Welcome to AI Chair OceaniX

ANR AI Chair OceaniX (2020-2024) “Physics-Informed AI for Observation-driven Ocean AnalytiX” (short presentation)

Summary.

Covering more than 70% of earth’s surface, the oceans, especially the upper oceans (e.g., the first few hundred meters below the oceans’ surface), play key roles for the regulation of the earth climate (e.g., climate change) as well as for human societies (e.g., marine resources and maritime activities). Despite ever-increasing development of simulation and observation capabilities leading to ocean big data, our ability to understand, reconstruct and forecast upper ocean dynamics and police maritime activities remains limited for key societal challenges (e.g., catastrophe monitoring, global current changes, fishery, energy production, illegal activities). Building upon the cross-fertilization of the cutting-edge expertise of Ifremer and Univ. of Brest in marine science and technology and IMT Atlantique in engineering/data science, OceaniX aims to explore and develop AI-driven strategies and frameworks for the next-generation of self-adaptive multi-platform ocean monitoring and surveillance systems and services with an emphasis on observability and sampling optimality issues for complex dynamics and processes, including extremes and long-term properties. This general objective will rely on bridging model-driven paradigms underlying physical sciences and data-driven learning-based approaches at the core of AI to learn novel computationally-efficient and physically-sound representations of complex dynamical systems. Supported both by institutional and industrial partners, we aim to develop a cutting-edge dual expertise in AI and Ocean science with a particular emphasis on academy-industry synergies and international attractiveness.

Keywords: data-driven representations, complex dynamical systems, deep learning, inverse problems, upper ocean dynamics, maritime activities, spacebone and in situ sensors, multi-platform and multi-source data, geophysical extremes, observability, predictability.

Institutional partnerships: ANR, CNES, ENSTA Br., Ecole Navale, ESA, Ifremer, IMT Atlantique, IRD, ISblue, SHOM

Industrial partnerships: ACRI-ST/ARGANS, CLS, Eodyn, ITE-FEM, Mercator-Ocean, Microsoft, Naval Group, ODL, OceanNext, Scalian

International collaborators: Prof. S. Brunton (Univ. Washington, USA), Prof. S Matwin (Dalhousie Univ., Canada), Prof. A. Mahadevan (WHOI, USA), L. Bertino (NERSC, Norway), F. Doblas-Reyes (BSC, Spain)

Role in the project: Principal Investigator (co-PIs: B. Chapron (Ifremer, LOPS), X. Carton (UBO, LOPS), L. Méméery (CNRS, LEMAR))

Operating budget: 2M€ over 5 years (including 9 PhD scholarships)

News

5. December 2023

PhD defense of Q. Febvre. .

16-20. October 2023

4DVarNet posters presented at the Int. Symp. on Data Assimilation.. Link here.

10-12. October 2023

Invited talk on end-to-end neural data assimilation for ocean reanalyses at the CMEMS workshop on Ocean Reanalyses.. Link here.

22. September 2023

New NeurIPS paper. OceanBench here.

15. September 2023

PhD defense of Simon Benaïchouche. OceanBench here.

12. September 2023

New MWR paper on the estimation of tropical cyclones' features.

12. September 2023

New JAMES paper on 4dvarnet and UQ Paper here.

1. September 2023

New preprint on 4DVarNet for the mapping of real altimetry data Paper here.

12. July 2023

AI4PEX project got funded by EU Link here.

... see all News