Advancing flood hazard modelling with the Surface Water and Ocean Topography satellite mission

Principle Investigator: Jeffrey Neal (Institution)

Co-Investigator(s): Paul Bates, Stephen Chuter, James Savage, Izzy Probyn, Guy Schumann, Paolo Tamagnone, Paul Bell

---

This project will use SWOT data to support a step change in the accuracy of the models that underpin flood hazard and risk analytics globally across the Insurance, Financial services, Engineering, ESG, Government and Humanitarian sectors.

For disaster risk reduction to be effective, an understanding of where and what might be exposed to flooding now and in the future is needed in the form of flood hazard mapping. It is widely acknowledged that hydrodynamic models are the most accurate method for simulating flooding processes and mapping flood hazard. Over the last decade hydrodynamic modelling methods have emerged that are primarily built from earth observation data, and are therefore applicable globally. However, although these models are useful, flood depth errors of several meters and flood extent errors in the kilometers are not uncommon. This can have substantial implications for risk estimates and prevents the data being used for community level planning and asset risk management because local scale errors are simply too high. It is well established that calibrating these models with river water surface profiles, via the estimation of river bathymetry (e.g. depth), can demonstrably improve their performance when simulating extremes such as floods. To date, such data have never been available globally or more specifically from an earth observation platform, however with SWOT this will now change.

The basic principles around how SWOT data could be used to do this, given various degrees of uncertainty in river discharge, are well established in the scientific literature, however the data have never previously been available to build a prototype flood model or validate its outputs. Questions remain over the best way to estimate river bathymetry in a practical and scalable manner and how well a SWOT-conditioned flood model will perform in relation to traditional terrestrial methods. Therefore, this project aims to develop a proof of concept and validate each stage of the modelling process. The step change in data accuracy will catalyze new applications for flood hazard data.