What is a Bicycle Street?

Bicycle Streets Beyond Europe is a weekly multi-part series exploring the “bicycle street”. This first post overviews the Dutch implementation of the bicycle street (“fietsstratten”) and the next several showcase some of the best examples of bicycle streets in Canada and the US. Join the Beyond the Automobile mailing list to stay up to date with the series.

Bicycle streets flip the natural hierarchy and make bicyclists the dominant user of the street, while motorists must act as “guests”.

When people riding bikes mix with motor vehicles, it is often in an inferior capacity: bicyclists ride at the sides of the road, and move over further (and ride single file) when a motorist approaches from behind. This established hierarchy really reveals itself when a cyclist refuses to pull over – often resulting in a motorist getting angry and honking or tailgating, and an anxious cyclist.

In mixed traffic the motorists are typically the dominant user and bicyclists pull over to allow them to pass.

This hierarchy is not always desired. In the case of a quiet residential street with low traffic volumes, for example, why should the cyclist always yield to the motorist? What if that quiet residential street actually forms a major route for the bicycle network of a community, making it a “bicycle arterial” of sorts? In this case, the street function for motor vehicles (local) is inferior to the function for bicycles (arterial), and based on this, people riding bicycles should be considered the primary user of the street.

This is the essence of a bicycle street: a street that accommodates both bicycle and motor vehicle traffic but reverses the natural hierarchy and prioritizes the bicycle users. It’s a concept that originated in Germany and has received much experimentation in the Netherlands and elsewhere, and has great potential for use in Canada and the US.

This post will summarize the bicycle street, its latest design guidance, and important lessons learned.

A bicycle street through a park, shown with the Netherlands’ iconic red asphalt and a typical “fietsstraat” sign.

What is a Bicycle Street?

The Netherlands is probably the best place to start for understanding a bicycle street (“fietsstraat” in Dutch) and its intended purpose. For this, we turn to the Dutch Design Manual for Bicycle Traffic (commonly known as “The CROW Manual”), arguably the world’s leading bicycle facility design manual. The CROW Manual defines a bicycle street as:

“a residential road for motorized traffic that forms part of the main cycle network, .. and which is identifiable as a bicycle street due to its design and layout, but has a limited volume of car traffic on it and that car traffic is subordinate to bicycle traffic”

In other words, bicycle streets are intentionally designed to treat bicycle traffic as superior to motor vehicles, on street segments that form a city’s main cycling network, and where car traffic is intended to be low because of the context. The function of the street dictates the hierarchy of users.

Source: Dutch Urban Index

While main cycling routes often coincide with main routes for car traffic (“arterials”), it’s also common for bicycle routes to be separate from arterials for cars, and often this is done deliberately (this concept, called “unbundling” or “disentangling” of routes can be helped by “filtered permeability”, as discussed in this Not Just Bikes video).

An example of a bicycle street that offers a direct route to the city centre while avoiding main arterial routes for cars (Source: Rick Delbressine, The traffic safety of bicycle streets in the Netherlands)

Separating main routes for bikes and cars can actually be advantageous because cycling on arterial streets is less pleasant and involves more conflicts with cars at intersections and driveways. The separated bike route must not compromise the directness of the route though, and if the arterial road contains destinations for cyclists, then cycling there should still be possible. An example of a main bicycle route that is both direct and separate from motor vehicle arterials is shown above.

The “Bicycle Dominance” Prerequisite

“The fundamental innovation of a bicycle street is that it changes the established power relationship between automobiles and bicycles. Accomplishing this involves more than simply putting up a sign or changing the color of the pavement.”

Matthew Bruno, The challenge of the bicycle street: Applying collaborative governance
processes while protecting user centered innovations

The CROW manual is clear that a bicycle street cannot be created just anywhere – road sections designated as bicycle streets must already carry a high volume of cyclists, both in absolute terms as well as relative to the volume of motor vehicles. This is referred to as “bicycle dominance.” Because of this, a bicycle street usually only aligns with a segment of the “main cycle network,” rather than a street providing only a local link for cyclists.

Specifically, CROW suggests that all the following conditions should be met in order to designate a bicycle street:

  • maximum travelled speed is 30 km/h (the default condition for all residential streets in the Netherlands)
  • at least 1,000 bicycles per 24-hour period (equal to about two bicycles per minute in the busiest hour)
  • Volume of bicycles should exceed volume of motor vehicles
  • No more than 2,500 cars per 24-hour period

In many cases, motor vehicle traffic can be intentionally reduced with physical measures, but where this is not possible, CROW recommends that another solution for the bicycle route must be sought. For context, a typical local residential street in Canada and the US carries less than 1,000 vehicles per day.

