Sustrans Cycle-Friendly Design Manual (Part 1)

Last spring Sustrans released their Handbook for Cycle-Friendly Design, a relatively short 35-page document which got a bit of a kicking from many people, including David Hembrow and the Cycling Embassy.

This year they’ve released a much longer document in 16 separate chapters, the Cycle- Friendly Design Manual (not Handbook!). This Manual is a whopper – well over 400 pages long, which makes it rather longer than the Dutch CROW Design Manual for Bicycle Traffic.

Given that the examples contained in this Sustrans Manual are almost entirely from the UK, you would be forgiven for leaping to the assumption that there’s probably a good amount of sub-standard stuff in it, to flesh it out to something that outweighs the CROW manual.

And you would be justified in jumping to that conclusion. Some good stuff is being built in the UK, but unfortunately there’s not a great deal of it, and basing your best practice examples entirely on what is found in Britain almost inevitably means you are going to fall short of actual ‘Cycle-Friendly’ Design.

Don’t get me wrong – there’s a great deal of genuinely good advice and guidance contained within these 400+ pages. Probably the majority of it is sound, and in the hands of an enlightened engineer or planner, who wants to do a good job, it could produce some quality cycling infrastructure. The problem is that the good stuff is often accompanied by advice and guidance that really isn’t very good; usually advice that less keen engineers or planners will automatically reach for when things get a bit tricky, or when compromises have to be made – which is, frankly, pretty much all the time, when you are attempting to build cycling infrastructure into a highway environment that has never accommodated cycling properly, ever before.

It’s also not clear what the actual purpose of this Manual will be, particularly at a time that we have a large amount of new stuff from TfL including the new London Cycling Design Standards that will (hopefully) be adopted by the Department for Transport as an England-wide replaced for the pretty dire LTN 2/08, as well as the Welsh Active Travel Design Guidance, and good guides being produced by campaigners.

Who is this Sustrans Manual for? How does it sit alongside the aforementioned guidance? This isn’t obvious.

Anyway, I thought I’d post some comments here on the opening chapters – it’s too big to take on all in one go.

Bear in mind that the stuff I’m picking out here is the bad stuff that has caught my eye. This isn’t comprehensive, by any means, nor is it an impartial review. I’m deliberately singling out things that should be changed, to make this a better manual, principally because (as I’ve already described) it’s the crap stuff that people who don’t care, or who have been forced to ‘compromise, will seize upon.

So Chapter 1, which is an overview – ‘Principles and processes for cycle friendly design’.

This is a pretty reasonable chapter, but it gets off to bad start – the opening lines, and Paragraph 2.13, tell us to

Design in line with cycle training – on-highway design should reinforce how people are taught to cycle in National Standards / Bikeability Level 2, in particular primary and secondary road positioning.

This is simply the wrong approach – in fact it’s completely back-to-front. Much contemporary cycle training, while worthy, involves coping mechanisms to deal with inadequate or flawed road and street design. For instance, the primary position is used to control driver behaviour at hazardous areas of the road – pinch points, for example. It also involves cycling well away from parked cars. So Rather than explicitly designing for a way of cycling developed to cope with hazardous road design, the hazardous design itself should be addressed. Don’t build pinch points. Don’t put cycling infrastructure outside car doors. And so on. (There is no ‘Primary Position’ in the Netherlands, because cycling infrastructure is designed in such a way as to make it unnecessary to unnaturally position yourself in the middle of the road).

This is followed up by some suggestions on the dreaded ‘different categories of cyclist’, where it is alleged that ‘experienced cyclists… place particular importance on directness’ because they cycle on the road. Of course, this group really only appears to place a greater importance on directness because other users are not willing to deal with the stresses involved in cycling on the most direct routes, hence opting for a circuitous route that purchases a little comfort at the expense of convenience. It’s not credible to assume that some people don’t mind being sent around the houses – Every user values comfort, safety, directness – choices between these options are only made in the current British cycling environment because it is so inadequate.

Closely related, we also have the advice

Where more confident cyclists choose not to use any facilities provided their needs should also be addressed with separate provision where appropriate; they should not be compromised by the design

Design should of course be good enough such that ‘more confident cyclists’ do not feel the need to avoid it. It is a mistake to provide two inadequate forms of provision for two different categories of user; if you find someone avoiding your design, you should be asking yourself why, not tinkering with another parallel approach somewhere else.

