridership is over 1.7m for the year) is not traditional commuter rail. It is somewhat of a cross between light rail and commuter rail, with short stop spacing, relatively small trains, and all-day frequent service (with "frequent" here interpreted in the SoCal way of "better than half-hourly").

MB94128 said...

The ridership numbers do stink on the lines you've listed. But you need to distinguish between the lines that are ab initio rail service and the ones that tie into an existing rail system (e.g. Northstar). The latter can serve as cracks in the barriers against rail expansion and lead to regional networks that are very valuable.

I suggest you add an update that describes the growth of the Capitol and ACE (both SFBay area) services. Also, one nasty factor in some of these lines is the near extortion of access charges by the freight railroad that is the dominant user.

jon said...

This is so relevant to WES. It should have been LRT but they cheaped out and now have really poor/low ridership, a target for TriMet critics, infrequent service, a required transfer to LRT to go downtown, high operating costs, poorly located stations, etc. A LRT line in this corridor would have been a lot more expensive to build but would have had many many more riders, lower operating costs and would been much more useful especially with flyovers/deviations from the freight rail line to hit nearby major destinations like Washington Square Mall, Kruse Way office park edge city and the Bridgeport Village shopping center.

Although I definitely wouldn't write off NM Rail Runner, and to some degree, SLC Frontrunner.

M1EK said...

Note that the shared lane streetcar systems that are squeezing light rail from the OTHER direction have ridership no better than these awful commuter lines.

Pantograph Trolleypole said...

Give me ridership on modern streetcar lines M1ke? Portland's gets 11,000 per weekday with a three mile route. It's also always been full of people when I'm on it. Outside of Seattle which gets about 3,000 on a single mile and will get more when its extended, where's your evidence that modern streetcar ridership is horrendous? Since these are the only two examples in the United States.

Scott said...

How about pre-CEI and post-CEI. Commuter rail can be set up to make CEI look amazing, but does not optimize for ridership.

If we care about ridership, the new starts criteria should go back to ridership as the prime metric for new starts.

Pantograph Trolleypole said...

Maybe pure ridership but not the BS they used before where you had the travel time savings over cost to make cost effectiveness. All that gives you is a line with no stops. Perhaps there should be a cutoff like you have to automatically have 10,000 riders or 15,000 riders before it will even be considered. Just thinking out loud.

Joe SF said...

One factor in new commuter rail lines that contradicts the goals of most serious rail advocates is that station area planning seems non-existent. The Northstar line stations, for example, seem to be merely parking lots on the edge of low intensity industrial zones far from any existing town center or residential areas. Ideally, new commuter rail lines would include agressive station area planning stressing relatively high density walkable districts with a mix of residential, commercial and business. But this would probably nix the project from the point of view of freight railroads which would not want complaints from new residents about noise and blocked streets.

Curt said...

How would 10,000 riders be qualified ahead of time? That sounds like just another regulatory hurdle that can be debated depending on your political viewpoints. Don't get me wrong, I agree with your intent. If there were a rock solid way of determining this say.... number of potential jobs served. Or... number of potential shoppers or tourists served it may be a better way of quantifying the value of a frequent stop & service route vs a low frequency & stop route.

Pantograph Trolleypole said...

Well then lets use 100,000 jobs as the metric. Zupan's book states that you need a certain amount of square feet to serve different transit modes.

Alon Levy said...

Ottawa's train line has 14,000 riders, and cost $21 million to construct. The issue isn't low ridership; it's outsized cost relative to ridership.

The Sprinter cost more per km than some greenfield high-speed rail lines. Faster lines of similar class have come in at an order of magnitude less cost in Germany (link).

Anonymous said...

I'd say the FrontRunner and RailRunner are still worthwhile projects, and they do have higher ridership. The key is to build a line that actually goes somewhere, specifically from the primary city to a secondary center, rather than just ending in a suburban field. Both the above systems have that, and that's where their somewhat higher ridership is coming from.

John Perry said...

To paraphrase Anonymous above, I'm surprised to see the Rail Runner in this discussion (I'm from Albuquerque, so I grant I may be biased in this matter), namely because many of the things you recommend were things the Rail Runner took into account - it serves a mainline route through the middle of the valley which several communities and urban centers built around, so it connects some major destinations (most notably Downtown Albuquerque and the central government/tourist district of Santa Fe). Yes, it was an easy route to choose, but it also made perfect sense given the regional context.

