Related Experiment Video
Updated: Jul 1, 2025
05:57
Long-term Video Tracking of Cohoused Aquatic Animals: A Case Study of the Daily Locomotor Activity of the Norway Lobster Nephrops norvegicus
Published on: April 8, 2019
6.8K
Validating hidden Markov models for seabird behavioural inference
Rebecca A Akeresola1,2, Adam Butler2, Esther L Jones2
1School of Mathematics and Maxwell Institute for Mathematical Sciences University of Edinburgh Edinburgh UK.
Ecology and Evolution
|March 5, 2024
Summary
Hidden Markov models (HMMs) accurately infer seabird behaviors from tracking data, validated by visual tracking. This improves conservation planning by providing reliable animal movement insights.
Area of Science:
- Marine ecology
- Animal behavior
- Conservation science
Background:
- Observing animal behavior in marine environments is challenging.
- Hidden Markov models (HMMs) infer behaviors from telemetry data.
- Validating these inferred behaviors is difficult due to lack of ground truth data.
Purpose of the Study:
- To investigate the accuracy of HMMs for inferring seabird behaviors.
- To validate HMM-derived behaviors using simultaneous visual tracking data.
- To assess the conservation implications of accurate behavioral inference.
Main Methods:
- Utilized a unique dataset of simultaneous boat-based visual tracking and seabird behavior observations.
- Applied Hidden Markov Models (HMMs) to telemetry data to infer animal states.
- Compared HMM-inferred behaviors against a 'gold standard' of manually classified visual tracking data.
Main Results:
- HMM accuracy ranged from 71% to 87% during chick-rearing and 54% to 70% during incubation.
- Model choice had minimal impact on accuracy, even with varying AIC values.
- Missed foraging bouts, critical for conservation, were identified as lasting only a few seconds.
Conclusions:
- HMMs reliably identify key conservation-relevant seabird behaviors when validated.
- Visual tracking data serves as a robust method for validating HMM behavioral inferences.
- Increased confidence in using HMMs for animal behavior analysis in conservation is warranted, with a call for integrated validation data collection.
