Categorizing Shallow Marine Soundscapes Using Explained Clusters

Natural marine soundscapes are being threatened by increasing anthropic noise, particularly in shallow coastal waters. To preserve and monitor these soundscapes, understanding them is essential. Here, we propose a new method for semi-supervised categorization of with further interpretation categories according to concurrent environmental conditions. The proposed methodology uses nonlinear mapping short-term spectrograms two-dimensional space, followed density-based clustering algorithm identify similar sound environments. A random forest classifier, based on additional data, used predict their occurrence. Finally, explainable machine learning tools provide insight into the ecological explanation clusters. This was tested Belgian part North Sea, resulted clearly identifiable that could be explained spatial temporal parameters, such as distance shore, bathymetry, tide or season. Classifying facilitates identification, which can useful policy making conservation programs. Soundscape categorization, this work, acoustic trends patterns space time might indicators biodiversity ecosystem functionality change.