Should we be building one-way or two-way cycle tracks? Should we be building cycle tracks at all? It's one of those tricky issues which attract lots of opinions, but for which there isn't always a clear answer. But yes, I have Opinions.
My own starting point really comes from two viewpoints which I think are reasonably practical rules of thumb;
- One-way tracks should be used on streets.
- Two-way tracks should be used on roads.
OK, there are subtleties between the definitions of roads and streets. In fact, there is also a a prerequisite of how we define the highway network as I think most people could understand a starting point which provides *no* cycle tracks.
The photograph above is an example of the base situation for cycling infrastructure on a street - none. Streets should be motor traffic-free to a point that anyone can use them for cycling in the carriageway. If we can get that right, then we've cracked a significant part of developing a cycle network because many cycle trips will be starting and ending where people live.
Having people cycling in the carriageway is obviously a cheap solution, but more than that, it does avoid the clutter that cycle tracks can bring to the visual and physical landscape, notwithstanding the fact that cycle tracks and most street clutter is there because of motor traffic. Quiet streets still allow access by motor vehicle for those who need it, but over time, we get all sorts of opportunities to repurpose space for other things.
Francis Road, Waltham Forest. No cycle tracks here,
but low traffic means more space for people.
Getting to that point is obviously a challenge in itself and this is why Low Traffic Neighbourhoods feature so heavily in network planning and design - they are vital. However, we know that some streets, and certainly roads, will be taking too much motor traffic for most people to feel safe mixing with it. So we still need cycle tracks.
The other thing to say is that we might need cycle tracks in places that if the World was sensible, we wouldn't, but until we do tame those places, cycle tracks are important to get people cycling in them. A good example is Kent Street, Birmingham (below). This should be a low traffic street without cycle tracks, but currently, it provides a connection from the city centre to the A38 cycle route.
There's lots of development going on in the city and so eventually these older industrial and commercial streets will become more residential in nature and so access for HGVs for older uses and the redevelopment will cease and the protection should no longer be relevant if the right decisions are taken.
I wish I could remember where I read it, but I am sure I read that around 80% of Dutch streets don't have cycle tracks and if you've ever nosed around the country in person or on Google Streetview, you can see why. In fact in common with the UK, the Dutch have lots of very old city centres within which it would be hard to fit cycle tracks, it's just that the cars have been cleared out making them unneccessary.
The Old City of Utrecht. There's no space for cycle
tracks here, but there's no space for lots of motors,
so it doesn't matter!
OK, at some point we do need cycle tracks, even though I have so far both dismissed their use and called them motoring infrastructure (which they are). One-way or two-way? For most urban applications on connector type roads (which connect local access streets to main roads), I want to see one-way cycle tracks. The main reason for this is around how we treat junctions that are not controlled by traffic signals; notably side streets.
Despite people driving having received training and with the vast majority thinking they are better than average, surely people should be able to cope with people cycling at side roads. Except they can't. This is both a design and a psychological issue whereby drivers entering or existing side roads either aren't expecting someone cycling to be there or they didn't see the person.
From a driver's point of view, you expect to see
vehicles coming from the right as you reach the junction.
From a driving point of view, emerging from a side road puts you in a position where you will expect drivers coming from your right. If there is an obvious cycle track, then it's easier to expect someone to be cycling there. With two-way cycling, an emergent driver isn't necessarily expecting people cycling from the left. You'll have realised of course that people walking along the footway will come from the left too, but they are not moving at cycling speed which mitigates the issue somewhat.
This always makes me remember a story that fellow designer, Brian Deegan, recounts about his Royal College Street scheme in Camden, London. The (one-way) street used to have a two-way cycle track on it's eastern side and there was a problem with driver vs cyclist collisions at the side streets. Being a one-way street, drivers didn't have to worry about motor traffic from the right at all and so looking right wasn't a priority.
Over the years, the warning information to drivers on the side streets was added to leaving a patchwork of brightly coloured surfacing supported by lots of signage, including flashing amber signals warning drivers of the cycle track. The image below is of Pratt Street in July 2008 looking towards Royal College Street.
Despite the money spent on all of this clutter, the collision problem persisted and eventually, the layout was changed to place one-way cycle tracks on each side of the street using planters and floating car parking. This dealt with the main collision problem (below).
Unfortunately, the planters eventually got wrecked by drivers and have since been replaced with kerbs.
