Sunday, October 26, 2014

What place for buses in a properly designed city?

One thing I noticed quickly when I found a document grouping together trip surveys for many Japanese cities was how low bus mode share was, everywhere, almost never more than 10%, even in big cities often only 3 or 4% of trips were made on buses. That led me to wonder why it was so low. 
Mode shares of Japanese metropolitan areas (not just central city)
One explanation is that the Japanese attitude towards transit, namely that it always must pay its own way rather than seeing it as a public service that deserves subsidies hurts buses a lot. For instance, recently the Tokyo Metropolitan government canceled a night bus route for having run a 34 000$ deficit in the last year... keep in mind the average subsidy per bus route in New York City is 7,3 million dollars a year. To pay back their operating costs, they either put fares that are too high, discouraging use, or have little to no service outside of peak periods. Which is a pretty bad news for the efficiency of buses.

Another may be that most Japanese cities have very few straight roads, and to offer good service levels (good speed and comfort), buses need the same type of road that cars need: straight, wide with high speed limits (the same kind of road we largely do not want in our cities). Yes, if you want fast buses, you need the same kind of streets that leads to fast cars. These are often rare in Japan... but there are exceptions, like Kyoto, which follows a grid of wide avenues built in an Imperial Chinese style, and the cities of Hokkaido (Sapporo, Asahikawa, Obihiro, etc...), and buses aren't much used there either. The lack of bus lanes certainly do not help, as I wrote about, congestion is the main transit killer as it lowers speed, makes it less attractive and makes operating costs higher.

Thinking about it more, I think I have come to a conclusion that some people may not like very much: buses simply do not have a large role to play in a properly designed city.

What is a properly designed city as I say it?

A properly designed city as I describe it is a city that has the following features:
  1. It has a relatively high density, from 6 000 to 20 000 people per square kilometer or more in residential areas (15 000 to 50 000 per square mile)
  2. It is comfortable and pleasant to walk, with a porous street grid so pedestrians can walk directly to their destinations and not have huge detours to do
  3. It has good mixing of uses so that businesses and job centers are located in optimal locations to reduce travel distance to the largest number of people as possible
  4. It has rapid rail transit along the most important travel corridors if it is big enough to warrant it

My demonstration, a mode race

My reasoning is based on, essentially, a race. My own thinking is that speed is the single most important factor in determining mode use. Some other factors may pop up: cost, availability of parking, perceived safety, etc... but speed is the single most important one.

So let me do a quick exercise here, a race if you will between different modes of transport. First, let's start between walking, at 5 km/h on average, a bus, supposing a high frequency of one bus per 6 minutes and frequent stops, going at a commercial speed of 15 km/h (about average for buses in urban areas) and a subway, going at an average of 35 km/h, at the same frequency as the bus. The transit modes both require walking to and from stops, which compound the wasted time in their trips. (divide by 1,6 for mph)

Time (in minutes) versus distance (in meters) for three different modes: walking, bus and subway
The three colors on the vertical scale represent the zones where each mode is faster than the two other modes. Basically, it means that below 800 meters (half a mile), walking is the fastest mode of all three, buses are fastest between 800 meters and 1 500 meters (roughly half a mile to a mile), for anything above that, subways, if present in the traveling corridor, are faster.

That is pretty dire situation for buses. They're almost perfectly squeezed out between just walking and subways, only the absence of subways can save them, but if the cities follow a transit-oriented development pattern along subway lines, it's not good for them.

Worse, in the following, I will add bikes in there, the green line:
Here, bikes completely squeeze out buses, and not just by a little. They also squeeze out walking. They do so largely because of the absence of waiting time or need to walk to a given stop and having roughly the same speed as buses.

The only thing that saves buses is that not everyone wants to bike or even can. Cyclists are exposed to elements and are rarely given adequate infrastructure to travel upon, and some people are disabled and unable to bike for long periods of time. Some people may find biking too exerting or dangerous, or hills can convince people not to bike.

But even if we ignore bikes it still looks pretty bad for buses. With buses being slower than walking for distances under 800 meters, that means that the walkshed of any given point, supposing a grid, would look like this:
Area where walking is faster than buses or subways
That area is 1,3 square kilometer. Supposing a residential density of 6 000 people per square kilometer, the lower end of what I considered a "well-designed city", that's about 8 400 people living in that area, and up to 26 000 people using the higher end of the residential density I described. If you have proximity stores or neighborhood stores like groceries, in a well-designed city made to reduce distances, you'd get a pattern of nodes of neighborhood stores looking a bit like this:
Kooriyama in Japan, with the red dots being the results of a search for "supermarket" (BTW, 45% of trips made on foot or on a bike, only 5% on transit)
If I were to draw the "catchment area" of each store, they would look like diamonds or hexes of semi-regular dimensions, so that everyone lived within about 1 kilometer maximum from the nearest grocery, which is barely out of the zone where walking is faster than buses. So the vast majority of people in Kooriyama can get to the closest grocery faster on foot (or on a bike) than they could on a bus, even supposing there was a bus line taking them straight to the grocery.

