Merging Multi-Source Data for Shallow Water Model Parameterization

Organisation/Company Aix-Marseille Université Department IUSTI Research Field Geography » Cartography Computer science » Modelling tools Environmental science » Earth science Researcher Profile First Stage Researcher (R1) Positions Postdoc Positions Application Deadline 28 Oct 2026 - 12:00 (Europe/Paris) Country France Type of Contract Temporary Job Status Full-time Offer Starting Date 28 Sep 2026 Is the job funded through the EU Research Framework Programme? Not funded by a EU programme Is the Job related to staff position within a Research Infrastructure? No

Offer Description

Context
The vulnerability of populations to flooding is exacerbated worldwide due to the combined effects of climate change and socio-economic factors (Lehner et al., 2006, Gosset et al., 2023). This vulnerability is more pronounced in the Global South due to a lack of protective measures, monitoring data and forecasting tools to anticipate extreme events.
Cambodia, for example, is regularly affected by significant floods from the Mekong and Tonlé Sap rivers. These floods have both beneficial effects on agriculture by providing water and enriching soils, and detrimental effects when they are exceptional and cause large-scale, long-lasting floods that pose challenges to human, health and food security. In this context, it is crucial to predict and characterize flood risk. Shallow Water 2D numerical models are well-suited for modeling free-surface flows. However, their application to risk management is still hindered by uncertainties, particularly due to the lack of knowledge about the topography of the study area. The floodplains in Cambodia are a prime example. The topography is significantly marked by drainage networks (Preks) set up for agriculture, which are poorly mapped and whose geometries are largely unknown, despite their important role in water propagation during rising and falling water levels.

In urban environments, land use maps provide information about buildings, roads and vegetated areas, but "hydraulically relevant" elements that can divert or store water, such as small walls, ditches, underground parking lots or retention basins, are generally not mapped.

Assigned Mission
The objective is to collect multi-source, multi-format data and integrate it into a geographic information system (GIS) suitable for incorporating relevant information into hydraulic models. The format of this GIS will need to be determined; indeed, while topographic information is typically represented in a raster-based digital elevation model, it would be beneficial to retain vector formats where possible, due to their lighter data management requirements.

The mission will initially involve collecting existing geographic data in "shapefile" format for the modeled areas, whether urban or along the Mekong, and extracting useful information for the hydraulic model.
Given the relatively limited and inadequate information sources, the next objective is to use remote sensing data from Earth observation. Synthetic Aperture Radar (SAR) satellite imagery is best suited to provide spatialized, large-scale information on flooded areas day and night, regardless of cloud cover. Optical images also provide relevant observations for better mapping of land use and numerous objects of interest. The second part of the mission will involve developing and implementing image processing algorithms to produce land use maps and identify drains and hydraulic structures (dikes, etc.) that are not represented in digital elevation models but are necessary for setting up a hydraulic model. The missions and objectives may be adapted based on the skills of the recruited individual.

Where to apply

E-mail carole.delenne@univ-amu.fr

Requirements

Research Field Computer science » Programming Education Level PhD or equivalent

Skills/Qualifications

• Proficiency algorithmic and Python.
• data processing
• Knowledge of geographic data would be a plus

Ideally, the recruited individual will have a Ph.D. in computer science (defended within 3 years max), with proficiency in the Python programming language.

Specific Requirements

PhD obtained less than 3 years ago

Languages ENGLISH Level Good

Languages FRENCH Level Basic

Additional Information

Selection process

The position is funded for 1 year, under the projects SWIFTS and ATLAS, and the recruited individual will be required to report both in writing and orally during project meetings in French and English. The