In this regard, Paragraph 4.9 in Chapter 3 of the Manual is more acceptable, in that it highlights how this kind of parallel provision should only be an ‘interim arrangement’ – ‘the longer term aim should be to design all routes as suitable for the full range of target users’, which is right, but leaves me wondering why the door is left open in this manual to councils opting for the easy option of dual provision, in the first place.

Chapter 3 is entitled ‘Placemaking’.

This is a troubling chapter for a ‘Cycle-Friendly’ manual because in many places it recommends sacrificing the comfort and safety of cycling in order to create ‘place’.

We are told that

Many urban streets are not wide enough to provide separate cycle facilities or have frontage activity that makes such provision impractical. Design for such environments needs to think beyond standard highway design, defining a slow speed highway environment where cycles, pedestrians and motorised traffic can safely integrate.’

and also that

In some streets there is no room to provide standard cycle facilities. Placemaking helps define a slow speed highway environment where cycles, pedestrian and motorised traffic can safely integrate.’

If streets and roads are genuinely not wide enough, or there is not enough room, then measures should be taken to reduce motor traffic volumes to an acceptable level at which it is comfortable to cycle on the carriageway – around 2000 PCU/day.

High traffic levels do not allow cycling to ‘safely integrate’ with motor traffic, particularly if there is a relatively high proportion of HGVs/buses. Many of the examples featured in this chapter – Kensington High Street, Exhibition Road, Ashford, Poynton – have uncomfortably high levels of motor traffic for cycling to be combined with it.

An example from the Sustrans manual. ‘Placemaking’ in action, but certainly not ‘cycle-friendly’

If there is not sufficient width to separate cycling from these traffic levels, then rather than attempting to integrate cycling into it with ‘placemaking’ features, the genuinely cycle-friendly approach is to reduce that motor traffic volume to a comfortable level.

It’s this kind of analysis that is missing from the Sustrans manual – although there are helpful speed/volume diagrams at the start of the manual, describing what kind of provision is appropriate, that approach appears to get jettisoned when the practicalities of designing for cycling on actual streets and roads comes to be discussed.

Indeed, this ‘placemaking’ chapter is essentially all about attempting to accommodate cycling on the carriageway on roads that are still carrying far too much through traffic for acceptable ‘sharing’ – what I have called placefaking, a fudging of the function of roads that are busy with motor traffic. A more helpful approach would be to employ the Dutch Sustainable Safety principle of Monofunctionality, which would involve moving every road and street into a particular category, either one for access (with low motor traffic levels, through design) or a distributor road that serves a through-function, and with appropriately-designed separate cycle provision.

A former through-route for motor traffic in Assen, too narrow for separate cycle provision. But instead of ‘safe integration’, the through motor traffic has been removed.

Chapter 4 – Streets and roads

This chapter sadly follows on from the previous one, with much of the same cycle-unfriendly advice.

In streets with high place function (e.g. high streets or town squares), segregated cycle tracks will generally not be a suitable provision because of the complex pedestrian movements and competition for space with other social activities and parking and loading requirements.

Again, we see – weirdly for a ‘cycle-friendly manual – that ‘place function’ trumps adequate cycle design, regardless of the amount of motor traffic a particular road or street is carrying.

Of course cycle tracks can and do work well on high street locations, and places with parking and loading requirements.

A cycle track on a high street location, with parking and loading and ‘complex’ pedestrian movements. Not a problem.

The elephant in the room here, however, is volume of motor traffic, just as with the previous ‘placemaking’ chapter. If motor traffic on particular street is above 2000 PCU/day, then separate provision for cycling should be provided, immaterial of the street context. If it is not practicable to achieve this – either due to the width of the street, or genuine complexity with other social activity, then motor traffic levels should be reduced below 2000 PCU/day, to create a genuine place. It is pretty ridiculous to suggest that high streets carrying large amounts of motor traffic can’t accommodate cycling infrastructure because that would interfere with ‘place’, but that appears to be exactly what this Sustrans manual is doing.

As it happens, paragraphs 3.2 and 3.3 in this chapter provide sensible limits for motor traffic levels for acceptable sharing with cycling (1500 vehicles/day, or 3000 vehicles/day, in slightly different contexts). However paragraph 3.4 suggests that sharing at up to 6000 vehicles/day ‘should be considered’ in locations with a high place functions. Such a level of motor traffic (600-700 vehicles per hour, or 10-12 a minute, in peak) pretty much renders any ‘place function’ moot.

