Bayesian Melding of the Dead-Reckoned Path and GPS Measurements for an Accurate and High-Resolution Path of Marine Mammals

• Activity Number: 302
• Type: Topic Contributed
• Date/Time: Tuesday, August 11, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics and the Environment
• Author(s): Yang Liu* and Brian C. Battaile and James Zidek and Andrew W. Trites
• Companies: The University of British Columbia and The University of British Columbia Marine Mammal Research Unit and The University of British Columbia and The University of British Columbia Marine Mammal Research Unit
• Keywords: spatial-temporal ; big data ; Dead-Reckoning ; Bayesian Melding ; Biologging ; Brownian Bridge

Abstract

With recent advances in electrical engineering, devices attached to free-ranging marine mammals today can collect oceanographic data in remarkably high spatial-temporal resolution. However, those data cannot be fully utilized without a matching high-resolution, accurate path for the animal, something that is currently missing in this field. In this paper, we develop a Bayesian Melding approach based on a Brownian Bridge process to combine the fine-resolution but seriously biased Dead-Reckoned path and the precise but sparse GPS measurements; the result is an accurate and high-resolution estimated path together with credible bands as quantified uncertainty statements. We also exploit the properties of underlying processes and some approximations to the likelihood to dramatically reduce the computational burden of handling those big high-resolution data sets.