Sunday, 27 March 2016

Cycle lanes

Cycle lanes are something often used in North America to try to encourage cycling. I say try because it's not working well. But the Dutch have several thousand kilometres of cycle lanes and yet they have lots of cyclists. So what do they do differently?

The width of the cycle lanes is a factor. Their cycle lanes are usually wider than ours. We sometimes have 1.2 metre wide cycle lanes in some areas, although 1.5 metre wide cycle lanes are also common, and in Edmonton, is the minimum width. 1.8 metre wide cycle lanes are getting more popular, and is the standard width for new implementations in Edmonton. 2.1 metre wide cycle lanes are sometimes prescribed depending on what is needed for comfort and how busy a cycle lane might be. We can also add a painted buffer, between .5 metres and 1.5 metres, and we can add this buffer between the parking lane and cycle lane if there is parking, between the motor vehicle lanes and cycle lanes, or both if there is parking.

They used to prescribe 1.5 metre wide minimums for bicycle lanes. Now that minimum has increased to 1.75 metres. The standard width is 2 metres, although some 2.1-2.5 metre wide cycle lanes exist. It should be normal to provide a .5 metre wide painted buffer (or possibly textured buffer) when possible, and a buffer between the parking lane if present should be used so as to prevent dooring.

We are able to colour our cycle lanes. In North America usually we pick green, and in the US this is required by their manual on uniform traffic control devices. The UK has a variety, some reds, some greens, some more bright than others. Not very uniform. in the Netherlands, they use red, consistently. In fact, modern guidance says that it is required on cycle lanes and where cycle paths cross roads, and many cycle paths use it for midblock sections as well. We tend to use paint or thermoplastic for the colouring, and generally apply it, if we even do, at intersections, either when crossing a right turn lane, a bike box (ASL), or a two stage queuing box, sometimes at the intersection crossing.

The Dutch when colouring their cycle lanes, do so by dying the asphalt. Red works with black asphalt because red is the only colour that works with non clear base asphalt. This dyed asphalt approach makes it much longer lasting, and also this approach is not too expensive, and by doing it once every time the surface needs repairing as opposed to once every few years that we usually need to do with our methods.

Parking next to the cycle lanes is avoided near commercial parking zones, due to the higher frequency that people will cross the bike lane to park or get back into the traffic, and more door openings and closings. Cycle lanes also are not used on roads where there is more than 1 lane in each direction, as this would mean a much busier road, and would be unsuitable. Beyond 50 km/h, cycle lanes are also unsuitable.

Sometimes you will see combinations of cycle lanes where they make up combined around half of the road and a single lane width black asphalt stripe between the two cycle lanes. These are traffic calming elements, used on access roads that are busier but don't need cycle paths.  They also help in encouraging a safe passing speed.

Cycle lanes are also supposed to be bypassed with the normal bus stop bypass method, this ensures that buses do not cause a conflict. Intersections also ensure that the cycle lanes are continuous across the junction. Much of the time here, our cycle lanes end, either in a right turn lane with sharrows if you're lucky or just disappear. Sometimes you might get a right turn lane design where the cycle lane has priority to cross into a cycle lane to the left of the right turn lane, sometimes you get a bike box. The Dutch would simply continue the cycle lane across the junction along with some elephants feet markings and make the side street into an access road gateway if the side street is a gateway, otherwise transition the cycle lanes into separate paths at distributor road-distributor road junctions, using roundabouts, sign control, the protected intersection or simultaneous green design.

A big difference that makes their cycle lanes work is that they are much less common than cycle paths, about 6-7 times as many kilometres of cycle paths as compared with cycle lanes. They are not considered the desirable option for a distributor road, even relatively low volume ones. We consider cycle tracks at volumes above 10k vehicles per day. They have a far lower threshold. Many cycle lanes are being redesigned as separate cycle paths. A few old distributor roads without any cycle provisions get cycle lanes, like what a suburb in Utrecht a few years ago got.

They do have an interesting design called the advisory cycle lane, well, advisory in the sense that cars are not prohibited from using it if it is needed for the vehicle to proceed and after checking such lane for cycle traffic first, and after yielding to them, you can enter the cycle lane. This allows their use on some narrower distributor roads where otherwise cycle provisions could not be created.

Also know that their network is much more dense than ours, and they have extensive 30 km/h zones to link up with cycle lanes, and many cycle paths as well to connect larger roads. This makes that you will be able to get from the cycle lanes to your intended destination without any gaps.

Their cycle lanes are much more comfortable to use (when properly designed) than ours are for the reasons listed above. But even the Dutch consider most cycle lanes old fashioned. And they do what they can to use cycle lanes only as a last resort. This I think makes the key difference in design between the Dutch and us. Why not  use their cycle lane design manuals for the few places where we do need cycle lanes?

Saturday, 26 March 2016

My new omafiets

My New Omafiets Video

Today, I finally got my new omafiets. Brand new, 3 speed shimano twist shift, internal gears, almost fully enclosed chain. Nice leather (don't know about whether it's real leather or not) saddle and grips. A step through frame, but it doesn't really make a difference whether I would be a man or woman in terms of who can operate it. The bike is tall, at least 8 cm, probably 10 or more, higher up on the handlebars than my previous bike had. 

A major change to me, is that it only has a single lever brake (which connects to a rim brake). Where is the second brake? In the transmission. It's called a coaster brake. Most people who've ridden a bike as a child used this at some point. It's a simple idea, but works very well. Including for around half of the bicycles in the Netherlands or so. This helps to make my handlebars very clutter free, and easy to remember once you get used to it. I will probably sometimes think to use my right hand to squeeze a lever for a few days, hopefully that will go away soon. It's also an enclosed brake, meaning that nothing gets in it. Not dirt, rust, water, anything. 

I handed my old bike off to my dad, who needs a new bike for himself after my brother grew too tall for his current bike and decided to start using my dads old one. 

A couple things were less than idea, but not many things. The frame turned out to be too small for my AXA defender lock and I doubt that I will be able to find any that will fit. Leave suggestions with the smallest ring locks you can buy. Because I don't have any ring locks that small, I decided to get a chainlock instead. The rear light is battery powered, although I have no idea how to turn this particular kind on. A small light like this will last a long time on battery, but batteries always find a way of going flat in the most inconvenient of times. 

The design is built like a tank, designed to last decades out in the open (assuming nobody steals it prior to that) even in harsh weather. It will be interesting to see how this rides in the rain and snow, but spring is often full of rain, and sometimes can have snow (it once literally snowed in July in Alberta). But it's much easier to ride than you might have thought one of these "old fashioned" bikes. It felt light, possibly because of my upright position. 

You have good posture, especially useful if you're over 40, and that also gives you good sightlines. You can have chats with your friends and family as you ride, assuming you are in an environment where you can ride relaxed. It's a very easy to use and understand bicycle. 

I only added a cupholder, which I switched over from my last bike. Such is the influence of the American car on just about every part of our lives. Including the beverage holding handlebars I have. 

Price: ~750 dollars, I got a discount to the tune of 150 dollars because the owner needed to get rid of the only bike he had of that kind.

I described some of the features that other similar Dutch bicycles have here in another blog post: I suggest you go and read that too. 

I suggest you get a good lock for any bike, especially one like this which has extra features. I also suggest registering it with a local bike registry, usually the police can tell you who does that where you live, as soon as possible. I will be doing the same once I can figure out what my serial number is. Insurance is another option you probably should look into, get it if you can if it won't be covered by other insurance plans you have. 

My bike was especially hard to find. I saw some decent but not quite bikes, often under the cruiser aisle, which had some non enclosed or hockey stick style chainguards, 3 speeds but no coaster brakes and rim brakes. They can be decent bikes for this purpose, but won't quite be weatherproof, and not having coaster brakes can be not what you want, especially when you want clutter free handlebars and a system that is usable when your hands are tied (not literally) up doing things like signaling and turning. No other accessories are included, no lock (then again, neither does mine), no lights, and sometimes fenders aren't even included. Look carefully, you just might find what your looking for in a crowd of thousands of look a likes. 

This bike also taught me a few things about practical design. Hills should be avoided when you can, especially when such hills are only aesthetic hills built as part of a drainage system. And do whatever you can to reduce the incline. Make sure that you have adequate width for turning, otherwise you will be artificially slowed even when there is no need for going so slowly, like when yielding to cross traffic at a roundabout. Make sure that you have a smooth surface to ride on, either smooth bricks or level concrete or asphalt. Otherwise whatever you may put on your bike might slide off or get bumped too much. Just a few things to keep in mind. 

This bike should keep me happy for a long time, and may very well replace any need for a car for many years to come. 

Thursday, 24 March 2016

When upgrading main roadways, don't forget to not make it a barrier. How through roads should look and function in Edmonton

Many main roads for motor vehicles exist, and many need to be updated to meet Sustainable Safety's standards. But when you do that, don't forget to make the main road not a barrier.

Large roads can easily cut people off from one another. The US and many other places have learned this the hard way when they tried to built urban freeways. But they don't have to be barriers.

The Dutch found that there are many ways of maintaining people's ability to stay connected on either side of a main road. I propose an example road to see how this would work in action.

75 St, down to Argyll Rd, and after 101 Ave North, it becomes Wayne Gretzky Trail (if you aren't from Canada or the US or are unfamiliar with ice hockey, Wayne Gretzky was a star hockey player (he retired, he isn't dead), broke many records, played in Edmonton for a long time), is the route I want to upgrade to meet how Sustainable Safety would organize the roadway. Let's see what can be done to it.

The functionality of the road is the first thing to identify. It has some autoweg like characteristics, especially north of 101 Ave, but it also has parts where it functions like a distributor road. I classify it into two parts. North of 98 Ave, it would be an autoweg, as close to 100 km/h as possible speed limit, and regulated like one. South, down to Argyll Road (where it connects with the rest of my proposed grid of through roads for the city), it would be a 70 km/h urban through road. Both types require as close to free flow of traffic as possible, and a focus on flow not only on links, but also at junctions (flow at least as 75 St is concerned, the side road can be delayed with traffic lights or roundabouts or a right of way junction). Roundabouts and traffic lights are possible on a non motorway, but should be avoided.

Let's make it work on the street itself. The two directions need a divide where there isn't one already, especially at 70 km/h, sometimes more than that in the grade separated areas.

Next, because we want to make it easier to cross the road for vulnerable road users, and we can easily do that with underpasses, we can raise the road level itself up about 2 metres where we are going to place an underpass, so that you only need to descend down 1 metre (the foundation of the cycleway and sidewalk doesn't count). It's called a semi buried approach. This needs to not be a barrier to the communities. This blog post on the Aviewfromthecyclepath website shows how it's done:

The underpasses need to be spaced frequently, about every 500 metres apart. You are never further than about 250 metres away from one. Someone on the blog did some calculations and (converting units into metric because that's what 95% of the world uses) to go from a stop to a speed of 18.5 km/h takes the same amount of energy as sustaining 18.5 km/h for 800 metres. This is partly where I got the indication that the underpasses need to be this far apart, or rather no further than this far apart. You would expend the same or less amount of energy stopping to wait for a traffic light. The other part of that the grid of cycle routes needs to be a very tight grid. Doing the math for the section between 101 Ave and Argyll Road, we would need between 7 and 8 underpasses. And remember, only some people would have the underpasses spaced at the distance limit to them, most people would be much closer.

On the north side of the 98 Ave, we will need a few updates to the cycle crossings we already have. The LRT crossing is almost perfect, it just needs widening and the construction of a separate sidewalk, plus some landscaping and good connections to cycle routes on either side of Wayne Gretzky Trail. Like how Assen not too long ago built a bridge for it's ring road over a canal, and a cycle path that paralleled that canal also was overpassed by the ring road, no change in elevation to the cycle path.

It is not as good to provide a crossing over the Drive at the Yellowhead. It isn't quite on the desire lines for pedestrians and cyclists, So let's create a crossing at Fort Road instead. We have the space to again raise the Drive 2 metres to create a semi buried approach underpass, and of course, return it to normal level at the interchange. It would work like just about any other underpass. Nothing really special. But what would be more interesting is the idea for a cycle bridge over the Yellowhead-Wayne Gretzky Drive interchange. It could be a location where we can build a turbo roundabout, and even if it wasn't, it would still be a much safer and more convenient crossing if we didn't have to interact with the motor traffic at all. Even at well designed signal intersections with this volume, there are still high rates of crashes. Motor vehicles have large and heavy metal shells. Cyclists and pedestrians don't. We can't build under the Yellowhead, at least not feasibly, so let's instead bridge it. We have enough room to provide a 3.5-4% gradient, which is very comfortable to take, and can be taken by all kinds of cyclists. Whether the interchange itself would become a turbo roundabout type, I don't know.
At 118 Ave, you can raise the Drive up a few metres, lets say 2 metres and make 118 Ave go a little bit down underneath, the roadway about 3 metres down, with the cyclists only needing to down 1 metre (trucks and other vehicles are restricted to 4.1 metres, so this will give some extra clearance. It would function probably like a diamond or possibly dumbbell interchange, given the space, I am guessing diamond with traffic lights. The Drive should continue to say up to cross 119 Ave with a motor vehicle and bicycle underpass. It is possible that an interchange with 118 Ave may prove unfeasible. If that is the case, then at least the cyclists and pedestrians should get an underpass.

Most of the little local businesses and collector roads and sometimes access roads can be closed off. We would replace the cycle and pedestrian routes with underpasses, and any non arterial road access that can be accessed via a different route should be closed. This makes it closer to the goal of monofunctionality, no local access, and it also makes cycle and walking routes shorter and more attractive.

Cyclists wanting to go parallel with the road would use the service roads on 75 St between 73 Ave and 101 Ave, that would be connected with short cycle paths where the service street ends. It would look and operate much like this street, except with a larger main road: North of 101 Ave, there would be a separate cycling path on one side of the road in my vision, probably the west side as it would lead to an easier route on the north side of the River. Crossing either in the River Valley or at 101 Ave are options to get from northbound on the east side to northbound on the west side. Southbound is always on the west side. Next to the University of Concordia (not the place where the Concord was invented), there is a cycling path that roughly parallels Wayne Gretzky Drive and will lead almost up to the stadium, and makes for a good route.

Crossing the river either means a new bridge that is completely separate, my choice, especially if it was something like maybe 50 metres away, or on an expanded combined bridge, which would require extensive noise barrier systems. To get back up to the correct level, long loops to allow extra length to climb and thus a less steep incline should be used.

There are quire a number of other main arterial roads that cross 75 St/Wayne Gretzky Drive. Ideally they would be full interchanges. Some of them can't be due to costs, lack of space, etc. 98 Ave may or may not have an interchange. If it doesn't, then it should have a turbo roundabout. In either case, the road would be a very important one for motor traffic, a through road too. An underpass is required under a turbo roundabout, and even for an interchange, an underpass is still a good idea. 101 Ave should just have a simply flyover under or over Wayne Gretzky Drive. A simple underpass of 101 Ave would just be parallel with the Drive with access cycle paths leading to the cycle paths that would be built there.

Now that I've talked about the cycle routes, let's see what can be done to the road itself to make it closer to the goal of part 80-100 km/h autoweg and part 70 km/h urban through road.

The Dutch created a manual for designing roads that would be like this that I (and others) can access for free in English. It talks about autowegen and dual carriageways, what Europeans usually use when they are talking about roads that have a central median in the middle and more than one lane in each direction. This is quite helpful, so I'll draw from that manual, which you can read for yourself here:

An ideal autoweg has a divide between the two directions, a hard shoulder, a large clear zone, preferably 10 metres wide (otherwise a crash barrier is required), plus full grade separation. If interchanges are not possible, then roundabouts are a good option. Traffic lights and sign controlled junctions are to be avoided when possible. The latter two have proven to be among the most dangerous kinds of junctions you can build.

The ideal dual carriageway intended for 80 km/h traffic (I don't have numbers for 70 km/h traffic, but I'm under the impression that it would be safe at 70 as well, although speed limiting measures would be needed) would obviously have a divide between the two directions, with a crash barrier and 3.9 metre wide median, no local access, a non continuous edge line, at least 2 lanes per direction, limited access to local streets, ideally none, is closed to mopeds and bicycles, they have a service road or a cycle path, and no parking. Junctions to be grade separated, traffic light or roundabout controlled if possible.

Thus, the ideal dual carriageway distributor road should look like this: This would also be quite close to how an autoweg would be built, although with 3.25 metre wide lanes and 10 metre wide clear zones on either side, 8 metres minimum without a safety barrier.

