A longitudinal study of changes in urban sprawl between 2000 and 2010 in the United States

The debate over metropolitan sprawl and its costs has been ongoing since the early 1970s in the U.S. To inform the debate, this study uses principal component analysis (PCA) and 2010 cross sectional data for large U.S. urbanized areas (UZAs) to operationalize compactness/sprawl in each of four dimensions-development density, land use mix, activity centering, and street accessibility. Higher values represent greater compactness, lower values greater sprawl. The four factors are then combined into an overall compactness/sprawl index. The study then applies factor score coefficient values for 2010 to the same variables for 2000 to create comparable metrics for 2000. Compactness scores for 2000 are compared to the same scores for 2010 to see which UZAs sprawled the most between censuses, and which sprawled the least or actually became more compact.Finally, the study validates the compactness index and its component factors against transportation outcomes for 2010, specifically walk mode share, transit mode share, and average drive time on the journey to work. If sprawl has any widely accepted outcome, it is automobile dependence and heavy automobile use. Consistent with this characterization of sprawl, this study finds that the overall compactness index bears a strong relationship to transportation outcomes. Generalizing across the entire universe of large urbanized areas in the U.S., compactness decreased and sprawl increased between the two census years; but only slightly. Several urbanized areas, however, have significantly different rankings in 2000 than 2010.