Generalizing the first-difference correlated random walk for marine animal movement data

Area of Science:

• Ecology
• Movement Ecology
• Biologging

Background:

• Animal telemetry data are frequently analyzed using discrete-time movement models with regular time steps.
• Marine animal telemetry data are often collected at irregular intervals, posing challenges for standard models.

Purpose of the Study:

• To introduce a novel time-irregularised first-difference correlated random walk model with drift.
• To generalize existing models by accommodating irregular time steps and including a drift term.

Main Methods:

• Development of a time-irregularised correlated random walk model.
• Application of the model to ringed seal telemetry data from the Argos satellite system.
• Comparison with related movement models through simulations.

Main Results:

• The new model provides accurate parameter estimates and movement path reconstructions for irregular telemetry data.
• It offers more accurate movement paths compared to regular time models.
• Simultaneous analysis of multiple animals is enabled by directly handling irregular data.

Conclusions:

• Accounting for irregular sampling in movement models is crucial for accurate analysis of marine animal ecology.
• The developed model enhances the understanding of space use patterns, vital for effective wildlife management.
• Efficient analysis of large-scale telemetry datasets is facilitated.