Not every place has enough room for that, so safety barriers are likely to be common along here. A noise barrier is needed anyway for much of the length of the road given how busy and fast, and noisy, it is. Adding lots of greenery too would help to shield the effects of this through road from the road. Here is an example of just how effective these noise barriers and visual screening can be:

Other interchanges also have to be built. At Argyll Road, it may be a standard interchange, like a diamond or dumbbell. Otherwise, a turbo roundabout could be built. There are many options for building roundabouts like these. Pretty much the only three remaining arterial roads would be 82 Ave, 90 Ave and Fort Road. Because 82 and 90 Avenues are so close together, and the environment they are in probably restricts interchanges, I think a turbo roundabout should be used. We could make each one into a three way roundabout, 90 Ave allows access to the East, 82 only offers access to the West. Fort Road could have a turbo roundabout or it could have traffic lights. It could be possible to remove the access and require traffic wanting to get on to use 118 Ave or other routes. It might not be possible, there are about 10K vehicles per day. Fort Road shouldn't be a through road and removing this access is one possible way to make it like that, but whether the other arterial roads will be able to handle the extra volume and the distances would still be acceptable, I don't know. Of course, bicycle underpasses are required in all of these cases.

On existing interchanges, the ramps often connect to traffic lights, only from eastbound on 101 Ave to northbound on Wayne Gretzky Drive are there no traffic lights. Many, and probably all, of them can be changed to dumbbell or dogbone roundabouts. They probably won't need a cycle and pedestrian grade separation due to the lower volumes and that you can use single lane roundabouts there. This makes it safer as well.

The road profile itself should change a little bit at other locations. Bus stops would need to be in dedicated bays  with tapers. The bus stops can be built into the noise barrier to make waiting for the bus more pleasant. Hard shoulders would be added in a few other locations on the areas classed as legally autowegen.

This design would eliminate practically every traffic light anywhere here. Maybe one at Fort Road, 90 Ave and 82 Ave, plus maybe Argyll Road and at the Yellowhead, but even if those couldn't be roundabouts or be closed, that would be a dramatic drop in the number of traffic lights. Between 9 and 14 sets of traffic lights could be removed, as well as a lot of the local accesses that constantly interfere with the traffic flow, and is very unsafe.

Electronic traffic control systems help to maximize the speed and safety, so that should be added too.

Why would I advocate for a road like this to be redesigned like this? Don't I want more cycling, walking and transit use? Yes, but this type of infrastructure can help enable that. By attracting more traffic to this road and making more capacity for it, it makes other roads have less volume, less need of four lanes or otherwise more than 1 lane per direction, more pleasant to ride next to, and with less requirement for traffic lights. Cars also by driving here, would bypass areas of the city where we really don't want motorized traffic light downtown. Their journeys may be longer, making it more likely that cycling and walking will be able to compete.

And I do drive, understand frustrations of drivers, and know how dangerous it can be. This road is also not sustainably safe, and by being a congestion prone area, there is more smog, as well as secondary accidents. By making the road safer and less congestion prone, but not making it harder for cyclists, pedestrians and buses to get around, it also helps to make the smog less likely to happen, as well as secondary accidents. And would you want your friend or yourself involved in a crash that could have been prevented by roadway design? I wouldn't.

By adding these noise barriers too, and cycle underpasses, people notice the road less, and have higher qualities of life without all of the noise and traffic right in front of you.

Overall this could be a very modern and capable road in East Edmonton, meeting Dutch design standards which have saved a lot of lives and prevented thousands of crashes and injuries over the years. Why not create a design that makes it so?

Wednesday, 23 March 2016

Removing clutter from the roads

In the suburbs the roads aren't too cluttered, except when you pass by the large businesses near arterial roads, but as you go closer and closer into the city centres, they do become much more cluttered. There are simple ways to remove much of this clutter.

There is one regulation change that is very important here. The establishment of zonal signed regulations. This is a crucial step in making things non cluttered. 30 km/h (and 60 km/h) speed limit zones can be established very quickly and cheaply. Signposting 70 km/h on a major arterial road isn't too cluttered especially given that you don't have many side distributors on them, but in the areas that should be part of the 30 km/h zones, you only need two signs, one on entry, and one on exit. And because you can restrict the number of side streets with filtered permeability, you would need even fewer signs. All you need to do to modify a normal regulatory sign into a zonal regulatory sign is to add a supplemental plate with the English word ZONE. The words looks similar in different languages if they use the Latin alphabet, and given how few words that is, just one, it isn't really clutter.

First, we can check for what signs are really needed. Are any parking regulation signs redundant? Have any businesses put up too many signs, or the wrong signs? These can be removed pretty quickly.

Next, we can check the signs that remain to see if they still make sense and if they can be simplified. Traffic signals that control the different turns independently of each other in Canada (except for a few examples in Quebec) either have a no turn on red sign, with the arrow pointing in different directions depending on what you need, or a sign saying "X turn signal". The right turns on red should be prohibited anyway to motor vehicles, so the turn on red sign is not needed at each signal. By using amber and red arrows rather than the amber and red circles, we have a more precise signal that makes it clear what it's function is that can be understood at a glance. And because we can get rid of the supplementary signs, and in Alberta and Saskatchewan, get rid of the second red aspect, the poles are less cluttered.

Another example is that we often put signs up dictating who goes first. This works on roads like access road-distributor road, but not on most access road-access road junctions. A simple raised table with nothing else works fine. It makes people more cautious, as they may need to give way at any time, it equalizes the streets, and needs fewer signs.

Speed limits should mainly be based on the default speed limits. Distributor roads should almost always be 50 km/h roads, except for a few distributors with limited access that would be 70 km/h roads, and because side roads are few and far between, the few 70 km/h limit signs are not clutter. The limit signs themselves could be improved. They could be simple signs, with just a circular sign with a red border and the speed limit in kilometres per hour, and in the transitional period, a sticker underneath that says km/h to make it clear what it's regulating.

Parking regulation signs are very common. However they don't have to be nearly so common. We can make two simple rules about how parking works. On 30 km/h urban or 60 km/h rural zones, or other areas where such a limit (or less, for example a woonerf) applies, parking is allowed unless otherwise prohibited. On all other roads, it is prohibited unless otherwise signed. Very simple to understand. The creation of parking pays, clearly delineated areas that only have space for parking where it is to be allowed, raised off the road a couple centimetres (about 5), connected with a rolled or angled curb, paved in grey brick with curb extensions covering everywhere that parking may be interrupted (intersection, fire hydrant, bus stop, etc), would make it easy to know where to park, and pretty much impossible to misinterpret.

The signs themselves can also be decluttered. A simple parking bay sign for a road that would ordinarily have parking prohibited could look like this:
  It is very easy to understand, and applies for all the bays between curb extensions. You can add supplementary plates to indicate that it is for loading or unloading, that it can come with a time limit, I suggest looking at this website to see how I got the idea: The guy who came up with the improvements for the UK uses very similar signs, except that we would use a slightly different means of indicating the parking part and we use white rather than blue.

Other means of decluttering the signs is to use more symbols rather than words. Often they are smaller and simpler, and of course are language independent, so the words can mostly go away. This goes for all signs.

Businesses that are located in the pedestrianized zones have a couple advantages. People can more easily walk in to shop. You can't do that easily with a car, you have to find a place to park your car, often pay for the parking, and of course parking itself takes time, as does going from the lot or space to the business. And because you are at walking pace in the pedestrianized zones, you have more time to read the signs. In other areas, where you see sign after sign asking you to drop into the parking lot to shop or buy something, you have more clutter to deal with, already while having the task of driving, let alone the act of deciding whether you want to buy something within the next exit or two, then having to decide what you might want, and this makes it more dangerous. I also see in the US on urban freeways, due to lax control over signs, they've popped up next to them, distracting drivers. When I was on vacation in Washington State, I saw loads of them. This happens in other places too. Another place to declutter.

We should minimize the number of objects we have near the travelway. This makes things simpler, and we don't have objects near moving traffic, including moving pedestrians, that often cause disruption. We have utility lines, but those can be buried underground. Good thing sewers are buried! We have lots of signs to place, and so consolidating signs, removing them when possible and otherwise getting rid of the need for signs also removed a lot of poles. We can put poles out of the way of pedestrians by creating boulevards, often in the space between cyclists and motor traffic on distributor and through roads, if it's of suitable width. By moving transit stop related things onto dedicated space, often between the cycle track and roadway, or sometimes between bus lane and the rest of traffic on some bus roads, we can reduce the clutter for pedestrians and cyclists.

Another thing that we can use is that Sustainable Safety demands that we have self explaining roads, IE, fewer specific regulations, laws and signs are needed to explain what to do because the road design already tells you that. In the 30 and 60 km/h zones, they are supposed to be low volume low speed roads, so you don't need signs to let you know that you should expect things like speed bumps. In Edmonton for some reason, we use a large diamond warning sign at each and every speed bump on most roads that have them, so that's not good.