My home suburb of Boucherville has 5 groceries for a population of 40 000 people, that's 8 000 people per grocery, so that means that the walkshed I illustrated above would have enough population to sustain one grocery store even at the lowest acceptable density of 6 000 people per square kilometer. Restaurants and convenience stores and other neighborhood stores could also be present as they require even less potential customers to subsist. The result is that most daily needs could be satisfied with trips of a distance of less than the 800 meters under which walking is faster than buses. Using transit would only be required for longer trips like for commutes or to go to malls with specialty stores (electronics, large clothing stores, etc...).

That's not even mentioning cars here.

So overall, it seems like buses are not really destined to be a major mode of transport in a well-designed city. For most neighborhood trips and trips to local stores and restaurants, walking would likely be faster than taking a bus, no matter how frequent they are. We can see this with the mode share of Japanese cities, or even with the mode share for internal trips in the Plateau-Mont-Royal neighborhood I spoke of in my article about streetcar suburbs:
Plateau-Mont-Royal internal trips' mode share
If you create an entire bike path network like what is seen in the Netherlands (or in Japan considering they shamelessly use their wide sidewalks on arterials as multi-purpose paths), local buses will never be widely used. That's the case in cities in the Netherlands that lack tram services and other forms of rapid transit like Eindhoven, which has a 40% mode share for bikes, but only 5% for public transit (source). However, rapid transit can still be dominant for long distance trips in such a case, especially for commuting.

That's not to say that buses are irrelevant for urbanism. They still fulfill a niche role of providing mobility to those who have trouble walking or during climatic conditions where people do not want to walk, much less bike, long distances. However, as I said in my earlier comparison between Sapporo and Montréal, a very high mode share for local buses is probably not a good sign in terms of urban planning and sustainable urbanism and likely a symptom of poor zoning and of an automobile-centered urban design that forces people to use motorized vehicles to get around even for short local trips. Buses can also suffice for small cities that are relatively compact so that they do not justify investing in rapid transit, but even then, investment in biking facilities could probably yield better results.

Monday, October 20, 2014

Highway-oriented development: know thine enemy

People in urbanist circles talk a lot about TOD, Transit-Oriented Development. I've personally described one clear example of it when talking about streetcar suburbs. But I think it's important to look at other forms of development too, especially the current type of development which might be called car-oriented development. It's very important because:
  1. It's important to understand how we've built cities if we want to reform them into something better, to see how they can be gradually migrated to a better form of development
  2. Car-oriented development has been amazingly effective at spreading itself, trying to understand how it propagates and sets the stage for its success can be useful to conceive of how TOD can hopefully replace it
  3. To understand its pattern of replication so as to know when we are doing it, so we can avoid that trap
You'll note that in the title I spoke of "Highway-Oriented Development" rather than simply of "Car-Oriented Development". I'll come to this later.

Basics of car-oriented development

  • Residential areas
Despite the association of sprawl with low-density suburbs full of single-family detached houses with quarter-acre lots, that is not an essential condition of car-oriented development.
These two images are from Swindon, UK - 85% car mode share

Mississauga, Toronto's car-dependent edge city
What COD requires is to have plenty of residential parking everywhere so each working adult can find a parking spot for his car. That means in general having at least two parking spots per house, with on-street parking to deal with the overspill. This requirement does cap density, but it still allows for some. For instance, a 6-meter wide (20-foot) townhouse on a lot 30-meter deep (100-foot) could have a driveway that could have 4 cars parked on it, given a large enough setback. Such a house could get a density of around 40 units per hectare. 40 units per hectare with single-family houses can yield a population density of around 9 000 to 10 000 people per square kilometer (22 000 to 25 000 per square mile). Even single-family detached houses can achieve decent density, the Japanese achieve 30 to 40 units per hectare with SFD housing, mainly by having shallow lots (around 15- or 20-meter deep lots, 50- or 70-foot).
To scale, 9 houses in Japan (3 deep plus a street, 3 across) vs 3 houses in Vancouver, Canada
With multi-family housing, density can be even higher. This often looks like towers in parking lots, but density can even reach 100 units per hectare (40 units per acre), or even more if you use storied parking, either in aboveground structures or underground parking lots.

Of course, if you want to avoid having cars everywhere, making areas ugly as sin, and don't want to use expensive underground garages, the density limit of parking requirements is even tighter.

Likewise, in consideration of the residents, there needs to be ways to force through traffic off of residential streets, often resulting in cul-de-sac and the like.

