Tuesday, 12 November 2019

North Cottesloe Primary School parking

North Cottesloe Primary School has a problem. A lot of parents use cars to deliver their children to school. The car parks adjacent to the school are busy and near full capacity at morning drop-off and afternoon pick-up times. The "kiss-and-drop" car park on the southern side of the school sometimes has drivers queuing onto Eric Street which can cause congestion problems for general traffic.

The car park on the western side of the school can only be reached by crossing Railway Street. Although Railway Street has a children's crossing with traffic warden on school days, many people cross unsupervised directly from the car park area.

Similar to most Australian schools, the amount of car traffic makes it risky and unpleasant for people walking and cycling. 

In 2015 the school board resolved to improve the situation with "the clear objective to create a more accessible and safe environment for the school, while attempting to effect improvements for the community" (Railway Street Car Park Design Proposal, M. Goodlet, 2017).

Railway Street Cottesloe 8:45am

The main idea has been to create a new "kiss-and-drop" area on Railway Street with enlarged car park. Over the past four years the concept and proposed designs have been controversial issues with local residents. The Town of Cottesloe has called for public comment on the latest proposal. Comments close on 12 November 2019 at 4pm, there is not much time left.  Here is a simple summary. 


The school is located on the corner of Eric and Railway Streets. The Fremantle railway line is nearby and there are many mature trees in the railway reserve.


The first proposal was a large new carpark with drive through "kiss-and-drop" facility adjacent to the school. The street would be realigned further to the west. This would eliminate the need to cross the street from the car park and would have doubled the number of car parking bays. 

This design was a typical car-orientated engineering solution. It seemed very functional but took up lots of space and would have required the realignment of Railway Street into the trees. The potential loss of up to 24 mature trees meant this plan was rejected.


This plan aims to retain almost all the trees and has a more compact design. 


Plan B is an improvement on Plan A, but we should be considering if we are on the right track. Is Plan B just doing-the-wrong-thing-better? The current budget for these works is $550,000 to $700,000. The Town of Cottesloe is not planning on providing all the money. They are planning on providing $350,000 from Cottesloe ratepayers, the rest they hope to get from the state government. In other words, the rest of us. Should we spend half a million dollars on encouraging the people of Cottesloe to drive their children to school? 

Everyone wants the streets around our schools to be safer. One of the best ways is to reduce the amount of car traffic. However, building a bigger car park and another "kiss-and-drop" facility is similar to adding another lane to a freeway. This is the school version of induced demand.

There is another option. 

The Town of Cottesloe should consider a modal filter on Railway Street. This would stop car traffic travelling all the way through the street and allow the existing car parking area to become much safer.  With less traffic and without high-speed traffic, the existing street could be used as a drop-off zone. This would be much cheaper, quicker and easier. 

The filter could be made in a variety of ways. Most commonly bollards are used, but those can be supplemented with a range of landscaping devices including trees to make the area an attractive addition.

The modal filter could be applied at all times, or just during the important school hours. The times when we are hoping to keep our children safe.

For those not convinced, the modal filter solution is simple enough to be done as trial before committing to the expensive (and unfunded) Plan B idea. 

A modal filter could save the Town of Cottesloe and Western Australian tax payers more than half a million dollars and achieve a better result. Instead of supporting and encouraging motor vehicle traffic around schools, we should be looking a ways to reduce it.

Sunday, 27 October 2019

Roe Street Perth. An opportunity for a better city.

The City of Perth is proposing to upgrade Roe Street. The "enhancement" is currently at the community and stakeholder consultation phase. Comments can be made here until 31 October 2019.

Roe Street indicated in blue
The City seems to be downplaying the significance of this change and the opportunity it presents. This will be more than an enhancement. It's a 1.2 kilometre stretch of inner city street that will be completely rebuilt between property boundaries. The railway reserve on the southern side is currently empty as a result of trains being put below ground. Instead of weeds and fencing, there will be commercial buildings. What has been a single-sided street for 100 years will become double-sided. The area between the buildings is a like a blank canvas and we can choose to make it whatever we want. This is a once-in-a-lifetime opportunity to knit the city together, to make it more walkable and better for cycling.

