Leveraging Distributed Acoustic Sensing for monitoring vessels using submarine fiber-optic cables

The global offshore cable and pipeline infrastructure is ever growing to fulfill our demand for renewable energy -and internet communication. At the same time there is an increasing amount of vessel traffic due to the growth of offshore activities, which also imposes an increased risk of damage done to seabed infrastructure. This highlights the need to develop methods for continuously monitoring vessel traffic in the vicinity of seabed infrastructure over its full extent. Recent studies have demonstrated that redundant optical fibers pre-existing in offshore cables and pipelines (i.e., dark fibers), can be used to monitor vibrations along and over long distances (>100 km) using a Distributed Acoustic Sensing (DAS) interrogator. Different vessel parts are inducing acoustic vibrations, that subsequently propagate through the water column and shallow subsurface and temporarily deform a fiber-optic cable present inside seabed infrastructure. We developed a migration-based source location method to automatically detect and locate vessels using dark-fiber data and applied it both to a dataset acquired on the Dutch North Sea and a dataset from the west coast of Oregon, USA. The track, speed and course of the considered vessels determined from the analysis of DAS data show excellent agreement with the Automatic Identification System data in the vicinity of the fiber-optic cable. The migration-based source location method effectively uses the high spatial-temporal density of DAS data by constructive summation of coherent waveforms over space and time. Furthermore, the method can be extended to construct and refine velocity models by iteratively modifying the velocity model until a maximum of the objective function is found—corresponding to an optimal fit between observed and synthetic travel-times—provided that the uncertainty in propagation speed can be decoupled from position uncertainties in the fiber-optic cable. Thereby the method allows both for optimizing the velocity model and inverting for the acoustic source location in a sequential manner, which makes it additional valuable for subsurface studies.