Regional Variations in Urban Fragmentation Among U.S. Metropolitan and Megapolitan Areas

Bradley Bereitschaft, Keith Debbage

Research output: Contribution to journalArticlepeer-review

25 Scopus citations


In this paper we investigate the regional variations in the fragmentation of urban landcover among 86 metropolitan areas and 19 megapolitan areas in the United States. Urban fragmentation was evaluated using nine spatial metrics that collectively quantified the continuity and shape complexity of urban landcover. Spatial metrics were calculated for each metropolitan and megapolitan area using a high urban threshold containing the urban core and surrounding suburbs, and a low urban threshold that further encompassed the outer exurban fringe. A principal component analysis was used to collapse the nine spatial metrics into two components of urban form: "shape complexity", which describes the porosity of the urban fabric, and urban "continuity", which represents the aggregation of urban patches or pixels. Urban "continuity", and seven of nine spatial metrics, varied significantly by U.S. census region. A hot-spot analysis further revealed a high degree of spatial autocorrelation, with metropolitan areas in the Northeast and South regions generally exhibiting a more fragmented urban landscape than those in the Midwest or West regions. Relative to metropolitan areas, megapolitan areas exhibited more complex and fragmented patterns of urban landcover, presumably due to the high level of inter- and intra-urban polycentricism at this broader scale.

Original languageEnglish (US)
Pages (from-to)119-147
Number of pages29
JournalApplied Spatial Analysis and Policy
Issue number2
StatePublished - Jan 2014


  • Megapolitan area
  • Metropolitan area
  • Spatial metrics
  • Urban fragmentation
  • Urban sprawl

ASJC Scopus subject areas

  • Geography, Planning and Development


Dive into the research topics of 'Regional Variations in Urban Fragmentation Among U.S. Metropolitan and Megapolitan Areas'. Together they form a unique fingerprint.

Cite this