Every weekend we dig into the archives. This post was originally published in August 2011.

The most commonly cited characteristic of urban sprawl is its low-density. In fact, density is often used as the sole way of determining whether a city is sprawled or not – and (following on from that) whether a city’s urban form is conducive to public transport or not. However, you only need to look at a comparison of the density and auto-use of many different cities around the world to see that things might be somewhat more complicated than that: Overall city density is actually, I think, a fairly poor indicator – except at the extremes – of whether a city is dominated by urban sprawl and whether it has a form that works well for public transport or not. The digram below illustrates this point fairly well, assuming that each dot on a map represents a certain number of people, you have two places with the same overall density which are actually vastly different environments: But even then, I think that what I consider to be “urban sprawl” has much more to it than simply what density an area is – even if what we’re looking at is a small part of a city.

A good post on the New Jersey Future blog highlights this issue further:

Low density is certainly one of the dysfunctions of New Jersey’s (and the nation’s) dominant development pattern since 1950, but it is not the only one. Separation of uses – keeping homes, stores, and workplaces each in their own segregated zones, distant from each other – and a lack of connectivity in the local street network (with lots of looping streets and cul-de-sacs and a lack of direct through-routes) also contrive to make it hard to get around without a car. These other two factors can force people into their cars for most daily activities even in neighborhoods with high housing density.

This, of course, means that we can have ‘non-sprawled’ urban areas, even where the density is not particularly high:

Conversely, a mix of land-use types (residential, employment, shopping, etc.) puts a variety of activities – not just a variety of buildings all housing the same activity – in close proximity, shrinking the distances among multiple types of destinations. And a well-connected, grid-like street network ensures that physical proximity actually translates into easy accessibility by offering multiple, direct routes among destinations. That is, it means short as-the-crow-flies distances are also short walking, biking, or driving distances that may not require a trip out onto the regional highway network. And of course, putting dense, mixed-use, well-connected neighborhoods near transit creates yet another option for getting to desired destinations that are farther away.

Los Angeles is given as a good example of a city characterised by large amount of high-density sprawl, with the following paragraphs coming from here:

But if we measure sprawl by population density, LA would not sprawl at all. In fact, it would be the least sprawling urbanized area in the country. How can Los Angeles be so dense and yet also exhibit so many characteristics associated with sprawl, including high levels of car travel (both in per capita and absolute terms) and low rates of walking, bicycling and transit ridership?

… density by itself—the simple ratio of population to square mile—is not a very useful way to measure sprawl. What matters is the distribution of density, or how evenly or unevenly an area’s population is spread out across its geographic area. If we look at the density distribution in Los Angeles, we notice that its suburbs are much denser than those of other large U.S. cities, such as New York, San Francisco or Chicago. These high-density suburbs compensate for the comparatively low density of LA’s urban core, and, in so doing, increase the average density of the area as a whole …

The LA region’s combination of high, evenly distributed density puts it in an unfortunate position: it suffers from many of the problems that accompany high population density, including extreme traffic congestion and poor air quality; but lacks many of the benefits that typically accompany more traditional versions of dense urban areas, including fast and effective public transit and a core with vibrant street life. Los Angeles has, to borrow a term coined by urbanist William Fulton, “dense sprawl.”

As Paul Mees reminded us last week, Auckland’s population density is relatively high – at least compared to other Australasian cities: Auckland’s higher than expected population densities might somewhat be the result of development in parts of the city like Ponsonby and Royal Oak that I wouldn’t consider to be “high density sprawl.” However, some of the more recent development in areas like Botany seem to fit the definition of high density sprawl almost perfectly:

 At least with these places there are some shops across the road. Pity the road’s a 6-8 lane megahighway, about as uninviting  environment for pedestrians as you’ll ever see. Just down the road in Flat Bush, we have arguably an even better example of high density sprawl, as here there are very few shops anywhere near these new apartment buildings. I tend to think the best definition of urban sprawl is the type of urban development that promote car dependency. And while Auckland may be reasonably high density compared to other cities around Australasia, we certainly have an urban form (as well as a transportation system, obviously) which significantly contributes to our auto-dependency. Solving that issue will involve a lot more than simply raising urban densities.

Share this


  1. Good post.

    There is also population density done well vrs not so well. Interesting I noticed last week that the Netherlands is next to Haiti (in the Wikipedia table if you rank it in Density order) being a tad more dense than Haiti overall. 418 vrs 416 pop/km2. NZ is way down on 18 per km2.

  2. If we’re going to pack detached housing in so close together; why not just build terraced housing or apartment blocks?
    Space might just be saved for more green space…

    1. The aerial images actually show terraced houses or apartments.

      But generally, yes, you can often find many detached houses with just 3 token strips of outdoor space between the fence and the back and sides of the house. What’s the point. Even a terraced house on a 200m² section has more usable outdoor space than many of those. (depending on how silly the front setback rules are)

      I think there is a reason why you don’t see it more, and that is aversion for body corporates. I don’t think I have ever seen a terraced house in Auckland which is not in a body corporate.

      Overseas it doesn’t work like that. Terraced houses are much more like their detached counterparts. One house on one section. They can be built, torn down and maintained independently of each other. Your house will front on a public street, not a shared driveway.

  3. The 100plus terraced 70’s townhouses east of Franklin Road in Freemans Bay do not have body corporates, although the some, that have a shared enclosed “common” outdoor area have a committee to manage and maintain only this shared common area.
    Works well.

Leave a Reply

Your email address will not be published. Required fields are marked *