“Quietway” https://t.co/wWrxSwY0UY

— Joe Dunckley (@steinsky) June 4, 2015

Again, at this level, some form of separation should be provided, and if it can’t, motor traffic levels should be reduced.

This strange fudging is repeated later in this chapter, under a section on Mixed Priority Routes –

Mixed Priority Routes (MPR) are streets with a mix of land uses (commonly commercial and residential frontages) that also carry high levels of traffic. MPRs have important movement and place functions and need to accommodate a diverse mix of road users – pedestrians, cyclists, passenger service vehicles and passengers, motorists – and parking and deliveries.

Streets that have a ‘movement and place function’ should be moved into one category or the other, as per Sustainable Safety. It really isn’t acceptable to mix in cycling with through traffic on streets that are alleged to have a place function; either the street should have motor traffic levels reduced below 2000 PCU/day, or cycling should be separated from that motor traffic.

Shared space naturally makes an appearance too in this chapter, but there’s far too much emphasis on this design technique as ‘cycle friendly’ without any reference to maximum traffic levels for ‘sharing’.

Shared space design principles can be applied to links and junctions, including junctions with significant traffic flows and HGVs.

I’m sure they can be applied, but is sharing space with significant traffic flow ‘cycle-friendly’? Almost certainly not.

Shared space environments can be convenient and attractive to cycle users. Although many schemes include narrow lane widths, cyclists can mix comfortably with traffic because of the very low speeds.

Poynton is famously invoked as one of these ‘low speed’ shared space environments, but I challenge anyone to argue that this kind of environment – slow or otherwise – is ‘friendly’ for cycling.

 There’s just too much motor traffic; it’s deeply unhelpful to include environments like this in this kind of manual without reference to motor traffic levels.

It’s really disappointing, especially when other stuff in this chapter – like cycle streets – are explained and described well, with clear limits (2000 vehicles per day) on motor traffic levels.

Good recommendations. So why not apply this rigour to other streets and roads?

Another intervention –  homezones – is described in a peculiar way –

The layout [of homezones] discourages through traffic and reduces vehicle speeds to less than 20mph

Homezones should be designed to prevent through traffic – ‘prevent’ should obviously be substituted for ‘discourage’.

There’s also a lengthy section on ‘Community street design’. While worthy, experience with these kinds of projects is starting to demonstrate that asking the community to make changes they want to see to a street won’t necessarily result in changes that are ‘cycle friendly’.

Community-led highway design. No accommodation of cycling. (An example from this Sustrans manual)

It’s pretty naive to expect outcomes from these kind of projects to be ‘cycle-friendly’ – so why include this approach at all in a manual that should be about high-quality cycling design?

There is, unfortunately other rubbish in here too. Pinch points –

The left hand side is a recommendation.

Cycle lanes arranged outside car parking, which should be a complete no-no on through routes for motor traffic –

A weird recommendation for a bus stop bypass that sends people cycling on the road onto a shared use footway, right at the bus stop –

… As well as a suggestion that ‘wide general traffic lanes’ are an acceptable way of passing stopped buses. (Again, it would be helpful here for some kind of motor traffic volume indication of when it is acceptable to direct cycling around the outside of stopped buses – presumably <2000 PCU/day).

And finally there are also poor examples of cycle (‘partial’, whatever that means) priority across side roads –

To repeat, this manual is mostly composed of good advice – you might not get that impression from what I’ve focused on here. But there shouldn’t be any place for this kind of inferior design, or substandard recommendations, in such a lengthy manual, because that is what will get picked out by councils who are not committed to doing a good job.

If a council is faced with a choice between reducing motor traffic levels to a genuinely acceptable level for sharing the carriageway, or a Sustrans recommendation that sharing is acceptable on ‘Mixed Priority Routes’, or that cycling can be ‘safely integrated’ on roads with heavy traffic – which will they pick?

If a council is faced with a choice between designing proper protected cycling infrastructure on the inside of parked cars, or painting a crap cycle on the outside of them, as per Sustrans guidance – which will they will pick?

If a council is faced with a choice between removing a pinch point and providing a safe convenient design for people walking and cycling, or painting a bicycle symbol in the middle of a 3.1m pinch – which will they pick?

And so on. The crap needs to go, because that’s the stuff that will be chosen.

More to come on the remaining chapters next week…