From City of Perth video. Proposed new Roe St building in pale blue.

Unfortunately, the City of Perth is proposing to facilitate motor vehicle traffic at current levels. Some positive aspects: the footpaths will be wide and there will be trees, but the proposal for cycling is inadequate. The cycling infrastructure within the Roe Street proposal is not best-practise by current international standards, but not only that, it doesn't even meet Austroads recommendations from 2011.

At first glance it appears the City is improving cycling. The proposal describes "protected cycle lanes" but looking at the detail, it's apparent there are many flaws. The existing street has a three-metre wide bi-directional shared path which stops at each intersection. It's not ideal, but it's suitable for people of all-ages and abilities. The proposed cycle lanes, which are separated from pedestrians, will only be suitable for experienced and confident cyclists. The lack of protection in key places will be a problem. The City is hoping to reduce conflict between pedestrians and cyclists but this  proposal will result in many people continuing to cycle on the footpaths.

Uni-directional cycle paths on both sides of the street, separated from pedestrians and motor vehicles could be great, but it needs to be done to a much better standard than currently proposed. 

Current shared-path


I urge everyone who wants a better city to contact the City of Perth and tell them this proposal is not good enough. You can look at some of the detail that follows but if you read no further, here are the key points:

  1. Reduce the proposed width of the footpaths and allocate the space to safer cycling. This will also make it safer for pedestrians.
  2. The cycling infrastructure needs to be suitable for all-ages and abilities. The current proposal does not meet this requirement.
  3. Tell them it is OK to make it less convenient to drive in the city. It is OK to remove a motor vehicle turning lane or two if it makes it a better place to walk and cycle. We want a city for people, not cars.


The City of Perth states the cycle lanes will be 1.5m wide, however once you read the detail in the drawings, it becomes clear the lanes will be 1.2m wide at intersections.

Even if the cycle lanes had a width of 1.5m all the way, this is not sufficient to provide safe side-by-side riding and it does not allow enough space for a single cyclist to pass a slower cyclist safely. A width of 2.0m is considered the minimum by international standards for a protected cycle track (e.g. CROW Design Manual for Bicycle Traffic and NACTO Global Streets Design Guide).  Even if you think "Perth isn't Copenhagen...those rules don't apply to us" we should at least work to the Austroads standards. The illustration below is from Cycling Aspects of Austroads Guides which specifies a minimum of 1.8m and recommends separation of 1.0m to provide clearance from car doors where kerbside parking is likely to occur. 

Cycling Aspects of Austroads Guides 2011 page 33

You don't have to read the street design manuals. Simply ride on any of Perth's bidirectional 3.0m wide paths and try passing a slower cyclist without crossing the painted centre line.


Although the City of Perth states the proposed cycle lanes are protected, they are not protected in all places. There is no protection on the east-bound lane where is passes the loading bays between Nick's Lane and William Street. At this point there is no protection for people cycling at all. The cycle lane passes between moving traffic and parked vehicles being unloaded. Also note that the loading bays are only 2.3m wide and most trucks are 2.5m wide plus mirrors.

There is also no protection when the cycle lane passes the two five-minute bays near the train station. At both the loading zone and the five-minute car bay areas will have a high turnover of moving motor vehicles. These are not the sort of places you would want to ride with a child. A protected cycle path should travel on the footpath side of these hazards.

At places where the cycle lanes pass driveways there is no protection, and not even green paint. As the proposed cycle lanes are at street level, people cycling will be getting no help at these conflict points. No green paint across driveways is fairly common in Australia, however it is interesting to note there is also no paint proposed across the four-lane entry and exit to the City of Perth's own parking building which has 573 car bays. This will be another high-risk area.

