Friday 25 July 2014

New infrastructure. A real improvement or making stop-start cycling even slower ?

This piece was originally titled "New British Infrastructure..." and the bad examples shown below are all from the UK. However, the exact same problems are seen around the world.

Cyclists often win "commuter races" because of their ability to get through traffic jams which hold up both motorists and public transport. Many existing cyclists enjoy the fact that they can make fast journeys have have predictable journey times. These give cyclists major advantages over using other modes of transport.

If cycling is to be spread wider through the population then other people need to see the same advantages. There is no point in giving people a choice between being able to ride slowly on inadequate cycling facilities or having to "take the lane" on the roads. Neither of these two options is good enough. Cycling must be both safe and convenient and we should not be asked to choose between one or the other of those two desirable characteristics.

Making stop-start cycling even slower

Unfortunately, many people who plan for cycling in the UK don't seem to understand the need for efficiency when cycling. Rather than creating cycling facilities which make cycling both convenient and safe, there is a consistent problem across the UK that facilities are designed to such a low standard that they are neither convenient nor safe.

From a Southend council website, the
Prittle Brook Greenway. Nice photo of
recreational use on a sunny day but this
is simply too narrow for a bidirectional
shared use path and its inadequate as
a main cycling route. Riders appear
to be approaching one of many places
where they will have to cross a road.
I wrote two weeks ago about Southend's lacklustre attempts at encouraging cycling. Quite apart from having used almost all their funds for two expensive schemes which are designed for motor vehicles and ignore the needs of cyclists, the blind corners, narrow shared-use cycle-paths and road crossings of astonishing slowness seen elsewhere demonstrate how the town is not working seriously to provide efficient go-everywhere routes for cyclists.

In response to my blog post there were several suggestions that I should especially have looked at the Prittle Brook Greenway because this is a highlight.

I decided in any case to look up what I could find about the Prittle Brook Greenway. The first item to come up was Southend's own information, a small part of which is reproduced here. The second item was a local campaigner's description of it as "a 2 mile long, 2 metre wide shared use path that involved 19 road crossings". Follow the link to see a surface which washes away in the rain on a path with blind corners and low social safety. A local civil engineer pointed out the lack of lighting and a previous problem with dog excrement as well as a later post discussing problems with crossings and with vandalism.

Would cyclists need this
advice to ride on Southend's
best cycle-path
if it were
built to a proper standard ?
The campaigner's description goes some way to explain why Southend council's official recommendations for this route suggest that under 12s should be supervised, that there should be "no racing" and that cyclists must "Look out for vehicles" and "take care" around pedestrians. I've seen paths like this all over the UK and I'm not sure I missed much by not seeing this particular example. It can be OK for an outing with children on a sunny day but this is not the sort of infrastructure which leads to mass cycling.

The truth is that I missed the Prittle Brook Greenway before because it didn't go anywhere that I was going. The most important result from Dutch research in the 1970s was that a dense grid of very high quality routes is required. i.e. Good infrastructure shouldn't be something that you have to look for, it needs to be everywhere.

Even if it is of very high quality, very direct and socially safe (unfortunately, Prittle Brook doesn't tick those boxes), a single path cannot achieve much on its own because it simply cannot serve every location. Southend has actually spent a lot of money. They could have transformed the town but they've instead directed it at the wrong projects. Neither a high enough quality level of facility nor the required dense grid have been provided for cycling and that's why cycling remains a niche mode of transport in the town.

Brand new infrastructure
in London's Olympic park.
Not an uninterrupted and
direct route. It's also very
obviously too narrow for
a bidirectional route.
Source: @twistandshout
I visited London last year. From a cycling point of view, I took two things in particular away from the visit:

Firstly, cyclists in London are a rare breed. Just as everywhere else where cycling is for a small minority, those who do it mostly fit into a particular demographic. School trips by bike are difficult to imagine taking place. You don't see a lot of pensioners or disabled people cycling either. Cyclists in London must be alert and constantly looking out for where the next threat will come from. Cycling isn't a way of getting about which anyone can choose as most people simply won't consider it in those conditions. Those who cycle in London have reasons to do so.

