A spatial statistical approach to examining sprawled urban growth patterns over time in the framework of Geographic Information Systems (GIS).

摘要

This dissertation aims at examining sprawled urban growth patterns over time by devising a quantitative measure in a spatial statistical framework. Urban sprawl is recognized as an urban form in which negative impacts of separated land uses occur in the economy, society, and environment, owing to the spatial structure of inducing excessive automobile use. As an alternative land use practice against sprawl, the smart growth (or new urbanism) paradigm has emerged over the last decade. One of the key components of this new direction is the promotion of land use mix through which automobile uses for daily movement can be considerably reduced. In this study, a separated land use pattern is highlighted to draw a typical image of sprawled urban growth processes countering mixed land use patterns.; Recently, the importance of devising quantitative measures that capture the spatial pattern change of urban sprawl over time has been emphasized in association with providing objective bases for formulating urban land use policies. Consequently, many quantitative measures in this regard have been propounded. This study proposes a methodological framework in which the spatial pattern change of separated land uses can be investigated. Since the key premise for this study is the complex nature of land use decisions over some locations and times due to many unpredictable factors, the formulation of methodology is built on the framework of spatial statistics that would properly reflect the spatially stochastic process of sprawled urban land uses.; The methodological framework consists of three related components: (1) the formulation of a separation index (SI), (2) the continuous investigation of land use separation by simply plotting a series of SIs for some periods, and (3) the visual exploration of a local counterpart of the separation index in an exploratory spatial data analysis (ESDA) scheme. The implementation of the framework is integrated in Geographic Information Systems (GIS) to provide an auxiliary, exploratory tool for drafting associated land use policy strategies. The framework is demonstrated using empirical data for Erie County for 1899-1995.; Keywords. Sprawled urban growth, spatial statistics, exploratory spatial data analysis (ESDA), Geographic Information Systems (GIS)