DRF: Thesis subject SL-DRF-21-0822
Past sea-level markers evidence episods of rapid sea-level drop of several meters in less than 1,000 years. Such glaciation speeds are not well understood and numerical models tend to under estimate them. Within this project, we will develop an ensemble of numerical methodologies and models in order to better reperesent the interactions between polar ice-sheets and climate in the view of understanding the speed of past sea-level drops. In order to do so, we will use a multi model approach to improve the simulation of the snow cover. In particular, we will use an innovatice Deep Learning method to create a snow model emulator with little computing cost. The ensemble of snow models developped will be included in a coupled climate model to study the mechanisms at play during sea-level drop over the Quaternary (last 2.6 Ma).
Modélisation du CLIMat
Start date of the thesis: 01/10/2021
Sciences de l’Environnement d’Île de France
Laboratoire des Sciences du Climat et de l’Environnement
Centre d’Etudes de Saclay, Orme des Merisiers
91190 Gif s/s Yvette CEDEX