We can even declutter what people wear. I don't mean your fashion, but we don't really need helmets or high viz on cyclists, and the only time when I would think a pedestrian should wear a high viz jacket is when they are a construction worker or in a similar profession or walking on a hard shoulder, setting up things like warning triangles or fixing their tire. Most people don't need them otherwise. It isn't really a bugbear with clutter, but it is an example of it.

We have a lot of streets with a lot of obstacles in them, pointless signs and things that can be easily and cheaply fixed (maybe except for the utility relocation). Why not make them easy to understand, and meet one of Sustainable Safety's demands for self explaining streets?

Tuesday, 22 March 2016

Trains, large trucks and buses and where they fit in Sustainable Safety

As a slight side break from my Sustainable Safety series, one thing that I didn't cover in my homogeneity segment was that of trains, trucks and buses, at least not in the detail that I wanted to cover. So let's do that now.

Trains are large and heavy objects that when allowed to, can move at very high speeds. Freight trains usually can go up to 70 mph (for some reason the railways in Canada also haven't moved on from mph) or 110 km/h. The Netherlands a while ago built a 120 km/h railway reserved for freight, complete with electrification, double tracking and no level crossings. Much faster than speed limits trucks often have, in Canada and the US this is often around 100 or 105 km/h, or 60-65 mph, and in Europe, trucks over 7.5 metric tonnes are often limited to 80 km/h.

But this blog post is about road safety of course, and at an at grade level crossings, if a train hits anything, there is going to be a big problem. If the vehicle is large enough then it might derail the train too. Many campaigns about road safety often cover level crossings. But as usual, without a natural reason to stop, people often take risks. By removing as many at grave level crossings as possible, there is no crossing to be risky, and provided that there is a fence on either side of the track (s), then it is very unlikely that there is going to be a crash. If you can't grade separate, then the next best option is to fully protect the crossing with flashing lights, bells, and quad gates that also block off the cycle path and pedestrian crossing if they are present, and that they lock into some sort of slot and is designed to withstand a car crashing into it. Adding a bend in the road to require slowing down for this also helps, especially in letting people know where the crossing is after a long section of ordinarily straight road.

This also has the advantage in that you are able to let trains go faster if they are prevented from doing so only by having unprotected level crossings. In the US you can go up to 110 mph (180 km/h) just by having the gate arms, bells and flashing lights (as well as the right signal protection), and up to 125 mph (200 km/h) with impenetrable barriers.

Some advantages of trains is that they don't induce demand for motor vehicles like roadway upgrades tend to. They are also easier to make automatic, and in fact, many trains (although as far as I know, only passenger trains) are automatic. Ever ridden Vancouver's Skytrain? And even if not fully automated, they can have many systems to make them run with minimal driver input, and also far smaller risks. China uses this a lot on their bullet trains. And you need less width for a double tracked railway than for a 4 lane motorway, and with a smaller impact. You can also make them run on electricity more effectively, as batteries are still fairly inefficient, at most giving you back 2/3 the power you put in. You can also run trains further into the city centres than you can with motorways and main arterial roads.

Trucks as identified in Sustainable Safety are not able to be used to fully meet the homogeneity standards unless and until these large trucks only take on the role of being used on the through trunk roads that are grade separated with a head on crash barrier if they are over about 7.5 tonnes in mass. And you can do even better by not having commercial vehicles over 3.5 tonnes used on urban streets. The main semi trailer trucks, and others over 3.5 tonnes, would turn off of the motorway almost immediately into a distribution centre, then distribute the cargo into vans and other light trucks to take the goods around the city. This also has the advantage in that intersections can be smaller, down to the size of buses and firetrucks as opposed to long semi trailers.

Buses also have certain similar problems. They are large, often around 12 tonnes for a non articulated vehicle, but at least they don't go as fast as a truck does on a freeway. But the best solution by Sustainable Safety's standard is to separate them just like cyclists are from motor traffic, but in this case, this is because otherwise they would be a danger to the other road users, not that they are vulnerable. Simultaneously this also preforms many functions that is asked of transit, to be reliable and fast and especially avoids congestion.

The next best option is to route them on distributor roads where the most vulnerable road users are separated in space, including at intersections, and time, at any remaining traffic lights, and to provide them with stopping bays at their stops, so as to not be a solid object that can be hit from behind. They should also make as few turns as possible, so as to help make their movements predictable and minimize the risk of a conflict in direction.

A few other things that can be modified in the realm of vehicle design includes skirt guards, so that other road users and vehicles can't as easily get dragged under, better cameras to cover more blind spots can really be an improvement. There are experiments going on to see how feasible systems to make it so that you can see ahead of a truck if you are behind the truck, useful for certain undivided highways where it could be dangerous to overtake. Trucks could get some elements of automation, at least on motorways and other through roads, as truck drivers often get sleepy due to tight schedules and long drives, and given that your primary job on such a road is primarily driving in a straight line at a speed of 100 km/h for the most part, that is a fairly simple thing for a computer to do.

Some vehicle regulations I propose mainly involve rest breaks. Combined with motorway and autoweg service stations and rest stops every 50 km to take a break, it would be easier to take the legally required breaks with less disruption to timetables. But another law is needed to make it would better. You must be given specific allotted times for breaks and that you are not allowed to arrive before the time is up, and no importer or the person paying the trucker can create any financial incentives in any way to avoid such rest breaks or create a program to that effect.

Large vehicles can be incorporated into the latest and best design standards and safety models, but it requires careful use of controls over routing, size and speed of vehicles, and by separating them from the other types of road users that are more vulnerable and do not match as much as possible. The Dutch still have plenty of commercial deliveries and large trucks, so opponents claiming that buses and trucks will suffer too much do not have a foundation on which to stand, and these improvements to roadway design, vehicle design, education and rules and regulations have saved many lives over the last 24 years since Sustainable Safety has been introduced and for 11 years since it was expanded. It works. Let's adopt it, also for large vehicles and train level crossings.

Sustainable Safety: A thorough look into each of the 5 principles. Part 4: Forgiveness

That news article inspired me to write this blog post so soon. It made me want to state that accidents on the roads are in fact mostly accidents, nobody wanted to commit the road crash. Very few times is anyone ever trying to cause a crash, and of those who do, often times it's for things like insurance fraud or something like hat.

However, the big difference is acknowledging that we make mistakes, frequently, regardless of what mode of travel we use, pretty much involuntarily, and that we design our roads and as much as possible, the vehicles we use, to account for those mistakes, and make them less dangerous, and not lethal, as the crash last weekend shows how bad a mistake can be.

The Dutch call this principle "forgiveness", well, obviously they would normally use the Dutch word for it, vergiffenis, and it's made a huge reduction in the number of crashes, injuries and fatalities on their roads on roads which have been redesigned or originally built to the Sustainable Safety standards.

As much as I and many others would want people to be always able to make the right choices on the roads, cycleways and walkways, if we can walk into lamposts or stub toes by accident, what hope do we have that driving a car at something like 50 km/h being incapable of error? It's a given that unfortunately we don't have the ability to remove.

Let's see what can be done to make our roads forgiving of errors and not relying on perfect behavior, and when crashes do happen, making them as non lethal non injurious as possible.

You'd be surprised how often cars manage to veer off the road and while in some cases you're lucky and all you'll encounter is grass, in a lot of cases you will hit something like a pole or tree, or other cars. This often happens on main interregional through roads, designed like motorways or divided highway. This is because then driving becomes boring, and especially here we design them to be quite straight and with little in the way of scenery. That makes a direct line, but it makes highway hypnosis a real danger, one that kills thousands each year in the US. It makes you nod off, or go to autopilot, that is pretty much only capable of holding the streering wheel in roughly a straight line and hold the gas pedal down. The other part of your mind is in la la land, dreaming about something else.

This can be less of a danger by using tactile feedback between lanes and especially on the shoulders, rumble strips have saved a lot of lives, by creating clear zones, wide shoulders on autowegen and motorways with a stabilized non paved shoulder, and if the thing that you might crash into is more dangerous than hitting a guardrail, then installing a guardrail. Extra campaigns to take breaks at regular intervals, and rest stops to take those breaks safely, as opposed to stopping on the shoulder for example, at regular intervals, around every 50 km, couldn't hurt either. However I should make it clear that these extra wide clear zones greatly increase speed and lower's drivers attention on the roads that don't need to be high speed roads, like on the 50 km/h urban arterial roads and especially the low speed low volume access roads.

