A machine learning approach for underwater gas leakage detection

TL;DR

This paper presents a machine learning-based method utilizing Passive Acoustic Monitoring and Hidden Markov Models to detect underwater gas leaks effectively, aiding early detection in Carbon Capture and Storage projects.

Contribution

It introduces a novel combination of classification algorithms and smoothing strategies for underwater gas leak detection using simulated acoustic data.

Findings

Classification algorithms achieved high detection accuracy.

Hidden Markov Models improved detection stability over time.

Method demonstrated effective performance on simulated Brazilian shore data.

Abstract

Underwater gas reservoirs are used in many situations. In particular, Carbon Capture and Storage (CCS) facilities that are currently being developed intend to store greenhouse gases inside geological formations in the deep sea. In these formations, however, the gas might percolate, leaking back to the water and eventually to the atmosphere. The early detection of such leaks is therefore tantamount to any underwater CCS project. In this work, we propose to use Passive Acoustic Monitoring (PAM) and a machine learning approach to design efficient detectors that can signal the presence of a leakage. We use data obtained from simulation experiments off the Brazilian shore, and show that the detection based on classification algorithms achieve good performance. We also propose a smoothing strategy based on Hidden Markov Models in order to incorporate previous knowledge about the probabilities of leakage occurrences.