An efficient multi-scale modelling framework for assessment of coastal flood events

Flood risk in coastal areas is projected to increase with climate change, affecting more coastal communities and their economies. An accurate estimation of risk is essential to reduce the potential impact of extreme flood events. Recent studies have focused on improving the simulation of extreme sea levels at large scales. However, the inundation mapping derived from those water levels is usually done using simple static approaches that can lead to an overestimation of flood extents. This research provides a step forward towards global-scale modeling of coastal inundation due to tropical cyclone events. We are developing a model framework to rapidly and realistically simulate flood hazard in any coastal region. The framework will incorporate three main improvements compared to previous approaches: (1) use of a multi-scale modelling approach that allows for the use of global models and datasets for local flood hazard assessments at high-resolution; (2) move from simple static inundation models towards more physically-based approaches; and (3) move from homogeneous return periods towards event-based modeling. For this purpose, we use the Oceanographic Multipurpose Software Environment which enables model coupling across different spatial scales and physics. Refined local models are nested in the Global Tide and Surge Model for a better representation of the extreme sea levels driven by tropical cyclones. Next, we will combine the water levels with SFINCS and the Synthetic Tropical cyclOne geneRation Model (STORM) to map inundation for a large number of tropical cyclone events.  Here we will present the validation of the framework for three historical events, that is Idai, Haiyan and Katrina.