An index for quantifying geometric point disorder in geospatial applications

Abstract

Many techniques have been developed to quantify different conceptualizations of self-interaction and patterns within spatial data. We propose a new metric and related algorithm that describes the geometric spatial disorder of geographic point sets, the “Index of Disorder” (IoD). The IoD algorithm was applied to synthetic and natural datasets and was shown to be able to differentiate between areas of high spatial disorder (randomly placed points) and low spatial disorder (e.g., curvilinear grids, wallpaper groups, and other repeating patterns). Because the IoD is a quantitative metric, it can be used on its own as an aid for identifying areas of unusually high or low spatial disorder or as enrichment for machine learning classification algorithms.