Doctor of Philosophy (PhD)


Geography and Anthropology

Document Type



This dissertation revolves around three issues on the urban area of Port-au-Prince, Haiti: the population distribution pattern, its estimation from remote sensing images, and its relationship with environmental quality. It follows a three-paper format. Paper 1 examines the population density pattern by the monocentric and polycentric models, based on the 2003 census data. The regression results show a poor fitting power of monocentric functions, and improved but less than satisfactory R2 by polycentric functions. A five-sector conceptual model is proposed to capture the urban structure shaped by the absence or lack of institutional enforcement of land use regulations and urban planning. Paper 2 proposes a population estimation model based on Landsat ETM+ images that are widely available. The subpixel vegetation-impervious surface-soil (VIS) fractions derived from the Landsat multispectral bands (the mean value of houses fraction image, the mean value of vegetation and the standard deviation of vegetation fraction image) are used as predictors for urban population density. The research indicates that the geographically weighted regression (GWR) model, which accounts for spatial non-stationarity, performs much better than its Ordinary Least Square counterpart. Paper 3 uses multiple factors to assess and map the urban environmental quality (UEQ). In addition to parameters typically considered in previous studies, this study includes natural hazards and other parameters unique to Port-au-Prince. Crowdedness, waste, lack of vegetation, presence of slums and water body pollutions are considered as the most critical factors (negatively) affecting the quality of the environment in Port-au-Prince. All are exacerbated by population pressure on the resources, i.e., population density. The scores for corresponding factors are integrated together by weights extracted from a panel of local experts. The overall UEQ results are validated by field surveys. Each paper discusses important implications of major findings for public policy and planning



Document Availability at the Time of Submission

Student has submitted appropriate documentation to restrict access to LSU for 365 days after which the document will be released for worldwide access.

Committee Chair

Wang, Fahui