  • Retail and job centers
Too often, people tend to look only at residential areas. There is almost a taboo around commercial ventures, as if the home was sacred but commerce was profane. Anyway, here again, the COD requires parking spots, plenty of parking spots. Buildings can be high, but the higher they are, the bigger the parking lot around them. The form these take then is highly hostile to non-drivers in most cases: buildings are spread apart, there is little density and parking lots are between the door of the building and the street to have car drivers walk as little a distance as possible to it and to clear up the street for signs to advertise to passing car drivers what business is on the side of the street.

The location of these centers is also very important, and tends to depend on three important factors:
  1. The catchment area
  2. The local traffic
  3. The capacity of the road network
The catchment area is simply the number of residents who inhabit close enough to the store or office or factory to consider patronizing the place or working there, forming a pool of potential customers or employees. This can be approximated by asking the question: "how many people live within a XX-minute drive from here?" This time varies depending on what store we're talking about, people may tolerate a 15-minute drive to go to a major mall and a 30-minute drive to go to work, because they expect a lot of benefits from those trips, but they will have less tolerance if they're just looking a buy a pint of milk from a corner store. Businesses will thus favor areas that are accessible to a greater number of people within as short a time as possible.

The local traffic is sought especially by small stores, because it opens the door to a lot of "pass-by" customers, people who may not make a trip just to go there but who will stop and shop there if it's on their way. Donut/coffee shops thrive on pass-by customers for example.

The road capacity of course matters because if it's insufficient to deal with the expected traffic, then some customers will turn back, or some employees will actively seek other jobs.

All of these factors combine to explain why stores locate on arterials, these areas tend to be accessible to the largest amount of people, to already have plenty of traffic and relatively high capacity.

  • The road network
COD depends on a road network that can support a lot of traffic. For instance, a grid of wide arterials with 2 or 3 lanes per direction, if not more. This can allow plenty of people in cars on the road at the same time. Residential streets can be narrow if you want, as traffic is channelized on arterials, as long as residential streets connect to arterials, it is fine.

Street-based car-oriented development

First, what does a car-oriented city based on regular streets look like?
Cities following a street-based COD actually develop not that much unlike traditional cities, in that they would tend to be relatively round, with a strong center. Why? Because if all streets go at roughly the same speed, say around 30 km/h on average (including stops), then growth will tend to happen in a concentric fashion to reduce the travel time to all other areas of the city, just as they would in traditional cities where most people walked. The main difference is one of scale.

Likewise, the center of a city is by definition the most accessible part of it. It is the part of the city that is closest to every other part of the city on average. Also, supposing an uniformly distributed population, the center also has the highest road capacity of all areas as exactly one quarter of the population is likely to access it by the north, by the east, by the west and by the south:
The center doesn't just reduce the maximum distance, but also distributes people coming into it to 4 directions equally, in this case, if every street in has a capacity of 2 000 vehicles per hour and there are 6 000 vehicles per hour coming into the downtown area, then each street is at 75% of capacity
Meanwhile, a retail and jobs center that would not be at the center would face a much higher traffic coming from one direction than from the other directions, risking congestion and becoming undesirable to go to for people who are stuck on that particular way in.
An off-center location results in a lot more people coming in from one or two directions, in this case, using the data from the previous image, the west access is probably at 150% of capacity, the north access is at 100% of capacity while the east and south access are at 25% capacity, so most people coming to the center would come through congested streets
Major stores would likely be located in the center, with some proximity stores on arterials all over the city.

As a city grows, it may come to limits on growth, the arterials leading to the center may become congested, the city may grow beyond what would be considered acceptable travel distance (in minutes) from the center, which can be dealt in a variety of fashion: creating new centers on the periphery, with new residential developments to follow, creating a polycentric city, investing more in transit to increase road capacity for peak hours, widening streets, etc...

A street-based car-oriented city is not so bad. Sure, residential and commercial forms give priority to cars rather than to humans, but since there is no high-speed road and speed is largely the same on all arterials, there is a focus on reducing distance and on being more compact rather than spread out. Transit can also be effective with bus lines on arterials following the same routes as cars. There is no street off-limits to pedestrians and cyclists, though crossing some streets may be long and unpleasant, it remains a possibility.

Highway-oriented development 

Highway have a big impact on traffic and development patterns. Since they are much faster than regular streets, people often have an advantage to take detours by the highway instead of traveling slower on city streets. The result of highways is thus a concentration of traffic on the highway, drawing them away from arterials. Highway interchanges also become the most accessible places with the biggest catchment areas of the entire region:
10-minute driving distance between an interchange versus an arterial far from highways
Consequences on development are dramatic:
  • Opening the way for more sprawl
One of the big effects of highways is to open up new lands for development. People, even those who want to be far from activity centers, want to be within reasonable distance (in minutes) from the rest of the region they settle in. If roads are limited to average speeds of 25-30 km/h (15-20 mph), then that limits how far cities can sprawl, some kind of effective informal urban growth boundary. By adding high-speed links with two or three times the speed, this opens up the amount of land that can be effectively developed without being too far from the rest of the city. This makes land cheap, and what happens when something is cheap? People feel free to use it a lot, in this case with large lots and big parking lots.