Secondly, there was an overwhelming sense of frustration and slowness. On Dutch cycle-paths, I expect to be able to get up to speed and to be able to keep cycling with the minimum of interruptions until I reach my destination. I rarely have to stop at traffic lights and am rarely inconvenienced by motor traffic. That's not at all how it is in London. Cycling in London feels quite competitive. People jostle for position in ASLs, there's a lot of sprinting from one traffic light junction to another, but overall progress is slow. Average speeds over a complete journey, even over short distances, are low. Compared with what I'm used to, I found cycling in London to be irritatingly SLOW and I was amazed by how long it took to cover even a short distance.

London has not learnt the importance of providing infrastructure of a high enough standard. The "superhighways" are well known to be not super at all but rather than copying from best practice and providing adequate funds to do a good job, London is still going its own way, thinking up new and absurd ideas, producing lots of press releases and trying to build the actual infrastructure on the cheap. For instance, less than a year has gone by since London proposed an absurd multi-stage right turn which demonstrated a complete lack of understanding what is required to make cycling both safe and efficient, and they've continued right up to the present time with designing and building infrastructure which is way short of the standard which is required for true mass cycling.

Amazingly, Bedford once had this
as a proposal. It's close to a proper
safe Dutch roundabout
in design.
But they scrapped that...
via aseasyasridingabike
The continuing "#Turbogate" scandal in Bedford, where they claim Dutch inspiration for a roundabout design which is fundamentally different from real safe Dutch roundabouts, dates back a long way. I was first involved in trying to advise a Bedford planner about how Dutch roundabouts really accommodate cyclists way back in 2011.

Sadly, rather than accept the widely published and very reasonable criticism of their flawed design, Bedford and their backers (including Sustrans) have not accepted the widely published criticism but have continued to push a design which divides cyclists into two groups: those who will "take the lane" on the busy roundabout and those who will be too timid to do this and will instead use pedestrian crossings to pass the roundabout. Dutch cyclists never have to choose between "safe" or "fast". With one minor change which might actually make the roundabout more dangerous, building of this flawed and dangerous design started this week.

No on-carriageway markings
Carriageway exit angle too sharp
Double Yellow Lines over c/w exit
Update 2015: A downloadable report is now available about this roundabout. One image, along with its caption, is shown on the right.

I recommend reading the report in full. Amongst other conclusions, it is said that "Arrangement of cycle crossings maximises possibility of conflict between cyclist and pedestrian"which shows the many problems caused by the design used in Bedford.

This mistake should not be repeated.

Northern Ireland
On-road cycle-lanes are amongst the more useless types
of cycling infrastructure and they should never be this
, but this is one of the better examples of
infrastructure in the DOE video. The width problem is
demonstrated in the video by the cyclist swerving out
before the grey car because she thinks the blue car won't
stop for her. via NIgreenways
Northern Ireland has its own problems. The nigreenways blog pointed out this week that the Department of Environment, "where local road safety promotion sits in isolation from the road builders", has produced a jaw-droppingly bad road safety video which gives ludicrous recommendations for how to stay out of trouble, such as that people should "avoid the busiest times" (i.e. don't use cycling as a practical means of transport for such things as going to work or to school) and makes one excuse after another for the terrible quality of the infrastructure which they have built.

It's the same problem as seen elsewhere in the UK. Northern Ireland's cycling infrastructure is not fit for the purpose of making cycling attractive and convenient for the entire population. That's why Northern Ireland, like the rest of the UK, has a very low cycling modal share.

I've written quite often about Cambridge because we used to live and campaign there, watching local infrastructure change at a glacial pace.

Cycling in Cambridge has perhaps the greatest help from demographics and local laws of any city in the world, yet it doesn't not have the highest cycling modal share of any city in the world. I suspect that if the city had infrastructure like Dutch cities then Cambridge could quite easily have a higher cycling modal share even than Groningen. Unfortunately, the current level of political commitment and investment prevents cycling in Cambridge from growing to its full potential.

