Category Archives: Blog

Understanding Sprawl

What is urban sprawl? We’ve all heard the term and seen the images of expansive subdivisions and big-box stores that it conjures up but do we truly understand the nature of the beast? Sprawl tends to be typecast as a low-density version of suburbia when in reality it can take on many different forms. Evaluating sprawl is more nuanced than simply discussing differences in population and building densities; questions of land-value, connectivity, metro shape, development pattern, and relative area must be included in any comprehensive analysis of urban sprawl. The following are some important factors to keep in mind when judging a city’s “sprawl factor.”

1.) Contiguous Development– In discussions of urban expansion and development, the importance of proximity tends to be overshadowed by debates over density, which often frame the goals of anti-sprawl proponents and low-density suburban dwellers as contradictory. The importance of proximity is lost in these discussions, which is a shame since many of the negative effects of sprawl can be remediated by addressing issues of proximity without necessarily forcing people into higher-density environments. In many cases, a metro area’s leapfrog development can be quite dense, creating a visually-misleading impression of the area’s perceived compactness. Examples of this form of urban development abound in cities the world over. Consider Montreal, overall quite compact by any measure, since the central city is constrained to an island. There is quite a bit of adjacent undeveloped land in between the river banks and the suburban corridors just beyond the periphery. Imagine if we could move the off-island suburbs of St. Jerome, Blainville, and St. Therese on the undeveloped land of Laval, directly adjacent to Montreal, without changing anything about their densities or general construction. That would mean an additional 147,631 people living a mere 12 miles from downtown Montreal without asking anyone to give up their suburban lifestyle and we’d still have an extra 4.02 mi2 to spare! The upshot of this would be that suburbanites have shorter commute times while still living in their desired setting, while the increased close-in population would bolster the services and businesses available to Montrealers and Lavallois and probably support a wider variety of amenities. Furthermore, with this setup, less of the landscape would be fragmented by suburban sprawl, allowing more of the region’s natural habitat to remain unified and uninterrupted.

Above: A crude estimation of the undeveloped land of northern Laval, let’s say 30.29 mi2.

Above: The bulk of suburb, St. Jerome’s, area (12.57 mi2) and population of 68,456.

Above: The majority of Blainville at some 9.98 mi2 and 53,150 inhabitants.

Above: The entirety of St. Therese in all its 3.72 mi2 and 26,025 population.

The general shape of the sprawl pattern matters a lot in terms of keeping any given point of the within certain range of every other given point in the urban zone. Obviously, the shape a city takes is determined largely by geographic constraints. Looking at the Bay Area, for example, which consists of two metropolitan areas (San Francisco-Oakland and San Jose-Sunnyvale), the mountains constrain development while the presence of the long, narrow bay encourages development in a North-South pattern, meaning that any two random points are farther from each other than they would be if the area was laid out in a concentric circle. Suppose that part of the bay between the San Francisco peninsula and East Bay was actually land and the cities of East Bay developed there. On average, the distance between the cities on the east end of the former bay and those on the western slopes would be reduced by 3-12 miles and drive times between destinations located on either side would be significantly reduced since traffic would no longer be forced to route itself around the bay or over strained bridges.

  2.) Relative Size– This is probably the most obvious measure of sprawl since cities with larger populations can take up more room than smaller populations before being classified as truly sprawling. But don’t be tempted to reflexively analyze this spatial spread using standard density measures which divide total population by total area. Such numbers are prone to give tempting but misleading impressions of an area’s true density. One of the main reasons using raw density as a measure of sprawl falls short is the inclusion of large swathes of undeveloped land in the land area count since metropolitan areas are defined by political boundaries (county lines) not the actual extent of development. Additionally, different urbanized areas will support different ratios of non-residential land use (low population density) to residential land (high population density) which makes comparisons of the distribution of population between different metro areas difficult. There are some statistics that do a better job of capturing the density that an average city-dweller experiences. Housing unit density is an important metric for determining how densely-built (not the same as densely-populated) an area is and, thus, how compact or spread out the development pattern in a given urbanized area is. Weighted-average density measures, are helpful for understanding how the average citizen experiences the city and the typical density level that they perceive and experience, since its weighted nature means that drastic discrepancies in population densities within a metropolitan region don’t distort the average too much. Obviously, population density will vary greatly within a metropolitan area, more drastically so in some than in others, which is why the weighted-average measure is so useful.

3.) Connectivity– A city is greater than the sum of its constituent parts. A cohesive city or metropolitan region has to be more than an amalgamation of developments or city blocks; there has to be a certain level of interconnectedness that extends across the entire urbanized area.This concept can be concretely evaluated in the street grid, the availability of mass transit options, and city-block and sizes. Intersection densities are a good, though not perfect, measure of interconnectedness at the street level. Streets are the tissues of cities and, as such, allow for the expansion of an urban area. A grade-separated, 10-lane freeway is going to spur a very different development than a width-restricted, pedestrian alley. A more difficult but no less important factor to measure is the effect of streets in unifying or bisecting the neighborhoods on either side of them. The size of city blocks or developments is directly correlated with the idea of land-value and accessibility. The greater the ratio of street perimeter to enclosed area, the more that can be serviced by transport and the more exposure it has to public space (streets are public spaces) as well as to other spaces in the urban street network. As for mass transit, the more a system extends far out into the ‘burbs and the relative level of enclosedness (i.e. the level to which the transit loops or intersects itself), the more interconnected any given space in the metro area is, and thus, the more restrained sprawl is. Of course, in some countries, transit assume the role of catalyst for suburban expansion that freeways serve in the U.S , so this can be a double-edged sword.  

4.) Value & Use-The relative spread of highest and best land uses is the ultimate measure of sprawl, since that’s essentially what sprawl is about: the inefficiency of land-use choices and the associated costs arising from those decisions. If you think about, land that is very productive in an urban setting is usually intensely used as opposed to land that is more lightly-utilized (an acre of land supporting a 70-story skyscraper versus a single acre supporting 3 McMansions.) Of course, a more dense build-out doesn’t always mean a higher or better use of the land, but it’s generally a pretty indicator of that. In practical terms, I think a good way to evaluate this is the presence of surface parking lots, probably the least value-adding and most resource-consuming urban land use out there. Floor-to-area ratios also provide good insight into the relative intensity of a parcel of land’s use and can be a proxy for the level of maximization of a particular plot of land.