Should Speed Limits outside schools be lowered to 30 km/h?

Variable speed limits of roads outside schools.

Below is a copy of my submission on the proposal to lower the variable speed limits outside schools to 30 km/h. I fear that the increasingly the seemingly perfect is becoming the enemy of the good.

The focus on speed limits at the school gate perpetuates the myth that crossing the road outside school is the main danger. This is well managed in NZ. Our safety record for children crossing roads outside schools is the envy of other countries. So the focus on the school gate is largely misplaced. It gives the appearance of caring for child safety, while achieving very little at significant expense.?Crash records clearly show that crossing the road outside a school is by far the safest part of a child’s journey to and from school.?It is on the rest of the journey home, and during play after school where the injuries happen. The focus on the school gate, is diverting resources that should be devoted to the surrounding area.?It would be much more important to invest in safer routes to school programmes, and neighbourhood accessibility plans that manage traffic speeds and provide for safer crossing facilities in the neighbourhood catchment of the schools. The biggest difference to the safety of my children walking to primary school was when the council built a pedestrian refuge and kerb buildouts across a collector street that my children had to cross, which was about 400 metres away from the rear entrance to the school. Permanent 40 km/h speed limits in the residential neighbourhood streets would be much more effective than at the school gate.

The road safety message needs to change from “slow down outside schools” to “slow down when children are about”?

30 km/h or 40 km/h school speed limits.

I was rather surprised that reducing the value of the variable school speed limit in urban areas from 40 km/h to 30 km/h was not in the survey monkey questions.

The case for 30 km/h is not as compelling as first appears.

I was NZTA’s subject matter expert on pedestrian safety, and have a deep understanding of the effect of impact speed on the risk of death and serious injury to pedestrians. The usually quoted figures on the risk of death at various impact speeds are wrong. The briefing to the associate Minister of Transport in 2018 of school speed limits was based on these risk of death curves that have?been known to be wrong since 2010. They quote a 10 percent risk of death at an impact speed of 30km/h. This is based on old research by Eero Pasanen of Finland who used data from an English study by Ashton of serious crashes analysed in-depth by that study.?So the sample was highly biased. Eero was on the OECD committee with me that published the Pedestrian Urban Space and Health report.?He told the committee not to use the old values.?Instead he emailed the chart below based on a much better German dataset by Rosen, that was corrected for bias in crash reporting rates.

Note;??See discussion below. The limitation in the German data is that it only included passenger cars, SUVs light trucks and vans. Impacts from heavy?vehicles are much more severe. It only included adult pedestrians.

The left chart is collision or impact speed. The 10 percent death risk happens at 50 km/h not the 30 km/h popularly reported.

The right chart is driving speed. The risk of death at 50 km/h is halved again. Note that the benefit from a drop in driving speed from 50 km/h to 40 km/h is much greater than from 40 km/h to 30 km/h. Any braking by drivers means that the impact speed is lower than the approach speed. Because at lower travel speeds a driver travels less distance while reacting,?then commencing to brake and there is also a shorter braking distance. With the law of physics also on our side as braking distance is proportional to the square of vehicle speed. The result is that where the driver has time to brake, a drop in approach speed results in a much greater drop in impact speed. So an approach speed of 40 km/h is actually quite good. However it should also be noted that in some of the “dart out” crashes, the driver does not have time to brake.?

The original English data from Ashton was re-analysed by Davis in 2001 and adjusted for the bias in the sample.?He also divided the pedestrians into different age groups.?The result quoted in an international review by Rosen in 2010 is shown below. Ref;?Literature review of pedestrian fatality risk as a function of car impact speed Erik Roséna,?, Helena Stigsonb,c, Ulrich Sander.?https://transportsafety.ir/wp-content/uploads/Courses/UrbanRoadsafety/Literature-review-of-pedestrian-fatality-risk-as-a-function-of-car-impact-speed.pdf .

Of particular note is that the risk of death curve for child pedestrians only starts to trend up significantly above 40 km/h. Old people have more fragile bones and it is their risk that increases after an impact speed of 30 km/h.??

A more recent review of the literature by Hamish Mackie of NZ was never published, but came to similar conclusions.?

I presented on pedestrian safety at an ACRS workshop in Melbourne. The specialist in crash reconstruction from the serious crash unit of the Victorian Police also presented and told me that out of over 50 passenger car vs pedestrian impacts they had no fatalities at impact speeds below 40 km/h.

So the basis on which 30 km/h has been chosen as the 'must have" speed limit for outside schools is based on outdated information that has been proven to be wrong, yet keeps on getting repeated by road safety authorities.??

I was on the team that first trialed 40 km/h variable limits outside schools in Christchurch and can recall spending hours with a laser speed gun measuring before and after speeds. The results are reported in a paper by Wayne Osmers; https://www.transportationgroup.nz/papers/2002/12_Osmers.pdf . It found that when the variable 40 km/h sign was displayed and children were not present, and school patrols were not operating, speeds dropped noticeably by about 10 km/h to about 46 km/h. Once the wardens or school patrol operators in their red fluoro safety jackets appeared, and where for a warden system they put out cones to reinforce the no-stopping zone, the speeds dropped to 40 – 42 km/h. However the two control sites on urban roads where variable 40 km/h sign were not installed, also had mean speeds of about 40 -42 km/h when the school patrol was operating, and the patrols were able to operate safely.?So there was no point in installing variable speed limits?at such sites. In my observation while measuring speeds at these sites, the way the schools operated the crossings outside the school, such as the wearing of bright jackets by the wardens and school patrol operators, and the use of traffic cones to mark off the no stopping areas near the crossing points, were much more effective than the variable speed signs. As a result of the research, variable 40 km/h speed signs were recommended only for locations where operating speeds during school arrival and departure times were higher, because these were the only locations where there would be a benefit. The priority for now should be to get variable 40 km/h school speed limits where children need to cross faster roads to get to school.

The 40 km/h school limits achieved good compliance, and the distribution of speeds narrowed closer to the new mean. My experience setting speed limits and monitoring traffic speeds over a long career, is that if you set a speed limit below what drivers consider reasonable, you will get compliance by a very law abiding group, but for others they will travel faster than if a higher but seemingly more reasonable limit had been set. It is these faster vehicles that cause the most risk, and research consistently shows that pedestrians underestimate the approach speed of faster vehicles. Reducing variable speed limits from 40 km/h to 30 km/h at the school gate may not achieve willing compliance so may bring little benefit and cold possibly be detrimental.??

??So while 30km/h might seem the perfect limit in urban areas from a safe system perspective, there are also practical considerations around willing compliance. 30km/h variable limits have not been trialed anywhere outside schools, so have not been demonstrated to have achieved corresponding reductions in mean traffic speeds near 30 km/h.

I recommend that the default 40 km/h school speed zone that exist now be retained with 30 km/h trials taking place at place at urban schools and thoroughly evaluated like we did when 40 km/h limits were introduced.?

Along with Brian Neill of Christchurch City, and Wayne Osmers of NZTA, I was one of the instigators of school speed zones. The original concept was of a zone around schools where speeds would be managed down by a variety of means in our tool box of which only one was variable speed limits.?When the officials and politicians in Wellington got hold of the proposal they turned them into a speed limits only concept, applied only near the school gate. It is good to see a broader focus on a broader suite of tools for speed management.?I support the flexibility in the definition of roads outside schools. However I fear that the implementation will revert to the school gate only and variable speed limits only focus.?

Tim Hughes, SafenSussed, July 2021