Friday, August 1, 2014

Is Good Urban Form Slowing Us Down?

There has been a lot of chatter recently on the issue of fast vs slow transit.  This week is the perfect time for this discussion as two major United States transit projects of differing stripes opened up; the Metro Silver Line in Washington DC and the Tucson Streetcar.

Last week, Yonah Freemark wrote a post discussing the benefits of fast transit specifically calling out the Green Line in Minneapolis for running 11 miles in about an hour.  Now, this line has parts of what people are always asking streetcars to have; dedicated lanes. "They get stuck!"  Yet this line, as well as the T-Third in San Francisco and others mentioned in the post are still "too slow".  Yonah goes on to discuss metro expansion in Paris leaving a discussion of politics and costs of rapid transit to the very end.

To me this points to the first place where urbanism and fast transit disagree with each other, block sizes and stop spacing.  By trying to maximize connections to the community, the transit line has to stop more often, slowing speeds.  And if built into a legacy urban fabric, this also includes negotiation with tons of cross streets where designers don't give priority to the transit line.  This happens in Cleveland on the Health Line BRT as well as the Orange Line in Los Angeles, even though it has its own very separated right of way.  The Gold Line Light Rail in LA and the Orange Line originally had the same distance, yet one was 15 minutes faster end to end. A lot of this had to do with less priority on cross streets given to the Orange Line, not because it was a bus or rail line.

We continue to talk as if dedicated lanes are magic, but its a suite of tools that helps speed transit along inside of our wonderful urban fabrics.  Transit is directly affected by urbanism, if we let it be.

But then there is the other side of this discussion.  Transit's effect on urbanism.  Some New Urbanists believe that slow transit is necessary for building better urbanism.  Rob Steuteville of New Urban News calls this "Place Mobility".  The theory goes like this:
When a streetcar -- or other catalyst -- creates a compact, dynamic place, other kinds of mobility become possible. The densest concentrations of bike-share and car-share stations in Portland are located in the area served by the streetcar. That's no coincidence. You can literally get anywhere without a car.
In Portland parlance, this is the "Trip Not Taken".  When you build up the urban fabric of a city, many usually induced trips disappear.  That car trip to the grocery store becomes a walk and that streetcar trip to Powell's Books might be a bike trip now.  Or in the world of the web, that trip might change hands, from you to the delivery truck.  In Portland at the time they calculated a 31 million mile reduction in VMT from the housing units built along the streetcar route.

To increase the viability of streetcars in a world dominated by a "cost effectiveness" measure dependent on calculations of speed, the "Trip Not Taken" was refreshing.  Many transit lines were being built without regard to neighborhood or were cheap and easy.  But they were fast!  You can see how the "cost effectiveness" measure intervened with elevated rail through Tyson's Corner (yes I'm still annoyed) or the numerous commuter rail lines on freight rights of way in smaller regions that probably should never have been built.  But they were fast!

Yes the streetcar helps with creating place in the minds of developers and urban enthusiasts, but no it doesn't do the whole job.  The Pearl District and Seattle's South Lake Union were perfect storms of huge singular property ownership, massive investments in additional infrastructure, proximity to a major employment center, lack of NIMBYs, and a strong real estate market.  But look at the results.  It's hard to argue that the streetcar didn't help develop this massively successful district in one of planning's favorite cities.  But it's also hard to give it all the credit.

The crux of the argument is that place making should be the ultimate goal and slowing things down makes things better.  And many cities see the streetcar as some sort of fertilizer that makes it grow and a reason to change zoning code. Because of very stringent local land use opposition (read NIMBY), this makes a lot of sense.  If a streetcar can lead to the restructuring of land use or the fulcrum of a district revitalization, I see that as a benefit. But again, don't give it too much credit.   

From a safety standpoint this slowing down idea makes sense.  The Portland Streetcar has been in collisions, but no one has died or been seriously hurt, unlike a number of high profile collisions in places like Houston, where drivers can't seem to follow the rules. Our society also puts up with over 30,000 deaths a year to get places faster on interstate highways as well.


Ultimately the base success of a transit line isn't in the amount of development it has spurred or the zoning it has changed.  It's the ability to get a lot of people where they want to go, in a timely fashion.  A commenter on Jarret Walker's Human Transit Blog says it best.
But the romantic impulse towards slow transit wears away quickly if you have no choice but to rely on it all the time! I don't have a car, so I rely on buses that travel excruciatingly slowly, wasting much of my time.
As someone who has gotten rid of my car and considers myself a walking, bike riding, transit loving (and sometimes zipcaring) urbanist, I find it very annoying that it takes an hour to go three miles here in San Francisco on the bus.  And if I need to get downtown, I take the Subway which is a half mile away versus the streetcar which is half a block away because time does actually matter.  We see this decision play out every day when people choose to drive cars over using transit.

