Missions

The engineer recruited will explore the potential of AI methods for processing spatial altimetry data in the coastal ocean band. Altimetry measurements, which have been used for over 30 years to observe sea levels and currents in the open ocean, are crucial for understanding and monitoring ocean dynamics and are now integrated into many operational oceanography applications. However, as they approach the coast, they undergo a number of disturbances that make them much more complex to interpret and process, generating errors and ultimately excluding them from conventional altimetry products. Given the vulnerability of coastal areas to climate change, these data, which are quasi-global and continuous for more than 30 years, are nevertheless indispensable. For more than 15 years, the LEGOS laboratory (https://www.legos.omp.eu/) has developed recognised expertise in the field of coastal altimetry processing (X-TRACK software) and now calculates and distributes several demonstration products (https://www.aviso. altimetry.fr/en/data/products/sea-surface-height-products/regional/x-track-sla.html; https://www.seanoe.org/data/00631/74354/).
In the processing of these innovative products, the editing phase remains critical and is based on statistical criteria that are known to be suboptimal for ensuring maximum data quality. This is particularly true for data from the most recent space missions, which are theoretically more effective in the coastal ocean environment. The engineer will initially be responsible for exploring AI methods suitable for editing coastal altimetry data and evaluating their performance compared to the statistical method currently implemented in the X-TRACK processing chain. The implemented model will then be used to better understand the factors that determine the quality of coastal altimetry data. These studies will be carried out using data from different generations of altimetry missions and in different coastal environments, with a view to ultimately preparing for operational implementation in the processing of long series (more than 30 years) on a global scale. The engineer will also be responsible for documenting the codes and analyses developed. This work will be carried out within the CTOH National Observation Service (Centre for Ocean and Hydrosphere Topography, https://www.legos.omp.eu/ctoh/fr/accueil/).

Activités

- Build a multi-parameter database
- Implement and test AI models for editing coastal altimetry data
- Compute performance diagnostics on the tested AI models
- Document, validate and disseminate the results and codes within the working group.

Compétences

- Scientific computing, Python language (scikit, numpy and pytorch or equivalent),
- UNIX/LINUX environment
- Knowledge of AI methods, data science
- Knowledge of space altimetry would be an advantage
- Document drafting
- Interpersonal skills: Independence, rigour, enjoyment of teamwork and communication

Contexte de travail

The Laboratory for Space Studies in Geophysics and Oceanography (LEGOS) is a joint research unit of UPS-UT3, CNES, CNRS-INSU and IRD, with a staff of 130. LEGOS conducts multidisciplinary research on the global environmental system, drawing on ocean physics, marine geochemistry and biogeochemistry, hydrology and glaciology. LEGOS is heavily involved in university education with the Université Paul Sabatier -UPS, Toulouse III and research-based training in France and internationally. At LEGOS, the position to be filled will be within the CTOH, a French observation service dedicated to satellite altimetry studies (Centre de Topographie des Océans et de l'Hydrosphère). The CTOH's objective is to develop new altimetry products and applications for oceanographic and hydrological applications, and then to assist in their transfer to operational services.