Where the cycle lanes travel on the inside of parked cars there is also a problem. The cycle lanes will have good separation from motor vehicles in the travel lane but there is no protection from cars moving into the parking bays. There is a 600mm buffer proposed but it is just paint. This paint will not prevent car drivers parking in the cycle lane or parking on top of the painted buffer. It will not prevent car drivers crossing into the cycle lane when they manoeuvre their cars as they park. Without a kerb to separate the car parking bays, you can expect it to happen... frequently!

The intersections are also not protected.  Head-start boxes are planned, these can be useful for confident riders but they are not recommended for people of all ages and abilities.


With the proposed design there is a high risk of car passengers opening their doors into the path of people cycling. This is because the 1.5m lane will have insufficient width for passing and faster cyclists will use the paint buffer area as a riding space. 


There are only two five-minute car parking bays along the entire 1.2km length. We can expect taxi drivers, including Uber and Ola, to block the cycle lanes at the driveway areas for their pickups and set-downs. As these cycle lanes are at street level and there are no barriers this will be the driver's easiest option. Obviously, taxis can be a problem for any cycle path, but the on-street design and lack of a continuous colour to help define the cycling space will increase the probability.


The City of Perth currently has a policy of using barrier kerbing (square) to define the change from motor vehicle carriageway to footpath. The Roe Street cycle lanes, by being at motor vehicle grade, will probably have the same kerbing as currently used in other recently upgraded streets within the city. These are difficult to mount and would discourage cyclists entering and exiting the cycle lane mid-block. 

This is an important detail because if semi-mountable kerbing of a suitable height is not used the effective width of a cycle lane (or path) is reduced and the safe movement of people cycling is restricted. Where the lane is enclosed on both sides and only 1.5m wide this presents a risk for rear-end collisions because people would need to come to a complete stop to lift their cycle over the kerb at a destination.


There appears to be no reduction in motor vehicle movement capacity along Roe Street. Nearly all intersections have the same number of travel and turning lanes. The exception is Lake Street where is enters Roe. That intersection will have an increase in capacity with an extra lane being added.

At the Beaufort Street end, Roe Street will still have four motor vehicle lanes. At this point the cycle lanes are reduced to 1.2m and the footpaths are narrowest. This is not the sort of allocation of street space found in good walking and cycling cities. 

This is currently one of Perth's busiest cycling intersections. The fourth lane enables drivers to enter the carpark building on the south side from both directions. It is not essential, there are multi-level carparks on both sides of the street! 

This additional turning lane creates an increased safety risk for cyclists and pedestrians. The Citiplace carpark on the southern side of Roe Street was built in 1987 and the turning lane would have been created at the same time. That was also the period when the City of Perth was using the advertising campaign slogan "Your car is as welcome as you are". That's out-of-date now, so is the turning lane.


Some of the pedestrian crossings have been moved away from natural desire lines. An example is the east-west crossing at Melbourne Street. This is annoying and usually results in non-compliance. People will walk along the shortest route whether it's the official crossing or not. People want to walk directly from A to B, not via C.


The elevation below can be found on page three of the "Roe Street Enhancement Revised Masterplan 2019". It is not to scale. The motor vehicle lanes have been shown much narrower than the widths  proposed by the City of Perth. This gives the impression there is not enough room to allocate more space for cycling.

In the elevation, if we accept the cycles lanes are 1.5m, the motor vehicle lanes have been drawn at about 2.3m wide. There is one metre missing from each lane.

If presented correctly, the motor vehicle space would be two metres wider.

Another misleading image can be found in the "Existing and Proposed Cross Section" document. Again, the proportion of street space allocated to motor vehicles is incorrectly presented. Here is the image in its original form.

This corrected version shows the motor vehicle lanes in true proportion and also demonstrates how wide some of the footpaths will be in comparison to the cycling space.

Although the City of Perth states there will be a "narrowing of vehicular carriageway allowing for dedicated uni-directional cycle lanes" and also "The Roe Street Enhancement aims to minimise the vehicular carriageway width".

