Daily Counts on the Custis Trail

CommuterPageBlog has more information about Arlington's trail counting technology and the results. The spike is Bike to Work Day and the lowest days all coincide with bad weather.

For the month of May there were 40,499 bicycle trips past this section of the Custis Trail and 13,322 pedestrian trips.

I can't wait to see year to year data. But then, I'm something of a nerd.

There's also a link to a USA Today story about the sensors. 

The increased use of high-tech sensors supplements a push for expanded counts by the National Bicycle and Pedestrian Documentation Project, which this September is overseeing censuses in about 150 cities, including Kansas City, San Francisco and New York City, Michael Jones says.

Jones, a planner and principal with the Portland, Ore.-based Alta Planning and Design, says he founded the count in 2004 after growing frustrated by the lack of consistently collected pedestrian and bicycle use data. He says about 10 groups conducted counts that first year.

Under the project's census, trained volunteers record the direction of each passing biker and pedestrian for two hours each on a weekday and weekend day in multiple locations, and then use around-the-clock tallies from automated devices placed on other nearby trails and roads to account for seasonal and daily weather variations, Jones says. He says it's easy to find volunteers to monitor riders on sunny days, but hard to find people willing to stand in the rain at night, even though cyclists are still out.

"It's a great relief to have robots out there counting … rain or shine," Patton says

Comments

[Washcycle]

Looking at the plot again, what's interesting is that the eight busiest days are weekdays.

[Wayne Phyillaier]

I'd like to learn more about their count methodology. I'm curious how they established that using volunteers to count traffic for two hours on a weekday and two hours on a weekend day was enough for an accurate count, even with the use of automatic counters to fill in the daily and seasonal variations.

I supervised the 2006 trail traffic survey on the CCT, online at www.cctrail.org/CCCTsurvey.htm, and we used volunteers for a minimum of 20 hours at each of five locations. We had no automatic sensors since we could find none at that time that would do a reasonable job. I think the automatic sensors are almost a necessity for a good count, since the variation for time periods in the day, and day to day, was very high.

[Washcycle]

Wayne, it probably has more to do with manpower availability than good science. They have enough trouble doing the two hour counts.

[Daniel]

Spike on May 21 is Bike to Work day. Cool.