Mighty Segregated bike lanes

May 21, 2013

…or perceived safety and objective safety of the cycle tracks

A study on Toronto and Vancouver (Canada) from [4]: the risk of bike infrastructure separated of traffic is under-estimated. Note the result carried for the cycle track is an aberrant and irrelevant one for reason explained in [8]

Usually, Urban segregated bike lanes (cycle tracsk) are perceived as safer than non segregated one, by many cycle advocates and public alike. Alas most accident statistics say otherwise, and most scientific studies conclude, consistently overtime, that segregated bike lanes impair safety by ~20% ([1] summarizes and complete previous studies, see also a list of studies at [9]), some older studies putting this number up to 4 time higher [2].


    Of course, it is possible to find some studies saying otherwise, but usually those studies show significant methodology shortcomings. To focus only on recent Canada centric examples: [5] draws conclusion on cycle track from a field study conducted in cities not having such infrastructure per sei, as seen in [8] and obvious selction biais discredit results from [3] (more critics here and there):

Montreal, QC: In (3), a separate bike path in a one lane residential street (rue Brebeuf) is compared to an up to 6 lanes thoroughfare (rue st Denis) on a 1km section (Rachel to Laurier), where St Denis has more intersection, and higher speed limit than Brebeuf...to conclude that separates bike lane improve cyclist safety! (no indication of motor traffic volume is provided) -

    The most recent study extended to the USA by the same authors, [10], seems to suffer similar flaws [11].


In urban area, most of the cyclist accidents are due to conflict with motor vehicles (85% in French cities according to the OSNIR), and most of them occur at intersection: In Canadian cities, 50% of fatal accidents and 72% of accidents resulting in serious injury occurred at intersections [12].

Thought, that a separated bike lane can remove potential conflicts along a road, and is recognized to reduce risk in such cases, it makes matter worse at intersections: This is mainly due to the fact cyclists, not on the road, tend to be overlooked by other road users, generating conflict at road intersections. The increased risk for cyclist is illustrated below:

According to some study, the cyclist could be up to 4 time safer on the right side of the street - credit photo (6)CycleRisk

According to (2), the cyclist could be up to 12 time safer on the right side of the street - credit photo (6)

Aware of this fact, Some transportation professional organizations don’t recommend separated bike lane: it is the case for the AASHTO in the USA, or the CERTU for urban area in France. A position supported by numeorus cyclist organizations, be in France (FFCT, Fubicy) or Germany (ADFC), which have been at best rather neutral on the development of segregated cycle track, in some case opposed, and consistently advocating against the mandatory use of it. That eventually became the case for most of the french cycle track, circa 2000. For this later purpose a new road sign has been introduced, and Germany is following track:

B22a_PisteCyclable_obligatoire

The cycle in a blue square sign has been introduced circa 2000: it indicates a recommended cycle track. The cycle in a blue disc indicate a mandatory cycle track ... except of course in UK Which has not ratified the Vienna convention on road sign, from which those signs are derived

An issue is that motorists tend to ignore the difference, and harass cyclists not using the cycle tracks

Traffic engineers, on their side, sometimes eager to remove cyclist of the road for their “good”, have worked to increase the safety of separate bike lane:

Reintroduction into general traffic at intersection

Rennes, France: Bike paths merging in general traffic at intersection, and resuming after it


bikeLaneEntranceBdArmorique Rennes, France (Armorique Bld): Cycle track merging in general traffic at intersection, and resuming after it

Treating cyclist as pedestrian at intersection

MapHongKongBikeLaneIntersec

Hong Kong (Along Ting Kok Rd, Kong Kong NT): Cyclists are expected to walk their bikes to the cycle track... and dismount at every intersections...what by the way is seldom respected in despite of the British style staggered pedestrian crossing! -credit photo left (16), right, Google

Cycling Commuters are generally not impressed by those treatments, which are just slowing down their commute, even when the obligation to walk the bike at intersections (Hong Kong case), is obviously widely disregarded by cyclists using such facilities.

The Copenhagen’s Treatment: Blue cycle crossings

Copenhagen, DK: An intersection where potential conflict zones are highlighted in blue

Copenhagen, DK: An intersection where potential conflict zones are highlighted in blue – credit photo (13)

It has been “invented” in Copenhagen in 1981: The basic idea is to mark the area of conflict between motor vehicles and cyclists so road users pay more attention to this conflict and cyclists have a lane marking through the junction area. Alas, while it is found effectively reducing the number of accidents (and injuries) with one line, it increases it with 2 lines or more, according to [13].

A reason for that is that, it becomes too much solicitation for the motorist than he can process – resulting in an increase of rear ending collisions and red light runnings; and provides a false “sense of safety” to the cyclists, becoming more complacent- not doing head check or using hand signals according to [14]- what is consistent with the “naked street and risk compensation theories.

