Proc. SPIE. 6421, Geoinformatics 2006: Geospatial Information Technology
KEYWORDS: Statistical analysis, Data modeling, Remote sensing, Error analysis, 3D modeling, Geographic information systems, Analytical research, Data centers, Earth observing sensors, Animal model studies
It has been widely recognized that spatialization of census data is of great significance to implement the interdisciplinary research in population, resource and environment. By means of associating census data with geographical factors such as urban land and rural habitation that intimately tie up with population distribution, adopting revised areal interpolation to extract sample points of census data base, and utilizing trend surface analysis of Kernel to research spatialization of census data, a grid model of census data (GMCD) that lies in the practical population distribution namely the location of habitation are put forward to spatialize census data. Taking Hubei province as study area, data input collected in the year of 2000, including the Landsat/TM images, the fifth national population census data and administrative division data of Hubei province. Application of GMCD simulates regularities of population spatial distribution of Hubei province in 2000, and implements the three dimensional model to represent population density. Study findings indicate that this model can improve accuracy for spatializing census data efficiently. With National Scientific Data Sharing Project (NSDSP) sharing multi-period land use and census data in different scales, this method can be definitely generalized for its simplify and convenience to spatialize census data.