Date of Award


Document Type


Degree Name

Doctor of Philosophy in Civil Engineering (PhD)

College, School or Department Name

Department of Civil and Environmental Engineering


William Sproule


Urban sprawl research generally fits into one or more of four realms including definitions, causes, components, and consequences. Although research on consequences continues to thrive, research on components is in its adolescence, primarily due a lack of consensus on definition. Recent studies such as Ewing et al. 2014 have narrowed the list of sprawl metrics to about 20 within four factors including development density, land use mix, activity centering, and street accessibility.

This main product of this research is a Sprawl Scorecard for small Michigan cities varying in size from Traverse City, nearly 50,000 people in the urban cluster, down to Saint Ignace, with only 2,500. 42 small cities are included in the study, with an even spread of cities across the state.>/p>

One of the limitations with sprawl research is the focus on large cities. There is good reason to study large cities. Large cities affect more people, have more economic influence, and suffer recognizable consequences of sprawl (e.g., traffic congestion). However, large cities have more confounding variables at play than small cities making it difficult to narrow down the components. Even assuming components could be measured well, large cities have more players making change difficult.

In small cities, sprawl may not affect everyone’s lives in the same magnitude (e.g., Houghton’s “rush-minute”), but sprawl does exist and is noticeable. Sprawl is easier to measure in small cities and if measured well, policy is much easier to change as there are many fewer players involved and less existing development.

The Sprawl Scorecard provides insight to local and regional planners to mitigate sprawl in their regions. This research also offers researchers several paths for future work in all four areas of sprawl research.

Included with the development of the Sprawl Scorecard is original software written in Python using ArcGIS. The first program generated Extended Urban Clusters, based on Extended Urban Areas developed by Wolman et al. The second calculated 21 sprawl metric scores for each city and the third used principal components analysis to combine the metrics into four component scores and an overall Sprawl Score for each city.