En poursuivant votre navigation sur ce site, vous acceptez le dépôt de cookies dans votre navigateur. (En savoir plus)
Portail > Offres > Offre UMR5243-HELOUR-029 - Postdoctorant.e: Développement de l'apprentissage automatique supervisé pour simuler l'évolution de l'anisotropie dans le manteau terrestre (ERC RhEoVOLUTION) H/F

Postdoctoral fellow: Development of supervised machine learning tools to simulate the evolution of anisotropy in the Earth's mantle (ERC RhEoVOLUTION) M/W

This offer is available in the following languages:
Français - Anglais

Date Limite Candidature : mercredi 12 octobre 2022

Assurez-vous que votre profil candidat soit correctement renseigné avant de postuler. Les informations de votre profil complètent celles associées à chaque candidature. Afin d’augmenter votre visibilité sur notre Portail Emploi et ainsi permettre aux recruteurs de consulter votre profil candidat, vous avez la possibilité de déposer votre CV dans notre CVThèque en un clic !

General information

Reference : UMR5243-HELOUR-029
Workplace : MONTPELLIER
Date of publication : Wednesday, September 21, 2022
Type of Contract : FTC Scientist
Contract Period : 12 months
Expected date of employment : 1 November 2022
Proportion of work : Full time
Remuneration : Between 2 690 and 3 821 € gross per month
Desired level of education : PhD
Experience required : 1 to 4 years

Missions

Develop an effective supervised machine-learning approach for accelerating by orders of magnitude the prediction of the evolution of mechanical anisotropy (elastic and viscoplastic) associated with crystal preferred orientations in geodynamic flows (plate tectonics and mantle convection). This postdoctoral fellowship is part of the ERC RhEoVOLUTION project, which aims to develop new approaches to simulate in a self-consistent way the localization of deformation at different scales in the Earth.

Activities

The aim of the postdoctoral position will be to:
- acquire a good understanding of the processes controlling the evolution of mechanical anisotropy during viscoplastic deformation in rocks (and also ice and metals) and master the existing methods to simulate this evolution – the senior researchers in the group are experts on the subject;
- define the supervised machine learning (AI) approaches best suited to the problem and the best strategies for the creation and formatting of the training databases, focusing on the evolution of mechanical anisotropy in the Earth's mantle;
- create the databases using codes already available in the RhEoVOLUTION team to simulate the evolution of crystallographic orientations and elastic and viscoplastic anisotropies of polycrystalline aggregates;
- train supervised machine learning (AI) codes;
- implement the AI codes for simulating the evolution of viscous anisotropy as subroutines within geodynamic simulation codes, which will be used for simulating the deformation of the solid Earth in collaboration with the other researchers of the ERC RhEoVOLUTION.
The extension of this approach to the simulation of the effect of mechanical anisotropy on glacial flows is envisaged in collaboration with researchers from the ERC RhEoVOLUTION at the Institut des Géosciences de l'Environnement in Grenoble. Applications in materials science are also possible, as similar processes take place during metal forming.

Skills

We are looking for highly motivated candidates with strong numerical and methodological skills. At the time of appointment, the candidates must have a PhD degree in geophysics, geology, physics, mechanics, applied mathematics, or a closely related field, a strong expertise in numerical modelling and programming, and excellent English skills. Experience in AI, in particular in machine-learning approaches applied to the resolution of regression problems, as well as knowledge in solid mechanics, in particular in the domain of crystal plasticity will be highly valued.

Work Context

This postdoctoral project is part of the ERC RhEoVOLUTION project, how the evolution of rock rheology controls strain localization at different scales in the Earth. To do so, we will develop a framework for modeling self-consistently strain localization in rocks deforming by ductile processes. We will design: (1) stochastic descriptions of the evolution of the rheology in time and space and (2) fast (supervised machine-learning) methods to calculate the evolution of its anisotropy, which we will incorporate in geodynamical and ice-flow models. The ERC RhEoVOLUTION project team is composed of researchers in Earth Sciences, Glaciology, Materials Sciences and Applied Mathematics working in Montpellier, Grenoble, Nice and Argentina.
The post-doctoral fellow will be based at Géosciences Montpellier, a joint CNRS & University of Montpellier research unit (UMR 5243), attached to the Mediterranean Research Observatory of the Environment (OREME) - http://www.gm.univ-montp2.fr. Research at Géosciences Montpellier focuses on the study of the physical and chemical processes controlling the evolution of planet Earth, through the observation of natural objects, experimentation and the modeling of complex associated processes. At Géosciences Montpellier, the post-doc will integrate the theme Deformation - one of the 6 major scientific themes of the unit for the 2021-2026 period, and will be part of the Mantle & Interfaces team which includes ~30 researchers, post-docs and doctoral students.

Additional Information

Project web site RhEoVOLUTION - https://erc-rheovolution.gm.univ-montp2.fr/en/rheovolution/

We talk about it on Twitter!