A Practical Guide to Studying Magnetic Alignment of Free-Living Animals: Lessons From Leopards

Abstract

Spontaneous magnetic alignment describes the tendency of animals to preferentially orient their major body axis in a non-random alignment relative to the magnetic field. Several well-controlled laboratory studies have reported this behavior, yet the majority of studies come from field observations in natural contexts. However, behavioral field data are intrinsically complex due to many potential factors influencing animal spatial behavior. Here, we use a published data set on leopard alignment, Yosef et al. (2022), to showcase potential pitfalls in studies of magnetic alignment, and offer guidance on avoiding such pitfalls. We provide an approach to test for spatial autocorrelation (with scripts available in R) which was validated in a small simulation study and offer a roadmap for magnetic alignment investigations. Our example demonstrates that spatial autocorrelation can be a complicating issue when analyzing field data, potentially leading to reduced power to detect real effects or overinterpretation of apparent evidence of alignment. Developing specific techniques to overcome spatial autocorrelation, without losing statistical power, will be an important challenge for future methodological development in circular statistics. Our paper will help researchers to more robustly evaluate evidence for field studies on magnetic orientation.