Data-driven and AI-guided multi-platform observing systems for poorly-resolved ocean processes

When:
01/07/2019 – 02/07/2019 all-day
2019-07-01T02:00:00+02:00
2019-07-02T02:00:00+02:00

Annonce en lien avec l’Action/le Réseau : aucun

Laboratoire/Entreprise : Lab-STICC/IMT Atlantique
Durée : 36 mois
Contact : ronan.fablet@telecom-bretagne.eu
Date limite de publication : 2019-07-01

Contexte :
Understanding, modeling, forecasting and reconstructing fine-scale and large-scale processes and their interactions are among the key scientific challenges in ocean-atmosphere science​. State-of-the-art approaches strongly rely on joint research effort in observing systems (e.g., in situ monitoring, satellite observations) and numerical simulations, especially ensemble simulation schemes [e.g., 6-7]. The ability to relate models and observation data, though significant advances in data assimilation, remain open questions for numerous processes (e.g., small-scale parameterization, ocean-atmosphere interactions, biogeochemical ocean dynamics, climate-scale dynamics) [e.g., 1-4]. Artificial Intelligence (AI) technologies, models and strategies open new paradigms to address these questions from the in-depth exploration of the existing observation and simulation big data [4,7-11].

Sujet :
The general goal of this project is to explore and develop these AI paradigms and their interactions with model-based approaches [5] for the design of future multi-platform adaptive ocean observing systems. ​It is widely acknowledged that no single-platform system may provide direct observations of all ocean processes and scales of interest. Sea surface winds, currents and waves are typical examples, for which for instance no in situ or space observing system can alone provide the direct observation of their dynamics at a synoptic scale even for the mesoscale range (i.e., up to horizontal scales of ~ten kilometers).
Synergies between different satellite sensors (e.g., scatterometers, SAR sensors, multi-spectral sensors), in situ networks (e.g., ARGO floats, buoys,…), airborne sensors (e.g., lidar sensors embedded on drones),… are clearly of interest. The rapid development of new embedded communication and processing capacities of such sensors further push for the design of ​context-aware systems for the adaptive and optimized deployment of multi-platform observing systems (e.g., acquisition or streaming of high-resolution satellite data conditionally to pre-analysis steps based on other observation/simulation data, adaptive routing of drone-based acquisitions based on synoptic observation and simulation data). This PhD will investigate the data-driven and AI-guided methods and strategies that we envision to be the processing core of these new systems.

Profil du candidat :
MSc and/or Engineer degree in Applied Math., Data Science and/or Signal Processing with a strong interest in earth science

Formation et compétences requises :
Computer skills, especially Python programming
Skills in deep learning (models and frameworks such as Keras, TensorFlow and/or Pytorch) would be appreciated

Adresse d’emploi :
IMT Atlantique, Technopôle Brest-Iroise, Brest

Document attaché : phd_proposal_rfablet201901.pdf