The bicycle dominance prerequisite ensures that the context of the street (lots of bikes visually) matches the design of the street (visual indicators that prioritize bicycles). Both of these conditions must be present in order for bicyclists to claim the position as the dominant user of the street (and for motorists to cede the dominant position).

Bicycle dominance is achieved in volume here, but you can see that cyclists still prefer to ride at the side of the road, which is unfortunately next to parked cars.

Geometry and Design

Design is the other half of the equation for enforcing bicycle dominance in bicycle streets: narrow lanes, combined with speed control measures and a street that visually looks like a two-way bicycle path that allows motor vehicles.

CROW suggests that the basic width of the travelled portion of the roadway should be 4.5m, which is wide enough to allow two pairs of cyclists passing each other while riding side-by-side. This is also just enough space for two passenger vehicles to pass each other (though bicycle streets may be one-way for cars). On-street parking, if provided, is in addition to this width. At points where cyclists may need to frequently turn, traffic islands are recommended to prevent motorists passing from behind.

The CROW manual presents three functional profiles for a bicycle street:

  • mixed profile (4.5m single lane, on-street parking not permitted)
  • cyclists at side of roadway (2.0m lanes) with centre border strip (0.8-1.5m)
  • cyclists in middle of roadway (3-3.5m) with border strips (0.5-0.75m) on each side

All of these profiles reinforce bicycle dominance, with a smooth red asphalt area that resembles a cycle track (either a 2m one-directional or 3.5m bi-directional) plus “border strips” that create a rumble surface for motorists and visually narrow the street.

In the latter two functional profiles, on-street parking can be provided at the edge of the roadway, provided it is separated from the edge of the through portion of the street by 0.5 metres. This means that, for a bicycle street with on-street parking on one side, the curb-to-curb width of the street is about 7 metres.

A bicycle street in Utrecht with on-street parking. Notice that a 0.5m edge strip is present between the parked cars and the bicycle travel area.

It is essential that the maximum travelled speed of vehicles on bicycle streets be limited to 30 km/h, and this must be reinforced with design, using narrow travel profiles, raised intersections, and speed humps. At intersections, bicycle streets should be provided priority over intersecting local streets by continuing the red asphalt surface through the intersection.

Do they Work?

At first glance, it’s easy to see the bicycle street as a concession to motorists: take the cyclists off the main street, and put them on a local street without disrupting traffic too much, and at a low cost. But in order to be successful, the bicycle street must be seen firstly as a facility for bicyclists. It must be designated at the network level based on its usefulness as a direct route, and measures must be taken to manage motor vehicle volumes and behaviour at the street level, to the point where the route is unattractive for motor vehicle through traffic.

Further though, a successful bicycle street is not measured on its design, but on the resulting behaviours of users. On a successful bicycle street, cyclists comfortably ride in the middle of the travelled area and do not yield to motorists. This behaviour is the most difficult element to produce and replicate, because it requires a reversal of the established hierarchy of road users.

The first bicycle street in The Netherlands had a median barrier that made overtaking in a safe way impossible. But drivers still tried to overtake people cycling, causing very dangerous situations. This experiment was short lived, the median barriers were removed and the ban on overtaking was dropped. (Picture: Utrechts Archief, via Bicycle Dutch)

Bicycle streets have not been consistently successful in the Netherlands (see this Bicycle Dutch post and this As Easy As Riding A Bike post), and researchers have documented important lessons learned from case studies that have informed new guidance. This includes:

  • the need for a consistent standard for the design of bicycle streets, to make them more recognizable to road users
  • a stricter approach to the “bicycle dominance” prerequisite; there must be visibly more cyclists on the street than motor vehicles in order to create the desired behaviours
  • a higher focus on speed management, with more speed humps and raised intersections
  • ensuring that bicycle streets are not pitched as a way to please all road users (which is sometimes the case on shopping streets); in order to be successful they must explicitly seek to advantage cyclists

When bicycle streets are implemented in residential areas, they tend to be more successful because residents also desire and benefit from reduced motor vehicle traffic. In commercial areas the interests of the bicycle user are at odds with those of the local businesses (who tend to want easy car access), and there are several examples of bicycle streets failing in the Netherlands because of this.

Toronto’s recently-updated Shaw Street bikeway. Is it is a bicycle street?

Bicycle Streets in Canada and the US?

Bicycle streets have been growing in popularity in the Netherlands, Germany, and Belgium since the 1990’s, but do they exist in Canada and the US? The answer is yes – to some degree – and under a different name. My next post will discuss examples of the North American adoption of the bicycle street and whether or not it lives up to the essence of a true bicycle street.

Don’t forget to check out the other posts in the Bicycle Streets Beyond Europe series below.



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s