Quantifying the Spatiotemporal Trends of Urban Sprawl Among Large U.S. Metropolitan Areas Via Spatial Metrics

Abstract

Spatial metrics have emerged as a widely utilized tool to quantify urban morphologies and monitor urban sprawl. Since previous applications of spatial metrics have typically considered only a single urban class, this study evaluates how deriving spatial metrics from multiple land use/land cover (LULC) classification schemes can help elucidate the spatiotemporal trends of urban sprawl. Specifically, the urban morphologies of the fifty most populous metropolitan areas in the U.S. were quantified in 2001 and 2011 using spatial metrics derived from two LULC classification schemes: the more common urban/non-urban binary and a non-binary that considered four urban classes individually. The results indicated that many of the spatial metrics were significantly correlated with existing sprawl indices, suggesting that they accurately quantified components of urban form associated with urban sprawl. More sprawl-like morphologies were typically located in the Eastern region of the U.S. although the regional variability of select spatial metrics was dependent on the LULC classification scheme. Over the 10-year study period, spatial metric-based sprawl indices that compared the relative abundance of low and high intensity urban development suggested that sprawl attributable to low-density single family residential suburbs generally decreased among most metropolitan areas. However, detailed case studies revealed that sprawling development was still likely increasing within particular metros in the form of strip commercial development. Overall, the findings highlight the importance of considering multiple classification schemes to maximize the utility of spatial metrics for urban morphological analysis and urban planning.