  • Draining traffic off of existing commercial arterials
High-speed roads, especially limited access roadways, will likely drain traffic by offering much faster travel. To some, it may superficially seem like a good thing, it is not. As I said earlier, stores want local traffic to have "pass-by" customers, if much of traffic is drawn away from the arterials, this will take away much customers from small stores that may then close, leaving arterials built into residential areas without much stores. Even if arterials can be seen from highways, highway traffic is fast and cannot easily stop.

  • Attracting commercial businesses and employment centers to highway exits
This is a particularly bad outcome. As I said, once you put in highways, the places with the biggest catchment areas become highway exits and interchanges, and highways draw in traffic off of arterials. Highways also have high road capacity per lane and many authorities will widen them significantly to avoid congestion. Big catchment area, plenty of traffic getting on or off the highway at a few exits, high road capacity... highway exits are prime locations for businesses, especially big box stores who need large catchment areas to have the customer pool needed to sustain themselves.

The result is that you will see malls and big box stores springing up at highway exits while arterial commercial streets will weaken. As much as we can find strip malls ugly, they are still a lesser evil compared to big box stores located at highway exits.

What is the big problem with highway locations? Well, first of all, they tend to be located far from population centers, at the periphery of urban areas, not at the center. Second, the highway itself creates a barrier imposing a huge detour to get to the area for those on the wrong side of the highway, including for pedestrians walking along extremely wide arterials with plenty of fast traffic. Third, unlike the earlier arterials, it is hard to serve an highway with transit. On the arterials, buses could stop frequently, and even with parking lots, people would still be in face of the businesses, but buses can't stop on highways in general, even if they did, they would leave riders in an eminently unpleasant location, far from everything.

So not only are these commercial areas distant from population centers, there are barriers imposing detours, walking or biking there is uncomfortable if not outright dangerous and transit service will usually be terrible. The only ones advantaged by locations at highway exits are car drivers.

Here is an example of highway-located malls in Atlanta to illustrate my point:

Note how commercial areas (that can be easily identified by the prevalence of gray color due to the parking lot pavement and the roof of big box stores and malls) congregate directly at the highway exits, with residential areas (identified by the green with small gray line patterns) further out. In the second case, the mall is even enclosed by two highways running parallel to each other for a while, forming a significant barrier to direct trips.

Now, let's see examples of malls in street-based car-oriented developments. Once again, I turn to Japan, because that's where this kind of development is most seen, even European cities use relatively higher speed arterials (though nowhere near the highways of North America).
An AEON mall in Obihiro, put inside a residential area, do not mistake it for a huge parking lot, it's just the parking on the roof that gives that impression

You can see the AEON sign in the background, with a 8-story residential building in the foreground

Another AEON mall marked by the red "F", in Asahikawa
What it looks like from the street
These two examples of malls in car-oriented Japanese cities. We may not like how they look like, but they are built inside residential areas and can be walked or biked to, unlike the Atlanta malls. The price of land compelled the owners to build parking on higher stories to be able to fit their stores in smaller lots. I'm not saying this is perfect, not at all, but they are way less bad than highway-oriented developments.

Conclusion: how to avoid HOD and what we can learn from it

Highway-oriented development is not just a form of development that favors cars, it is one that essentially punishes any other mode of travel. A street-based car-oriented development at least allows for other modes of travel to exist and even the possibility of conversion if need be. HOD has no redeeming features whatsoever for sustainability and multi-modality. To avoid it, there are two main solutions:
  1. Do not build any highways inside urbanized areas, or even high-speed arterials (what Charles Marohn at Strongtowns calls Stroads). Build highways so that they go around cities and do not penetrate them, with very few exits very spread apart, this will reduce the catchment area of highways and avoid them being used for local trips. Local car traffic should be dealt with by a tight grid of lower-speed arterials with frequent intersections (high capacity, low speed, low cost).
  2. Toll highways high amounts, 10-20 cents per km (15-30 cents per mile). This can be justified by the high cost of building and maintaining them. Tolling highways will significantly reduce demand for it, and commercial developers will want to avoid locations that require people take highways to access stores and job centers, because the tolls will be a constant irritant to potential customers.
As to what we can learn from HOD... there is a strong analogy to make between highways and rapid transit. Both offer much higher speeds and higher capacity than usual modes of transport, both have limited access (RT has great stop spacing, highways have few exits and entries) and both tend to draw users away from other axes of transport (RT eats local transit ridership, highways drain traffic from local arterials). So if you have a form of rapid transit like subways, you should try to replicate a bit the form of development seen at highway exits: large stores and malls at stations for maximum catchment area and pass-by potential customers. With the difference that residential areas must be much closer for RT than for highways.