While the city has the highest cycling modal share in the English speaking world, it doesn't make a useful model for other places because the causes of the relatively high cycling rate of Cambridge are not things that can be duplicated. Much of the cycling in Cambridge takes place despite the infrastructure rather than because of it.

Since we left the city there have been changes, but they've not made a very big difference. There is still no sign of the comprehensive grid of high quality infrastructure required to get everyone cycling. For instance, it took 12 years of discussion and a lot of campaigning before the busy Gilbert Road was changed, but the outcome even on this one km length of road wasn't nearly what it could have been.

It's part of the plan to deliberately
make the cycle lane narrower when
it passes the bus stop, specifically to
slow cyclists down. That creates
danger for cyclists and should not
be a feature of a real cycling facility.
Source: Cambridgeshire CC
More recently there have been proposals to change Hills Road and Huntingdon Road. These proposals have been a long time coming and were eagerly awaited. I used Huntingdon Road regularly when we lived there so I looked at this part in detail and explained in a long comment on the Cycling Embassy of Great Britain website the many problems with the design for this short length of road. Please read that link to find out more.

To summarise, they're planning to change just one side of approximately one mile of road (you're on your own on the return journey). There are many problems packed into this distance, which will make for a far too varied and "exciting" mile of cycling. Consistency may be boring but everyone knowing what comes next is a safety feature. Just one of the issues with the plan is enough to illustrate the gulf which exists between what Cambridge really needs to enable cycling for everyone and what the plans which currently exist in the city promise: "the plan is for this already narrow cycle facility to be even narrower and even more dangerous at bus-stops. Here it will shrink to a totally inadequate 1.5 m, and this is to be done deliberately in order to slow cyclists down. If this were a cycling facility designed to actually facilitate cycling then it would try to make cycling faster and more convenient, not to make cycling slower."

After the bus stop design which makes
you slow down you then come to this
junction at which you'd better speed
up and "keep your wits about you".
A lethal junction design. New for
Cambridge but exactly like junctions
not seen in Assen since the 1970s.
That's just one of the problems. There are other faults such as being too narrow for safe passing and lethal junction designs and that it gives up entirely before the biggest junction. But in my view the most damning thing is that even this much awaited very short length of facility is not actually to be optimized for cycling at all but will instead include features deliberately designed to make bikes slower and therefore also less attractive.

Rather than seeing the existing level of cycling in Cambridge as good for the city, good for the people of the city and as something to encourage more of, planners in Cambridge view cyclists as if they are a problem which it is somehow their job to solve. This is precisely the attitude which has prevented Cambridge from achieving its potential as a cycling city.

Leeds / Bradford
What happens at large traffic light and
roundabout junctions in Leeds now ?
BBC Radio Leeds rode along a cycle-
path new enough not even to be open
yet and found many problems in a
short length of cycle-path. Amongst
them are that the new facility simply
ceases to exist at major junctions,
where it is needed most.
The Leeds-Bradford cycling "superhighway" proposal was announced with the customary blitz of publicity surrounding all new cycling schemes in the UK and achieved some considerable press already, but like most new infrastructure in the UK it's designed in such a way that it will be far from convenient to use.

Its telling that the official blog of the project (they have their own strange blog software so it's not possible to link to the correct page) includes statements like the following:

"Cycling in a segregated cycle lane [...] might be slower, for instance when they are congested, or when we are required to make two-stage right turns, waiting for signal phases." i.e. they're planning to make the cycle-paths too narrow so that it's difficult to pass safely and to use junction designs which are already known to be both inefficient and unsafe for cyclists.

"Many current cyclists are used to speeds of 20 miles an hour or more (going downhill) and a lot of current cyclists are, for want of a better phrase, speed freaks. It will not be easy to maintain this along all of the segregated routes." i.e. existing cyclists for whom journey time is important (that's everyone who cycles if they're late for an appointment, for work or for school) as "speed freaks". Rather than being seen as a group of users who need to be supported and assisted to make their journeys in safety, they're seen as a problem. Rather than planning for the entire population, these planners want to design infrastructure for a mythical person who is never late, never in a hurry and has all day to meander around and to work out how to use their complicated design. They are asking people to choose between "safe" on their cycle-paths or "fast" on the road. No-one should be asked to choose between those two things.

