Projet
OWFSOMM
Offshore wind farm surveys of marine megafauna: standardisation of tools and methods for monitoring at farm scales
- Durée : 36 months (2020 - 2023)
- Budget : €1,447K
Motivation & background
Digital solutions for monitoring marine megafauna from aerial surveys have undergone significant technical development and are now being used as part of the environmental monitoring of French offshore wind farms. It is essential to ensure that the data acquired digitally is commensurate with data obtained from aerial surveys using on-board observers. In addition, the use of instrumented platforms at sea to monitor marine megafauna is increasingly being considered, and holds great promise. However, methodological developments are needed to automate future monitoring and extract the relevant ecological information from the multimodal data obtained.
Objectives
- To provide an operational guide for robust intercalibration of visual and digital aerial monitoring of marine megafauna in offshore wind farms
- To develop artificial intelligence tools for the identification and characterisation of marine megafauna
Main achievements
Completion of 14 aerial monitoring campaigns of marine megafauna over 5 offshore wind farm sites and several seasons
Calculation of inter-calibration indices and statistical analyses of their variability according to site, species and observation conditions.
Development of automated detection algorithms for birds and marine mammals
Methodological and practical recommendations to facilitate and harmonise the monitoring of megafauna, both from aerial surveys and from instruments deployed at sea.
Resources generated
- Multi-source databases: visual and digital observations from aerial surveys, passive acoustics, 360° HD videos, etc.
- Technical reports and recommendations on the intercalibration of visual and digital approaches, and on the development of deep learning algorithms
- Development of a standardised data format for the processing of images from digital aerial surveys
- Algorithms for automated detection of marine megafauna from videos and passive acoustic data
- R package ‘sismow’ for simulating the distribution and abundance of birds and marine mammals, and testing the effectiveness of different sampling plans
Conclusion
OWFSOMM has highlighted the difficulty of comparing abundance estimates obtained by visual and digital methods from aerial surveys, showing that it is preferable to opt for the same monitoring method throughout the life cycle of an offshore wind farm. As offshore wind farms have to be flown over at higher altitudes for security reasons, digital aerial surveys should be favoured from the environmental baseline onwards. By developing promising automated algorithms and implementing a multimodal data acquisition approach, OWFSOMM have laid the first operational building blocks for integrated, continuous and long-term monitoring of the various compartments of the marine megafauna, an objective currently being pursued in the DRACCAR-MMERMAID project.
Partners
This project was led by CEFE, IRISA (Université de Bretagne Sud) and France Energies Marines.
Funding
This project received French State funding managed by the National Research Agency under the France 2030 investment plan. It also benefited from financial support of the Direction générale de l’énergie et du climat (DGEC) and the French Biodiversity Agency (OFB).
Accreditation
This project was certified by the maritime cluster Pôle Mer Méditerranée.
Pour aller plus loin
Ajoutez votre titre ici
Lorem ipsum dolor sit amet, consectetur adipiscing ut elit tellus
Ajoutez votre titre ici
Lorem ipsum dolor sit amet, consectetur adipiscing ut elit tellus
Ajoutez votre titre ici
Lorem ipsum dolor sit amet, consectetur adipiscing ut elit tellus