Authors: Bin Li*, Central Michigan University
Topics: Spatial Analysis & Modeling, Cartography, Temporal GIS
Keywords: geovisualization, spatial structure, Moran spectrum
Session Type: Paper
Start / End Time: 1:10 PM / 2:50 PM
Room: Marshall North, Marriott, Mezzanine Level
Presentation File: No File Uploaded
Eigen-decomposition of the projected spatial weights matrix results in a set of eigenvectors, with each corresponding to an elemental map pattern and a Moran coefficient, forming the Moran spectrum. In a given geographic landscape, the spatial distribution of a geospatial random variable is often controlled by certain spatial structures persistent through time. Such a persistent spatial structure can be represented by the linear combination of selective eigenvectors, which is similar to the spectral signature in remote sensing. The individual eigenvectors can be identified and corresponding weights can be calculated, and visualized as a series of choropleth maps, with each being the decomposition of the spatial structure controlling the geographic distribution of the variable, enabling in-depth analysis of spatial autocorrelation and heterogeneity. The paper reports the latest findings with application examples.