Austroads have published a new research report related to cycling, AP-R380/11 Effectiveness and and selection of treatments for cyclists at signalised intersections.
“The provision of cycle facilities at intersections, especially traffic signals, is becoming morecommon across Australia and New Zealand, particularly in cities that are popular for cyclists.There is currently very little local research on the effectiveness of these facilities in terms of safetyand cyclistsâ€™ perceptions. Research from New Zealand by Turner et al. (2009) indicates that thereis a safety benefit of at least 10% by installing mid-block cycle lanes, and that the presence ofother road features, such as flush medians, influence cycle safety. Turner also reviewed a numberof international studies which indicated that the provision of cycle lanes reduce crashes by around10% and the installation of an advanced limit line at intersections for cyclists (storage facility)reduces crashes by around 27%.”
The full summary of the report is provided below along with the download link.
The benefits of cycling as a mode of transport are widely understood and recognised across theglobe. Cycling forms an integral element of a transport system, and provides a genuine alternativetravel choice to cars while contributing to healthier lifestyles. Many governments, including those inAustralia and New Zealand, have set targets to increase the levels of cycling in their respectivejurisdictions. Yet, given the benefits of cycling, it accounts for only a small proportion of all dailytravel.
The provision of cycle facilities at intersections, especially traffic signals, is becoming morecommon across Australia and New Zealand, particularly in cities that are popular for cyclists.There is currently very little local research on the effectiveness of these facilities in terms of safetyand cyclistsâ€™ perceptions. Research from New Zealand by Turner et al. (2009) indicates that thereis a safety benefit of at least 10% by installing mid-block cycle lanes, and that the presence ofother road features, such as flush medians, influence cycle safety. Turner also reviewed a numberof international studies which indicated that the provision of cycle lanes reduce crashes by around10% and the installation of an advanced limit line at intersections for cyclists (storage facility)reduces crashes by around 27%.
This study focused on the safety impacts of providing cycle facilities, in combination with a numberof other features, such as width of approach kerbside lane, at traffic signals. A review of facilitiesprovided and information available was undertaken across each of the Austroads jurisdictions. Itwas necessary to locate sites where cycle facilities have been installed for at least five years (toallow before and after assessment) and where adequate cycle and motor-vehicle exposure (flow)data was available. A survey questionnaire was also sent to each jurisdiction to determine keycycle facility types and data available for each. A workshop was held with representatives from anumber of jurisdictions to understand which cycle facilities are preferred, to discuss cycle safetyconcerns and better understand what data was available in each city.
Based on the feedback from jurisdictions, study sites were selected based on the availability ofexposure data in Christchurch and Adelaide. A total of 383 approaches at 102 signalised crossroadswhere included in the analysis. The major crash types were right turn against (right turningvehicle hitting opposing through cycle), right angle (both travelling straight through), same direction(on the approach), left turn side-swipe (left turn vehicle cutting off straight through cyclists) andâ€˜otherâ€™ (most of the remaining crash types).
Crash prediction models were developed for each crash type, relating crashes to a number of roadfeatures and exposure. In addition to the cyclist facility type and cycle and vehicle volumes, otherkey predictor variables found to influence cycle crash rates included total intersection approachwidth, depth of advanced stop boxes, intersection depth, number of through lanes, cycle lane widthand kerbside lane width. While a combined crash prediction model was developed for each crashtype across both cities, different constants were developed for each city, which enables acomparison to be made on crash risk between the cities and allows predictions that are specific toeach city. A before and after study of a smaller sample set of sites was also undertaken for allcrashes and separately for each of the major crash types.
The key conclusions from the ˜before and after and cross-sectional analyses are as below:
- The overall effect of cycle lanes was neutral. However, more specifically, cycle lanes built tohigh standards improve cyclist safety and those built to lesser standards can reduce cyclistsafety.
- The total space available at a wider kerbside approach lane, or the total width of cycle laneand adjoining traffic lane, is more important for safety than whether a cycle lane is markedwithin the available space.
- Cycle lane width had varying impact depending on crash type. Overall, wider cycle lanesover the range of 1 metre to 1.8 metres were more beneficial.
- Better driver behaviour was observed in previous studies at coloured cycle lanes. This studyshows that the provision of coloured cycle facilities result in substantially safer outcomes.
- Sites with shared left-turn and through lanes have higher initial crash rates. This report indicates that they may benefit from coloured cycle lanes and advanced storage boxes.
- Sites with exclusive left turn lanes are much safer for cyclists than those with a sharedthrough and left turning lane. Any cycle lanes provided need to use colour from the transitionacross the diverge area to the limit line.
- Deep intersection distances appear to increase right turn against (LB) crashes and reducecrossing crashes.
- The overall benefits of providing advanced storage, could not be separated due tocorrelations. However for treated sites greater storage depth (up to 4 metres) was clearly beneficial.
Cycle lanes are known to improve cyclist perceptions of safety to a greater extent than theimprovement in actual crash risk observed in these and other studies. If we are to avoid merelyproviding a false sense of security, the provision of coloured cycle lanes, of good width leadingfrom the transition to the advanced limit lines is of critical importance. Such facilities will improvecyclistsâ€™ perceptions of safety even more and encourage more of them to ride.
The models clearly indicate the safety benefits of providing good quality cycle facilities. However,the models do not fit well enough to be certain of the size of the safety benefits of most of the typesof facilities studies. Future studies using the same methodology should use a sample size at leastdouble the number of sites used in this study.
Other key recommendations from the study are outlined below.
- South Australia should bring their police reported crash recording system into line withAustroads best practice guidance. A lot of manual effort was required to compile theappropriate crash movement types. The absence of Police diagrams for all of the crashesalso reduced the reliability of coding. Having the full Road User Movement codes wouldspeed up the compilation of crash data by type.
- Christchurch and Adelaide were the only jurisdictions that routinely count cyclists atintersections. These counts should be routine in all jurisdictions.
- The Christchurch experience highlighted that when manual cycle counting is conducted atthe same time as motor vehicle counts the number of cyclists is seriously undercounted,especially when the site becomes busy. It is recommended that a separate person countscyclists, and also pedestrians. This is consistent with minimising delay to people – not justmotor vehicles.
- The Adelaide analysis revealed an increase in cyclists struck by motor vehicles whencrossing from the footpath using the pedestrian crosswalk. The effect of the Australian roadrule changes permitting footpath riding for cyclists under 12 years of age (Australian RoadRule 250-1, 1999) should be studied.
The reference for the publication and download link are:
Turner, S., Singh, R., Allatt, T., Nates, G., and Beca Infrastructure Ltd. (2011). Effectiveness and selection of treatments for cyclists at signalised intersections. Sydney, NSW: Austroads Ltd. Available for download as a PDF here.