[bouldercouncilhotline] Hotline: Safe Streets Report

[cmosupport at bouldercolorado.gov]

Sender: Bill Cowern

Hi KC,

Your Hotline request concerning the city's "Safe Street Boulder Report" was forwarded to me for response.  Thank you for taking the time to contact us with your concerns.

The scope and intent of the "Safe Streets Boulder Report" is to identify the most common types of collisions, discuss behaviors that cause them, and identify locations where collisions occur most frequently.  A primary objective is to highlight trends in an effort to identify potential countermeasures that would reduce the number of motor vehicle collisions involving bicyclists and pedestrians.

The report does detail a breakdown of the total number and percentage of all motor vehicle collisions involving a bicyclist or pedestrian by different types of crosswalks, including that 6 percent of all motor vehicle collisions involving a bicyclist or pedestrian (during the 40 months studied) occurred at flashing crosswalk locations.  The report does not state, nor are those statistics intended to imply, that flashing crosswalks are universally safe or even safer as a treatment than other crossing treatments because of that percentage.   It is only intended to mean that a small percentage of collisions occur at such crossing treatments.  For staff, this  is a significant finding because it means that any actions we take that address only flashing crosswalk treatments (such as the recent ordinance change requiring persons crossing them to use the warning device) have the potential to impact a relatively small number of bike/pedestrian collisions; whereas, actions taken that can impact all crosswalks (such as the recent ordinance change requiring a bike to enter and traverse at 8 mph) have the potential to impact a much larger number of bike/pedestrian collisions.

It would be fair to criticize conclusions drawn about the safety of a treatment based solely on the percent of collisions that occur citywide.   There are thousands of legal crosswalks in the City of Boulder, most of which are low-volume streets with very little pedestrian crossing activity and even less possibility for conflict.  A comparison of such locations to higher speed, high-conflict locations typical of signalized intersections and flashing crosswalk locations would be erroneous.  Appropriately, the "Safe Street Boulder Report" provides no conclusions about appropriateness of any crossing treatment based on the percentage of collisions occurring at different crossing treatments.  Overall, staff is confident that the information provided in the "Safe Street Boulder Report" is appropriate and accurate.

The most pertinent information concerning flashing crosswalks and their relative safety as a crossing treatment in the "Safe Street Boulder Report" can be found in the "Countermeasures" section (Page 12), in which it is stated: "While the safety of these crossings has been questioned, this analysis shows that flashing crosswalks are not a major collision concern when installed in appropriate locations." This conclusion is drawn from the analysis that generated the high collision location diagram on the prior page.   This conclusion is not based on the percentage of bike/ pedestrian collisions that occur at all flashing crosswalks as a category, but rather at specific locations.

This evaluation looked at all locations, regardless of treatment type, that had higher numbers of bike/ pedestrian collisions.  There were 15 locations that had seven or more bike/ pedestrian collisions (during the 40 months studied).  Of these locations, one location was an existing flashing crosswalk location, and one other was the location that used to be a flashing crosswalk but was removed.

There are 15 locations using the flashing crosswalk treatment today.  Fourteen of those locations have so few bike/ pedestrian collisions that they do not make our list of high collision locations.   The evaluation used to generate the high collision locations is based upon a rate of collisions per year.  This is the most common criteria used as a first step to determine whether improvements or mitigation should be considered at a location.  A rate of collisions per year (and more importantly, collisions that can be mitigated per year) speaks directly to the potential to mitigate or reduce high numbers of collisions over time.  In the case of this study, if the majority of the high collision locations were flashing crosswalk locations, that would be concerning, and staff would not have drawn the same conclusions.  However, since only one existing flashing crosswalk location showed up on the high collision location list, it does not point to the treatment type being a specific problem in general.

Determining locations with high numbers of collisions is only the first part of a crosswalk safety evaluation.  Next, staff reviews the types of collisions and determines whether there are trends that can be mitigated.  For instance, the location with the highest number of collisions on the list is the intersection of Colorado and Regent.  A detailed review of those collisions (provided on Page 1 of Attachment A to the report) shows that there are three different collision trends occurring at this location, each with different potential mitigation.  The point to all of this is that there is a lot that goes into the evaluation of a location to determine if improvements are needed, including whether the appropriate crossing treatment is being used.  In other words, we review each location on a case-by-case basis, rather than how it compares to the percentage of collisions citywide.

