Detecting collective behaviour in animal relocation data, with application to migrating caribou

• Published in: Methods in Ecology and Evolution
• Main Authors: Dalziel, Benjamin D., Corre, Mael Le, Côté, Steeve D., Ellner, Stephen P.
• Other Authors: Matthiopoulos, Jason
• Format: Article in Journal/Newspaper
• Language: English
• Published: Wiley 2015
• Subjects: caribou; Rangifer tarandus
• Online Access: http://dx.doi.org/10.1111/2041-210x.12437; https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1111%2F2041-210X.12437; https://onlinelibrary.wiley.com/doi/pdf/10.1111/2041-210X.12437; https://onlinelibrary.wiley.com/doi/full-xml/10.1111/2041-210X.12437; https://besjournals.onlinelibrary.wiley.com/doi/pdf/10.1111/2041-210X.12437

Summary Collective behaviour can allow populations to have emergent responses to uncertain environments, driven by simple interactions among nearby individuals. High‐throughput ethological studies, where individual behaviour is closely observed in each member of a population (typically in the laboratory or by simulation), have revealed that collective behaviour in populations requires only rudimentary cognitive abilities in individuals and could therefore represent a widespread adaptation to life in an uncertain world. However, the ecological significance of collective behaviour is not yet well understood, as most studies to date have been confined to specialized situations that allow intensive monitoring of individual behaviour. Here, we describe a way to screen for collective behaviour in ecological data that is sampled at a coarser resolution than the underlying behavioural processes. We develop and test the method in the context of a well‐studied model for collective movement in a noisy environmental gradient. The large‐scale distribution patterns associated with collective behaviour are difficult to distinguish from the aggregated responses of independent individuals in this setting because independent individuals also align to track the gradient. However, we show that collective idiosyncratic deviations from the mean gradient direction have high predictive value for detecting collective behaviour. We describe a method of testing for these deviations using the average normalized velocity of the population. We demonstrate the method using data from satellite tracking collars on the migration patterns of caribou ( Rangifer tarandus ), recovering evidence that collective behaviour is a key driver of caribou migration patterns. We find moreover that the relative importance of collective behaviour fluctuates seasonally, concurrent with the timing of migration and reproduction. Collective behaviour is a potentially widespread dynamic property of populations that can, in some cases, be detected in coarsely sampled ...