As bad as it is for sustainability and as wasteful as it is, HOD is incredibly successful at imposing itself, so taking inspiration from it for TOD makes all the sense in the world.

Thursday, October 9, 2014

An example of how neighborhood design matters: Lasalle vs Plateau-Mont-Royal

I spoke earlier about streetcar suburbs, using the Plateau-Mont-Royal borough as a prime example of one.  It was designed in a particular way, namely commercial areas were built around main streets on which the streetcars used to run, with high-density housing around the commercial area. 

Plateau-Mont-Royal basic design
The Plateau is also a high-density neighborhood with multiplex housing that have around 20 000 people per square kilometer and over 100 units per hectare (40 per acre).

Residential street of the Plateau, full of multiplexes
The result is a neighborhood that shows strong multi-modality, with transit and active modes of transport (Walking and biking) having very high mode shares.
Plateau-Mont-Royal mode share
Plateau-Mont-Royal internal mode share
Now, this was an early 20th century development, but what I've realized is that the neighborhood I currently live in, Lasalle, is essentially an adaptation of the Plateau.

Lasalle: an automobilized Plateau-Mont-Royal

Like the Plateau, Lasalle is built with a commercial zone on its main streets, with residential areas flanking the commercial zone. There is also a hgih-frequency bus line that runs down main streets, for the biggest street, the Newman Boulevard, that line is the 106, that has around 250 buses traveling its two directions every day. During the peak hour, the buses have an headway of 3 minutes, for around 1 000 passengers per hour per direction in capacity. There are also additional bus lines on side streets.
Lasalle's basic design mirrors the Plateau's, with a commercial zone (in yellow) following its main streets and residential areas (green) on the side streets
The Plateau's form of housing is Québec's traditional multiplex with outside stairs and each unit having its own door to the outside. Well, Lasalle's main housing form is a form of the multiplex, but adapted to the era of the car, with front-loading or side-loading garages in order to fulfill minimum off-street parking requirements (of 1 per unit).
Lasalle's modern multiplexes with side-loading garages in the basement

View from the front

Rows of duplexes with front-loading garages built in the basement

A view from the sidewalk of the duplexes with plenty of external stairs
Whereas the density of the Plateau's multiplexes was about 20 000 people per square kilometer, the density of the blocs of Lasalle multiplexes varies between 12 000 and 15 000 people per square kilometer. So a reduction in density, but not by that much, this is still quite sufficient to support walkable neighborhoods.

So there are plenty of similarities:
  • A basic design with commercial zones following main streets, surrounded by high-density residential areas
  • Surface transit lines running these same main streets
  • Typical housing being similar to traditional Québécois multiplexes seen in the Plateau, but with garages built under the buildings
So how does Lasalle fare in mode share?

Not that well actually. I mean it's still pretty good in a North American context, but it's nowhere close to the Plateau:
Mode share, Plateau vs Lasalle

Mode share, for internal trips only
So Lasalle has twice the car use rate as the Plateau, and three times as high for internal trips. The number of transit trips is much closer, but walking and biking take a severe drop.

Exploring the reasons why Lasalle fares much poorer than the Plateau is important to underline the point that, in terms of urbanism, urban design matters. It doesn't suffice to have the same blocks to achieve the same result, the interaction between these blocks are vital.

Why Lasalle is no Plateau

 One of the biggest differences between Lasalle and the Plateau is the street network. The Plateau famously had residential streets going straight to the commercial main street without any detour. Lasalle on the other hand has a street network built to favor speed on the main streets by reducing the number of intersections, and so the porosity of the street network. It also created streets to discourage through traffic in residential areas. In fact, on average there is one intersection every 50 meters (170 feet) on commercial streets in the Plateau, but only one every 250 meters (800 feet) in Lasalle.

Most streets provide no direct path to the commercial street, people have to make detours
There is also the issue of the width of the commercial area. In the Plateau, the commercial "zone" is only about 80-meter wide.

In Lasalle, it is three times as wide, at around 250 meters, with little or no housing therein.

The Newman Boulevard is also a 6-lane boulevard with a median, which is hard to cross for pedestrians, unlike the Avenue Mont-Royal in the Plateau, which adds another barrier to pedestrians and cyclists. The result of this is that there are very few people who reside within 300 meters of any business in Lasalle. Even if you have high-density housing, if it's located beyond comfortable walking distance, you're not much better off in the end. 

This is crucial because the decision to walk is very distance-sensitive. People within 300 meters (1000 feet) of their destination are unlikely to do anything but walk to it. But beyond that, walking mode share falls very quickly. Having residential areas built right next to commercial destinations and having residential areas built at 300 meters from commercial destinations, supposing a density of 20 000 people per square kilometer, is probably the difference between having 10 000 people who would opt to walk and having just 5 000 people.

