Cartographic visualization of Saskatchewan’s population using dasymetric mapping.

Date
2012-07
Authors
Krahnen, Anne
Journal Title
Journal ISSN
Volume Title
Publisher
Faculty of Graduate Studies and Research, University of Regina
Abstract

Maps of population density are commonly displayed using the choropleth method. Although choropleth maps are popular in mapping population density, strictly speaking, they are often incorrect due to non-conforming boundaries of populated places and administrative areas. This thesis looks at the dasymetric concept as an alternative to the choropleth map and investigates how it can be applied to display Saskatchewan’s population. Using ancillary data, the dasymetric method can redistribute statistical population data exclusively to areas identified as populated. As a basis for the research, 12 existing dasymetric methods were reviewed and evaluated for their applicability to Saskatchewan. Based on this evaluation, three methods (Binary Method, Intelligent Dasymetric Method and the Chinese Population Distribution Model) were selected and applied to a test area (Saskatchewan’s Census Division 6). It was assumed that the test area’s geographical attributes were representative for the province. The ancillary dataset used for the dasymetric mapping was chosen based on the results of regression analysis. Different file formats (raster and vector) were considered when producing the dasymetric maps. It was observed that, due to the regular sizes of raster cells, dasymetric maps using raster data are generally more accurate than dasymetric maps using vector data. In the mapping process, 88 dasymetric maps were produced for the test area. All maps were evaluated subjectively (visually), as well as quantitatively by calculating the variations from population densities of Dissemination Blocks (the most accurate population data available for this project) using the Mean Absolute Error. In both analyses, the binary method using raster data was determined to be the method producing the most accurate maps for the test area. Based on the results, three dasymetric maps were produced for the Province of Saskatchewan. The binary method using raster data was applied at three different resolutions (250m, 1,125m and 2,500 m). The resulting maps were therefore optimized for three different scales (1:5,000; 1:2,250,000 and 1:5,000,000). Keywords: Cartography; GIS; Population Maps; Dasymetric Concept; Binary Method; Intelligent Dasymetric Method; Chinese Population Distribution Model; Mean Absolute Error; Saskatchewan

Description
A Thesis Submitted to the Faculty of Graduate Studies and Research In Partial Fulfillment of the Requirements for the Degree of Master of Science in Geography, University of Regina. xv, 124 l. : maps (some col.)
Keywords
Citation
Collections