Cyclists hit the curb more often than you might think. Having a splay curb, angled at between 30 and 45 degrees, or more preferably, having a flat curb (this would be used when the asphalt of the cycle path needs to be contained), and having wide clear zones as well, preferably grass and soil ones rather than concrete and metal ones, helps as well. The curbs must be low, between 5 and 7.5 cm for an angled curb, and there must never be an upstand, always flush. Even a few millimetres can grab a tire and through you off.

Intersection design as well can really help contribute. When you provide wide areas between the things you need to do, it also simultaneously increases reaction times. When you have as few things as possible that a road user is capable of doing, you have less ambiguity about what they will do next, and if it becomes apparent that they are making a mistake, you have a lot more room to slow or stop or maneuver to avoid the problem if need be. Roundabouts are perfect for this, provided you have the non annular cycle path design, as they give you very long reaction times, they give you very good angles, and they make doing things like driving too fast around a roundabout much harder to do, especially by accident. And if a mistake does happen, the speeds are low enough that you can react to those mistakes, and if the worst happens and a crash does happen, the speed and angles make crashes far less dangerous than the often 50 and 60 km.h, often with traffic going 60-75 km on many arterial roads, speeds we use at other arterial to arterial intersections.

The very existence of cycle paths makes it much less likely that any mistake that a driver does is going to affect cyclists due to the space and physical protection between them. Many drivers veer over a painted line, but it is much harder to veer over a curb.

Vehicles themselves can become better. We have far safer cars than we did back in the 1950s, when if you were crashing into something with a 1 tonne car you would have about the energy of a stegosaurus dropped from a three story building, but we could do better. We could slightly increase the length of the crumble zone, perhaps up to 75 cm, which could be possibly done by shrinking the engine by adopting electric vehicles that need less engine space. We can add skirt guards around the bottom of large trucks, the kind with the space underneath the box part, so that if a truck does drag someone underneath the results should be less dangerous, though certainly not desirable. And we could potentially mandate that each motor vehicle shall have a collision avoidance system, at least on freeways and main roads where highway hypnosis is likely. These things already exist, and Mercedes recently got some government to close off a motorway for a little bit to test out their automatic truck. It had a human driver ready to step in if needed, but it does show that at least for the environments we have on the long distance roads, it shows considerable promise to be feasible very soon.

We have a lot of ways to improve our roads, but the tragedy of that pedestrian killed, all people killed or injured in crashes, is almost always avoidable if we had roads that forgave error and met the other principles of Sustainable Safety. We cannot ignore the elephant, or rather the car, in the room any longer. How much longer until we realize that the Dutch have figured out the way, and that it can be exported anywhere that is willing to receive it? Many people, myself included, and many professionals at universities and roadway design engineers, people who devote their lives to this, often know this. Humans are fallible, and few people will openly admit that they themselves are. It's natural to think this way, but we can control that bias. If you want to go and oppose this, go to the nearest hospital and look around for the nearest victim of of a serious road crash. See whether you can still tolerate your principles making this crash far more likely.

Monday, 21 March 2016

Sustainable Safety: A thorough look into each of the 5 principles. Part 3: Predictability

There are a few ways that roads can be more predictable and have more predictable traffic of all kinds using them, and when you know what is expected of you and what you can expect of others, then everyone will know what to do, when, or at least know it better than they do now.

The roadway design must make it predictable what you are going to find on a road and what sort of regulations and behavior is expected. Many times some collector roads look like arterial roads, and the other way around. 40 Ave is hard to distinguish from an arterial road, especially if few are parking on the road when you are using it. Many times collector roads look like large access roads. Other times, you can't even tell the difference between a built up area and a rural area because of a lack of a clear definition between them. A median that requires that you go in a sharp S curve, go over a speed table and then you pass a sign that says "name of municipality. Speed limit: 30/50 km/h", then you are on a road with concrete curbs on either side, makes it very clear what kind of street you're on.

By making only full motorways having a wide paved shoulder, then you can more accurately predict whether you are on a wide autoweg or are on a motorway. Often times I have no idea about whether the roadway is like this or not. Other examples include that many cities in Alberta and most other countries post either 60 km/h or 35 mph speed limits on urban arterial roadways, but some roads that are slightly older, 51 Ave for example in Edmonton, don't have 60 km/h speed limits. How am I supposed to know which kind of limit to apply where if for some reason I miss the sign? The speed of traffic is no help, it is very easy to speed in Edmonton and I have no way to know whether the traffic is going at the speed limit or whether it is speeding without knowing the speed limit, going at it and pacing the traffic.

The Dutch tie a lot of things together. If you are on a road with brick paving, car parking in dedicated bays, you don't see a separate cycle path and sometimes neither a sidewalk, and there are buildings that affront the roadway, you know that this is a 30 km/h area and should expect to see cyclists mixing with traffic and that you must go no faster than 30 km/h. If you are on a roadway that has at least 2 lanes in each direction, a guardrail and a verge between the two directions, a hard shoulder at least on the right side and you had to get on the road via interchanges, you know that you are on a motorway, even if you didn't see the sign. If you are on a road without much development nearby, separate cycling paths or a service road, there are a pair of continuous centre lines and a dashed white edge line, and are in the rural area, you know that you are on a rural distributor and can go up to 70 km/h (80 in the Netherlands).

When you approach an intersection, the approach will be very clear. Often it will be a roundabout if you are on a distributor road, exiting an interchange or are exiting the 30 km/h zone (indicated by a change in surface from brick to asphalt, the cyclists having a transition into a cycle path, and a few other clues). If you do encounter traffic lights, then the design will also be very clear. A single signal head above each lane, separate controls for the different directions you can go in, and a physical design that closely resembles how a non traffic light sign controlled intersection would look (if the traffic lights go out for any reason, then how else can you expect the intersection to work at all?). As few intersection designs as possible are used, so as to make it easier to have a mental list of what you need to do at each kind.

The behavior of other road users is assisted by this predictability. By creating natural reasons to obey the rules that apply to each road user, for example someone that is expected to give the right of way to someone else will usually encounter things like an increase or decrease in elevation and sharp corner radii. Stop signs are rarely used, partly because using them on mass and places other than where the visibility is so low as to require them creates a lack of predictable behavior because you have no idea whether the person facing the stop sign will obey it or not. When safe behavior feels like natural behavior, safety is greatly improved.

Laws must be logical and easy to understand why you would create such a law for the purpose it was created for. This also applies to signage and regulations. It is easy to understand why you should not go 130 km/h in a 30 km/h school zone. When you face a speed table, tight corner radii, clear yield signs and the markings and physical design from the main road continuing on as if nothing happened (think how a driveway is usually built over a sidewalk in new developments in Edmonton), then yielding becomes logical and easy to understand. This again creates predictable behavior.

The physical design has other ways of making behavior predictable. The roundabout has another thing to make it easy to understand, both by cyclists and pedestrians and also by motor vehicles. When you are in a car, approaching a single lane roundabout, you only ever need to look to the left to see if there is traffic that you must wait for. And you only ever need to look to the right to see if the traffic will yield to you. With wider median refuges and tight corner radii on the exit of roundabouts as well, it is easier to know whether traffic will exit or not, even if they don't use their turn signals. This makes it far more predictable as to who is doing what, when and where. Cyclists and pedestrians also benefit. It is easy to know what direction to look at to see where the traffic is coming from. You look towards the roundabout to see the traffic coming from the roundabout and you look at the approach arm to see where the traffic from the approach is. And being able to do this fairly far away from the main roundabout and on completely separate paths with plenty of room to divide all of your tasks into easy to follow algorithms makes it easy to know what to do, when, and where.

People often talk about cyclists being unpredictable. Well, this is because the rules applying to them and where they should cycle is not often logical, is not easy to explain and yes, education could be improved. If you don't provide cyclists with a separate path when they feel intimidated by the traffic, where would any reasonable person expect them to be if there is a sidewalk? On the sidewalk away from motor traffic in almost all cases. And if you are a motorist and expect this, only to find a cyclist on the road by surprise, this can be a very dangerous situation. When the traffic light delays are excessive and pointless, and especially when the "flow of traffic" is being prioritized, then people will cross on a red light. If you give a cyclist a sidepath but don't make the intersection design the way the Dutch do, with cyclists having their own specific crossing, signals if applicable and signage, markings and other cues like continuous surface of the cycleway over the side road, then expect unpredictable behavior because there is nothing to suggest what to do if the sidepath suddenly becomes a sidewalk for whatever reason.

We have a lot to do if we want to make roadway design, laws and actual behavior predictable. I recommend this article for another overview and for more information: I hope to see you back for part 4: forgiving road design.

