UACE logo

Ocean sound model predictions validation with deep-sea field data

Underwater Acoustics Conference and Exhibit 2025


L.O. Júnior, H. Chognot and S.M. Jesus, LARSyS, University of Algarve, 8005-139 Faro, Portugal


Comments: download presentation pdf.
Ref.: in UACE'2025, June 2025.

Abstract:
Before industrial activities of exploitation of mineral resources of the deep seas begin, the EU Horizon project TRIDENT is actively working on a comprehensive study to fully characterize the processes at play. The project's main objectives are to develop observation methodologies and associated tools required to monitor such activities for impact assessment and forecast.
The apparatus associated with deep-sea industrial activities can be expected to generate at least two important consequences: the generation of sediment plumes and acoustic noise. In the present work we present the details of the development of an ocean sound modeling tool to monitor anthropogenic noise from deep-sea exploitation activities and preliminary validation results; this tool will integrate different input models and real time data from various platforms to provide objective predictions and support decision making.
Gliders and free drifting arrays provide acoustic observation at different ranges and are used for model validation, calibration and to trigger alerts based on predetermined thresholds. Noise observations provided by hydrophones installed on the machinery can be expected to provide source levels to be used as input to an acoustic model. Shipping noise is accounted by transforming AIS information into source levels using the JOMOPANS-ECHO model.
Data acquired during a sea trial in June 2024 at the Tropic Sea Mount resulted in the generation of baseline sound maps. Preliminary validation shows that model predictions are bounded by far-field glider and hydrophone array data for each observation depth range. In general, for the low frequency band (< 200 Hz), model predictions tend to match glider observations, while above that frequency a better match is obtained for the fixed hydrophone data. Future work will be devoted to accounting for self-noise due to both the glider motion, flow noise and surface waves, and to produce excess noise level maps for alert trigger.