Postdoctoral researcher in oceanography / data science – Institute for Environmental Geosciences
The Institute for Environmental Geosciences (IGE) is looking for a post-doctoral researcher to develop inverse methods for processing observational data in oceanography. He/she will particularly focus on the applications of machine learning methods in the context of data assimilation in oceanography.
The work envisaged is part of the ANR MeLODy project (https://anr-melody.github.io) and will be carried out in collaboration between the MEOM group at IGE (Grenoble) and the Lab-STICC (Brest). It will focus on the development of new approaches for the exploitation of oceanographic data, at the interface between data assimilation and machine learning. The context of application could be that of ocean forecasting systems (Copernicus Marine Services) or the processing of space observation data (in particular satellite altimetry). The experimental framework will use Observation System Simulation Experiments based on numerical simulations performed by the MEOM group. The selected candidate will work in close collaboration with Julien Le Sommer, Emmanuel Cosme, Pierre Brasseur and Ronan Fablet.
The expected candidates will hold a PhD in oceanography, meteorology or data science. Their research background should demonstrate their competence or interest in the fields of physical oceanography, data assimilation and machine learning. They will master the implementation of the main software libraries in machine learning (in particular PyTorch and TensorFlow). They should speak and write English fluently and be able to interact in a multicultural environment. They will need to demonstrate curiosity, autonomy and initiative.
The recruited researcher will work at the Institute of Environmental Geosciences (IGE) within the MEOM group in the framework of the ANR MeLODy project. The IGE (http://www.ige-grenoble.fr) is a joint research unit bringing together staff from the UGA, CNRS, Grenoble INP and IRD who are interested in the study of the climate and the anthropisation of our planet. The MEOM group (http://meom-group.github.io) brings together 25 researchers, engineers, students and postdocs in the field computational oceanography. The MeLODy project (2019-2023, PI: R. Fablet) “Bridging geophysics and MachinE Learning for the modeling, simulation and reconstruction of Ocean DYnamics” (https://anr-melody.github.io) is a research project supported by the French National Research Agency (ANR) which aims to accelerate the development of data-driven approaches in physical oceanography, particularly in the context of the reconstruction of upper ocean dynamics.