A month ago, Martin posted this amazing video (watch it fullscreen with the audio off) from STLTransit. Since then (and before, too I guess), STLTransit has posted similar videos for a great number of cities around the world in their youtube feed. The Seattle one is impressive, but taken together STLTransit has really made something. I’ve watched all of them, and I have learned a few things comparing them that I’d like to share with you below the fold.
It is really interesting to see how geography effects an area’s transit system. The Seattle video gives a pretty clear picture of how North-South the majority of routes in Seattle are, which is obvious from the geography of the city. Then there’s Auckland, with everything funneled through a narrow strip of land around the CDB. There everything is North-South out of necessity.
You can also get an idea of how awkward our city’s grid is by comparing to other cities. Compare to Vancouver, Chicago, or San Francisco (itself Hilly with grids intersecting at angles). If you have a look at New York’s video, you can see Manhattan is probably the place most friendly to this approach. More or less flat and long, you get a pretty nice grid despite being relatively old.
When I visited Toronto, I remarked how transit lines run more or less straight down streets without turning. You can really see that in the visualization. One of the consequences of this for Toronto is that even though they have only a few subway lines, they are able to get a lot of impact out of them by funneling bus riders into subway stations. Since Seattle’s not going to be getting a ton of Link lines, this is something that we may be able to do here to increase the impact of our transit systems.
One last thing that comes out clearly in these visualizations how important the CBD of a city is relative to its size. Dallas has a reputation for having a tiny CBD relative to its size, and has a map that fits that. An the other hand, Los Angeles has a tiny CDB considering how massive a city it is, but because of its amazing but small subway, it doesn’t appear quite that way in its visualization. DC, Toronto and Seattle have a lot of gravity around their central areas, which seems about right.
Anyway, I could watch these sort of things for ages. It’s interesting to see how different cities are oriented and how they have decided to move their people around via transit.
If this were an image of coronary blood flow, then Puget Sound would need a by pass.
It makes it so obvious that the “center” is in the wrong place. Imagine of those dots were clustering around Renton or Bellevue or both! Seattle downtown would make a good nature park, but it is a terrible location for transport.
Manhattan makes even less sense but that hasn’t stopped them.
Business fled Manhattan long ago.
It’s peak of population was just after 1900 with 3 million.
It dropped to 1.6 million in the 1970s and rose just a bit up to 1.8 million with the “new urbanism”.
Manhattan makes a little more sense because New Jersey is a de facto “sixth borough”. In any case, Manhattan’s importance isn’t so much because of the people as the jobs.
Tid bit on Q13 this morning. The most expensive city to live in for foreigners in North America is Vancouver B.C. It ranked 35th in the world, Manhatten was right behind in 36th. I’m not positive but I think San Francisco slotted in around #45.
Seattle is in a much better location for ship transport, which is why it ended up where it did…
Yeah Complaining about Seattle’s location is like complaining about Rome’s location, but to a much, much, much, much (x100) extent.
No, it’s very valid to discuss the location of Seattle, because there is still so much growth and infrastructure planning being discussed that will happen over the next 30 to 50 years.
My contention is that we have been making our costs and problems more difficult than they have to be by not seeing the big picture of where we could put business, residences, density and transit and by having a regional, nay, a statewide viewpoint rather than a Seattle-centric perspective.
Admittedly, Bellevue would much better justify the Ship Canal’s existence…
Actually Bellevue lost it’s main working waterfront when the Army Corp dug the ship canal, lowered the lake and created what is now the Mercer Slough Nature Park.
HT to ~JL this morning on a related subject of streetcars in Vancouver and some interesting comments by Jarret Walker on the subject, which ties in nicely to this post.
http://pricetags.wordpress.com/2012/12/10/the-broadway-line-ransford-vs-walker/
Joni’s report today shows that Parsons will be doing the 2 mil streetcar/rail study to Ballard, along with Jarret’s response above that streetcars are pretty much useless unless they have a big time advantage to walking the corridor.
