Modelling animal movement as Brownian bridges with covariates

Movement Ecology

  ¹ 

[1] 0000 0004 1937 0650, grid.7400.3, Department of Evolutionary Biology and Environmental Studies, University of Zurich, Winterthurerstrasse 190, CH-8057, Zurich, Switzerland;grid.452577.6, Kalahari Meerkat Project, Kuruman River Reserve, P.O. Box 64, 8467, Van Zylsrus, Northern Cape, South Africa;

关键词: Animal tracking;    Brownian bridge covariates model;    Brownian bridge movement model;    Meerkats;    Movement ecology;    Suricata suricatta;    Utilization distributions;   

DOI  :  10.1186/s40462-019-0167-3

来源: publisher

【 摘 要 】

BackgroundThe ability to observe animal movement and possible correlates has increased strongly over the past decades. Methods to analyze trajectories have developed in parallel, but many tools fail to make an immediate connection between a movement model, covariates of the movement, and animal space use.MethodsHere I develop a novel method based on the Brownian Bridge Movement Model that facilitates investigating and testing covariates of movement. The model makes it possible to flexibly investigate different covariates including, for example, periodic movement patterns.ResultsI applied the Brownian Bridge Covariates Model (BBCM) to simulated trajectories demonstrating its ability to reproduce the parameters used for the simulation. I also applied the model to a GPS trajectory of a meerkat, showing its application to empirical data. The value of the model was shown by testing the interaction between maximal daily temperature and the daily movement pattern.ConclusionThis model produces accurate parameter estimates for covariates of the movements and location error in simulated trajectories. Application to the meerkat trajectory also produced plausible parameter estimates. This new method opens the possibility to directly test hypotheses about the influence of covariates on animal movement while linking these to space-use estimates.

【 授权许可】

CC BY