There's nothing "super" about these "superhighways". The criticisms which have come up are not of segregated cycling infrastructure in and of itself (there are no speed limits on Dutch cycle-paths) but of the poor designs which are being pursued for this project.

The City Connect project in Leeds and Bradford has designed down to extremely low standards (to read more, see The Alternative Department of Transport's extensive criticisms of the plans). By building cycle-paths too narrow for safe use at speed, and by implementing copies of junction designs known to be both dangerous and inconvenient rather than copies of the safest and most convenient designs of roundabouts and traffic light junctions, they are guaranteeing that the most efficient place for a cyclist to be along this route will remain the road and not the cycle-path.

But aren't these places "making a start" ?

The King's Hedges estate in Cambridge
is one of many examples of where not
quite good enough infrastructure for
cycling was not improved upon or
linked with other areas to create a
useful grid. The result is that these
paths are not heavily used because
they do not lead to many destinations.
This could have been "a start" over
30 years ago but it was not.
When examples like these are criticised, it's quite common that someone will suggest that "it's a start". People have a hopeful idea that building substandard infrastructure now will somehow lead to better infrastructure later. They'll say that "everything has to start from somewhere", "something is better than nothing" or suggest that providing infrastructure like these examples works to allow people to "gain experience" or as a "gateway drug" to more cycling in the future.

Unfortunately, none of this is true. There is no "tipping point". i.e. no level of cycle-usage from which only growth is possible and no level of cycle-usage from which an increase in funding for cycling in inevitable.

What's more, none of this is new. There already exist dozens of existing examples of older inadequately designed infrastructure which wasn't quite good enough when first built and which remains not good enough decades later.

We don't need to keep repeating the same experiment and achieving the same result. We have known for decades that isolated islands of not quite good enough infrastructure do not work as this was the most important result from Dutch research in the 1970s. Islands cannot create the comprehensive good conditions which are required for true mass cycling and they cannot demonstrate the potential for such a comprehensive grid. All that is achieved by repeating these errors is that time passes without real progress being made. That is why the UK and other countries are more than forty years behind the Netherlands.

The Dutch already avoid these mistakes

When you're working towards a high cycling modal share, it's important to make cycling as convenient, as pleasant and as safe as possible. All these things must go together. To encourage everyone to cycle, facilities must enable all types of cycling.

While planners in the UK spend time trying to make cyclists slow down, the Dutch CROW guidelines emphasize the need to speed cyclists up. The Dutch work to remove those things which could slow cyclists down or lead to discomfort when cycling as these can put people off cycling.

In the Netherlands, having to stop frequently when cycling, having to cycle slowly, having to ride single file, not having ridght of way and not having enough easy turnings all count as nuisances. These nuisances are recognized as working against cycling and must be reduced.
In the Netherlands, cyclists are seen correctly as already a slow means of transport. Even the fastest cyclist does not reach speeds which are easily reached in a car. A requirement for "low speed at conflict points" is met by slowing down motorised vehicles not by slowing cyclists. It is recognized that forcing cyclists through narrow gaps, pushing them into conflict with each other or with other modes and limiting efficiency of junctions all have negative effects on cycling. Cycling needs to be made faster so that it competes with motor vehicles. Having to slow down, even rarely, is a "nuisance" and nuisances are designed out of the cycling network.

Wide high quality cycle-path in
Assen. It's good for relaxed cycling
and also for fast cycling. Everyone
is safe here.
Consider what this paragraph implies: "Cyclists have no standard traits. On the contrary, the Dutch cycling population is characterised by a broad composition in terms of age, gender, physical capacities and reasons for cycling. In certain conditions, the fast commuter cyclist is indicative for the design (in terms of design speed, for example). More often than not, however, the group of older cyclists with limited physical capacity will determine the limits (in terms of gradients and crossing times, for example). In still other cases, the design will be geared to young, inexperienced, sometimes careless cyclists (in terms of eye level, red light discipline and the complexity of intersections, for example)."

