Description du projet
Des modèles optimisés pour simuler la glace de l’océan Arctique en été
La glace de l’océan Arctique diminuant rapidement en raison du changement climatique, la demande de prévisions précises et opportunes concernant les glaces se fait plus pressante. Divers problèmes liés au traitement des données limitent toutefois la disponibilité des observations estivales, ce qui crée un important déficit d’observation pour la recherche polaire. Financé par le Conseil européen de la recherche, le projet SI3D entend combler ce déficit par le biais de l’apprentissage automatique profond et de la modélisation de la réponse des altimètres radar. En intégrant les données satellitaires de plusieurs missions de l’ESA, SI3D se propose de créer un enregistrement de haute précision de l’épaisseur de la glace de mer sur une période de 15 ans, incluant les mois d’été. Ce jeu de données unique permettra de mieux comprendre les mécanismes de fonte saisonnière et d’améliorer les prévisions concernant la glace de mer. Les résultats du projet pourraient contribuer à transformer la science du système arctique, en améliorant la modélisation, les prévisions, le bilan de masse et les études biogéochimiques pendant les mois critiques de la fonte estivale.
Objectif
Arctic sea ice is diminishing with climate warming at a rate unmatched for 1000 years. As the receding ice pack changes rapidly and becomes increasingly mobile, the demand from academic and commercial stakeholders for accurate and timely sea ice forecasts is intensifying. Forecasting accuracy is enhanced with the assimilation of sea ice thickness observations from satellite altimetry. However, these data are currently unavailable during summer when they would be most valuable for stakeholders, owing to significant data processing challenges. This has been identified as a key observation gap for polar research by the IPCC.
SI/3D will address this gap, harnessing deep machine learning, modelling of the radar altimeter response, and dedicated field campaigns, to overcome the processing barriers. I will integrate satellite data from multiple ESA, EU-Copernicus, and NASA altimetry missions to produce the first 15+ year high-accuracy record of pan-Arctic sea ice thickness without interruptions in the summer. With this unique dataset, I can achieve the following goals. (1) To close the Arctic sea ice volume budget, pinpointing the mechanisms driving seasonal decay and breakup of the ice pack and the feedbacks of sea ice loss on Arctic temperatures. (2) To upgrade seasonal sea ice forecasts from state-of-the-art modelling systems by assimilating summer ice thickness observations.
SI/3D will create a new discipline of sea ice research by using altimetry to study the Arctic summer, which is risky. Having led the first published pilot research in this field, however, I am ideally placed to carry out the project and will leverage the expertise I have gained at three international Arctic research institutions to make it happen. On completion, this work will transform opportunities for Arctic system science, improving sea ice modelling, forecasting on timescales from weeks to years, mass budgeting, and biogeochemistry studies during the critical summer melting months.
Champ scientifique
- engineering and technologymechanical engineeringvehicle engineeringaerospace engineeringsatellite technology
- natural sciencesearth and related environmental sciencesphysical geographyglaciology
- natural sciencescomputer and information sciencesartificial intelligencemachine learning
- natural sciencescomputer and information sciencesdata sciencedata processing
- natural sciencescomputer and information sciencessoftwaresoftware applicationssimulation software
Mots‑clés
Programme(s)
- HORIZON.1.1 - European Research Council (ERC) Main Programme
Thème(s)
Régime de financement
HORIZON-ERC - HORIZON ERC GrantsInstitution d’accueil
9019 Tromso
Norvège