…and more often that not:

Separated bike lanes come with a panoply of restrictive sign

All, in the name of cycling safety of course…

Left, Bideford UK; center, Harlow UK (now dismantled); right Vancouver, CA - credit photo resp (5),(unknown),(16)

But at the end, it is sometimes better to give-up

…than to cut the trees:


ClosCourtelOld

Rennes, France (Clos Courtel Street): A once mandatory segregated bike lane, has been replaced by a painted bike lane, allowing much better visibility of cyclists by other road users - credit photo Google

Should we be Against the separated bike lane?

or…Should we support the helmet law under evidence of greater safety provided by the helmet

Both generate passionate debates, and unfortunately, both generate biased scientific literature too.

  • Supporters of the helmet laws are because they are concerned by the safety of existing cyclists, they will be obviously against separated bike lanes for the same reason. Not surprisingly, most of the anti cyclist lobbyist will fell in this category
  • Supporter of the helmet laws supporting separated bike lane are not logical with themselves and probably grossly misinformed
  • Opponent to the helmet laws, will explain that, while the safety of existing cyclists is important, it is not paramount- One have to take a more holistic view to assess the benefit/drawback of such safety tool than the existing cycling population- and opponent to the helmet laws, without necessarily denying the positive safety effect of the helmet on an individual, will oppose to a law on the ground that it discourages sufficiently cycling to have a general negative effect for the society.
    Same logic apply to the cycle tracks: there is no need to deny their negative effect on road safety, or to produce biased studies to try to counter evidence, to support them: that is only conductive of complacency with poorly designed cycle tracks which do no good for cycling. Former Vancouver Planning Director, Brent Toderian was able to implicitly recognize the safety issue and supporting it [17]: What is important is to produce evidence that the positive effect they induce outweigh their negative ones

  • [1] Traffic safety on bicycle paths – results from a new large scale Danish study, ICTCT workshop Melbourne, 2008

    [2] Signalreglerade korsningars funktion och olycksrisk för oskyddade trafikanter – Delrapport 1: Cyklister. Linderholm, Leif, Institutionen för trafikteknik, LTH: Bulletin 55, Lund 1984

    [3] Risk of injury for bicycling on cycle tracks versus in the street, Anne C Lusk, Peter G Furth, Patrick Morency, Luis F Miranda-Moreno, Walter C Willett and Jack T Dennerlein, Injury Prevention, February 2011. doi:10.1136/ip.2010.028696.

    [4] Route Infrastructure and the Risk of Injuries to Bicyclists: A Case-Crossover Study, Teschke K, Harris MA, Reynolds CC, Winters M, Babul S, Chipman M, Cusimano MD, Brubacher JR, Hunte G, Friedman SM, Monro M, Shen H, Vernich L, Cripton PA., American Journal of Public Health: December 2012, Vol. 102, No. 12, pp. 2336-2343.

    [5] Safe Cycling: How Do Risk Perceptions Compare With Observed Risk?, Meghan Winters, Shelina Babul, H.J.E.H. (Jack) Becker, Jeffery R. Brubacher, Mary Chipman, Peter Cripton, Michael D. Cusimano, Steven M. Friedman, M. Anne Harris, Garth Hunte, Melody Monro, Conor C.O. Reynolds, Hui Shen, Kay Teschke, Injury Prevention, Canadian Journal of Public Health , Vol 103, No 9, 2012

    [6] Bicycle Quaterly

    [7] Gary James

    [8] Conclusion of both [4] and [5] are drawn from a study carried from May 2008 to Nov 2009 in Toronto and Vancouver. To the bets of our knowledge, it was no “cycle track” in Toronto, and the only ones able to qualify in Vancouver, were an experiment started on July 2009 on Burrard Bridge, with no intersection along the ~1km cycle track segment, and a ~300m segment in one direction on a quiet street (Carral street) with ~300 cars at peak hour with only one very quiet intersection (Keefer street) featuring ~120 car at peak hour (From City of Vancouver’s 2006 traffic count) what is barely representative of a typical cycle track: The result provided for the cycle tracks is hence certainly irrelevant, and that is the reason it stands as an outlier.

    [9] Bicycle Infrastructure Studies review by Ian Brett Cooper

    [10] Bicycle Guidelines and Crash Rates on Cycle Tracks in the United States, Anne C. Lusk, Patrick Morency, Luis F. Miranda-Moreno, Walter C. Willett, Jack T. Dennerlein, American Journal of Public Health, July 2013

    [11] [10] draws conclusion by comparing current crash rate on some cycle tracks with some numbers collected, sometimes in specific situation- like a study on Boston’s bike messengers- more than 10 years ago, without correcting them of external factors, like significant general crashes reduction rate in the last decade, and well documented safety in number effect affecting more particularly the cyclists. Furthermore, one could argue that the “crash rate” is a very poor, if not uncorrelated, proxy, to qualify the safety of a road infrastructure: Roundabout are well-known to increase the rate of crashes, vs a signaled intersection, but they are also well recognized to reduce the risk of serious injuries, most of the crashes being limited to fender-bender type. In other word, a crash rate ratio is not representative of the safety social cost of an infrastructure…what ultimately matter. More awkward [10] suggests that “The AASHTO recommendations may have been influenced by the predominantly male composition (more than 90%) of the report’s authors” without being able to substantiate this assertion, showing that we have here more a opinion paper: attacking the gender of authors to disqualify their works, seems pretty petty at best!

