Algorithm for analyzing randomness in point patterns

Résumé fourni par la source

Various tests can be used to assess whether the spatial distribution pattern of a set of points is random, dispersed, or clustered. These tests typically compare the expected and observed distances among points, assuming no barriers. However, what is deemed ``random'' in point spatial patterns may be influenced by socio-environmental factors such as wetlands or transportation networks. This tool introduces a sequence of spatial analysis procedure and a statistical testing to evaluate the correlation between observed point patterns and potential spatial determinants (polygons). If a determinant influences the observed point pattern, the classification of the distribution as random must be reconsidered. We implemented this algorithm in Python as a QGIS script with two main steps: the first handles overlay operations and preliminary calculations for the chi-square goodness-of-fit test with and without Bonferroni correction in the second step.•A detailed step-by-step procedure for analyzing randomness in point patterns in a processing toolbox Python script for integration into the open-source software QGIS.•Automated scripts for structuring data, calculating expected and observed values, and applying the chi-square goodness-of-fit test for statistical analysis.•Advanced spatial analysis using chi-square goodness of fit coupled with and without Bonferroni correction, providing deeper insight into the study of the effect of spatial phenomena on the distribution of point events.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.