Lasalle is also not comfortable to walk, as you can see from the pictures above of the typical Lasalle housing, streets are generally very wide, with pedestrians stuck on narrow sidewalks between cars parked on the street and driveways on which cars are sometimes parked, with few trees around. As to the commercial street, it is nothing short of a disaster in Lasalle, with huge parking lots in front of 1-story buildings where each unit tend to be wide but shallow, the opposite of the narrow but deep commercial units of the Plateau.
A view from Newman Boulevard in Lasalle
Avenue Mont-Royal in the Plateau
St-Denis street, also in the Plateau
So not only is the commercial street harder to access, but once there, walking alongside it is a chore, not a pleasure. The presence of plenty of free parking also incites people to take their car, whereas parking in the Plateau is rarely free.

Some commercial malls are also built with barriers separating them from residential areas, like I talked about in my most recent post.


What I hoped to illustrate in this blog post is that good urbanism requires good urban design and to provide an example of the consequences of the barriers and detours between uses and of the dendritic street network that favors vehicular speed over providing direct access and short distances. Despite the Plateau and Lasalle having many basic similarities in terms of design and housing, due to the differences in street network and in the fine details of the urban environment, the walkability of Lasalle is much inferior and results in the dominance of the car as a mean of transport.

So it doesn't suffice just to build dense residential areas or to put them adjacent to commercial areas and providing transit lines along commercial avenues. Design matters.

Bad habits of North American cities: magicking distance from proximity

One fundamental principle of 20th century planning has been use separation, the idea that buildings used for certain purposes must be separated from buildings that have different purposes. Some of these usual uses are residential, commercial, industrial, institutional. Often, cities will have absurdly big single-use zones that are square kilometers in size.

Nowadays in Montréal, we have a metropolitan plan to increase density (simplistically defined as residential density) which results in residential zones taking much less place than before and therefore being much closer to commercial zones than before. Said like that, it seems like a very good thing, after all, density requires proximity to really achieve its potential. Yet, the mentality of use separation and car-centric planning still leaves its mark on newer neighborhoods.

How? Well, use separation in North America has become an art over the years. It's not just about keeping X away from Y in distance as a bird flies, it is also about deploying the magical art of making that which is near, far. An art that employs the street network to make direct paths as rare as possible.

Here is one example of this, in Laval, a suburb to the north of Montréal, near the Sainte-Rose commuter rail station.

You'll note that I identified the grocery, which I personally consider the most important store in a walkable neighborhood. In fact, if I were to make a checklist of criteria for valid TODs, I'd put in at number one:

"There is a supermarket or grocery at the station or within easy walking distance of it, with high-density residential areas within 5-minute walking distance, otherwise AUTOMATIC FAILURE"
Anyway, this area is pretty typical, with a commercial arterial street and residential side streets.

Just to the west of the commercial area, there is a new residential development with quite respectable density, which includes:
Relatively small single-family detached housing

Older semi-detached housing

New town houses built in groups of 3 or 4
Newer semi-detached houses
And some condos to top it all off
So this is some respectable density, and some of it within 10-20 meters of the grocery, great, right?

Except look at that street grid again, there is no direct link between this new neighborhood and the grocery. There is an unbroken fence all along the border between the commercial zone and the residential zone. So all the people who live in the high-density neighborhood have to take a detour by the north, through an inhospitable parking lot or on the sidewalk of a large street with poor enclosure and no buffer from cars to get to the grocery.

You have some housing that are built right next to the grocery, that they can probably throw a baseball on it, but there is a minimum of a 300-meter detour (1 000 feet), or 3-4 minutes of walking distance, in order to get to the front door.

This is a combination of two things.

The first is the psychological obsession with separating uses, even when they are right next to each other, we often build a fence or a barrier between the two to impose detours and make what is near, far. To make sure that they do not mix, no matter how near they are.

The second is a classic street network built all around the car, which may have sidewalks but which is still built with the car in mind. We can call it a "stream to river" model, with the small, sinuous residential streets going nowhere except onto collectors, which are a bit straighter with less stops, which flow onto arterials which have much higher speeds, a limited amount of intersections, most of which with traffic lights to favor speed and fluidity. This street network channelizes traffic onto high-speed arterials and keeps it off the residential streets. So the residential streets are like small, slow streams that flow into progressively faster and larger rivers. And as the arterials are where most of the traffic is and locations around it are the most quickly accessed locations, that's where businesses will want to settle.

The goal of this model isn't to reduce distances. That is inconsequential to the street network. What it seeks to optimize is vehicular speed, residential streets flow quickly onto collectors which flow onto fast arterials.

And indeed, here is an example of this at work, this is the quickest path as calculated by Google Maps:
1,3 km and 3 minutes
And this is the most direct one:
950 meters and 4 minutes
In many cases, the direct path doesn't even exist, like in the case of the newer neighborhood to the west.