Dutch racing cyclists and other fast
use cycle-paths even when
riding in large groups. And yes,
they all got across in one green light.
So the Dutch view is that there is no standard cyclist, and everyone should not and cannot be reduced to a common level. Rather, the infrastructure for cycling needs to support all cyclists, whatever their ability or degree of strength. Designs may be varied in some locations, for example to make conditions extra safe for children near schools, but that doesn't mean abandoning the other cyclists at that location. The only way to achieve true mass cycling is to accommodate the entire population's needs.

Far from seeing cyclists as "speed freaks" who need to be slowed down, the design speed for corners on Dutch urban cycle facilities is set as a minimum of 20 km/h for minor cycle-paths rising to 30 km/h for through routes because "people must be able to cycle at high speed". Yes, the intention is to enable people to safely cycle as fast as they wish to. That is how to make cycling speeds competitive with other modes and therefore make cycling attractive.

Another Cambridge example. At the
end of the decline from a bridge there's
a bend which hides dark coloured
bollards just before cyclists must come
 to a stop for a road. This does not
come close to Dutch requirements for
"free deceleration space".
The Dutch recognise that speeds of "35 to 40 km/h" are not at all uncommon downhill and this is why there should be "plenty of free deceleration space at the bottom of the incline, with no intersections, sharp bends or other obstacles in the way". The requirements for speed in the relatively flat Netherlands are way above the speeds that Leeds and Bradford see as problematic due to "speed freaks" going downhill and clearly above the speeds that Cambridge planners think will be achieved along Huntingdon Road. And of course its not just those two. Look at Bedford, Northern Ireland, Southend or London and you find similar constraints, low standards and treatment of cyclists as a problem.

Low aspirations lead to a lower cycling modal share. The full potential of where you live can only be achieved with the very best infrastructure. Copying from and starting to emulate the best examples from the leading nation is the most efficient way to make a real start. Until you have caught up, there's no reason to try to re-invent the wheel.

Avoid repeating mistakes: take a Study Tour

We run regular cycling infrastructure study tours to demonstrate to politicians, planners and campaigners from all over the world what the best quality cycling infrastructure in the world looks like.

The frequency with which the same mistakes are repeated elsewhere demonstrates that there is a widespread lack of knowledge of how to progress. We can help with this, but only if people seek our advice.

My daughter went to a music festival over the weekend. She sent me this photo of the cycle-parking: bicycles as far as the eye can see. This is what true mass cycling looks like, but photos like this sometimes mislead. This volume of bikes is not here as the result only of providing cycle-parking. The most important part is the comprehensive grid of extremely high quality infrastructure which goes everywhere, which accommodates all types of cyclists whether or not they're in a hurry and makes everyone safe. That's what enables huge cycle parking facilities across the Netherlands to fill up.

It's startling how many countries have cycling design guidelines which include ideas which are deliberately designed to slow cyclists down.


Anonymous said...

Thanks for posting the link to the Leeds video. That's so awful that I wouldn't believe it if Leeds City Council didn't already have a reputation for building such crap!

So there's 0.3 miles of narrow slightly segregated cycleway taken from the footway, next to a busy multi-lane road with unused central reservations. Great.

The council may say that it's finished, but the junctions look like they've been completed, and they're awful. No amount of last-minute painting will fix that.

Anonymous said...

Great post as always David. Coincidentally I've just published a post about plans for a new piece of infrastructure in Glasgow which provides another example of what you're saying here:

Glasgow has only a 1% cycling modal share and is just beginning to respond to the demand for segregation - there are some segregated lanes up and running and more are planned, including a network in the city centre.

However, the detail of all the designs they're coming up with makes it clear that, while political will to do something may at last have materialised, they really don't have much of an idea of what they're doing.

It also has to be said that the situation is much worse everywhere else. Edinburgh has slightly more cycling than Glasgow, at 2%, but the local council is completely clueless about how to increase this figure. Despite investing 7% of their transport budget on cycling, a figure which has been consistently rising and is now higher than perhaps anywhere else in Britain, they are still messing around with advisory cycle lanes, which is beyond a joke.

