Winter 2000 — Volume 2 Issue 4
|
|
Wrestling Sprawl to the Ground: Defining and Measuring an Elusive Concept
By George Galster, Royce Hanson, Hal Wolman, Stephen Coleman, and Jason Freihage
Introduction
Urban "sprawl" is one name for many conditions. It has been attached
to patterns of residential and nonresidential land use, the
process of extending the reach of urbanized areas (UAs),
the causes of particular practices of land use, and the consequences
of those practices. Sprawl has been denounced on aesthetic,
efficiency, equity, and environmental grounds and defended on grounds
of choice, equality, and economy. Sprawl has become the metaphor
of choice for the shortcomings of the suburbs and the frustrations
of central cities. It explains everything and nothing.
Much of the confusion about sprawl stems from the conflation of ideology, experience, and effects. A term so widely used cannot be easily dismissed as too vague for serious discussion. Many policy makers claim to know it when they see it and make important policy judgments based on what they see, or think they see. As a first step toward developing policies to deal with its causes or consequences, it would help both critics and apologists if agreement could be reached on what sprawl is, and how to measure it empirically and compare its occurrence across a large number of urban areas.
Definitions of Sprawl
An analysis of definitions of sprawl in the social science and planning literature suggests that definitions of sprawl can be grouped into six general categories:
1. Sprawl is defined by an example, by reference to the average density of a particular urban area, such as Los Angeles.
2. Sprawl is used as an aesthetic judgment about a general urban development pattern.
3. Sprawl is a cause of an externality, such as high automobile dependence, isolation of the poor in the inner city, the job-housing spatial mismatch, or loss of environmental qualities.
4. Sprawl is the consequence or effect of some independent variable, such as fragmented local government, "poor" planning, or exclusionary zoning.
5. Sprawl is defined as one or more existing patterns of development. Those most frequently mentioned are low density, leapfrogging, distance to central facilities, dispersion of employment and residential development, and continuous strip development.
6. Sprawl is defined as a process of that occurs over some period of time as an urban area expands.
As a noun, sprawl implies a condition characterizing an urban area, or part of it, at a particular time. If sprawl is to be measured as a condition of land use, it must be from other conditions that may well be related to it. Based on the descriptions of conditions characterizing it in the literature and amplified by observation and experience, the following conceptual definition is suggested:
Sprawl (n.) is a pattern of land use in a UA that exhibits low levels of some combination of eight distinct dimensions: density, continuity, concentration, compactness, centrality, nuclearity, diversity, and proximity.
Eight Dimensions of Sprawl
The eight dimensions of sprawl are:
1. Density: the average number of residential units per square mile of developable land in a UA.
2. Continuity: the degree to which developable land has been developed at urban densities in an unbroken fashion.
3. Concentration: the degree to which development is located in relatively few square miles of the total UA.
4. Compactness: the degree to which development has been "clustered" to minimize the amount of land in each square mile of developable land occupied by residential or nonresidential uses.
5. Centrality: the degree to which residential and/or nonresidential development is located close to the central business district of an urban area.
6. Nuclearity: the extent to which a UA is characterized by a mononuclear (as contrasted with a polynuclear) pattern of development.
7. Diversity: the degree to which two different land uses exist within the same micro-area, and the extent to which this pattern is typical of the entire UA.
8. Proximity: the degree to which different land uses are close to each other across a UA.
Testing
Our Definition: Measuring Housing Sprawl in 13 Urbanized Areas
We tested our operationalized definitions of dimensions of sprawl
by applying them to a small number of U.S. UAs. We selected 13 large
areas from different regions for our prototype test. Because of
constraints of both resources and time, our test is confined only
to residential uses; we thus examine only housing sprawl, using
housing units as the land use for our analysis. For that reason,
we could not test our operationalizations for the inter-use measures,
continuity and diversity. In addition, given our constraints we
were unable to separate out developable from nondevelopable land;
as a consequence, all land is considered developable. We applied
the sprawl measures to these UAs: Atlanta, Boston, Chicago, Dallas,
Denver, Detroit, Houston, Los Angeles, Miami, New York, Philadelphia,
San Francisco, and Washington, DC. We found New York City to be
the least sprawling and Atlanta the most sprawled. (See table.)
Next Steps
The prototype test of housing sprawl is encouraging. When extended to include measures of nonresidential sprawl, we believe a conceptually clear and coherent set of sprawl measures can be made available. The next step is to gather data for nonresidential as well as residential land uses on developable land, thus permitting use of all eight dimensions, and the data gathering and analysis to the 100 largest UAs in the United States. We will then not only rank these areas by the degree of each dimension of sprawl; we will apply analysis to determine whether some dimensions are closely associated and whether it is possible to distinguish among several types of sprawl. It should then be feasible to undertake an empirical analysis of the causes and effects of each type or dimension of sprawl.
----------
George Galster is Hilberry Professor of Urban Affairs at Wayne State University. Royce Hanson is Visiting Professor, Policy Sciences Graduate Program, at the University of Maryland, Baltimore County. Hal Wolman is Director, George Washington Institute of Public Policy, and Professor, Department of Political Science, at The George Washington University. Stephen Coleman and Jason Freihage are Graduate Research Assistants, Policy Sciences Graduate Program, at the University of Maryland, Baltimore County.
This article summarizes research conducted by the authors under a grant from Fannie Mae Foundation. More detailed findings are expected to be released in 2001.