FLOTTE: understanding urban flooding

Researchers from Inria at the Inria Antenna at the University of Montpellier and the Inria centre at the University Grenoble Alpes, the Pontificia Universidad Católica de Chile (UC) and the Centro de Gestión Integral del Riesgo de Desastres (CIGIDEN) collaborate in this Associate Team.

We spoke with its leaders, Antoine Rousseau, lead researcher at the LEMON project-team of the Antenne Inria of the Université de Montpellier in France, and Cristián Escauriaza, researcher at the Department of Hydraulic and Environmental Engineering (DIHA) of the UC Faculty of Engineering, to learn more about their work and progress.

Regarding the origin of FLOTTE, how did you first start collaborating?

Antoine Rousseau (AR): The LEMON project-team and UC have a long collaboration since my visit to Inria Chile during 2016. Together with Rodrigo Cienfuegos, director of CIGIDEN, we have been working on coupling algorithms for coastal ocean modeling, and we have also built the TsunamiLab numerical platform, which has been installed in Santiago de Chile and at the Terra Numerica Museum in Sophia Antipolis, in France.

Cristián Escauriaza (CE): Long-term collaborations with the LEMON team, on issues related to renewable energy and natural hazards.

What are the specific areas of research for each of you? What scientific questions are you seeking to answer or have already answered with FLOTTE?

AR: At LEMON, we are specialists in multi-scale urban flood models. In particular, we developed a numerical platform called SW2D-LEMON, in which innovative shallow water type models, which we call "porosity models", are implemented. The main objective of these models is to be able to compute shallow water flows at low resolution (fast computation), taking into account the variation of the terrain below the grid.

CE: As part of our research, we have conducted field observations and developed numerical models to understand the main factors controlling flood risk in mountain rivers in mediterranean and semi-arid environments. These systems are characterized by flash floods with high sediment concentrations. We are interested in advancing this knowledge and developing predictive tools to provide better estimates of potential hazards in urban areas.

AR: The objective is to include debris and sediment transport in SW2D-LEMON, and also to evaluate the impact of jam formation on the flow itself. Several directions are identified:

• Coupling of transport and shallow water models: large time step strategies should be considered to account for varying propagation velocities.
• Transport model extension: development of an extended model for modeling water-sediment mixtures.
• Sensitivity analysis: linking the required accuracy of debris characterization to that of flow/risk expected by operational stakeholders.

CE: I would also add the educational objective of preparing new researchers at the graduate level in these resilience-related topics.

Pablo Rogat/Extend/Shutterstock

AR: In general, these models can be applied in river hydraulics.

CE: Assess flood hazards in urban areas prone to flash flooding.

How does the work between the French and Chilean teams complement each other?

AR: Both teams are experts in numerical hydraulic modeling, this is our common language. The Associate Team is a good opportunity to collaborate around improved porosity models (LEMON experience) and sediment transport (UC experience).

Any other thoughts you'd like to share?

AR: If you have specific data on sediment transport (by the Mapocho in Santiago or any other river), do not hesitate to contact us!