It does utilize a mainline track that is shared with Amtrak and freight, but the freight traffic is so low along the route that it doesn't disrupt the Rail Runner service much (and the state purchased the tracks, just to make sure the Rail Runner would have priority). Plus the route is almost 100 miles long, which is a distance that just doesn't make sense for light rail.

Now I'm not going to say the Rail Runner is perfect - it does have low frequency and that certainly contributes to its relatively low ridership. But there's also the matter of the weak public transit systems in the cities it serves (indeed, the construction of the Rail Runner provided the motivation to actually start public transit service in some of the smaller communities along the way). The Rail Runner has been trying to address this by expanding the park and ride lots along the way, but they can't keep up with the parking demand.

Yes, it has problems, but the Rail Runner made a lot of sense given the regional context and it's done so much for public transit in Central New Mexico. We've got a long way to go, but that train finally got the ball rolling. I can't speak to the other systems you named (Nashville, Austin, etc.) but I think given the circumstances, the Rail Runner was the right choice.

Loren said...

I think that a possible counterexample here would be the Seattle area. Its commuter-rail line, the Sounder, was built before other recent urban-rail systems there, and it now has 10,000 riders/day. Chronology:

2000: Sounder Seattle - Tacoma
2003: Sounder Seattle - Everett, Tacoma Link streetcar
2007: Seattle's South Lake Union Streetcar
2009: Seattle's Central Link light rail

Dallas's commuter-rail line was opened half a year after its first light-rail line. Los Angeles's Metrolink was opened 2 years after its Blue Line and a year before its Red Line.

Pantograph Trolleypole said...

I'm not saying some of those projects aren't worthwhile at all. It's just harder from a political standpoint if these types of lines go first. If we think about the Hiawatha and Northstar lines in Minneapolis, do you think that the Central Corridor would be going through if Northstar was the only game in town right now? It would make things much harder.

An Alon I see what you're saying about lines that are much less expensive, but if regions in the United States have a choice, and they can go with a line that is frequent and connects major destinations, a city should go that way first to guarantee the argument isn't going to be about whether expansion should take place, but rather where.

We also need a discussion about cost because they are totally out of control in this country. LRT should not cost 85 million per mile.

neroden@gmail said...

I strongly agree with the thesis of this post -- that urban rail comes first, suburban rail afterwards. It's quite noticeable that the moderately successful "commuter rail" lines (Minneapolis, Salt Lake, Dallas, LA) are linking to massively successful urban rail which came first.

New Mexico is an exception but it's an unusual situation. Albuquerque to Santa Fe is the obvious first rail route in the region, the one which connects major destinations; there is no other route which would be more popular. Neither of the cities at either end is really that large -- and so neither really has enough people to make a local urban rail route more attractive. Furthermore, it's designed with an all-day schedule, and in fact it gets very substantial proportion of its traffic on the weekend.

Seattle is the only city which acts as a counterexample to your thesis. Sounder did start before Link (the urban rail), it *is* on a commuter-only schedule, and it has been quite successful. I wonder what the special characteristics of Sounder are? Perhaps there's simply more intercity demand between Seattle and Tacoma than one might imagine. Being "anchored on both ends" definitely helps, as it does with Albuquerque-Santa Fe or Dallas-Fort Worth, as opposed to Minneapolis-Big Lake (where?).

Cameron Slick said...

Northstar is the Minneapolis - St. Cloud corridor that got totally screwed up in the mid-2000's due to Gov. Pawlenty wanting to fund NOTHING, the northwest suburbs being apprehensive, and worst of all, BNSF running single track from Big Lake to St. Cloud. Had this been a true intercity line, with a few extra trips, it would be running at 6,000-8,000 trips daily, much better given the investment.

Unknown said...

It's funny to see how much higher the ridership numbers are for other Commuter Rail systems. From what I've heard from the Utah Transit Authority, they are actually a little sheepish about how poorly it has done compared to expectations. While the effects of the Legacy Parkway (a parallel freeway) were modeled for ridership, the housing crash and recession really put a dent housing development in Weber County, which had been growing at an explosive rate, and had been expected to fuel a lot of future ridership.

Mark said...

Previous commuter rail systems were often built as complete systems. While the incremental approach may work for light rail and streetcars, I think commuter rail needs to work as a full system to meet expectations.