Spatiotemporal Modelling of Animal Movement: Challenges and Opportunities

November 18, 2016 (3:30 pm - 4:30 pm)

Joanna Mills Flemming Associate Professor
Department of Mathematics and Statistics
Dalhousie University

Spatiotemporal Modelling of Animal Movement: Challenges and Opportunities

Animals move in order to maximize their probability of survival and reproduction. The movement of an animal therefore reflects its response to its current physical needs and available environment. In the marine realm, where direct observation of animal movements is often impossible, researchers typically employ satellite telemetry positioning systems to obtain series of estimates of locations of animals in space through time. Each series resembles an animal path or track. Inferring behavioural states from animal tracks is possible by reasonably assuming that different types of movement, and therefore behavioural state, can be reflected by a change in characteristics of an animal path. For example, while foraging can often be characterized by a tortuous track, a more directed path may suggest travelling between foraging patches. State-space models represent an ideal framework for accounting for both measurement error and process error in these tracks. I discuss novel formulations that allow us to model these tracks robustly and at the same time infer animal behavioural states. Information gained from these models can be used for proper management of both species and ecosystems.

Bio: Joanna Mills Flemming is an Associate Professor in the Department of Mathematics and Statistics at Dalhousie University. She is the Graduate Coordinator for its Division of Statistics, Associate Director of the Canadian Statistical Sciences Institute (CANSSI) and Incoming Chair of the National Sciences and Engineering Research Council Mathematics and Statistics Evaluation Group. She currently leads collaborative research projects for both CANSSI and the Ocean Tracking Network. The primary goal of her research program is to develop statistical modeling and inference methodologies that are essential for answering relevant scientific questions. Her particular interest is in marine ecology, and more broadly, environmental science, which links naturally with her expertise in working with complex data exhibiting spatial and/or temporal dependencies. The motivation for her research lies in its application and necessitates an interdisciplinary and collaborative approach.

Go back