Bayesian State-Space Models for Acoustic Telemetry Data

• Activity Number: 125 - A New Era of Mark-Recapture: Advances in Bayesian Methods for Modeling and Inference
• Type: Topic-Contributed
• Date/Time: Monday, August 9, 2021 : 1:30 PM to 3:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Saman Muthukumarana* and Inesh Munaweera and Darren Gillis
• Companies: University of Manitoba and University of Manitoba and University of Manitoba
• Keywords: Acoustic Telemetry; Animal Movements; Bayesian Inference; Markov Chain Monte Carlo

Abstract

Bayesian state-space models provide promising approaches for modeling individual-level animal movements in ecological applications. These models have been widely used for studying animal movements since they can account for both, process variation, which is the natural variation of the underlying movement process and the observational error, which is the difference between the observed position and the true position of the animal. In this talk, I will discuss Bayesian state-space modeling approaches with different smoothing methods including kernel smoothing, and cross-validated local polynomial regression and a new approach to reconstruct the animal movement paths. The approaches will be highlighted using data obtained from a telemetry receiver grid in Lake Winnipeg.