The existing travel lane is 3.2m wide and the proposed travel lane will still be 3.2m. The saving of space will be made from a reduction in the width of the parking lanes, not the travel lanes.

Sunday, 5 May 2019

Reid and Tonkin Highway Interchange

Perth's highways keep growing. They get longer and wider each year and consume a lot of our tax dollars. One of the newest sections is the intersection of Reid and Tonkin Highways. It's considered Western Australia's "biggest free-flowing freeway-to-freeway-interchange" and is part of the $1.12 billion NorthLink WA project.

As usual, the promise of the new construction is to have reduced congestion and improved freight capacity, efficiency, productivity, road safety, plus better amenity for the community!

There are alternative arguments: should we invest less in motorways and more in other forms of transport to achieve better results? I'll save that for another time. This is just a first-look at the cycling infrastructure that's being built as part of the interchange.

In the last couple of years, it's become Western Australian State Government policy to incorporate shared-paths to a width of four metres in most new major projects. This has been a great move. Previously the standard was three-metre paths, which were good quality but designed on the principal that cycling is a solo activity. Officially, people were required to ride in single file. Obviously, many people ignore this rule so they can talk with friends or their children while riding. By increasing the width to four metres, the new paths enable much safer side-by-side cycling.

4m Principal Shared Path, Kinross area north of Perth

Although officially shared-paths, they function well as cycle paths because there is usually a very low volume of pedestrian traffic. The four-metre width also allows the opportunity for people cycling to provide more space when passing any people walking.

The new interchange was recently opened to motor traffic but the landscaping and several of the paths are still being built. I visited on a public holiday and there were no fences or signs prohibiting access. I am sure during a normal working day it would be more restricted because of safety issues due to moving machinery.

It's likely to be quite a while before everything is finished. This is a big area. How big? Fun fact: you can fit five of Perth's new 60,000 seat sport stadiums within the interchange, and still have room left over for half-a-dozen high-rise apartment buildings.

Perth Stadium
A five-stadium interchange
The interchange is so big, the areas within the road loops have potential. A lot of the motor traffic is elevated and quite distant, there is not as much noise as you would expect. Perhaps these areas could have some life instead of becoming neglected scraps of bush. Is it crazy to consider a village in there? Perhaps a pop-up bar with live music or a concert venue; with cycling and walking access only. 

It's good to see the path is being connected directly to this sporting ground (below). We have a number of older PSPs (Principal Shared Paths) that lack permeability. Our PSPs should not only be for long-distance commuting, they need to function well for short local trips as well. Providing these additional connections is probably difficult. Not because of the engineering, but because of politics. These are the places where the paths need to leave the state government land and cross the boundaries into areas controlled by local government. 

The four-metre wide standard also includes the bridges. This one below helps to connect the residents of Beechboro with the Lightning Park sports ground and recreation centre.

This is excellent cycling infrastructure. There is no doubt our state government is on the right track. They are achieving great results on new projects where they have control of the land. However, the challenge for Western Australia is to get local governments to lift their standard to a similar level. Our state government can provide high quality PSPs next to motorways and railway lines, but the potential of these major links will not be realised until we are able to leave the safety of these paths and enter local streets with a similar level of confidence. Painted lanes next to motor traffic travelling at 40 - 70 km/h is not good enough.

Monday, 8 April 2019

Are we ready for protected cycle lanes?


One of our inner-city local governments, the City of Vincent, is currently modifying the northern section of Oxford Street in Mount Hawthorn.

Google Streetview 2018
This 300 metre "town centre" section has looked unloved compared to the southern section of Oxford Street and the Scarborough Beach Road town centres. It has no street trees and probably had no significant improvement since the 1960s. The City's publicity states the planned works will include:
  • 38 street trees for enhanced shade and beautification
  • A red asphalt shared-space to denote the town centre area
  • A central median for pedestrian refuge and improved walkability
  • 2 new motorcycle bay areas to accommodate 4 bikes
  • 12 bike racks with locations to be determined in consultation with businesses
  • Low profile speed-humps to accommodate a potential 30 km/h zone which will improve safety and town centre amenity
As usual, with local government street upgrades, the aim is to "improve visitor, pedestrian and cyclist amenity". I have no doubt adding 38 trees will improve the beauty of the street and the experience for pedestrians, but will these changes be better for people cycling?