But if we are going to spend so much money, we might as well figure out a way to transport the most people possible. Sometimes that might be streetcars.  Other times it's not.

But back to urbanism and transit.

In Portland, dedicated lanes on the North/South parts of the line wouldn't make as much difference because it has the same issues we mentioned with the Green Line above and narrow streets.  Streetcars have to deal with urbanism.  I think streetcars are ok as a circulator in downtowns, because these are the trips that help people get around dense places that are proximate.  You can bring your groceries on when its raining and disabled folks can load their wheelchairs with dignity. Tourists like the certainty of the tracks and little kids love the ride.  We see that even on 20 minute headways, 13,000 riders are on the line every day.  It's hard to argue with that, given it's more riders than many first choice bus lines in some major cities without rail. 

However for linear route based transit operations, we need dedicated lanes and signal priority to at least make the expenditure worthwhile and play nice with our urbanism.  Once you get outside of a district, people want to get places.  I like subways and wish we had more, but it seems politics and money seem to get in the way like Yonah mentions above.  Some might even argue that before we even think about building fixed guideway lines, we should focus on our buses.  Perhaps we should have a threshold system ridership before putting in rail, to determine whether all options for increasing ridership have been exhausted.  Houston's new network plan could be a good guide.  And personally, I don't think BRT should be special. It should be the norm. Luckily the new 5339 bus facilities funding guidance could allow for BRT and Rapid Bus funding (they are NOT the same thing). 

But there's a new report out which discusses which factors drive ridership for fixed guideway transit once we decide to go that route.  Employment and residential density around transit lines, the cost of parking downtown, and grade separation were found to be the most effective measures when put together to drive ridership according to a recent TCRP report released earlier this month. Individually employment had an r squared of .2 while the others had negligible impacts.  Only taken together as a whole did these measures drive the most ridership as seen below.

The report goes on to say "The degree of grade separation is likely influential because it serves as a proxy for service variables such as speed, frequency, and reliability that may lead to greater transit ridership."

But determining success is hard.  In fact, its so hard that of the transit projects surveyed, the only thing that transit agencies seemed to agree on (it has dots in every project below) was that the line would be cheap!  We discussed this briefly above. 
"Provide fixed guideway transit in corridors where inexpensive right of way can be easily accessed"
Which is many times why we end up with slow transit.  It's cheap. We're cheap. Streetcar costs are below that of light rail or subways and since its in a mixed traffic right of way, it will be cheaper politically than BRT.  Commuter rail on freight rights of way is the best to them though even though its the worst at creating ridership.  To me it's is even cheaper because it usually ignores the chart above with the focus on employment and residential density.

So all of this is to say that Streetcars are not the worst transit ever and urbanism will affect transit, and transit will affect urbanism.  We just need to decide what the appropriate ways are for intervention such that we maximize people's ability to get to the places they want to go and build great communities.  Let's not swing the pendulum too far to either side, it might tip the balance against us. 


Beta Magellan said...

I’m glad you wrote this—though I’m still of the opinion that mixed-traffic streetcars are a waste of cash, your Gold-vs.-Orange discussion really cuts to the heart of why a lot of the Green Line discussion bothers me.

That said, I’m still not convinced the Green Line is too slow, either. From what I understand, the areas along University Avenue aren’t that slow—it’s total end-to-end travel time, which includes traversing both downtowns, that does it. Anecdotally, there’s either weak or no signal priority for trains within downtown Minneapolis, and I’d bet that the signal priority along University could be improved, too. People have to remember that priority isn’t just lanes—it’s signals too, and the US is very bad at giving transit the green light.

It does frustrate me that all this discussion is happening independent of thoughts about cost-effectiveness. I’ve been following Minneapolis-area transit discussions for a while and there seems to be a persistent itch for subways. I’m sure a grade-separated would get more ridership than a surface line in most cases in Minneapolis, but I’m skeptical that the incremental increase in ridership would be enough. A subway’s no deal if it doubles the ridership at four times the cost. Minneapolis isn’t Vancouver, which was lucky enough to have convenient grade-separated ROW (the initial line didn’t even require a new downtown tunnel) and where geography conspires to make subways and new elevated lines cost effective relative to other alternatives. It’s more like Calgary, which has flat, sprawling geography and has built a lot of transit mainly through keeping costs down, which in practice means avoiding grade separation when possible (of course Calgary’s success is also largely the result of deciding not to build urban expressway and highly centralized employment, but I still think many mid-sized American cities would stand to benefit from their example).

Beta Magellan said...