Of course, this approach doesn't design out left and right hook movements by drivers turning in from the main road, but again, cycle traffic will be coming from the "expected" direction. Hills Road, Cambridge (below), has one-way cycle tracks and despite being a main road (A1307), there are plenty of side streets and private accesses which makes one-way cycle tracks an appropriate choice for the local conditions.
The other thing in favour of one-way cycle tracks in urban areas is that for those accessing premises to and from them, they don't have to cross the road to immediately access/ leave the cycle track. However, this also means that regular crossings of the road are needed otherwise people will simply cycle the "wrong" way, affectionately known as "salmoning".
In the UK, all cycle tracks are two-way in law unless modified by a traffic order and frankly, many highway authorities don't bother making an order because they either don't realise it or they don't know how they will sign it! So "salmoning" is largely a design issue than a legal one in my view.
For people crossing, one-way cycle tracks give more comfort that they are only having to contend with one "traffic" direction which is helpful in tight locations where people cross opportunistically from the cycle track or if they are crossing to access a bus stop (above); there is less going on to have to cope with.
For example, this crossing (above) on Lea Bridge Road in Waltham Forest gives an opportunity for a u-turn by people wanting to easily cross the road from one one-way cycle track to the other on the far side of the road.
Two-way cycle tracks (by design) are better for the very large urban roads or rural situations because opportunistic crossing is going to be much harder and is probably only going to be possible at signal controlled junctions.
The above photograph is Vondellaan in Utrecht which has a section of dual carriageway with a one-way cycle track on the north side and a two-way cycle track on the south side. I don't know the full story, but the layout means that people from the north wishing to head east don't have to cross the road.
It runs next to the railway and so essentially provides a route for people coming from the north of in the older part of the city where there wasn't space for cycle tracks. To the east, the two-way cycle track enters a large residential area as a direct route away from traffic. The design is a response to the local situation and is more about disentangling the cycling and motoring networks.
Above is a two-way cycle track on Burgemeester de Meesterstraat in Harderwijk. Again, I don't know why a 2-way cycle track was provided, but it is only on one side of the road, so maybe it was a space issue. There is residential development on the eastern and so people have to cross the road to reach it. There is no development on the western side, but there is land ready for building.
The existing residential development seems to have been there some time and the addition of the cycle track is relatively new (2016) when buildings on the western side were demolished, so I think it was probably a result of the space and the phased redevelopment.
In the UK, two-way cycle tracks have been quite popular in recent years. For example the C3 and C6 (below) cycle routes in Central London have large two-way sections which have to some extent operated tidally which is space efficient which has minimised the space taken from general traffic lanes.
They are also easier to thread through signal controlled junctions as can be seen below on C3 on Lower Thames Street at Fish Street Hill where there is a right turn ban on general traffic leaving the main road.
In situations like this, the main road can run on a green signal at the same time as the cycle track which simplifies the operation of the junction. Two-way tracks also require less construction work (in general) than two one-way tracks and where kerbs are paid for by the metre, this can be a small fortune saved.
They are also favourite choices along rivers and seafronts such as Marine Parade, Great Yarmouth (below), because of the general lack of busy side roads which need to be catered for. Of course, access to the attractions on the other side of the street are difficult to get to.
There's also the matter of rapid deployment, an issue which has been in our minds over the last year or so. For a wide street, a two-way cycle track can very easily be deployed, although there still might be works needed at junctions, side streets and kerbside features such as bus stops. Bradford Street in Birmingham (below) was rapidly deployed, but wasn't very good at bus stops (although it is being redesigned).
Where a choice has been taken for urban two-way tracks on one side of the street, the risk at side streets can be mitigated by providing some waiting space for drivers so that they can tackle crossing the cycle track without having to worry about stopping on the main road such as St Georges Road, Lambeth (below) and with some turns being banned.
In rural situations (between settlements), a single two-way cycle track on one side of the road is enough because it's less about capacity and more about utility and connecting villages to towns. In most cases, there will be very few pedestrians and so shared-use paths are appropriate such as on the island of Fanø in Denmark (below).
Of course, these might skirt small settlements where a few people might catch a bus and so some waiting space off the cycle track can be useful as below) near Amsterdam (although wider space with a seat and shelter would be more useful!)
So, is there any official guidance? LTN1/20 - Cycle Infrastructure Design gives some help from Paragraph 6.2.15 onward and it includes a handy table (6.2) summarising the issues and opportunities that two-way cycle tracks present (reproduced below).
So, we need to think about the type of street or road, what's happening in the wider highway network (for general traffic as well and cycling) and there's complexity and cost considerations. Hopefully this discussion will help you to understand and maybe challenge some of the decisions being taken where you are