Project description
Safeguarding the Coral Sea’s marine mammals, sea turtles and sharks
It is challenging for the EU to safeguard pelagic (open ocean) biodiversity in its overseas countries and territories. Techniques to assess the variety and quantity of pelagic megafauna (marine mammals, sea turtles, sharks) are lacking. The EU-funded MEGAFAUNA project will introduce a network that protects megafauna and keeps fisheries profitable in New Caledonia’s Natural Park of the Coral Sea. The network will inform the marine park’s decision-makers while helping to meet the International Union for Conservation of Nature’s target of protecting 30 % of the world's oceans by 2030. The project will use image-based surveys, deep learning and fishery economic data to detect pelagic megafauna, and predict and protect hotspots in the Coral Sea.
Objective
The EU faces the challenge to protect pelagic biodiversity over immense Overseas Countries and Territories. However, cost-efficient, safe and reproducible methods are still missing for assessing the diversity and abundance of pelagic megafauna (marine mammals, sea turtles and sharks). We will adapt image-based surveys and deep learning—a new method based on artificial intelligence—to pelagic megafauna. A Coral Sea Nature Park (CSNP) was recently created in New Caledonia Overseas Territory but this area still lacks reserves in pelagic ecosystems. We will propose a reserve network that optimises the trade-off between critical habitat protection of pelagic megafauna and fisheries economic profitability in this area. This network will directly advise decision makers of the CSNP and help fulfil IUCN target of 30% of oceans protected by reserves by 2030. Our objectives are: RO1—Detect pelagic megafauna, RO2—Predict pelagic megafauna hotspots, and RO3—Protect hotspots in the Coral Sea. To reach RO1, image-based surveys will be conducted and megafauna will be automatically detected on images using deep learning. To reach RO2, habitat models will be used to predict megafauna diversity and abundance as a function of environmental variables. To reach RO3, fisheries economic data will be incorporated to propose a reserve network optimising both conservation and economic needs. This project will be jointly hosted by the Marine Biodiversity Exploitation and Conservation Laboratory (MARBEC) and the Montpellier Laboratory of Informatics, Robotics, and Microelectronics (LIRMM), both part of the University of Montpellier. Supervised by D. Mouillot (MARBEC) and M. Chaumont (LIRMM), I will receive the best appropriate interdisciplinary training. After an enriching research experience in the U.S I aim to integrate a long-term research position in Europe. This fellowship would be a superb opportunity to develop an independent and innovative research project in marine numerical ecology.
Fields of science
- natural sciencesbiological sciencesmarine biology
- natural sciencesbiological scienceszoologymammalogycetology
- natural sciencesbiological sciencesecologyecosystems
- natural sciencesphysical scienceselectromagnetism and electronicsmicroelectronics
- natural sciencescomputer and information sciencesartificial intelligencemachine learningdeep learning
Programme(s)
Funding Scheme
MSCA-IF-EF-RI - RI – Reintegration panelCoordinator
34090 Montpellier
France