Let’s have an updated post on FHSC, so I can do some time calcs without getting bonged by the [ot] monster.
A geologist friend of my wife’s asked me to think about why so many seaport towns are prey to earthquakes. Answer is how often harbors are caused by tectonic faults. Port of Renton could probably go away in a lahar too. Port of Lynnwood? CT can’t even keep the 113 on time from Mukilteo to Ash Way due to traffic.
Like San Francisco, fact of being a confined place in a beautiful and profitable location gives transit some civil engineering requirements- reason I’m for Ben’s Seattle Subway no matter what it costs.
Also don’t mind being told I have to keep my car off of certain street lanes at least at rush hour, so transit isn’t just one more set of machines trapped in a linear parking lot.
The things I like about Seattle, like the views, didn’t cost anything. So I’m already money ahead just living here.
Mark Dublin
“The things I like about Seattle, like the views, didn’t cost anything.”
Do you think views in Seattle should be protected? Or do you think tall view-blocking buildings should be allowed right next to shorelines, for example?
“The things I like about Seattle, like the views, didn’t cost anything.”
How do you feel about the great views from the viaduct, which were (and still are, to some extent) enjoyed by close to 200,000 people per day, being lost to those people so that a much smaller number of wealthy people can enjoy those views from their very expensive condos or apartments?
How do you feel about the 200,000 people a day driving at high speeds in rush-hour traffic looking out at the view instead of watching the traffic? ;)
32,000 of those people could be enjoying the view, if 1.2 people are in each vehicle.
How do you feel about the great views from the waterfront, which are currently “enjoyed” only by clueless tourists who have to deal with an ugly and noisy highway directly over their heads, being made pleasant and accessible to anyone in the city, including all of those who use public transit?
Notice how on the LA video, you can discern the grid pattern of LA streets while this is not visible on the Seattle one. just goes to show how infrequent and spastic out transit routing is!
Yeah, I know exactly what you mean. I think it shows our geographical challenges to some extent.
When I visited Toronto, I remarked how transit lines run more or less straight down streets without turning…Since Seattle’s not going to be getting a ton of Link lines, this is something that we may be able to do here to increase the impact of our transit systems.
Except that we actually can’t, because we put miles between every Link station, did our best to site stations in violation of what grid we have, and ensured that feeder access will forever involve crappy labyrinthine bus routes that can’t hold a candle to the network connectivity of a TTC-style subway.
Whoops.
We could do it at Northgate or at Bellevue or at Lynnwood. It’s much harder at say, Capitol Hill, but that doesn’t mean we have to give up on it completely.
You said we could replicate what Toronto has done. What Toronto has done is a grid.
Most Seattle feeder trips will involve multiple right angles, one (or two) unreliable buses, and a great deal of parallel running to the so-called “spine”.
We cannot replicate what Toronto has done. That ship has sailed.*
*Of course, we could try to not to keep doubling-down on our past errors. And <a href="Hyperlink Code“>we could do this to earn at least a bare minimum of gridded effect. But the smart money says we’ll learn nothing and keep doing everything wrong. Ridership levels at 1/10 of Toronto’s will be our reward.
*(we could do this to earn at least a bare minimum of gridded effect…)
Now I know what you mean. Yes, there’s no way to put Roosevelt on 65th or Northgate right on 105th. Sigh. But we can do more than what we have done so far (see Mt Baker)
But we can do more than what we have done so far (see Mt Baker)
True that. It would be hard to do less.
But our track record of learning from mistakes is less than stellar.
“Most Seattle feeder trips will involve multiple right angles, one (or two) unreliable buses, and a great deal of parallel running to the so-called “spine”.”
To the extent that’s the case, it’s not Link’s fault. Try to run a straight east-west bus on 65th. Green Lake is in the way, a stone’s throw from the station too; the only way to get around it is to get within five blocks or so of another crosstown route. If Northgate Station were closer to Northgate Way (it already is at 105th, that’s the problem), the Northgate crosstown route would bounce from the heart of Lake City near 125th, to 110th where Northgate Way is, to 105th, to Holman Road. A 125th crosstown route would spend the majority of its run on 130th and might want to continue on Sand Point Way, and so on. Don’t even get me started on an 85th crosstown route (well, if we had a stupid Maple Leaf station that is).