50 Bollards Game

There is a game going around in a few locations asking people where they would put up to 50 bollards in their neighbourhood to prevent shortcutting or otherwise to channelize the flow of traffic. I am going to play that game. Link here to the concept:

Here is a map of my neighbourhood:

Fairly small neighbourhood. Let's see where the bollards can go. I indicated a bollard using a red dot and a number.

As you can see, we don't need that many bollards to prevent shortcutting or to create filtered permeability. Only 7 bollards needed. It may be that more than one bollard is needed to make a road look like you can't cheat and go between the bollards in a car, but we could use up to 7 (evenly) to make this happen if we really needed to here. 

Here is my basis for why each bollard exists. 

Bollard one: This removes an access from a large distributor road, creates an unraveled route and simplifies the intersection. It also makes it so that traffic cannot shortcut on Bowen Wynd. 

Bollard two: The school drop off zone exit is going to be located here, so simplifying the intersection would be a good idea, from a future 4 way to a 3 way. It also removes a potential congestion avoidance route once the school is opened. It also creates another example of filtered permeability. 

Bollard three: This makes the whole Barnes Way and Barnes Link area into an area where no shortcutting is even possible, simplifies the collector road and removes a reason to turn so close to the Ellerslie Road intersection. And of course like all the others, creates filtered permeability, creating shorter routes by bike and walking than by car. 

Bollard four: This one is a more interesting one. This bollard makes it so that the part of Blackmud Creek Drive just north of the pond is a residents only area, no traffic that could otherwise be using the Crescent will be using this residential area that would be downgraded to a 30 km/h zone. It also simplifies the intersection to the west. And also, it makes it less likely that traffic will shortcut through the neighbourhood in general by requiring a longer route if you aren't on a bike or foot. 

Bollard five works in conjunction with the bollard just to the west. It makes it so that it is harder to make a mistake and assume that the collector road is still a through route for cars. It also requires the use of using the little triangular bit that makes crossing to the collector road a 90 degree angle one, slowing speeds down. And of course, more filtered permeability. 

Bollard six makes the area to the right a no go zone, just a system of access roads. No motor traffic uses this without needing to be on those very streets. 

Bollard seven: The final bollard(s) I would place in this neighbourhood goes here to simplify the crossing with the collector road and to make the access roads no through zones for motor traffic. 

These bollards also channelize traffic onto the collector roads here, which would be simplified into a single collector road that just happens to take a pair of 90 degree bends. Motor traffic does not really suffer. An extra minute to your trip is nothing to worry about in a car. You say how comfy the car is, all the things it can do, and you don't want to spend another minute in it? Emergency services can have keys to unlock and fold down bollards if they need to. Many of these measures can also be done without bollards. The bollards are just to naturally enforce a rule that would say that you can't come this way. They can be replaced over time with curb extensions to make only about a 3 metre wide access for cyclists and emergency services with forgiving curbs (the 30-45 degree angled kind), and only if that is proven in that location that even with the curb extensions, motorists still come through, would the bollards be used. They can also often be the flexible kind, so crashing into a bollard shouldn't be too big of a problem, and with well lit and striped bollards, visibility should not be an issue. 

Why not play this game for yourself? Where would you put up to 50 bollards in your neighbourhood to make the streets more livable, create filtered permeability, closing off shortcutting or rat running routes, whatever reason? 

Note that this is just the start of making streets more livable and human scale and people friendly. Many more things are required, low volume 30 km/h access roads, cycle tracks/cycle lanes on distributor roads, and safe intersections to link them. 

Sunday, 20 March 2016

Comparing road deaths and injuries with other forms of more alarming ways to be dead

Joseph Stalin may have once said "One death is a tragedy, a million is a statistic." As we are about to see in this blog post, that is very true of our roads.

People remember things like 9/11 as tragedies, with posters and slogans saying "Never Forget", very well because they involve startling and uncommon ways to kill and injure. But more mundane and things we think are safe, really aren't. Note that this post is in no way denigrating the effects of any casualties. I recommend you read this blog post:

Each year in the US alone, about 70000 people die every 2 years as a direct result of road collisions. Quite a number. Let's see how many the atomic bomb Little Boy dropped on Nagasaki, Japan, killed. Around 40-80 thousand. So we are quite close to the number killed. Why does the atom bomb feel so dangerous and yet cars seem so safe? 

The town of Millet has about 2000 people living in it. Assuming someone strikes at a time when no visitors are present, and kills everyone, we would have about the same number dead as those who already do in traffic collisions. 

If we eliminated 4 schools with 500 people in them each, then we would also have pretty much the same casualty rate as the national traffic death number. 

And let's go worldwide. Around 1.2 million people (as of 2012) die each year in traffic collisions. That would be like wiping off the entire city of Calgary off the map. As much as Abu Bakr would like that to happen, it would be the same number as the worldwide road death count. 

Let's see what would happen if someone were to go postal and kill about 30 people each year, going around to random locations in the city, that would be the same number as Edmonton's current count from traffic fatalities. About as many as those murdered. 

Let's go to injuries. How how about if we maim about 50 million people, the same as living in the following cities combined: Philadelphia, Tokyo, Los Angeles, London (UK), NYC? Same number as those injured by traffic each and every year. 

One in every classroom of students in each school in the UK will be killed in a traffic collision by around the age of 20. Would anyone tolerate the idea of one in every classroom of students being murdered? 

Vsauce has a video about how people perceive risk. Link here: It talks about how startling and surprising means of casualties are often those that are the worrying kind. Airplane crashes, despite how infrequent they are, are generally seen as more of a problem than cars do. Emissions from planes do cause problems though, but directly, cars are more dangerous. 

Edmonton adopted Vision 0 in September. However, they have not adopted the required changes to roadway design manuals and standards, or even any guidelines. This makes it that it looks like education seems to be the only thing that might change, and probably in only small ways, if any. But it is always the mistakes that lead to crashes and casualties. Education will only do a small amount for preventing mistakes. It does nothing to rectify the results of the mistakes, other than perhaps educating people about how to provide medical aid or something like that. And this goes for everywhere that claims to adopt Vision 0 but does not redesign roads like Sweden and especially the Netherlands does. 

Drivers and other road users are put in the unfortunate position that because their mistakes are not prevented or forgiven by roadway design, they become killers and maimers just because of the bad luck of being in the wrong place at the wrong time. This is why I have said elsewhere and here that only those who are being reckless and those intending to cause a problem should feel the force of law, and the media and families pouncing all over them. But what our laws, families and media should be focusing on is preventing the crash from happening again. So long as our roads are the same, the same crash is almost inevitable. We can't change the past, but we can affect the future. 

The Dutch and the Swedes realized this around the mid-late 1990s, By changing their streets systematically and before waiting for a bad crash record, they have achieved a low number of traffic collisions. By investing in cycle infrastructure, the Netherlands has the safest streets in the world for cyclists. They compete for lowest traffic deaths, but only because they use the right mentality. 

I find our lack of caring disturbing. Do I have to force choke you to realize this (those who've watched Star Wars knows what I'm talking about)?

40 Ave

Last year, 40 Ave had bike lanes. Then, in the second half of that year, they were gone. And now the city intends to build new cycle infrastructure in this area. Let's see each of these steps close up.

The bike lanes actually met Dutch guidelines for ordinary cross sections. 2 metres with a 1 metre wide buffer. Yet this did not entice people to ride a bicycle much. However I was one of those who did ride my bike often, as it was a route between Harry Ainlay High School and the library, which I use to get books and movies. Among the reasons why it did not entice people to ride was that it did not provide a network. A couple connections to a couple sidepaths and bike lanes on 106 St, but they are substandard paths (by Dutch standards). It was missing a coherent network to the rest of the city. It also lacked a connection to ordinary homes. It lacked 30 km/h side streets with low volumes (or wide bike lanes on certain collector roads). 

Another reason was that it did not feel safe enough to ride here. Cars passed at speeds over 50 km/h, and the bike lane was just that, a lane, that was not protected. It also completely failed at intersections, with a bike box at 106 St and right turn mixing zones at other intersections. 

The perceived lack of use resulted in their removal by city council vote. This was a terrible failure. I got completely cut off, there was nowhere else to ride that was anywhere even remotely close to as subjectively safe (it was not too difficult for the enthused and confident, and anytime you cut off a route, you MUST create an alternative that is safe and fast. The city never did that when the bike lanes were removed. No temporary path, nothing) as 40 Ave was. As did a number of other cyclists. Also it made it feel like the 4 lanes were essential, despite only about 6-7 thousand motor vehicles per day. EASILY low enough that 1 lane per direction will work. It also felt like a confirmation that cars could win over active transport, and that anger pointed at an outgroup works. 