    [12] Vulnerable Road User Safety: A Global Concern, Transport Canada, 2004.

    [13] Safety effects of blue cycle crossings: A before-after study, Søren Underlien Jensen, Accident Analysis & Prevention, Volume 40, Issue 2, March 2008

    [14] Evaluation of Blue Bike-Lane Treatment in Portland, Oregon. Hunter,W.W., Harkey, D.L., Stewart, J.R., Birk, M.L., Transportation Research Record 1705, 2000

    [15] The finding of [13] seems in fact to suggest that the increase in accident and injuries are mainly among motorists, and eventually moped: so that in fact the blue line could effectively be not than “unsafe” for cyclists. Unfortunately, the paper doesn’t provide detailed break down of the injuries according to the transport mode. In any case, the measured global effect is a negative one

    [16] www.vivendesign.com

    [17] Vancouver Embraces Bikes, Adds Lanes, Tim Newcomb, Planning;, Vol. 77 Issue 2, Feb2011

    About these ads

    7 Responses to “Mighty Segregated bike lanes”

    1. Tessa Says:

      It’s important to note that Vancouver’s cycle tracks do have painted green lanes in the intersection, so one can’t take your results of an increase of accidents at intersections in that one study and apply it to Vancouver exactly.

      • Voony Says:

        True: Notice, that the study [13] recognizes that a painted lane, improve safety if there is only one in the intersection, what is the case in Vancouver. Furthermore in Vancouver, many conflicts are prevented by bylaw (right turn prohibited….), and when allowed, it is at the expense of an extra traffic signal cycle : That can probably work only in a certain extend, where there is not too much cycle tracks…but yes, for the time being most of the highest risk situations have been prevented on the Vancouver cycle tracks.

    2. mezzanine Says:

      WRT bike helmet laws, I would also argue there is a social standard with risks that we can accept to cyclists that trumps studies, vis-a-vis comparing motorcycle helmet laws in the USA.

      http://www.iihs.org/laws/HelmetUseCurrent.aspx

      FWIW, i support relaxing helmet laws for cyclists and improving cycling infrastructure.

    3. Andrew K Says:

      The Dutch have already solved this by the way they engineer their cycle tracks. They never make bidirectional cycle tracks for the reason shown (bidirectional cycle tracks along city streets are very inconvenient too, and shouldn’t be built.), they always provide clear lines of site at intersections and the cyclists are in an advanced position, so that cars can see them. In addition, it is written into law that they have priority over driveways, and minor side streets and it is really obvious that they do as the cycle track continues visually.

      There would indeed be teething issues changing over to this, but it shows that it can work very well in the end.


    4. […] Compare With Observed Risk?” [6] (affidavit [7]). A paper we have already studied in part here. What is important to retain for the case under trial is that this paper states that a “bike only […]

    5. arnoschort Says:

      Voony states:
      [5] draws conclusion on cycle track from a field study conducted in cities not having such infrastructure per sei,
      And he lists some of the separated paths in [8]. The list provided is incomplete, as the entire length of Carral Street was part of the study as was a section of the Seaside Path adjacent to Beach Ave. There may also have been a section of 1st Ave and the Dunsmuir bikeway may have been completed before the end of the study period.

      Also, if segregated bike lanes are so dangerous, I wonder why the Netherlands, which has mostly separated paths and a cycling mode share of about 30% also has one of the best cycling safety records in the world. I also wonder why in Canada, which has few separated paths, only about 2% of people regularly ride bikes.

      • Voony Says:

        I have listed Carral street, and the seaside path was all likely in the mixed-used path category.
        In any case the sections along Beach avenue are free of intersections (at the difference of a “real” cycle track such as Dunsmuir or Hornby which was not completed by the end of the study period), so it can be considered as a valid example.

        Regarding, the Netherlands safety issue:
        Is it because the Netherlands has an extensive bikepath network it is safe for cyclists or is it because 30% of people cycle it is safe for cyclists?

        people tend to draw some suprious correlation, and the object of this post was to show that.

        bike helmet can be considered as increasing the safety of an individual cyclist, but in fact it is not making cycling safer if it is a deterrent to cycling.

        On the opposite:
        …cycle track could be not safe, but it is possible that because they encourage cycling, that unsaefty issue (well recognized by scientific litterature) is compensated by “safety in number”.


    Leave a Reply

    Fill in your details below or click an icon to log in:

    WordPress.com Logo

    You are commenting using your WordPress.com account. Log Out / Change )

    Twitter picture

    You are commenting using your Twitter account. Log Out / Change )

    Facebook photo

    You are commenting using your Facebook account. Log Out / Change )

    Google+ photo

    You are commenting using your Google+ account. Log Out / Change )

    Connecting to %s

    Follow

    Get every new post delivered to your Inbox.

    Join 50 other followers

    %d bloggers like this: