Supervision : Dominique BÉRÉZIAT

Co-supervision : CHARANTONIS Anastase

Downscaling of oceans fields by fusion of heterogeneous observations using Deep Learning Algorithms

This Ph.D. thesis focuses on reconstructing satellite images of the ocean surface from sparse and noisy measurements. Our objective is the reconstruction of the Sea Surface Height (SSH), an important variable used to estimate surface currents. It is retrieved through nadir-pointing altimeters, leaving important observation gaps due to their remote sensing technology. Complete SSH maps are produced using linear Optimal Interpolations with low effective resolution. On the other hand, Sea Surface Temperature (SST) products have much higher data coverage, and SST is physically linked to geostrophic currents through advection.
This thesis explores deep learning algorithms to estimate SSH fields. Relying on years of data from the simulation and observations, deep neural networks are able to learn complex relationships between SSH and SST variables. Using these algorithms and SST observations, we first enhance SSH mapping from a downscaling perspective on a physical simulation. Then, we tackle the SSH interpolation problem on simulation and observation data, with a particular focus on how to transfer the learning in operational settings. Finally, we adapt our method to produce near real time and forecast estimations.

2022-2024 Publications

• 2024
  • Th. Archambault : “Downscaling of oceans fields by fusion of heterogeneous observations using Deep Learning Algorithms”, thesis, phd defence 10/04/2024, supervision Béréziat, Dominique, co-supervision : Charantonis, Anastase (2024)
  • P. Garcia, Th. Archambault, H. Bull, A. Charantonis, D. Béréziat : “Transfer-learning and data fusion for forecasting ocean surface currents”, RFIAP 2024, Lille, France (2024)
  • Th. Archambault, A. Filoche, A. Charantonis, D. Béréziat : “Pre-training and Fine-tuning Attention Based Encoder Decoder Improves Sea Surface Height Multi-variate Inpainting”, VISAPP 2024 - 19^th International Conference on Computer Vision Theory and Applications, Roma, Italy (2024)
  • Th. Archambault, P. Garcia, A. Charantonis, D. Béréziat : “Deep Sea Surface Height Multivariate Interpolation”, European Geosciences Union General Assembly 2024 (EGU24), Vienne, Austria (2024)
  • Th. Archambault, A. Filoche, A. Charantonis, D. Béréziat, S. Thiria : “Learning Sea Surface Height Interpolation from Multi-variate Simulated Satellite Observations”, James, Journal of Advancing in Modeling Earth Syst, vol. 16 (6), pp. e2023ms004047, (AGU) (2024)
• 2023
  • S. Thiria, Ch. Sorror, Th. Archambault, A. Charantonis, D. Béréziat, C. Mejia, J.‑M. Molines, M. Crépon : “Downscaling of ocean fields by fusion of heterogeneous observations using Deep Learning algorithms”, Ocean Modelling, vol. 182, pp. 102174, (Elsevier) (2023)
  • Th. Archambault, A. Filoche, A. Charantonis, D. Béréziat : “Multimodal Unsupervised Spatio-Temporal Interpolation of satellite ocean altimetry maps”, Proceedings of the 18^th International Conference on Computer Vision Theory and Applications, Lisboa, Portugal (2023)
• 2022
  • A. Filoche, Th. Archambault, A. Charantonis, D. Béréziat : “Statistics-free interpolation of ocean observations with deep spatio-temporal prior”, ECML/PKDD Workshop on Machine Learning for Earth Observation and Prediction (MACLEAN), Grenoble, France (2022)
  • Th. Archambault, A. Charantonis, D. Béréziat, C. Mejia, S. Thiria : “SSH Super-Resolution using high resolution SST with a Subpixel Convolutional Residual Network”, Climate Informatics, Asheville, NC, United States (2022)
  • Th. Archambault, A. Filoche, A. Charantonis, D. Béréziat : “Unlearned Downscaling of sea surface height with Deep Image Prior”, IA for Earth Sciences Workshop The International Conference on Learning Representations (ICLR), Virtual conference, United States (2022)