Offer Description
– work environment: The hosting team is a world-leading group in satellite hydrology, with recognized expertise on SWOT and on discharge and bathymetry estimation from SWOT and multi-mission altimetry. Permanent members have been part of the SWOT Science Team since 2016 and contributed to mission preparation and product design, in particular to discharge estimation algorithms; the Project PI currently serves as SWOT Hydrology Lead. This sustained investment has produced four key assets that directly enable the present project: (i) robust discharge retrieval methods operational at global scale and integrated into the official mission product ; (ii) river-bathymetry approximation from SWOT observations and/or digital elevation models and is applicable worldwide ; (iii) SWOT processing and filtering tools for river products ; (iv) a spatio-temporal densification framework that increases the effective spatial–temporal sampling of satellite observations. The team’s broader skill set covers hydraulic modelling, data assimilation, large-scale data processing, and statistical analysis, and is supported by established international collaborations with CNES and NASA/JPL (SWOT processing and densification), the University of Stuttgart together with the DAHITI and Hydroweb teams (altimetry time series), USGS and GRDC (gauge archives), and Météo-France (meteorological forcing).
– main mission: Use satellite data (mainly SWOT plus other altimetry missions) to measure how major climate patterns (El Niño/La Niña, Indian Ocean Dipole, North Atlantic Oscillation, Madden–Julian Oscillation) affect rivers. Produce improved and recovered long-term records and maps of river water level and flow, explain the impact of these climate modes on river basins, and turn the results into simple flood and low-flow risk information that could be useful for water managers.
– activities: The main steps of the project involve:
– Process SWOT hydrology data (water-surface elevation, width, slope) with rigorous quality checks and apply our multi-temporal densification method to reduce gaps and improve coverage.
– Estimate river discharge for the selected basins using the SIC4DVAR algorithm, and validate results against available in-situ measurements.
– Build deseasonalized anomalies of WSE and discharge, define appropriate seasonal windows and response lags per basin, and fit statistical models that include multiple climate indices (ENSO, IOD, NAO, MJO) and meteorological covariates (rainfall, temperature, soil moisture) to isolate each mode’s partial effect. Investigate Deep learning approaches.
– Produce pre-SWOT reanalysis of WSE, width, and slope from long altimetry time series using the SWOT dynamics catalogue together with the densification approach; derive reanalysis discharge from these SWOT-like profiles with SIC4DVAR, and validate against historical gauge records.
– Analyze extremes (floods and low flows) using methods that allow risk to vary over time and with climate phase; generate dominant-driver maps and climate-conditioned hazard metrics for decision support.
– Disseminate results, including at least one peer-reviewed paper, a short technical note/user guide, and well-documented, reproducible code and datasets.
To achieve this, the candidate will use and enhance the in-house tools developed for SWOT studies. This includes maintaining and extending the pre-processing chain; applying and refining the SIC4DVAR discharge algorithm; and adapting it into a reanalysis version that efficiently ingests long-term, multi-mission altimetry time series. The candidate will implement statistical methods for attribution and extremes and build workflows for large datasets. They will also produce basin-to-global visualizations using GIS (e.g., QGIS) and collaborating with mission partners where needed.
Where to apply
E-mail: hind.oubanas@inrae.fr
Skills/Qualifications
– PhD in environmental sciences (hydrology or atmospheric sciences) with a strong grounding in environmental systems.
– Comfortable with statistical time-series analysis (multi-predictor models, basic extreme-value methods) and uncertainty quantification.
– Strong Python skills (NumPy, pandas, xarray, statsmodels, matplotlib; PyTorch/JAX a plus), Git, and reproducible environments (e.g., Docker).
– Experience with satellite altimetry (SWOT, Sentinel-3/Jason; Hydroweb/DAHITI) is desirable but not required.
– Basic GIS skills (QGIS).
– Strong scientific writing and communication in English (French is a plus) and ability to work effectively in an international team (e.g., CNES/NASA/JPL, universities, data centers).
