Source: Dissertation Abstracts International, Volume: 31-05, Section: B, page: 2760. / Thesis (Ph.D.)--The Florida State University, 1970.

Urban growth or sprawl has been an interesting research topic for contemporary urban studies. The availability of remote sensing and GIS techniques facilitate a large number of empirical and practical studies in addition to traditional theoretical research. From definition, to spatial measures, to exploration of the driving forces and modeling/forecasting of urban growth or sprawl, this research topic has received increasing attention in multiple disciplinary fields, such as geography, urban planning, public policy and administration, environment science, and public health etc. The subtopics of urban growth or sprawl are broad and multidisciplinary. In my dissertation research, I focused mainly on spatial aspects of urban growth and sprawl through examining their patterns and investigating driving forces. My approach integrated data from multiple sources at various geographic scales (ranging from parcels at individual housing scale to land-use policy survey data at city scale) and utilized GIS techniques and statistical methods. The first two dissertation chapters provide an introduction of urban growth/sprawl issues and literature review of previous research. The remaining chapters present four major studies, with the first two focused on devising new urban sprawl measures dedicated to utilizing information-rich, fine-scale parcel data for urban sprawl assessment at aggregated scales. Census population/housing and remote sensing land-cover data have been used to characterize urban growth and urban sprawl effectively but not without limitations. These data at the aggregated level usually remove detailed information on housing types (such as single vs. multi-family residence), unit size, and land-use purposes (such as commercial vs. residential uses). With the increasing availability of parcel data especially in the United States, scholars began to explore and utilize these fine spatial/temporal data with full-detailed attributes (such as land-use type, total living space, number of residential units, the actual year built, and etc.) for urban growth research. To take advantage of the information-rich fine-scale parcel data, two sets of new urban sprawl measures were designed to characterize urban sprawl patterns from different perspectives. The two sets of sprawl measures are introduced in the dissertation as two relatively independent studies, given the variation of measures as well as distinctive study areas. For each set of sprawl measures, three indices were created (with some level of overlap) to capture urban sprawl from the aspects of housing characteristics (development density or housing size), land use diversity, and accessibility to business hubs. These measures of urban sprawl are based on fine-scale parcel data; and are able to capture patterns of sprawl at the city and metropolis levels. Our measures are easily transferable to cities of different development patterns and allow comparison across cities of various dimensions. They may also be used to compare growth of a city or metropolis in time sequence. The third and fourth studies explore urban growth drivers that integrate factors such as socioeconomics, environment, and sustainable urban development policies using Geographic Weighted Regression (GWR) as well as statistical multi-level modeling approaches. Single-level linear regression model is a common approach to examine the relationships between urban growth and associated driving factors. In the third study presented in this dissertation, Geographic Weighted Regression (GWR) analysis, single-level model that takes into consideration of spatial adjacency and variation, was used to explore each driving factor’s influence on urban growth. The potential driving factors were first examined by a global OLS (Ordinary Least Square) model to identify their global influential trends on urban growth across cities in Florida. Then, a local GWR model is applied to detect local variations of these urban growth drivers for cities at different locations. In the fourth study presented in this dissertation, a multilevel linear regression model frame was developed and applied to exploring impacts of urban growth driving factors on urban development, attempting to capture influences at both city and finer geographic scales. First, city effects were examined, and then block-group level variables were included, and lastly city level variables were integrated. Compared to the traditional single-level linear regression model, multilevel modeling is a relatively new method to be used in analyzing urban growth and the associated driving factors. Overall, the entire dissertation work enriches research of urban growth and urban sprawl, in particular the measurement and modeling perspectives from the geography stance. The first two studies present an innovative research attempt that suits well with the era of big data, which geographers can provide unique contribution given the nature of the data we constantly handle. The third and fourth studies target on the unknown relationship between fine-scale empirical observation of urban growth (based on remote sensing data) and meso-scale land-use regulation (based on survey data). This makes these studies unique in terms of integration of knowledge gained in geography, urban planning, and public administration. Finally, the results of our analysis benefit urban planners and policy makers through quantitative assessment of levels of urban growth/sprawl, which provides a knowledge base for their planning or design of sustainable urban development in the future. They may also benefit from our integrated assessment of urban growth drivers, in particular the effectiveness of individual policies on curbing urban growth. / A Dissertation submitted to the Department of Geography in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Summer Semester 2017. / June 19, 2017. / Multilevel Modeling, Parcel Data, Urban Growth Driver, Urban Sprawl, Urban Sprawl Measure / Includes bibliographical references. / Tingting Zhao, Professor Directing Dissertation; Richard Feiock, University Representative; Xiaojun Yang, Committee Member; Christopher Uejio, Committee Member.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.