Andy R said...

I drive the section of highway shown in the Leeds photo every evening.

Unsurprisingly, given the double yellow lines follow the kerb line, some drivers assume this is a left-turn flare at the traffic lights, thereby blocking cyclists using the track.

BTW, there are a number of bollards in the kerbline separating the footway from the cycleway, with signs to diagram 957 of the TSRGD on them, telling users which side is for which mode. Unfortunately, most of these signs are actually the wrong way 'round, suggesting cyclists should use the track furthest from the carriageway and peds should therefore walk on a track between fast-ish moving cyclists and faster moving motor vehicles, which then disappears leaving them on the carriageway!

Snagging - who he?

ktenness said...

Great writing. This site is instrumental to my education as a bicyclist advocate. Do you have any example designs for how to accommodate kids bicycling to school with high-speed bicycling commuters? That is my current issue here in Walnut Creek, California. I would love to raise some money to send some of our city staff to one of your cycling study tours!

David Hembrow said...

Ktenness: That problem is avoided in the same way as most others to do with capacity: there his to be an incredibly comprehensive grid of cycling infrastructure so that there is always a choice of routes, and everything has to be built to a good standard so that its possible to overtake safely where people do use the same routes.

Sometimes there are boasts made about ccl-paths having large numbers of cyclists on them, but actually that could be been as a sign of failure. It's better that people have a choice of routes and that all routes are of good enough quality for everyone to use. Removes conflict and makes journeys more convenient - therefore leads to more cycling.

Hope to see women from your end on one of our tours in the future!

highwayman said...

This blog entry should be one of your highlighted entries. Designing Nuisance Out of Cycle-Paths is a basic premise that deserves a highlight.

Onto something else: I was in Brossard (Québec), Canada, and there, they're actually trying to do right by cyclists. The suburb itself is nearby Montreal, so the mode vélo has spilled over some. It also helps that Vélo-Québec has the same kind of clout within the Province as the Fietserbond has within the Netherlands. Still, much remains inadequate.

Many of their segregated facilities do reach the 3.5 metre width to be adequate for cycling. Many paths (and painted lanes) do reach desired destinations. Still, there are too many narrow paths & lanes, too much dependence on sharrows, and when paths & lanes do go through junctions, too much zig-zagging is required if you do not want to "take the lane".

The Brossard network of cycle facilities is part of a greater network of paths & lanes in what is officially known as the Agglomeration de Longueil. (Metro-area of Longueil in our language). They would do well, though, to have a path or lane every 800 metres. They don't have that, but given the money, maybe they would do that.

Then there's consistency; it's lacking. I've ridden for 2 to 3 kilometres on one side of the road, and suddenly!, I have to change sides. Boulevard de Rome in Brossard is one such example. I see this elsewhere in the Agglomeration.

Maybe send out a missive (in French & English) to bicycle advocates (and planners) from both Longueil and Vélo-Québec to go on one of the study tours. The infrastructure may be substandard, but it seems that they want to do better. The political will seems to be there. They might be a very willing audience for you.

PS: Yes, Québec is my usual vacation spot.

Unknown said...

I've seen how it should be done right here in Cheltenham...

Proper separated infrastructure that has priority over side streets is possible in the UK with existing signage and road markings, here is an example in Cheltenham on Princess Elizabeth Way:

Mind you it is shared use paths, but the width is decent AND it has priority... pity it's only on this one street and not everywhere...

Cycling Punk said...

Ie 'Southend Prittle Brook Greenway'
I think you can sum it up by looking at this map. At this point the path is at its closest to Southend Hospital, one of the biggest employers in the town, and Chase High School, nearly a 1000 students. Yet does not have a direct cycle route to either from the greenway. To get to either of them involves a 150m No Cycling path.
I have given up believing anything I hear about cycling in Southend.,0.6871252,16z

houseofholland said...

Thanks for such an interesting and informative review of cycling in my hometown. We have just returned from a short vacation in Helsinki and what a difference we saw there - where cyclists are so much better accommodated!