In addition, staff's conclusions about the appropriateness of flashing crosswalks as a possible crossing treatment to be used in the City of Boulder are based on more than just the conclusions of the "Safe Street Boulder Report."  The City of Boulder has been using flashing crosswalks for more than 12 years, and the relative safety, effectiveness and appropriateness of these devices have been studied.   We have examined not only the total number of collisions and rate of collisions per year, but also the rate of bike/pedestrian collisions per pedestrian crossings and the number of rear-end collisions per entering vehicles.

For example, in December 2009 at a Transportation Advisory Board meeting, staff presented materials detailing the evaluation of flashing crosswalks specifically.   Evaluations of collision data at flashing crosswalk locations was compared to other high collision locations both by rate of collision per year and by rate of collision per entering pedestrians.  Attachment E of that December 2009 TAB memo provides a summary of the collision evaluations.  A copy is attached to this Hotline response.  This 2009 report concluded that most of the flashing crosswalk treatments operate in a relatively safe manner, with a frequency and probability of accidents that are consistent with other types of crossing treatments and acceptable when compared to the enhanced mobility these crossing treatments provide to bicycles and pedestrians.  However, based on the findings in the 2009 study, staff identified one location where the flashing crosswalk treatment type was not deemed appropriate, and that treatment was removed and replaced with a traffic signal.   This experience has also helped staff to modify the criteria used to determine where flashing crosswalk treatments should be used in the future.

Based on the information above, staff believes the statements made in the "Safe Streets Boulder Report" are appropriate and does not recommend making changes to the report.  Should there be additional questions concerning the appropriateness of flashing crosswalks as a crossing treatment or staff's conclusions in the "Safe Streets Boulder Report", we would be glad to address them.  Boulder's transportation system remains a source of pride for our community members.  The recent positive news coverage was a nice opportunity for us to share some of our research with them.  In addition, thank you for sharing your concerns with us.  We welcome the feedback and are happy for the opportunity to further explain our evaluation process.  Please feel free to contact me if there are other questions I can help answer.

Bill Cowern
Transportation Operations Engineer
City of Boulder
303-441-3266


From: Becker, KC
Sent: Friday, February 10, 2012 10:13 AM
To: HOTLINE
Subject: Safe Streets Report

The Safe Streets Report was included in our most Recent Information Packet. This study analyzes more than 8,000 collisions in Boulder - involving motor vehicles, bicycles and pedestrians - over a 40-month period.

The Daily Camera reported on the study last week, but I've just had a chance to review the report, and read the article and some of the comments I've received about the report.

As a press release from the city states, "safety of flashing crosswalks has been an underlying community conversation, but the results show that collisions in these crosswalks account for less than 1 percent of all collisions." The report states that crosswalks of all types account for 44% of accidents involving pedestrian and 56% of accidents involving bike, but only 6% of those took place in a flashing crosswalk.

I'm concerned that this statistic is misleading and I'd like clarification from staff, and if necessary, a revision to the report.

I believe that there are 18 flashing crosswalks in Boulder.  There are of course thousands of non-flashing crosswalks at intersections across the city.  It could be argued that collisions in flashing crosswalks are rare because flashing crosswalks are rare, while collisions in crosswalks with intersections account for a higher percentage of the total accidents because there are so many crosswalks at intersections.

In order to understand whether the flashing crosswalks are more or less dangerous than crosswalks at intersections, we need data on the collision rates data on the collision rates comparing the types of crosswalks.  This analysis is important for when we consider whether installing more of this type of flashing crosswalk makes sense. Looking at it this way, I think the statistic would show that the accident rate for intersection crosswalks is lower than the accident rate for flashing crosswalks. Assuming this, I'd like to know from staff if and how this revised look at the statistics should inform us about the safety and future use of flashing crosswalks.
Thanks!
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Attachment E_Accident study of PAFS locations.doc
Type: application/octet-stream
Size: 169984 bytes
Desc: Attachment E_Accident study of PAFS locations.doc
Url : http://list.ci.boulder.co.us/pipermail/bouldercouncilhotline/attachments/20120217/94a369c1/attachment.obj