Sensitivity of codispersion to noise and error in ecological and environmental data

TL;DR

This study evaluates how noise and measurement errors affect the reliability of codispersion analysis in ecological and environmental data, using simulations and real datasets to identify robustness thresholds and improve interpretation.

Contribution

It provides the first empirical assessment of codispersion's sensitivity to data contamination, offering practical guidelines and a method for imputing missing spatial data.

Findings

Codispersion estimates are robust with less than 15% contamination.

Sensitivity increases significantly with higher contamination levels.

A new method for imputing missing spatial data enhances analysis robustness.

Abstract

Codispersion analysis is a new statistical method developed to assess spatial covariation between two spatial processes that may not be isotropic or stationary. Its application to anisotropic ecological datasets have provided new insights into mechanisms underlying observed patterns of species distributions and the relationship between individual species and underlying environmental gradients. However, the performance of the codispersion coefficient when there is noise or measurement error ("contamination") in the data has been addressed only theoretically. Here, we use Monte Carlo simulations and real datasets to investigate the sensitivity of codispersion to four types of contamination commonly seen in many real-world environmental and ecological studies. Three of these involved examining codispersion of a spatial dataset with a contaminated version of itself. The fourth examined differences in codisperson between plants and soil conditions, where the estimates of soil characteristics were based on complete or thinned datasets. In all cases, we found that estimates of codispersion were robust when contamination, such as data thinning, was relatively low (<15\%), but were sensitive to larger percentages of contamination. We also present a useful method for imputing missing spatial data and discuss several aspects of the codispersion coefficient when applied to noisy data to gain more insight about the performance of codispersion in practice.