Statistical Models for the Movement of Halibut in the Gulf of Alaska

• Activity Number: 31
• Type: Contributed
• Date/Time: Sunday, July 31, 2016 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Margaret Short* and Andrew Seitz and Julie Nielsen
• Companies: University of Alaska Fairbanks and University of Alaska Fairbanks and University of Alaska Fairbanks
• Keywords: halibut ; Bayesian state space model ; Markov chain Monte Carlo ; simulated tempering ; hidden Markov model

Abstract

Popup archival tags are used to study the fall migration of halibut, an important commercial species inhabiting the Gulf of Alaska. Halibut are demersal fish, mainly staying near the ocean floor.  Each tag records maximum daily depth, which is transmitted via satellite when the tag pops loose at a preprogrammed time. We implement a Bayesian state space model along with a detailed bathymetry map to estimate the path, and we attempt to estimate other model parameters. We discuss two other approaches -- particle filtering and hidden Markov models. These methods apply to other demersal fish such as cod and pollock.