Improving the marine mammals monitoring
In order to study the effects of an ORE project on marine mammals, the frequentation of species in the areas concerned must be characterised. To this end, aerial monitoring is organised during the construction, operation and dismantling of the farm, which means over a total period of about 30 years. Observations are classically based on aerial overflights by specialised naturalist observers. However, recent scientific and technological developments offer new prospects for radically improving the cost-effectiveness ratio of such monitoring. Digital vision has undergone a significant evolution with the development of deep learning, in the form of deep convolutional networks. However, their application to aerial images for marine mammal observation purposes remains a challenge because the elements of interest are represented by a small proportion of pixels. Moreover, the animals evolve at sea, an environment characterised by a highly variable visual content that hinders the performance of deep networks. Finally, the learning data are insufficient in volume and variety.
The automatic characterisation of marine mammals by aerial imagery therefore remains an area open to R&D work. France Energies Marines is a partner in a project funded by the French Ecological Transition Agency and initiated in 2019, which aims to demonstrate the relevance of software solutions for processing and analysis of aerial photographs to ensure the automated census of marine megafauna (SEMMACAPE and OWFSOMM projects).
In collaboration with several entities, the Institute is also working to set up a high-frequency observatory for the marine ecosystems of the Gulf of Lion where marine mammals are numerous. The objective is to learn about the state and evolution of these ecosystems in interaction with high-power floating wind farms (ECOSYSM-EOF project).
Taking into account and preventing risks of collision
Marine mammals have the potential to collide with ORE structures and service vessels or become entangled in the mooring lines of floating systems. This risk is increased during the construction phase of an offshore farm due to increased vessel traffic in the area. France Energies Marines, its members and partners have been working for several years to establish and update a state of the art knowledge in this area. A methodological guide was published in 2013 on the assessment of the environmental impacts of tidal turbines (GHYDRO project). The risk of collision with the ORE systems for marine mammals is widely described. More recently, this aspect has been integrated into an environmental acceptability assessment module in the software suite designed to design and optimise tidal and wave energy systems at a farm scale (DTOCEANPLUS project).
Evaluating the noise impact of a ORE farm
Marine mammals have a way of life that relies heavily on the emission and reception of sound waves. Anthropogenic underwater noise can potentially disturb them for feeding, communication, reproduction… It is therefore crucial to assess the risks associated with the sound impact of a ORE farm beforehand and then to monitor them during the various operational phases of the project. At present, two challenges are emerging:
- the fine characterisation of the levels emitted by the energy recovery systems (including moorings),
- the difficulties of acquiring in-situ measurements in high-current environments where the instrumentation is subjected to severe loads and the background noise generated by strong currents is significant.
Recommendations for the implementation of offshore wind farms resulting from a collaborative R&D project set up and piloted by France Energies Marines was published in 2019. A sheet in this document is dedicated to the work carried out on the impact of noise on the four main species of marine mammals present in the Channel (TROPHIK project). In order to carry out an initial survey on a ORE site or to estimate the noise emitted by the turbines once installed, France Energies Marines has a multichannel underwater acoustic recorder allowing measurements over a wide frequency band. It thus becomes possible to characterise all ambient noise in the marine environment resulting from physical phenomena and anthropic or biological activities (BENTHOSCOPE and BENTHOSCOPE2 projects). The Institute also has instrumented measurement cages, in particular with hydrophones and acoustic receivers, specially designed for deployment on high-current sites (PHYSIC project).
Photo credit: Barni1 / Pixabay
Understanding and monitoring of ORE impacts on the benthic compartment via a measurement platform dedicated to passive acoustic
Monitoring and study of marine megafauna in wind farms by automatic characterisation
Modelling the role of offshore wind farms in modifying the functioning of coastal food webs and cumulative impact
Advanced design tools for ocean energy systems innovation, development and deployment
Prefiguration of an observatory of marine ecosystems in interaction with floating offshore wind farms in the Gulf of Lion
Offshore wind farm surveys of marine megafauna: standardisation of tools and methods for monitoring at farm scales
Acoustic index of benthic resources on hard substrates for the assessment of impacts from tidal stream turbines
Guide to the environmental impact evaluation of tidal stream technologies at sea
Marine life monitoring
Training in the field of offshore renewable energies