The Fundamental Challenge in Measuring Sprawl: Which Land Should Be Considered?

Lack of agreement on how to define and measure sprawl has hampered development of policy related to its causes and consequences. We question previous work for two reasons: the use of study areas that overbound or underbound sprawl landscapes, and the failure to account for land unavailable for development. We formulate "extended urban areas," based on housing density and commuting patterns and argue that they represent a preferable geographic basis for measuring sprawl. We operationalize with satellite imagery a way for measuring land unavailable for development in these areas. We then compute five measures of urban development using the National Land Cover Data Base and decennial census data to assess the extent of sprawl in the extended urban areas of Atlanta, Baltimore, Boston, Los Angeles, and Washington. Our sensitivity analyses reveal that the measurement of sprawl critically depends on which land area forms the basis of the analysis, and, to a lesser degree, how one accounts for land unavailable for development.