So at least provide a temporary cycle path on 40 Ave, all it takes are some orange barriers filled with water and some signs. The Dutch do this any time that a cycle path or cycle lane needs to be rebuilt. 

The removal came with a promise to have better bike routes on 40 Ave. But now the city (and Michael Walters) is backpedaling on that promise because they are considering having a route on the 43 Ave/pipeline corridor. By offering this route as a replacement as opposed to an addition to a route on 40 Ave, you backtrack on your specific statement that you will have a high quality cycle route on 40 Ave, and that it disregards the fact that the people on 40 Ave, plus the businesses and schools too, will lack a safe route to even get to 43 Ave. 

The Dutch cycle grid of routes goes everywhere. And I mean everywhere a person could be expected to want to go to. Every single home, every single school, every single business, house of worship, park, and other destinations. All of them. 40 Ave has those destinations on those, and thus requires a cycle route on it. Not a couple hundred metres away. On it. 

43 Ave would be a decent route actually, it would require a 3-4 metre wide cycle path and a 2 metre wide sidewalk, an underpass under 111 St due to the barrier that the LRT creates, and priority crossings over side roads, with no wheelchair ramps, the road raises up instead to cross it, and would need to be in a straight line. But the city doesn't want to do this. They propose a 3 metre wide shared use path, no sidewalk, and no discussion at all of an underpass under the LRT and 111 St despite the huge barrier they are (all roads that big are barriers). I don't know how they intend to build the side road crossings, but given how the city has designed shared use paths before, and how they continue to design them, it is almost certain that they will build them in the way they have before, no priority, visual or legal, a wheelchair ramp design going down to cross the road, and only sidewalk connections to roads that cross 43 Ave or the path. 

Even if 43 Ave gets a cycle path and sidewalk like I described, it must only be a complement, not a replacement, to 40 Ave. 

40 Ave is described as having the option to have a 3 metre wide shared use path on one side (which isn't described), or cycle tracks. How wide the cycle tracks is never described, but I have a feeling like they would be 1.5-2.1 metre wide cycle tracks, with a narrow curb protecting it, and probably something resembling a mixing zone at intersections, and no discussion of a safe crossing of 111 St. How the cycle tracks will be designed at intersections is something I can't know for sure, as the city has never built a cycle track-intersection before. But it will probably resemble the complete streets guidelines.

But a key element is that cycle tracks are only an option, not the one the city wants to build. They are proposing a shared use path. And of course this is only assuming that 40 Ave is chosen. And 40 Ave in any case is being proposed to retain 4 lanes despite the lack of warrants to require it. Wide roads provide extra dangers, a much higher chance of speeding, sideswipes, making it harder to cross the road, especially when walking or cycling, and it just wastes space. Plus by keeping the 4 lanes (and the parking), it makes it very hard, possibly impossible, to provide a cycle route on it without requiring property aqusition. 

Much of the claims of designing for 8-80 cyclists often divide cyclists into fast and experienced and slow and novices, commuters and children, and similar groups. This is a HUGE mistake. There are fast cyclists, but what about them makes them always opposed to cycling on a well designed cycle path? Experienced gives us no help here in terms of who were are supposed to be designing for. All it can say is that they are more used to cycling in general, not that they are any more desiring of mixing with motor traffic. A grandmother may have decades of experience cycling but do you think that she would be interested in cycling with motor traffic just because of that? Novice cyclists are only more likely to make mistakes that would be avoided if you knew how to cycle better. Cyclists must never be artificially be speed controlled. If there is a hill that isn't artificial because we don't have magic wands to wave and make everything flat, but when we can design for cyclists, why not design the cycle speed to be 50 km/h, just like the cars nearby on the same road? Not many cyclists would go that fast, but if there is a clear enough cycle path with not too many cyclists in front, if you want to, you should be able to go at 50. If it's good enough for 50, it will also be good enough for something like 20 km/h that an 8 year old might ride at. 

I also at that meeting at the Royal Gardens asked how the city is going to not alienate drivers by making it seem like that is an eternal conflict between them and cyclists, and by not alienating or dividing and conquering cyclists by dividing them into groups between fast and furious and slow and novices? All I got was that it was a good beginning of a discussion for healing the relationship, but not any answers as to how. I also asked about making sure that cycle infrastructure does not appear to be dependent upon a pretend lease that requires perfect behavior. I used my mom as an example of how everyone, cyclist, motorist and pedestrian, is fallible, by driving at 80 km/h on a 60 km/h arterial road. 

Here is my 40 Ave proposal. The representatives at the meeting said that you would check it out, leave a comment below telling me what you have learned by reading this to prove that you have checked it out. 

On 40 Ave, at 120 St, to the west of this block, Aspen Gardens becomes a 30 km/h zone. There are no collector roads or main arterials here, just some roads that lead nowhere for through traffic. So a 4.5-5 metre wide shared 30 km/h low volume roads or 3-4 metre wide bidirectional road lane plus 1.75-2.5 metre wide cycle lanes (designed like advisory cycle lanes) is used to calm the traffic. At 120 St, the cycle lanes start if they haven't already and they turn into a cycle track with a good transition over something like 10-20 metres. At 119 St, the road design is like this: 

It provides cycle tracks on both roads, with the protected intersection design. The traffic light phases are likely to be separated, at least from the westbound side. Right on reds are prohibited. 119 St as it is also low enough in volume would be reduced to 1 lane per direction, but given how that traffic flows during the hours, it requires traffic lights. 

40 Ave is a 1 lane per direction roadway, with this sort of design. Parking lanes 2 metres wide on one side of the street is acceptable if required, but not on both. 1.5 metres of a curbed verge must in any case protect the cycle track. 

Crossings of roads looks like this, although having a divide between the two directions would be very useful here, due to the volumes of traffic. 

This continues all the way until about 100 metres before 111 St. Given the sheer size of the road, I propose an underpass for cyclists, mopeds and pedestrians on one side of the road, with a 2 metre wide sidewalk and 4 metre wide cycle path. Because I propose one way cycle tracks on both sides of the road, how can you cross safely? You can have a median refuge allowing for a transition between sides of the road. Crossing the road at right angles, with good sightlines, crossing where the turning lanes for 111 St haven't started yet, and where they won't for a number of metres, with a wide median refuge, even with cyclists yielding, the volume is low enough that such a crossing won't be a problem. In any case it would be a shorter amount of time than to wait for the traffic lights. With a gradient of no more than 4%, it would be an easy incline for everyone. Elevating 111 St would be a good idea if that is possible, up to 1-2 metres if possible. 

At places where there are bus stops, the road can look like this: Buses stop in dedicated bays, not obstructing traffic, with a smooth transition back into the road, and with cyclists always well protected. Looking like the bus stops in this video:

At 106 St, I propose a roundabout. The roundabout circle for cars should be about 32 metres in diameter. Cyclists have bidirectional cycle paths, at least for the part of the intersection where you haven't picked which arm to exit on, and are thus 4 metres wide, given that these are primary routes. All of the approach arms would have single direction paths 2.5 metres wide each. Cyclists yield to motor vehicles (given that Edmonton drivers aren't used to cyclists in general and even fewer would expect that they should let a cyclist have priority while crossing a distributor road, and few are used to roundabouts, not having priority is even more important than in the Netherlands, where it is proven dangerous to have priority, at least on the annular cycle path designs), but have 90 degree crossings, crossings between about 6-10 metres away from the outer edge of the roundabout for cars, cars having to have their tires pointed towards the centre of the inner island, not flared out, on a speed table with wide medians and volumes under 1500 vehicles per hour, this will work well. 

Another advantage of the non annular cycle path design is that you can bypass the roundabout by going from eastbound to southbound without stopping and you can make right turns to and from 106 St without even stopping or giving way at all. Roundabouts have proven to be very safe and traffic lights here are a big time waster and are more dangerous than roundabouts (given the right design). Pedestrians have priority over cars in any case, also on their own raised table with a zebra crossing. 

This design makes for very convenient and pleasant cycling, with the biggest three delays, 111 St, 106 St and 119 St not a problem, ignoring 111 St altogether with an underpass, having a pleasant cycle path around 106 St-40 Ave roundabout with 30 km/h cars on the circle, and with a protected intersection on 119 St. It would also be sustainably safe by Dutch road design, minimizing conflicts, and reducing the speed at the remaining conflict points. It will be a very safe and attractive design should it be picked, and a well used route if it forms part of a grid, and connected to every home around here with 30 km/h low volume zones, even using filtered permeability to make cycle routes shorter than car routes. Underpassing 111 St also avoids the biggest delay to cars if you ride a bike, walk or ride a moped like a friend of my brother does. 

