The DeuteroNoise dataset: An open, calibrated, multi-basin resource for vessel noise and natural soundscapes in European coastal waters

Abstract

Understanding underwater soundscapes is essential for assessing the impacts of maritime noise pollution on marine environments. Characterizing these soundscapes is crucial both for evaluating current noise impacts and for developing future mitigation strategies. Several datasets have been published focusing on the acoustic signatures of common vessel types; however, most remain restricted to a single location, are not fully open to the public, and lack scalable storage or dissemination tools. The DeuteroNoise Dataset addresses these gaps; an open-access corpus that pairs long-duration, calibrated hydrophone recordings with time-synchronised Automatic Identification System data to document coastal vessel noise and contrasting natural soundscapes at three European sites: the Catalan coast, the Venice Lagoon, and the western Black Sea. Six short-term fixed-station campaigns conducted since December 2023 have produced nearly 700 h of publicly available continuous audio with more than 11 h of labelled audio. Each recorded and identified event is correlated with the ship’s identity, position, and speed metadata; the dataset therefore spans cargo vessel traffic, workboats, leisure craft, and non-anthropogenic background sounds. Vessel types are categorized and linked to their acoustic signatures, facilitating analyses of soundscape dynamics and ecological impact. Built on PostgreSQL with a FastAPI backend, and served through an interactive web interface, the dataset offers a scalable platform for large-scale retrieval and exploration. By integrating calibrated, multi-basin recordings with vessel metadata in an openly accessible, scalable framework. The DeuteroNoise Dataset represents the first resource of its kind in Europe, enabling robust cross-regional comparisons, supporting the development of AI-based classification models, advancing ecological research on anthropogenic noise, and setting a new benchmark for underwater soundscape monitoring worldwide.