So this street network actually works relatively well: people can still get where they're going fast thanks to the high-speed arterials (even when they have the same 50 km/h or 30 mph limit as residential streets), and traffic is turned away from residential streets. So you have speed and traffic-calmed residential streets, what's the problem?

The problem is that pedestrians and cyclists don't go faster on the arterials than on the residential streets. So these detours which don't matter to cars because they can get higher speeds have no benefit to pedestrians or cyclists, they walk or bike at the exact same speed no matter where they are. Worse, the high-speed arterials are often very wide, leading to very uncomfortable walking environments, or biking environments, unless bike paths are provided. Arterials also often lack crosswalks, allowing crossing the street only at rare intersections, where pedestrians need to wait one or two minutes for their signal after pushing the button. This forms a barrier to pedestrian trips in and of itself. Crossing even a moderately sized arterial is often the equivalent of walking another 100 meters or so (300 feet).

So this "stream-to-river", street hierarchy street grid is terrible for pedestrians and cyclists. It seems like a good idea only to people who assume that every trip is made in a car and that other people don't exist, or don't matter if they do exist. And talk about a self-fulfilling prophecy, on such a street network, most trips WILL be made in cars because that's the only mode of travel that is really advantaged by the street grid. Buses that follow the arterial can also be faster, but buses are not good for short trips because of the wait time, the fact is that you're usually better off walking to your destination if it's under 1 kilometer rather than taking the bus.

The result

So what's the result of this? We do have proximity between high-density housing and neighborhood stores, but that proximity is "car proximity", not "pedestrian proximity". It would have been the easiest of things to open up a pedestrian path to the front of the grocery, allowing all residents of the neighborhood to travel pleasant residential areas up to the grocery rather than taking a detour through a long parking lot.

A proper grid is generally best for pedestrians, if you already have a suburban-style street network, opening pedestrian paths to replicate a grid (or quasi-grid) for pedestrians would be a great idea.

Just for comparison's sake, here is a supermarket in a residential area of Obihiro, in the northern island of Hokkaido in Japan. Why Obihiro? Because it's one of the most car-oriented cities in Japan, but even there, walking, even if not that great, is certainly a viable alternative.
The supermarket is marked by the red bubble. Note first how compact its parking lot is and how they integrated the supermarket in the residential street grid rather than creating barriers between uses.
The contrast is striking, whereas in Laval the closest house to the grocery was about 300 meters away, in Obihiro, there are maybe 300 or 400 houses within 300 meters of distance of the grocery's front door, people who are very likely to walk there. They even added a pedestrian path in the rather longish blocks to the south of the grocery, even if it's hard to see on this image. And these are mainly single-family houses of no higher density than the new developments in Laval, in fact the condos are roughly 2-3 times as dense.

Density matters, proximity matters... and street network design matters. Unfortunately, many planners in North America have yet to shake off this mentality of separating uses and of putting barriers between them if they can't simply use distance as a buffer.

Wednesday, October 1, 2014

To setback or not to setback, that is the question

As a kid growing up in a suburb, in a bungalow on a roughly 650-square-meter (around 7 000 square feet) lot, I learned young that the house I wanted in the future was one without a lawn. I thoroughly hated mowing the lawn and saw it as completely useless. Especially the front yard, that I saw as "too public to be an effective private space, and too private to be an effective public space". Yet bylaws made sure every house had to have a large front yard because of a mandated minimum setback, a distance between the building and the curb, and a duty to respect a building line to maintain harmony so that houses' facades are built on one single line.

From the point of view of creating a good cityscape, massive setbacks are counter-productive. Not only do they often reduce density by imposing deeper lots, but they go against the feeling of enclosure that makes walking and even just being in a place comfortable.
Sapporo, a low-rise residential area with great enclosure

Vaudreuil-Dorion, weak enclosure due to large setbacks and absence of mature trees
So, I'd like to ask a question:

To setback or not to setback?

First, what might the point of a setback be? When looking at the front yards of most North American homes, the answer seems to be: nothing. They are extremely rarely used for anything. Back yards are more useful and often house pools, parties with friends and the like, but the front yard is most often an empty space, just a bit of lawn, maybe one or two trees, and a driveway.

However, setbacks do serve some purpose even if they're not actively used.

For one, they provide a space between the street (public space) and the building to protect people's privacy. If you have no setback and you reside on the ground floor, you will be compelled to close the shutters in order to protect your privacy. This is seen especially in France where it's not rare for single-family homes to have essentially no setback despite having deep lots.

The result is that houses are almost built like besieged castles, with windows, at least on the ground floor, shuttered to protect the occupants' privacy. The walls are often bare of details. This is especially true in places where pedestrians are limited to sidewalks, thus being forced to hug the walls of buildings. With shared, narrow streets, this issue is less important as pedestrians can walk farther from buildings and are not forced to walk within arm's reach of the house's front wall.