As for parallel running, would you want Link to have stops spaced so closely together on MLK as to make the 8 unnecessary (and again, the lack of crosstown service isn’t Link’s fault; I doubt Rainier would be any better), and would a Graham St stop be sufficient for that?
Grids need not be perfect to work: San Francisco’s geography hardly allows perfect griddedness either. But to the extent that San Francisco tries, it makes spontaneous anywhere-to-anywhere travel a simple and workable proposition.
Roosevelt station is on the grid, and unless Metro continue to be total fucking idiots, the services converging from all angles on 15th NE will be replaced by a single east-west line that continues through.
Northgate, as you correctly observe, is screwed for being so far off the grid. Pretty much all access will be laborious — foot, feeder bus, highway bus, kiss-and-ride. It will be politically impossible to get bus priority on the labyrinth of streets one uses to get in and out of the parking lot. If the station were simply on top of Northgate way, you’d put a single bus lane in either direction on that east-west street; cross-transit would simply stop, load, and keep going. That’s a grid.
But far more than the station locations, it’s the station gaps. Build a line around 2/3-7/8 mile intervals, and feasible cross transit will work itself out. Build it around 2-3 mile gaps, and look forward to the 43 and the 67 all of their obscenely unpleasant perpetuity!
The 49 would have been a better choice for a bus that should be redundant but isn’t, as the 43 could and should still be cancelled if the resulting transfer allowed it, but in any case, do you really think there would be no need for a bus on 5th at all if there were a Maple Leaf station?
Toronto is no model to follow, anyway. Toronto has been in dire need of a “downtown relief line” for, oh, 30 years approximately, but no politician has been seriously willing to consider it.
Again, Toronto’s problems are bureaucratic much more than geometric or infrastructural.
And it wouldn’t need “relief” anythings if it didn’t have a transit modeshare on par with New York, Mexico City, and much of Europe. From the vantage point of bare-minimum-service for bare-minimum-demand Seattle, hitting capacity on a very-high-frequency grid seems an enviable problem to have.
(Curious, though, if you’re talking about an east-west downtown relief. Like, say, the originally planned Queen Street subway…)
Morgan, I agree on the 49.
An urban-focused Link routing would be sub-surface under Broadway/10th, allowing for shallow-ish stations at reasonable intervals and making the 49 obsolete.
And yes on your Maple Leaf question. A single Maple Leaf station and the north-south shadow on 5th/Roosevelt becomes unnecessary. An east-west bus on 85th/80th expands its reach (and doesn’t hit 50 stop signs on Green Lake/Ravenna).
I used to live in Toronto and the TTC is one of the last transit agencies we should have to learn anything from.
The TTC is bureaucratically troublesome. The gridded route network is incredibly solid.
Toronto has the third best transit modeshare in North America, after New York and Mexico City (and beats Montreal’s transit modeshare, even though the TTC subway has slightly lower ridership than the Montreal Metro).
[Ot]
Your gonna need wheel sanders to get any traction on that one, but good luck.
The Seattle visualization shows how big of a hub the U District is. It’s really only second to downtown, and much more of a node than I had thought.
Here’s another interesting (to me, at least) and entirely different type of visualization, showing geolocated requests over a one-month period on Ottawa’s open transit data API:
http://www.ocbustracker.com/d/Statistics.aspx
Click on a bubble to see what routes were requested.
Wow, that Ottawa one is really nifty. If you zoom in you can see that the open data has been used all over the city. It says 1.5 million in a month. Not bad for a city of only 800,000.
It would be interesting to do something similar here with OBA. Admittedly, OBA apps generally don’t report the user’s location back to the server, but I’d imagine its safe to assume a roughly 1:1 relationship between user location and stop location.
Hmm…