Related Experiment Video
Updated: Jun 5, 2026
09:01
Handling and Tagging Techniques for Implanting Juvenile American Shad with a New Acoustic Microtransmitter
Published on: June 14, 2024
Advancing passive acoustic ship detection in the Western Canadian Arctic: Signal processing and deep learning
Farid Jedari-Eyvazi1, Fabio Soares Frazao2, William D Halliday3,4
1Department of Mathematics & Statistics, Dalhousie University, 6316 Coburg Road, Halifax, Nova Scotia, B3H 1Z9, Canada.
The Journal of the Acoustical Society of America
|June 4, 2026
Summary
Arctic vessel traffic is increasing, impacting marine mammals with underwater noise. New methods, including a CNN model, effectively detect ship noise in passive acoustic monitoring data, aiding conservation efforts.
Area of Science:
- Marine Biology
- Acoustics
- Data Science
Background:
- Arctic sea ice retreat is increasing vessel traffic and underwater noise.
- This noise interferes with marine mammal communication and poses risks to their habitats.
Purpose of the Study:
- To develop and compare novel techniques for detecting ship noise in passive acoustic monitoring (PAM) data.
- To assess the effectiveness of these techniques in supporting conservation efforts for Arctic marine mammals.
Main Methods:
- Proposed two ship noise detection techniques: a modified Frequency Amplitude Variation (MFAV) method and a convolutional neural network (CNN) model.
- Trained and tested the models using PAM data from the western Canadian Arctic.
- Compared detection performance using peak F1-scores.
Main Results:
- The CNN model demonstrated strong generalization to new sites, outperforming MFAV and FAV by 1%-8% (F1-scores >91%).
- MFAV significantly improved detection of smaller boats (up to 22%) and larger ships (6%).
- Both developed methods are available as an open-source tool.
Conclusions:
- The CNN model offers a robust solution for ship noise detection in the Canadian Arctic.
- MFAV provides valuable improvements for detecting various vessel types.
- These advancements support acoustic vessel monitoring and marine mammal conservation in the Arctic.