Most agree that urban sprawl is at least partly related to low densities of development. The previous post discussed population-weighted density as compared with normal density, referred to in that post as simple density, which is just the total population or number of housing units in an area divided by the total area. So this raises the question of which measure of density is most appropriate to use when discussing urban sprawl.
Discussions of sprawl frequently focus on the negative effects or costs of sprawl. And the explanation for the cause of the negative effects is very often the low density of development. So this provides a point of departure for considering whether population-weighted density or simple density is more appropriate. It should be whichever is most directly related to a particular negative effect and could be seen as a measure of the magnitude of the effect of density. It turns out that the measure of density that should be chosen depends on the negative consequence of sprawl being considered. For some, it’s clearly population-weighted density, for other costs, simple density is more relevant, and for yet other costs of sprawl, both could be related.
Before discussing specific negative effects of urban sprawl, a quick review of the key difference between the two density measures. Population-weighted density emphasizes the role of density in relation to the population. It is a measure of the average density experienced by persons in their census tracts or other subareas. Simple density, on the other hand, emphasizes the significance of density in relation to land area. This becomes clear because simple density is equivalent to the average of the subarea densities weighted by their areas.
Then for negative effects of sprawl where that effect results directly from the effect of the low density on the individual, population-weighted density makes the most sense. This is because it is the average density experienced by persons and hence a measure related to the significance of low density for the negative effect. The lack of a walkable urban environment due to low densities would be one example. The density experienced by the individual in his or her neighborhood is key to walkability, at least in the area near one’s residence. So lower population-weighted densities would tend to be associated with more people experiencing less-walkable environments.
However, many of the costs that have been associated with sprawl depend directly on the amount of land impacted by urban development. For these impacts, simple density, which emphasizes the importance of land area, is most appropriate. We can start with the effect of sprawl in causing the loss of agricultural land and environmentally sensitive areas. The effect here depends upon the amount of land devoted to urban development. It does not depend on how people are distributed on this land and how many experience higher- or lower-density neighborhoods. And for any given population, the lower the simple or overall density, the greater the amount of land devoted to urban development and potentially the greater the amount of agricultural and environmentally significant land lost.
Discussions of the costs of sprawl describe the increased costs associated with providing various public services in lower-density areas. For linear infrastructure such as streets and water and sewer lines, the larger the urban area to be served, the more miles of such infrastructure are required. Again, this depends solely on the extent of the urban area. And this will also be true for some other services such as emergency response services where proximity and response times are of fundamental importance.
Now for a major negative consequence associated with sprawl, increased automobile use and the negative environmental consequences associated with such use. First of all, effects will depend not only on population densities but also on the distribution of other destinations within the urban area. So population density, however measured, is only one part of the puzzle. And it becomes easy to see that both types of density measures can be relevant to travel behavior.
Consider public transit. Higher population-weighted densities suggest that more people are living in areas that can be within reasonable walking distance of transit stops. On the other hand, lower simple densities will require more transit lines and stops if the goal is to provide at least some level of transit access throughout the urban area.
Higher population-weighted densities can be associated with more walkable environments in which some automobile trips can be replaced by walk trips. On the other hand, lower simple densities means the urban area covers a greater area and some people (and likely other destinations) will be farther from some others than in an area with higher simple densities. So it’s very complicated and ambiguous for transportation.
Bottom line: The choice of a density measure should depend very specifically on the particular question you wish the density measure to inform.