Report by Stephen Donnelly on Cycleway 9

Safety issues of two-way track 'have not been adequately addressed'

The full text of the report by Stephen Donnelly Chiswick resident, architect and cycling campaigner, to Hounslow Council.

It is abundantly clear that a fundamental error was made at early design stage, to progress a two-way track design in a location where it is unsuitable, and there has been a concerted cover up ever since, as the more a project progresses it becomes increasingly difficult to make major changes.

There has been no strategic or objective risk assessment by either TfL or Hounslow Council Officers, both of whom have continually ignored warnings over the past two years that the design is unsafe. Seeking to justify their stance, Hounslow Officers have now produced a highly defensive and unobjective report, with a very belated attempt at risk assessment that is seriously flawed.

There is very little evidence that either TfL or Hounslow Officers fully understand the two-way track safety issues. It is clear they care more for policy objectives than for the safety of local cyclists who will be guinea pigs on this scheme, which is unprecedented in the UK. No other high street has a two-way cycle track with cycle priority over multiple side road junctions, of which there are 30 on the CS9 route.

Many other municipalities, mainly overseas, have made the same mistake when they first start providing segregated cycle lanes. The same type of unanticipated or underestimated collision patterns repeatedly occur, often leading to a sharp increase in the number of serious injuries to cyclists within the first few years, which then necessitate the schemes being ripped out and undergoing significant modification.

This happened in Royal College Street and Gordon Square in LB Camden within the last ten years, and these locations had far fewer side roads than CS9. This is why design guidance was developed, warning against using two-way tracks on roads with multiple side roads. The UK guidance is largely based on overseas guidance, particularly from the Netherlands. For further information, please see the attached document listing some of this guidance, along with my own risk assessment. Now we have this guidance, there is absolutely no reason for Hounslow to make the same mistakes - we simply need to heed the guidance.

Both TfL and the Officer's Report ignore the world expert opinion secured by Chiswick's Redesign CS9 group from expert Dutch designers and safety auditors (see attached) which concludes that the design should be reassessed and the safer option of one-way tracks on each side of the road revisited.

Based on the Officer's report, it is clear that the only responsible decision Cabinet can make to exercise its duty of care is to defer any approval of the current design until it has been objectively assessed by an independent expert. As there is insufficient experience of this design in this country, this should be done by an expert Dutch consultant such as DTV Consultants are employed to do this in collaboration with TfL and Hounslow Officers. DTV have designed and audited cycle networks in Amsterdam, Rotterdam and Utrecht, regarded as the best in the world, and are collaborators on the internationally renowned "CROW Design Manual for Bicycle Traffic..

I want to make it clear that in all of this I do not seek to frustrate the project, I fully support the principles, but am motivated by the need to protect local people from needless injury, particularly those who may be lured to cycle on the track because it will look safe particularly for those too frightened to cycle now such as people with children. I also want to ensure that the wider promotion of cycling in London is not held up as a result of the provision of discredited infrastructure that then needs to be rebuilt at considerable expense and with considerable disruption.  And I want to ensure my local High Road remains renowned for its boulevard style pavements, shops and cafes, and not for yet another failed cycling scheme in London.

The Hounslow Officer's Report is seriously flawed on safety issues (paragraphs 3.71 - 3.137) for the following reasons:

It ignores the warnings against two-way tracks in principle on roads with multiple side roads, instead jumping to the detailed section on junction design in an attempt to divert attention from the fact that the design is fundamentally non-compliant with the design standards

It then seeks to create the impression that the design is in accordance with the London Cycling Design Standards (LCDS) by referencing certain detailed requirements of the standards, often in error, for example:

Para. 3.88 seeks to justify the lack of the recommended 5 metre long waiting space between the cycle track and road in TfL's design, by asserting that the risk is low because the number of vehicles turning across the track at side roads is low. They incorrectly refer to a table in Section 4.1.6 which should be used to choose the type of cycling facility along a road, and claim this can be applied to turning movements across a track. In fact the most relevant part of Section 4.1.6 is the warning "The model of using segregated two-way tracks on one side of a street should be applied very selectively." which the report completely ignores.

The erroneous reference to section 4.1.6 of the LCDS should in fact refer to section 5.3.4 (page 25) which does not state that omitting the 5 metre waiting space is acceptable as the report claims, but that it "is possible but brings with it various risks...it is not recommended unless traffic speeds and volumes are very low....even then subject to site specific risk assessment. Closing side streets to motor traffic is likely to be the only reliable way of dealing with these risks." No side streets are to be closed. This is substantially different advice to that conveyed by the Officer's report. It also does not require only turning movements across the track to be "very low" but traffic volume in general, as drivers reactions and the time they give to scanning for cyclists when turning into or out of side roads is hugely influenced by how busy the main road is.

The report also uses a report by WS Atkins engineers on two-way cycle tracks in Kingston Upon Thames which apparently justifies up to 370 vehicles per hour turning across cycle tracks, which can hardly be described as "very low". This is a strange point of reference since neither WS Atkins or Kingston are recognised authorities on this subject. A comprehensive study of Helsinki's network of two-way tracks demonstrated the two-way safety problem at side road junctions was evident even where the tracks only crossed private driveways to private properties. The Atkins report appears to be at odds with wider international experience.

Para. 3.80 demonstrates a common misconception about side road safety on two-way tracks, suggesting the safety issues result from drivers not being aware of the track or that cyclists have priority, and thinking that problems will abate once people become more aware. Lack of awareness is a factor, but there is widespread evidence, included in the Dutch CROW design guidance, that lack of visibility and the unintuitive layout that is the major factor. The Helsinki study showed collisions were proportionately more likely among drivers and cyclists that use the route regularly, demonstrating that intuitive behaviour rather than awareness is the crucial factor, as drivers instinctively react to the largest perceived risk. At Royal College Street in Camden railway level crossing style warning lights were installed to make drivers more aware of the two-way track, but serious injuries continued, and the track had to be rebuilt as one-way tracks. 

