Thumbnail Image

Modelling Bird Migration with Motus Data and Bayesian State-Space Models

Bird migration is a poorly-known yet important phenomenon, as understanding movement patterns of birds can inform conservation strategies and public health policy for animal-borne diseases. Recent advances in wildlife tracking technology, in particular the Motus system, have allowed researchers to track even small flying birds and insects with radio transmitters that weigh fractions of a gram. This system relies on a community-based distributed sensor network that detects tagged animals as they move through the detection nodes on journeys that range from small local movements to intercontinental migrations. The quantity of data generated by the Motus system is unprecedented, is on its way to surpass the size of all other centralized databases of animal detection and requires novel statistical methods. Building from the bsam package in R, I propose two new biologically informed Bayesian state-space models for animal movement in JAGS that include informed assumptions about songbird behavior. I evaluate the models using a simulation study in realistic conditions of data missingness. One of these models is generalized to a hierarchical version that fits population-level movement through joint estimation of movement parameters over multiple animal tracks. To apply the models, I then employ a localization routine on a Motus data set from migrating songbirds (Red-eyed Vireos - Vireo olivaceus) from the Eastern coast of North America. This allows me to apply the new hierarchical model and its predecessor to estimate unobserved locations and behaviors. Migratory flights were observed to occur mostly in the evenings along the coast and directed migratory flights were detected over water over e.g. the Bay of Fundy, the Long Island Sound and the New York Bight. Area-restricted searches were confined to coastal areas, in particular the Gulf of Maine, Long Island and Cape May.
Research Projects
Organizational Units
Journal Issue
Publisher Version
Embedded videos