Missions

The candidate will be responsible for implementing bathymetry calculation methodologies within processing chains to obtain a "hydrocompatible" digital terrain model, i.e. one that enables accurate hydrodynamic simulation. Most of these methodologies have been the subject of scientific publications and/or specific scripts.

The work will be carried out in collaboration with researchers from CECI and INRAE, as well as engineers from CNES and those responsible for developing algorithms dedicated to floodplains.

Activities

To carry out this main task, the candidate will be required to:
1 – Review the current state of the art in bathymetry inversion methods and identify those that are most suited to the final objective. These methods are largely inspired by work carried out for the SWOT mission (DAWG group).
2 – Modify the architecture of existing workflows to integrate these methods. In particular, multi-mission data management will need to be incorporated.
3 – Develop scripts for calculating inversion methods based on existing scripts (SICFLOW, CEPHEE and hydro-MNT). The scripts will be written in Python.
4 – Deploy and validate the various approaches in two or three test areas where field measurements are available (Garonne, Ohio, Severn).
5 – Evaluate the methods by numerically simulating flood events in these areas using Telemac 2D software.

Skills

Master's degree, engineer
Knowledge in the field of hydraulics, hydrology or Python programming with geospatial data
Space remote sensing: radar, image processing

Work Context

The CECI Natural Hazard team is composed of senior researchers and early-career researchers, and is supported by a team of highly qualified research engineers with extensive expertise in climate and environmental modelling, high-performance computing (HPC), simulation workflows and data management. We conduct cutting-edge research on climate variability and climate prediction, oceanography and polar sciences, air-sea interaction, detection and attribution of climate change and its impacts, extreme events such as droughts, and environmental risks. We use a wide range of numerical models, from large-scale simulations to global Earth system models, as well as associated algorithms (data assimilation, uncertainty quantification, machine learning) to address our scientific challenges.
The CECI hydrology team is working with CNES on the use of spatial data to improve digital flood simulation. Spatial data has the advantage of providing a two-dimensional representation of floodplains as well as information that is independent of ground surveys. The type of data that can be used is either RADAR and altimetry measurements provided by the SWOT satellite or high-resolution optical data from the Pléiades satellite (and soon CO3D). Two previous CECI projects, each focusing on one type of data, have shown the limitations of each approach but also their complementarity. Thus, in order to correctly simulate flood flows and build a digital twin of flooded areas, it seems necessary to combine multi-mission data based on expert analysis of hydraulic equations and data collection. More specifically, combining data sources and their processing algorithms will be useful for reconstructing river bathymetry capable of defining overflow flows and thus improving flood warning systems and socio-economic damage estimates.
The work will therefore focus on detecting minor riverbeds and possible banks using optical imaging, then adding equivalent bathymetry derived from hydraulic model inversion and free surface elevation measurements. To do this, we will use the results obtained in the TOSCA-Hydros project by INRAE, as well as those from the CNES/CERFACS hydro-mnt project. The ultimate goal is to obtain a digital terrain model (DTM) that is both accurate and measurable using only remote sensing data.