This is a big opportunity Edmonton, don't screw it up! 

(You still have time to respond to the survey. Help by complaining about their "fixed" options and make them use Sustainably Safe designs)

Safe Enough?

One thing I want to caution you about is whenever any road authority (provincial department of transport, the DfT in the UK, etc) says that something is safer, ask them to explain what they did, how they believe it is safer, and if the end result is safe, not safer. A decrease from say 100 injuries per year to 99 injuries per year at an intersection is not "safe" in my opinion, just that it is technically "safer".

And when they say that it conforms to their guidelines, ask them to produce a copy of their guidelines and get them to defend the guidelines. Ask them whether they ever considered using the Dutch guidelines given that it has proven to make streets safe in many cases and far safer in many others. If the end result does not meet Vision 0, as in, no serious injuries and no deaths and as few minor injuries and crashes, preferably 0, per year has not been achieved, then ask them why they don't adopt Vision 0. If an audit or other study about safety has been done, ask for a copy. If they are so confident in their design, then they should gladly give you a copy.

If they cannot defend their plan, then it must not be done and it must be changed if it has already been implemented. I encourage you to read the rest of these blog posts, to read up about Sustainable Safety and see how it has make streets livable and safe, and to criticize plans that fail by Sustainable Safety standards.

Friday, 18 March 2016

Sustainable Safety: A thorough look into each of the 5 principles. Part 2: Homogenity

The second (out of 5) core principles of Sustainable Safety is that to the greatest possible extent, the differences in masses, speeds and directions on any particular thoroughfare whether it be road, cycleway or footway, be as close together as possible.

It is fairly intuitive as to why. You know by intuition that a large object heading your way at high speed is not going to result in a good outcome on your end and probably the other object's end as well.

It is literally not the speed that kills, but the crash, or at least one that causes problems. So preventing the crash is the ultimate goal, and the secondary desire is that any crashes that do occur, occur at low speed low conflict environments.

I know technically it's acceleration, but when you have a net acceleration (which oddly enough also includes slowing down) of 0, you can be said to effectively have speed rather than acceleration.

Let's see what we can do to make the roads homogeneous in terms of mass.

We can look at each type of road user and see what masses they have. Pedestrians have around 50 kg of mass, give or take about 45 kg depending on the person. Cyclists have a mass of around 100 kg, depending on the bicycle, it's rider and whatever the cyclist is carrying with them. Mopeds and scooters have a maximum mass in terms of the machine of 55 kg, and the rider is the same as with a normal person, for about 150 kg or so. Mobility scooters have up to about 410 kg depending on the rider and model. Cars can have a mass of anywhere between around 750 kg and 1250 kg depending on the model. Newer cars tent to be heavier. Light trucks and vans can be up to 3500 kg. A bus or truck may be 12000 kg or more.

So the ideal road should put pedestrians and mobility scooters (up to a certain speed) on the footway, cyclists, mobility scooters being ridden above the sidewalk speed, and mopeds. The roadway should combine anything larger, although in some cases buses may have their own busway due to the fact that a 12 tonne bus and a 1.2 tonne car means that due to the squaring law of magnitudal increases in kinetic energy, that a collision is going to be 100 times worse with a bus than with a car. This also is often why buses and trucks tend to have lower speed limits on freeways and expressways in many countries. Buses and trucks should not be present at all if possible on a road where small road users are present without physical protection and thus are vulnerable.

To look at how to homogenize speeds, we can look at actual travel speed for the various types of road users and see how they may conflict. Pedestrians and cyclists while they can in low enough volumes and speeds mix reasonable well, at least in terms of how they travel when going parallel to each other (IE not at pedestrian crossings), they do not mix well at normal travel speeds of cyclists and in the normal volumes you are likely to find them. Thus they must be separated except in the pedestrianized zones and low volume access roads.

The difference in mass and speed between cycles and cycle like objects (mopeds, scooters and mobility scooters) and motor vehicles must now be considered. A number of studies have shown that the maximum safe speed for mixing (assuming you have low volumes, thus low amounts of mass, of motor vehicles) is 30 km/h. More and the differential rapidly and predictable causes many more casualties, because of the fact that doubling the speed squares the energy.

Swedish and other research has shown that the maximum speed that a well designed vehicle is capable of handling in a T bone crash is 50 km/h. As you may expect, the Swedish default speed limit for their version of distributor roads is 50 km/h, due to the number of side roads, including other arterial roads.

Also by Swedish research, the maximum speed for a head on collision to be taken safely by a well designed vehicle is 70 km/h. As such, modern practice indicates that a road should have no more than a 70 km/h speed limit without a divide between the two directions. Of course this doesn't apply on a one way road.

But back to cyclists and pedestrians. In the rural area, access roads are limited to 60 km/h. The roads where cyclists and pedestrians share with motor vehicles are very low volume. They don't provide any use at all except by those who live on that very road. They are usually 4 metre wide asphalt or concrete roads with some space on either side as a clear zone. Or they may not even be paved at all, just some crushed seashells or grass, as this prevents mud from forming so easily when wet. On the busier access roads, like the route between two villages or small towns, there is usually a separate cycling path. One way or two way, it doesn't matter so long as they suit the functional requirements of that very road.

On roads where the head on and side crashes are not a problem and are prevented by physical design, the speed limit can be higher. On roads where there is no hard shoulder and not at least 2 lanes per direction, and interchanges cannot be assured on every junction (or no direct access to the through road), there is a maximum of 100 km/h in the Netherlands. These roads are called autowegen. On roads where there are at least 2 lanes per direction, controlled access, hard shoulders at least on the right side and a crash barrier between the two directions, then it can be called a full motorway. Due to the problem that motorways have very few ways to enforce their speed limit by design as things to physically prevent speeding are unheard of, nobody ever puts a bump on a motorway for example, the speed limit is hard to enforce without the speed being selected by the road users themselves. This doesn't apply to vehicles with limiters in their engines/motors.

So what can be done about this situation? Well, as best as can be done, the 85th percentile speed standard can be brought back as conflicts are already removed, and the problems arise when you have vehicles traveling at different speeds, some at the legal speed limit, some going at their comfort speed, which tends to be between 125 and 135 km/h. In Germany this is usually around 180-240 km/h. But we aren't ready to autobahns if we ever will be, and 240 km/h uses a lot of power. So a speed of between 120 and 140 km/h is a good speed. 130 km/h is the normal motorway speed limit throughout much of Europe. And if it isn't, it's usually 120. So adding these to the speed limit options will ensure that speed is not a problem on full motorways as much as possible.

The final major conflict has to do with direction. If you have a different direction to that of anything in your path, then there is likely going to be a big problem because you will suddenly change their direction as well.

The Dutch, and a considerable chunk of Europe and Australia (and NZ), even Carmel Indiana have found out that roundabouts are great at reducing the direction problem. Especially when you combine it with the lower speed. It also makes that you can easily predict the direction that other traffic will come from, the opposite of the driving direction (from the right in the UK, from the left in the Netherlands and here in Canada). They have been found to be very good at reducing crashes where side conflicts are possible. The Dutch even use them on 100 km/h autowegs, although they are less preferable to full interchanges.

Roundabouts of the annular ring with priority design have also proven dangerous due to the poor sightlines, the inability of cyclists to assure their own safety and drivers having to do too many things at once. The non annular, and unless experiments prove otherwise, non priority, designs are the way to go, unless you grade separate the roundabout, in which case at grade crossings are not any problems as they don't exist.

When this is not possible, on non through roads it is generally better to have intersections controlled by yield signs (and sometimes stop signs), with generous divides between the different things you need to do, (let cyclists go first, then traffic from the left, then the right, etc). I covered this in a blog post about doing one thing at a time. Something like 4-6 metres will do the job.

On access roads, no control is required. The volume will ensure that there is no gridlock by having to yield to the right, and the low speeds mean that crashes are less likely and bad results even less likely.

Uncontrolled crossings where there is no divide between the things you must do, as in you must do everything at once, but one road has legal priority over another, has proven to be very dangerous, often the site of fatal crashes and injuries, especially when cyclists and pedestrians are involved. So these designs should be minimized to the greatest possible extent.

Interchanges have the highest capacity, and the highest speed allowable, but they also are costly and need a lot of room, and thus should be limited to through roads as much as possible on limited corridors. However they are among the safest designs when roundabouts are not used, as they have similar speeds between vehicles and similar angles, with no crossing traffic. They are safest when traffic does not weave, which is why cloverleafs tend to not be used anymore for systems interchanges.

The next topic is predictability. Stay tuned for that.