A fractal approach to explore australian urban form and its impacting factors at neighbourhood scale

Abstract

This chapter demonstrates an application of the correlation fractal method for exploring place diversity at neighbourhood scale. An empirical case study was performed on 130 neighbourhoods in the state capital city of Brisbane in Queensland, Australia. We used Google Maps as the data source to capture the neighbourhood building footprints for fractal analysis and compared the fractal results (fractal dimension and α) of Brisbane with those from similar studies of European cities. Spatial correlation analysis was employed to explore key factors impacting place diversity including different characterisation of density, metrics of land-use mix and accessibility. Results show that the urban form in Brisbane lacks place diversity and is homogenized, with no remarkable difference between inner city and suburban neighbourhoods, and its fractal dimension is lower than European cities. The fractal results of neighbourhoods are influenced by modernist planning principles including low-density housing, functional zoning and hierarchical street networks. Our application confirms that the correlation fractal method is suitable for describing urban form at the neighbourhood scale in the Australia context.