Para. 3.120 correctly concludes that reference to design guidance and site specifics is the correct way to assess the likely safety of a design, yet the analysis in the Officer's report chooses to ignore the most relevant parts of the design guidance. Instead it relies on an erroneous reference to an irrelevant part of the standards and an obscure report by organisations with very little experience of a design unprecedented in this country.

Para. 3.120 acknowledges the detailed work by Redesign CS9, but completely ignores the Dutch expert's recommendation that the design should be reassessed and the safer option of one-way tracks on each side of the road should be revisited. The Officers suggest academic journals claim the design standards may be incorrect, and assert that our analysis that shows the design is likely to be 3 to 4 times more dangerous than the existing conditions "may be incorrect". Firstly, the Dutch data is not from an academic journal, it is from the Dutch CROW Manual - the world leading design guidance. Nor is the assessment by an academic - it is by a leading Dutch engineering consultancy (DTV) with extensive real world experience of designing and auditing the best cycle networks in the world.

Secondly, claiming the data "may be incorrect" implies significant doubt that demands a full objective assessment to dispel. The Officers state data from overseas may not be applicable to West London, particularly because mitigating measures have been provided. There is no reason why Chiswick High Road is any different to other similar locations overseas, and it is precisely the kind of street that is least suited to a two-way track. The design guidance makes it clear that mitigating measures are just that - they only partially mitigate the risks. If the Officers assertions were true, given that design guidance is generally written from a pro-cycling basis, it would read something along the lines of "two-way tracks may be better than no cycling facilities if mitigating measures are provided" but it does not. It recommends against two-way tracks, unless in exceptional circumstances where they can be justified by risk assessment.

Para. 3.93 states the design has already passed a Stage 1 Safety Audit and therefore it is not possible to change to one-way tracks. I have obtained the Safety Audit from TfL via a FoI request and found it to be wholly inadequate. It lacks any form of strategic risk assessment, and is instead a technical audit carried out after the design was completed when it was too late to make major changes. Most of the auditors recommendations for greater cyclist safety were turned down by the designers because of a lack of space.

While technical audits may be adequate risk assessment for well established highway designs, they are clearly inadequate for unprecedented designs because they are completed after the design work, and do not inform the principles of the design. Given there is so much design guidance against two-way tracks on urban streets with side road junctions, it is not unreasonable to expect that TfL/Hounslow would have carried out proper risk assessment and have referred to design guidance from countries with experience of such designs, and that is widely available in the English language.

It should also be noted that TfL's London Cycling Design Standards are the most permissively worded when it comes to two-way tracks even though there is very little experience of them in London. This is in contrast to other design standards such as those published by the Department of Transport, or overseas guidance such as the Danish Collection of Cycling Concepts or Dutch CROW Manual, which form the basis of the UK standards.

Para. 3.121 claims one-way tracks may not be acceptable. I have looked at this and it is possible within the same footprint if a more flexible approach is taken, possibly using lighter segregation at pinch points, and could maintain junction capacity. The design could also save two trees that would be lost, which the Conservation Area appraisal describes as framing important views into the Area at the junction with Chiswick Lane. It should be noted that to provide a two-way track with the recommended 5 metre long waiting space (in the exceptional circumstances where a two-way track can be justified) takes more road width (at junctions) than to provide safer one-way tracks on both sides of the road. Council Officers did not take up my offer to meet to discuss this option, or to seek advice from the Dutch consultants.

Perhaps I should also outline that there are other quality issues, not always specific to safety, with the current CS9 design:

The lack of the 5 metre waiting space between the track and road will mean vehicles entering and exiting side roads will frequently obstruct the track, making faster cyclists less likely to use the track

TfL's report states the route will make cyclists journeys slower because of the additional signalled pedestrian crossings, again faster cyclists will avoid using the track or be tempted to jump red lights

Eastbound cyclists will need to cross the road twice - to get to and from the track on the other side. This increases injury risks and many will avoid using the track

The track will be narrow for its intended use, with frequent changes of direction around trees and changes of level - it will be far below the standard of the recently built Super Highways in Central London, making it unattractive to faster cyclists

There are more collisions between cyclists on two-way tracks than on one-way tracks - these account for around 15% of cyclist injuries in the Netherlands and are a problem on the newly constructed Super Highways in Central London, including a fatality

The Dutch expert advises the two-way safety issue is also a problem for pedestrians crossing two-way tracks, although there is little published evidence

The segregated track will abruptly end at the CHR junction with Heathfield Terrace, mixing novice cyclists with traffic using the road to get to the A4 via Sutton Court Road

Cycling and walking conditions will be made worse on Kew Bridge by the introduction of an extra traffic lane, removing the wide southbound lane, narrowing the western pavement and banning shared use cycling on it

The segregated track through Watermans Park in Brentford will be a personal safety risk at night and will detract from the amenity of the park with fast cyclists passing close to children playing and dog walkers - TfL cost cutting should be reversed to provide a safe route alongside the road 

Clearly the only responsible decision Cabinet can make now to exercise its duty of care is to defer approval pending an objective and independent reassessment of the design. 

It is unfortunate but telling that TfL and Hounslow have sought to avoid giving the safety issue the scrutiny it deserves, by denial, avoidance and now seeking to hurry the decision through over the summer holiday period. To date we have sought to address this issue collaboratively and discreetly with the authorities, but in recent weeks this has not been reciprocated and we will now seek to give the issue scrutiny via the media, who have expressed a strong interest in this story.

  August 30, 2019