Bike counters under surface of Golden Gate bike lanes
Marking the Baker Street bike lane for counter placement
You're getting counted every time you pass over the diamond
Every time bicyclists spin along the bike lanes on Golden Gate Avenue and Baker Street through NOPA, they get counted automatically. Late last year the San Francisco Municipal Transportation Agency (SFMTA) installed underground counters in the 1900 block of Golden Gate, between Lyon and Baker, and the 700 block of Baker, between Golden Gate and McAllister.
The sensors are placed from one to three inches under the road surface and can distinguish between a bicycle, a pedestrian, or a vehicle passing over them. The system records a count once it detects an "electromagnetic signature" from the bicycle. The counters require a minimum of maintenance and are cost-effective, using batteries that last a year. All that is visible from the street is a diamond shape in the bike lane and a line connecting it to a similar sensor in the opposite bike lane across the street. The SFMTA first experimented with the automatic counters in March 2009 on Fell between Scott and Divisadero. The agency found the system accurate when compared to manual counts undertaken during the same time period.
The bike sensors in NOPA are among the 22 that the SFMTA intends place at 13 locations in the city by the end of the year. Eventually, the agency hopes to position the sensors at all 33 key intersections now included in the annual citywide bike count conducted by interns. The automatic counters will provide city planners with continuous and more comprehensive understanding of bicycle traffic patterns, use of specific lanes, and fluctuations in frequencies of trips during different time periods and weather conditions.
The 2010 bike count found a 3% increase in the number of observed cyclists since 2009, as previously reported. Other locations in and near NOPA also saw increases from 2009:
- 10% rise at Fell and Scott
- 26% at Golden Gate and Masonic
- 7% at Masonic and the Panhandle Path
Another automatic counter is planned for the Panhandle Path approaching Masonic.
Well, the obvious problem is that this is only counting bike routes where there is a separate lane thus leading to a form of observer bias. If there were far more bike lanes this might not be such a problem, but as it is this can't possibly provide enough in the way of useful data. Changes in routes might be more easily explained by switches to non-laned routes or even routes that aren't officially approved.
ReplyDeleteI also have to question how the "electromagnetic signature" works for various frames. It seems like this would cause problems for non-ferrous metals and possibly be completely useless for carbon fiber. A minority, true, but I still find it unlikely that even just between steel and aluminum it would be equally accurate.