Fractal analysis of multiscale spatial autocorrelation among point data

Environment and Planning A



The analysis of spatial autocorrelation among point-data quadrats is a well-developed technique that has made limited but intriguing use of the multiscale aspects of pattern. In this paper are presented theoretical and algorithmic approaches to the analysis of aggregations of quadrats at or above a given density, in which these sets are treated as multifractal regions whose fractal dimension, D, may vary with phenomenon intensity, scale, and location. The technique is illustrated with Matui's quadrat house-count data, which yield measurements consistent with a nonautocorrelated simulated Poisson process but not with an orthogonal unit-step random walk. The paper concludes with a discussion of the implications of such analysis for multiscale geographic analysis systems. -Author
Publication type Article
Publication Subtype Journal Article
Title Fractal analysis of multiscale spatial autocorrelation among point data
Series title Environment and Planning A
DOI 10.1068/a230545
Volume 23
Issue 4
Year Published 1991
Language English
Larger Work Type Article
Larger Work Subtype Journal Article
Larger Work Title Environment and Planning A
First page 545
Last page 556
Google Analytic Metrics Metrics page
Additional publication details