Update: 11/20/09
The slides are now online.
Original Post: 11/5/09
I’m attending the Congress for New Urbanism Transportation Summit here in Portland this week (where I’ve had the opportunity to meet several other transportation bloggers!).
For me the highlight presentation on opening day on Wednesday was about the safety affects of different street network types.
The study was based on looking at all cities in California with population of 40,000 or greater. The surprising finding was that cities built before 1950 are safer (in terms of both serious injuries and fatalities for all classes of users: auto drivers/passengers, cyclists and pedestrians) than cities built after 1950.
The differences appears to be in the type of street network. Compact street grids seem to be safer, compared to the arterial-collector-local street ‘tree’ style of street network popular in post-war development.
This explains a lot here in Portland, particularly the difference in traffic safety between inner neighborhoods new parts of Portland near or east of 82nd Avenue (illustrated by an example from current news).
I’ll post a link to the research report when it’s available online.
17 responses to “Cul-de-sacs Kill!”
How does a new report about a death on an essentially grid-style street illustrate your point?
How does traffic calming affect things? One problem with grids is that side streets are frequently used as thoroughfares, unless calmed. Does the increased accidents in the postwar style happen on dead-end streets, or on the collectors and arterials?
Looking at the intersection it seems to me that the grid is beginning to fracture by this point. The very ‘diagonal’ nature of Foster is one of the disruptions to the grid.
The current example you gave did not occur on a typical suburban street. The issue with cul-de-sacs and residential safety has a lot to do with backing up, which is a very dangerous activity.
I disagree with the diagonal being the issue. Washington DC is based off of this very design. It really does not look like it disrupts the grid like Broadway does in NYC, where you essentially have a 6-way intersection at some points. That’s just my quick glance at it.
Any sources for these presentations? It would be great to look over.
OK, I’ll concede that I may not have selected the best example, I was trying to link this to something topical.
Here’s a link to the presenter. I’ll get a link up to the presentation as soon as it’s available.
I have to wonder if some of it comes from the street widths. Using San Diego as an example, most of the arterials in newer neighborhoods are 3 lanes each way, up to 9 lanes to cross. Many have 55 mph speed limits, and only have a safe pedestrian crossing every 1/4 to 1/2 mile.
In a grid, you have a fairly safe crossing point every 200 to 500 feet, giving drivers more of a sense of needing to pay attention. Traffic speeds are lower, and the widest streets even in San Diego’s urban core are 5-6 lanes, but most are 3 lane one ways so you’re less exposed to traffic both due to the lower number of lanes, lower traffic speeds, and a one way grid is just simply safer for pedestrians.
In downtown and other grid-type areas there were some fatal accidents while I lived there, but most were in the newer (70’s and later) suburbs.
I’m curious if the people doing the study accounted for the safety benefits gained from having a one way traffic grid. I know I’ve seen it cited in articles before.
There was a Transportation Seminar at PSU covering this topic a few weeks ago. He really blew my mind with his regression analysis showing how different types of urban forms (in addition to streets) can impact the crash and fatality rate of a road.
http://www.cts.pdx.edu/seminars/index.php
Oct. 16th ‘Urban Form and Traffic Safety”
Tree networks are also less efficient than grid networks. I can’t find the study, but trees nearly always require a long trip to get from any one point to another, whereas a grid has a straightforward path. These designs tend to discourage walking and bicycling not just from a safety standpoint but from a convenience one as well.
Hey Chris. Unfortunately I wasn’t able to make it to the conference! Please keep us informed of the cool things that come up!
Foster actually pre-dates the street grid. It was an old farm road.
But back to cul-de-sacs:
http://www.planetizen.com/node/38700
Seconding Daniel’s comment above – looking forward to a full report, thanks!
Re the example:
It looks to me like there are a few factors affecting the safety of the crosswalk, judging from the video at the top of the page and the map at the bottom.
From the video, it’s pretty clear that though a traffic island is installed, there appeared to be very little thought given to how the design of the island would help calm the traffic. Rather, it appears it’s designed to not interrupt the smooth flow of traffic.
While the speed limit there is 35 mph, the road appears to be treated more like a suburban arterial. Motorists are encouraged, through design, to drive faster.
Take a look at the Google Street view of the crosswalk — nearly everything about that area is car-oriented suburbia.
“In a grid, you have a fairly safe crossing point every 200 to 500 feet, giving drivers more of a sense of needing to pay attention.”
>>>> Really? SE Hawthorne, for example, has a number of places where it’s much greater than this between signaled crossings.
I have to wonder if the data is tainted. California DMV is known to:
Given that California’s licensing standards are more lax than Oregon’s, what does this have to do with Portland transportation?
I have to wonder if the data is tainted. California DMV is known to:
* Issue driver’s licenses to drivers who pass the written portion with a high enough score, regardless of the score on the driven portion, and
* Provide answer keys for the written portion on request while taking the test
I got my first license there, and I’ll ask for a citation. That sounds like made up stuff to me, and doesn’t match with my experience or that of others.
Since presumably the same drivers, however they are licensed, drive on both types of street networks being studied, it would seem that driver quality is not the determining variable.
[personally directed comment removed]