Repository logo

A Time Segmentation Approach for Estimating Time-Varying Parameters in Northern Fur Seals

dc.contributor.authorAlghamdi, Omar Ahmad
dc.contributor.copyright-releaseNo
dc.contributor.degreeDoctor of Philosophy
dc.contributor.departmentDepartment of Mathematics & Statistics - Statistics Division
dc.contributor.ethics-approvalNot Applicable
dc.contributor.external-examinerDr. Haroon Mouhamed Barakat
dc.contributor.manuscriptsNo
dc.contributor.thesis-readerDr. Bruce Smith
dc.contributor.thesis-readerDr. Théo Michelot
dc.contributor.thesis-supervisorDr. Ammar Sarhan
dc.date.accessioned2025-10-09T16:19:07Z
dc.date.available2025-10-09T16:19:07Z
dc.date.defence2025-10-07
dc.date.issued2025-10-07
dc.description.abstractUnderstanding marine animal behaviour is essential for ecological studies and conservation. Traditional models assume animals switch between a few fixed behavioural states, but behaviour often changes gradually, requiring more flexible methods. This thesis introduces a time segmentation framework for estimating time-varying behavioural parameters from high-resolution northern fur seal tracking data. The approach divides trajectories into short, overlapping windows and applies simple statistical models to estimate movement persistence and variability, enabling dynamic behavioural classification without predefined states. Three studies demonstrate its flexibility: a state-space model with a Kalman Filter for horizontal movement, autoregressive spectral analysis for vertical diving patterns, and a three-dimensional continuous-time model incorporating ocean drift. The framework identifies multiple behavioural modes while avoiding the complexity of switching models, providing interpretable, time-resolved insights into animal movement and offering a practical foundation for future research in movement ecology.
dc.identifier.urihttps://hdl.handle.net/10222/85466
dc.language.isoen
dc.subjectState-space models
dc.subjectAnimal movement
dc.subjectTime segmentation
dc.subjectNorthern fur seals
dc.titleA Time Segmentation Approach for Estimating Time-Varying Parameters in Northern Fur Seals

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
OmarAAlghamdi2025.pdf
Size:
17.21 MB
Format:
Adobe Portable Document Format

License bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description: