22/06/2022

Postdoc - DEEP MELT (Parameterizing ocean-induced melting under Antarctic ice shelves using deep learning).

This job offer has expired


  • ORGANISATION/COMPANY
    Université Grenoble Alpes
  • RESEARCH FIELD
    Environmental science
  • RESEARCHER PROFILE
    Recognised Researcher (R2)
    Established Researcher (R3)
    Leading Researcher (R4)
  • APPLICATION DEADLINE
    22/07/2022 21:00 - Europe/Athens
  • LOCATION
    France › Grenoble
  • TYPE OF CONTRACT
    Temporary
  • JOB STATUS
    Full-time
  • HOURS PER WEEK
    35
  • OFFER STARTING DATE
    01/10/2022

OFFER DESCRIPTION

Structure Description:

The work will be carried out at the IGE (Institut des Géosciences de l'Environnement) which brings together about 250 people around research on climate, the water cycle, the cryosphere and natural and anthropized environments: http://www.ige-grenoble.fr. The IGE is a joint research unit whose supervisors are the UGA, the CNRS, the IRD, and Grenoble-INP. The IGE is part of the OSUG (Observatoire des Sciences de l'Univers de Grenoble) which is a structure federating 7 research units, 5 associated research teams and 2 service units, under several tutelles.

Team description (N+1 and colleagues):

The work will be performed under the supervision of Nicolas Jourdain (CNRS research fellow), and in collaboration with the CryoDyn (including Fabien Gillet Chaulet and Gaël Durand) and MEOM (including Julien Le Sommer) teams of the IGE. This position is proposed in the framework of the DEEP MELT project (Parameterising ocean-induced melt at the base of Antarctic ice shelves with deep emulators) funded by the IDEX Université Grenoble Alpes, and whose participants are : Bruno RAFFIN (DATAMOVE, Grenoble), Masa Kageyama (IPSL, Paris), David Salas y Mélia (CNRM, Toulouse), and Dan Jones (BAS, Cambridge, UK).

 

Project summary and estimated completion date: The greatest uncertainty in sea level is related to the uncertain fate of the polar ice caps, in part because the ice caps are not yet represented in climate models. The goal of the project is to improve the integration of ice sheet models into climate models by using neural networks to represent the interface between the ocean and the ice shelves. This approach should be able to emulate the fine spatial scales and polar processes that are absent in climate models but essential for the ice sheet model. This project will last until the end of 2023.

Missions / functions performed: The person recruited will first be in charge of a research work aiming at developing a neural network able to learn the links between the Southern Ocean temperature and the melting under the Antarctic ice shelves as simulated in an ocean circulation model coupled to an ice flow model. This will include quantifying the importance of several input variables, and then evaluating the behavior and uncertainty of the neural network under climate change.

Main activities: The person recruited will work daily on the programming and analysis of neural networks, and will monitor the literature on the application of neural networks in the field of oceanography and glaciology. He/she will participate in weekly team meetings and seminars of the IGE, and will attend at least one international conference during the project. The person will organize a visit to the British Antarctic Survey (UK) to initiate a collaboration with their "AI Lab". Finally, he/she will be in charge of writing at least one scientific paper on the research work undertaken.

Result at the end of the mission: The person recruited is expected to provide a neural network configured to be used directly at least at the interface between the IPSL-CM6-LR climate model and the Elmer/Ice-Antarctica cap model.

Evaluation and monitoring of the results: The supervisor N. Jourdain will meet at least every two weeks with the recruited person in order to check that the progress of the work is in line with the project objectives

 

More Information

Additional comments

The successful candidate will be hired as a Research Associate on a fixed-term contract. He/she will ideally start on October 1, 2022 and will be recruited for a period of 1 year.

Offer Requirements

  • REQUIRED EDUCATION LEVEL
    Environmental science: PhD or equivalent

Skills/Qualifications

The person recruited will ideally have expertise in learning methods, in particular neural networks, as well as expertise in polar oceanography or glaciology. A strong technical expertise in Python is expected, especially with the TensorFlow or PyTorch modules as well as Xarray and Dask. A good command of English is required, both written and oral. Applications demonstrating an ability to collaborate, to coach, and to communicate results to diverse audiences will be preferred.

The successful candidate will ideally be constructive and inclusive in their professional interactions. This person is expected to provide occasional assistance within the team, particularly to students and PhD students. Sharing of results, data and tools within the IGE will be strongly encouraged.

 

Specific Requirements

The individual will have a mandatory PhD, ideally in environmental science, fluid dynamics, or data science. Previous experience in academic research will be appreciated.

Work location(s)
1 position(s) available at
Institut des Géosciences de l'environnement
France
Grenoble
38058

EURAXESS offer ID: 802876

Disclaimer:

The responsibility for the jobs published on this website, including the job description, lies entirely with the publishing institutions. The application is handled uniquely by the employer, who is also fully responsible for the recruitment and selection processes.

 

Please contact support@euraxess.org if you wish to download all jobs in XML.