City of Vincent's "before" and "after" images

People cycling can enjoy the additional shade and more natural environment as much as pedestrians, but adding trees to the centre of the street can make riding more hazardous because they prevent drivers passing safely. As detailed in a previous blog post, raised medians with trees are becoming common in Perth but local governments seem unaware of the negative consequences for cycling. Rather than improving conditions, the central obstructions usually force people cycling to take to the footpaths in order to feel safe. Only the bravest or thick-skinned will ride in a narrow 50 km/h lane in front of motor vehicles.

The photograph below shows concrete medians have been installed ahead of the tree planting. As of today, fourteen central trees have been planted and more trees will be planted between car parking bays very soon.

Oxford Street March 2019


The City of Vincent is calling this a shared-space and will be adding large cycle symbols to the centre of the travel lane. If this was a genuine attempt at shared-space, the City of Vincent would be removing road markings, signs and kerbs. In this situation, the use of the term shared-space seems to be an attempt to make everyone feel better about the poor provision for cycling.

There are some elements of the shared-space concept that are useful when mixing pedestrians and cycling (e.g. Amsterdam Centraal Station), but I am yet to see an example of a shared-space design that works well when the dominate mode of travel is motor vehicles. 

Even London's relatively famous shared-space Exhibition Road is to be remodelled after the council finally conceded it presented safety problems for vulnerable road users.

Oxford Street will not be a shared-space. It will simply have cycle symbols in the travel lane and some green paint. 

Oxford Street March 2019


The City of Vincent are hoping to convince Main Roads WA, who control all speed regulation, to reduced the posted speed limit on this section of street from 50 km/h to 30 km/h. Even if that happens, most people will still not feel safe to cycle on the street. Although a 30 km/h speed limit is considered appropriate when mixing cars with cycling, the volume of each mode must also be considered.

The traffic volume on the southern section of Oxford Street is around 12,000 vehicles per day and this section is probably similar. That's far too many for a successful "bicycle street", or what the Dutch call a Fietsstraat. (CROW guidelines recommend less than 2500 vehicles per day and 1000+ cycles).

Another important design aspect of a successful Fietsstraat is that the street must have sufficient width to allow drivers to pass people cycling.

Oxford Street March 2019

Our Austroads guides also warn against creating travel lanes with widths that create safety problems for cyclists.
"Practice should be that lane widths are either designed to be wide enough in all instances to allow the safe passage of a cyclist and a vehicle side by side (3.7 m or more) or narrow enough to permit the passage of a vehicle or bicycle only (3.0 m or less). Widths in between these two extremes create squeeze points and result in conflicts.” 

Guide to Traffic Management Part 8: Local Area Traffic Management (2016 Edition, p126)
The width of the motor vehicle lanes in the proposal are 3.2 metres with a 1.8m central median, and 2.2m parking bays. However, when measured this week, it seems the concrete median has been narrowed to 1.5m, the travel lanes are 3.5m and the parking bays about 2.1m. Either way, both the 3.2m and 3.5m width are in the specific range that Austroads warn against

Unless the person cycling boldly takes the centre of the lane, car drivers will be tempted to squeeze past. Passing will be possible but life threatening.


A solution to the safety problem is obvious to anyone who seen what progressive cities are doing around the world. Protected cycle lanes. To achieve this, the on-street car parking could be removed and cycle lanes put beside the kerb on each side of the street. The City of Vincent know about protected cycle lanes, just around the corner from Oxford Street, a section of Scarborough Beach Road has protected lanes that were installed a couple of years ago. The City should be congratulated for this work, it was a good achievement.