The second purpose of the front yard is as a public facade. It allows people to participate in shaping the environment, for instance by decorating for Halloween or Christmas, or creating pleasant facades of their houses, like creating a display for themselves, some personalization of their home. I think it may be a very good thing to have in order to create a feeling of community and openness.

So I see the point for a setback for single-family houses, since there is a single owner of both the house and lot, the front yard is theirs to arrange how they see fit. For multi-family houses, setbacks aren't as important, most people tend to live over the ground floor and so do not require a setback to protect their privacy and the status of the front yard as a shared property or as the property of a non-occupant owner means that if there is one of some size, it is likely to be completely unused (a reversed tragedy of the commons if you will).

However that leaves the question about how large should a setback be?

Front yard or buffer?

It seems to me that most of recent setbacks are at least twice as deep as they need to be to fulfill their actual role. Looking at old single-family houses in North America built before the car, we can see that setbacks were much, much smaller, even when lots were very deep.
Burlington, Vermont
Charlesbourg, near Québec City
Ste-Agathe-des-Monts, a village in the Laurentides
Winston-Salem in North Carolina
In all these previous examples, we can see similarities. There is a setback in most cases, but it's about 3 meters on average (10 feet). Many of these houses also have porches as a physical buffer between the public realm (the sidewalk) and the private realm (the house).

 Now let's look at more recent developments:

The setback has basically tripled, from 3 meters to 10 meters on average (10 feet to 34 feet). What's interesting is that, though the space has increased, people will very rarely use that additional space, when they do some work to make their facade look good, they will concentrate on the area right next to their door and leave the front half of their front yard completely bare, or with only one tree.
The owner has planted flowers and shrubbery near the stairs to the door, but left alone all the rest of the front yard, apart from a single tree planted in the middle

What happened?

I think the car happened. This increase in the setback is no longer there to protect our privacy or to provide some space to build a facade, it exists to create a buffer between the building and the traffic on the street. Sometimes, they are used to park cars, but many houses have driveways built not in front of the house but to the side.

But still, cars happened in Japan and Europe too, yet they tolerate much smaller setbacks than we do, why? And why does it seem not to bother them?

The Netherlands

Sendai, Japan

United Kingdom
What do we do that is different from these countries and that leads us to build these huge buffers?
I think the main reason is that we have frequently built residential streets to the standards of highways, with lanes 3,5-meter wide (12-feet) plus shoulders, which is absolutely ludicrous in areas where there is plenty of off-street parking and therefore cars will only rarely be parked on the street. Shoulders as a whole do not belong in urban areas.

So we have too often designed residential streets as through streets rather than as access streets. They are built to allow easy car movement through an area by providing enough space to eliminate friction with other vehicles and street users. Without friction, cars move much faster, and fast-moving cars are more dangerous and noisier than slow-moving cars. It is natural that we would prefer to be further away from them. On the residential streets of Europe and Japan, cars generally go at speeds of 30 km/h and lower (20 mph and lower) because of the narrow streets and the presence of objects close to the roadway. 

Propose such streets in North America, and many people panic... yet, we have plenty of access streets that are also shared spaces in our cities, with narrow lanes and solid objects surrounding them. Want to know what I am talking about? This:

If this were a public street, it would be illegal, but it's a private access alley to a parking lot, so it's okay! Am I the only one who thinks that with a couple of trees, this would look great?

This entry into a parking lot is less than 6-meter wide, with curbs and fixed objects right next to the travel lanes

The average parking lot alley is about 6-meter (20-foot) wide, with cars right at the extremities, leaving a narrow corridor in which two cars passing each other will have to slow down. This corridor is also shared with pedestrians, making parking lot alleys a shared space. Yet in most cities, proposing 6-meter wide streets with trees right on the other side of the curb, that cars have to share with pedestrians, would likely give a few planners and engineers heart attacks. But these same people would likely have come to the meeting in cars, driven and walked in such "streets" without noticing when they parked their cars and walked to the building.

Granted, there are quite a few accidents in parking lots, but almost all of them involve cars parking or getting out of parking. Otherwise, the area is safe enough that people do not feel afraid of walking through parking lots, else no one would be driving anywhere!


Human-scaled setbacks are probably a good design for single-family homes in order to provide more privacy and a customizable space to the owner to allow them to participate in making the public realm more appealing, including participation in community celebrations (Halloween, Christmas, National days, etc...). But in most North American cities, most setbacks are much too deep as their true purpose is to create a buffer between traffic on the street and where people live. Much could be gained by narrowing the streets to make them access streets rather than through roads and reducing by half or more the current setback. Though hard to apply to built areas (well, you can narrow streets, but reducing setbacks would involve rebuilding every house), this needs to be taken to heart for new developments.

If anyone says such narrow streets are unacceptably dangerous, point out to parking lot alleys which are designed in almost the exact same way, but which no one seems scared of.