Forgot to mention—cost-effectiveness is where my skepticism of mixed-traffic streetcars comes from. Here it is useful to look to Vancouver, which has built strong bus ridership in part by using smaller, less-expensive-to-operate buses to boost frequency. I think that’s a useful approach to dense urban areas. One of the issues with new streetcar lines is that they are fairly low-frequency for short urban circulators (fifteen to twenty minutes)—it’s not just speed, but wait time that handicaps the new American streetcars against walking, much less biking or driving. You don’t have to build as much infrastructure, and while operating cost-per-passenger is lower for streetcars, cost-per-vehicle is higher, which is a liability if you’re trying to build frequency. The fact that so some of these new streetcar lines are operated by someone other than the transit agency opens up the possibility for new driver contracts for smaller buses, which is an impediment to operating smaller buses at many established agencies (in Vancouver bus wages are proportional to vehicle size).

Again, I don’t see a lot about this (using Minneapolis as an example again, I see suggestions for streetcars and subways down Nicollet and Hennepin a lot more often than additional bus service). I think it’s a similar issue to signaling—it’s less visible so less on the radar of advocates, serious or casual (and online transit fans tend to be infrastructure-fetishists of one type or another, which makes the online discussion further skewed). On the other hand there’s just weak knowledge of transit from other countries, even Canada.

Alon said...

A couple things:

1. Portland also upzoned right around the streetcar route, so it's not really correct to say the streetcar promoted development. It's more correct to say that the city legalized development concurrently with the streetcar.

2. Vancouver's community shuttles are a small fraction of its transit ridership. The main workhorses are SkyTrain and the trunk bus routes, which run on something like a grid in the city proper.

3. The examples Yonah is giving for fast transit are not situated in some horribly modernist anti-urban setting. On the contrary! Paris is the sort of city urbanists of all factions claim as their own. And finding the Metro too slow to be useful for suburban expansion, it built the RER, which averages about 45 km/h, and is now building the even faster Grand Paris Express.

Jonathan P said...

I think what most "fast transit" advocates want is for "slow transit" backers to simply acknowledge that--far from being the end-all-be-all--speed matters, and to stop dismissing claims that they don't. I can't think of a single "fast transit" advocate who has argued that we just give up on surface transit and only build subways or els everywhere, but I have heard "slow transit" advocates argue the converse regarding streetcars. There's plenty of room between projects like the Silver Line and Portland Streetcar. The problem is that for many of the CNU/architect/slow transit set, if they acknowledge the chink in the armor of their ideology--that speed does matter--it undercuts the foundation of many of their pet projects. I think that's why they try to avoid admitting this.

Let's get back to focusing on solving problems instead of copying and pasting solutions. If we're going to build streetcars (or BRT for that matter), then let's get as much out of them as we can: maximum dedicated lanes, signal priority, off-board payment, etc. And if we see streetcar projects that don't offer mobility benefits, let's call them out in the same way we call out BRT creep. Let's talk about the intrinsic differences in rail vs BRT when we evaluate projects rather than citing subjective arguments like 'it's about creating an experience.' Let's ask the community to weigh the trade-offs among economic development, travel time, impacts, etc.

As for cost effectiveness, there's also a lot of room between T-SUB and the current policy. I prefer Jarrett Walker's calls for measuring travel time savings in percentage terms rather than absolute ( That way, we don't encourage long commuter oriented projects at the expense of shorter distance urban projects.

My broader point is I think fast transit advocates are conceding a lot of room in the middle for more effective, place-based, urban transit projects that the slow transit advocates are unwilling to admit.

Nathanael said...

The main problems with the Green Line in MSP relate to lack of signal priority.

It is remarkably difficult to get city street departments to turn on signal priority; Toronto has been fighting with this on its streetcar system for literally decades. It seems best to demand crossing gates upfront.

Unknown said...

At the very least, it's ridiculous that light rail lines don't have complete signal priority. There might be over 100 people sitting on that train, waiting for three cars to take a left in front of them. That's absurd, the feds should not pay for that ever. I imagine it's a function of LOS planning, which thankfully we're starting to rid ourselves of here in California.

Anonymous said...

Unfortunately, North America is still auto-oriented. It is easier, and within a shorter time-frame, to build freeways than it is to build any rail based public transit. Also, automobiles (with 1.3 people inside) making a left turn are still given priority over any public transit vehicle (with 40+, or even 200+).

Christof Spieler said...

I think signal priority is key in the Minneapolis discussion. Houston Main Street Line averages 15 mph with signal priority and a station spacing pretty similar to Minneapolis. There is a big different in trip time (and what a transit rider can access in a reasonable time) between 8mph and 15 mph. Signal priority costs virtually nothing and has no negative effects on walkable urbanism. I'm at the APTA sustainability conference and I was talking to a speaker from the Munich transit agency. She was flabbergasted that we don't do signal priority for light rail. They're doing it not just for streetcars but for local bus.