Boontore, A; (2014) Sprawl and Distance Travelled: Evidence from European Metropolitan Areas. Doctoral thesis , (UCL) University College London.

Preview

Text Boontore.pdf Download (5MB) | Preview

Abstract

The fact that populations tend to agglomerate in urban areas during a time when world population is increasing and cities are exerting an ever greater pull on migrants makes city growth management one of the most challenging issues the world needs to confront. The urban phenomenon which has been called ‘sprawl’ for over a century has been identified as causing various key problems e.g., loss of resource land, amplifying dependencies on the automobile, generating air pollution, and so on. However, there is controversy, even amongst scholars, surrounding the advantages and disadvantages of this phenomenon since there is no common agreement principally on the definition and characteristics of sprawl. This research was sparked by the idea that measurement is a significant step to gaining knowledge of the fundamental structures and processes at work in city systems which generate sprawl. The main purpose of this study was to create an aggregate sprawl index using distance travelled made by private motorised vehicles as a proxy for sprawl, and taking European metropolitan areas as case studies. With respect to independent variables, quantitative characteristics of land-use patterns in relation to sprawl were developed across eight operational dimensions primarily using concepts from landscape ecology. Twenty-three indicators according to these dimensions were reviewed. Then, numerical data on land area, perimeter and distance were extracted from the land cover maps of various European metropolises, and used in calculating the aforementioned indicators principally using MATLAB© scripting. In identifying the most suitable indicators for each operational dimension of sprawl, various criteria were applied, including stability tests of these indicators. Finally, multiple linear regression analysis was employed to model sprawl. As a result, a linear regression model of urban sprawl for these metropolitan areas was formed using a single independent variable, the eta squared index (ETA), which is characterised by a mixture of land uses, thus introducing the concept of ‘exposure’. This model is capable of explaining 32.5 percent of the total variation of the degree of sprawl with respect to private motorised vehicle distance travelled. Complemented by the model’s effect size index which is also measured by Cohen’s f2 statistic (with a value of 0.48) gives some confidence in the variables that have been identified in this study and which form the basis of the exposure index.

Download activity - last month

Export as CSVXMLJSON

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as CSVXMLJSON

Archive Staff Only

View Item