Statistics for Animal Tracking Data

Abstract

Advances in technology are paving the way for researchers to remotely track wild animals and collect massive, high-resolution animal movement data sets with temporal and/or spatial structure. However, the rate at which data are becoming available is outpacing the development of statistical methodology that can adequately analyze them. In this article, we cover the most widely used modeling approaches for the analysis of animal movement data and various extensions that have been proposed for each modeling framework, as well as challenges that remain. There are several newer statistical challenges that researchers have tried to tackle in recent years, such as modeling data streams collected at vastly different temporal resolutions from multiple devices to study animal behavior and incorporating physiological processes as drivers of animal movement. We conclude with additional statistical challenges and opportunities that remain to advance the study of animal movement.