Scarborough Beach Road, Mount Hawthorn

The Scarborough Beach Road lanes were done in an area with low levels of on-street parking, Oxford Street will be more difficult. Removing parking bays in front of commercial businesses is a tough job. The City of Vincent obviously does not want to take it on, at least, not this time. The publicity on their website has a lot of detail to reassure people that car parking will be prioritised.
"The upgrade requires the loss of 3 car parking bays north of Wilberforce Street and 2 car bays south of Wilberforce Street. To mitigate this loss, the City will negotiate the use of redundant No Stopping Zones to accommodate additional car bays. The City will also review the use of the Oxford Street Car Park and on-street parking along Wilberforce Street to ensure parking efficiencies are maximised and parking availability for town centre visitors is prioritised."
If a city chooses to prioritise car parking it will often be at the expense of safe cycling infrastructure. That's what is happening here.

One strategy for Oxford Street would be to acquire vacant property nearby and provide off-street car parking in exchange for the loss of on-street bays. If the old taxi depot was utilised, the walk from car to shop would be 40 - 180 metres. That's similar to many successful suburban shopping centres.

Former taxi depot, Oxford Street
Karrinyup Shopping Centre

It would be great if local councils got tougher about car parking and more concerned about safety. Imagine. When discussing potential street upgrades with businesses they could offer more trees but explain the car parking needs to move further away. 
 "You can have a beautiful street, which will improve your business, but it also needs to be a safe street. We're a local government, we can't do anything that's not safe. If you want trees, we have to do protected cycle lanes at the same time."
Imagine if creating a street that was unsafe for people cycling was illegal. 


The design of the Oxford Street upgrade is nothing new. It's basically a repeat of the Mount Hawthorn town-centre section of Scarborough Beach Road nearby. The lane width is the same and it has a slower 40 km/h speed limit.  It looks great, the businesses seem successful, it's quite good for walking, but cycling is shit. How do we know this? It's simple, just go there and watch what people do.

Scarborough Beach Road, Mount Hawthorn

Even on a quiet Sunday afternoon, 90% of people riding are doing it on the footpaths. Even though it's legal for a person to slowly ride a bike in front of a car driver. People want to feel safe and they don't want conflict.

By painting cycle symbols on the street, the City of Vincent are encouraging people to cycle in front of motor vehicles and hoping to convince drivers to slow down from 40 km/h to 15 km/h. 

They're dreaming. Footpath cycling is legal in Western Australia. We have a low percentage of people cycling and where footpaths are not crowded it makes sense to ride on the footpath. That's what people are doing.

People are cycling to the Mount Hawthorn town centre, but it's not because there are yellow symbols on the road or heart-shaped bike racks, it because the street has destinations they want to reach.

Footpath riding is a great way to increase cycling participation in developing cycling cities. If the cycling mode share is 1% it will help to add a few percent. It's been great for Perth and should be legal throughout Australia.

Footpath riding is also a barrier to increased participation beyond low percentages. It's often slow and awkward to dodge around cafe tables and street furniture. Crossing side streets can be difficult. Once a town centre becomes a popular destination and pedestrian traffic increases, people cycling have to dismount to avoid conflict. If the City of Vincent (and Australia) is genuinely wanting an increase in cycling we can't rely on footpath riding.

Very soon, we will need protected cycle lanes to grow the number of people cycling. Conversely, unless we have a lot of people cycling, businesses and the general community will not understand why we need to move car parking to build the protected cycle lanes. Changing how cycling is perceived will be a challenge, but it is possible.


Protected cycle lanes are not going to happen on Oxford Street now, but hopefully it can be done in the future. Eventually other factors might help, car parking might not be such an issue and cycling can increase. We should be planning for it. Maybe it will take ten years, maybe longer. 

What's more difficult than moving car parking to build a cycle lane? Moving car parking and also cutting down trees to build a cycle lane!

One thing we can be sure about. If trees get planted this week on the sides of Oxford Street between the car parking bays, we can say goodbye to the protected cycle lanes for another fifty years. 

Oxford Street, March 2019