Tracking ships with submarine cables: Unsupervised arrival time detection and event localization using distributed...

Area of Science:

• Marine geophysics
• Ocean acoustics
• Fiber optic sensing

Background:

• Maritime safety and seabed infrastructure protection necessitate effective marine traffic monitoring.
• Existing methods may have limitations in continuous, real-time surveillance.
• Fiber-optic telecommunication cables offer a potential sensing network.

Purpose of the Study:

• To investigate the efficacy of distributed acoustic sensing (DAS) for ship detection and localization.
• To analyze ship-generated acoustic signals using persistent homology.
• To assess the performance of DAS against Automatic Identification System (AIS) data.

Main Methods:

• Utilizing a seabed-trench fiber-optic cable in Trondheimsfjord, Norway for acoustic data acquisition.
• Applying persistent homology to analyze continuous acoustic signals from ships.
• Extracting and inverting arrival time curves for ship traveltimes and localization.

Main Results:

• Persistent homology successfully identified arrival time patterns in DAS data.
• Ships were accurately localized up to 1.6 km from the fiber optic cable.
• DAS detected a vessel not present in AIS records, demonstrating added value.

Conclusions:

• DAS is a promising technology for real-time marine traffic monitoring.
• This method enhances maritime situational awareness and protects critical seabed infrastructure.
• The approach can contribute to collision avoidance systems, including whale-ship strike mitigation.