Enhanced Kinematic Hierarchical Filling-&-Spilling Algorithms for Pluvial Flooding: Synthetic and Real-Case Applications with Comparative Analysis to Fully 2D models

Pluvial floods, intensified by short-duration and high-intensity storms, are becoming increasingly frequent and severe due to climate change and urbanization. SaferPlaces addresses this with a digital twin platform that integrates high-resolution geospatial and climate data from sources such as Google Earth Engine (GEE), Open Street Map (OSM), Microsoft Planetary, Amazon, and Copernicus. These datasets are automatically integrated to construct detailed, multi-layered urban digital twins, enabling real-time flood hazard and risk modelling. As part of the SaferPlaces platform, Safer_RAIN, a fast-processing Hierarchical Filling-&-Spilling Algorithm (HFSA), combines spatially distributed rainfall input and infiltration simulation through a pixel-based Green-Ampt model (see https://saferplaces.co/) and enabling building-by-building flood risk modeling across large urban areas. Leveraging the platform’s cloud-based infrastructure, Safer_RAIN can efficiently run computationally intensive simulations at high resolution, delivering real-time results that support effective urban planning and climate resilience strategies. Comparisons with traditional 2D hydrodynamic models revealed limitations in Safer_RAIN, such as underestimation of flooded areas due to the lack of hydraulic backwater effect and single flow path constraints. We present significant enhancements of the SaferPlaces platform that were recently developed to address these challenges. These include: (1) incorporating a weir equation using a simplified kinematic approach to account for backwater effect, (2) introducing a travel-time distribution for water within watersheds and (3) implementing flow path flood extension using a Height Above Nearest Drainage (HAND) approach. These improvements are tested in a synthetic case study and applied to a set of real flooding events in urban areas of Pamplona, Spain. With its ability to run scalable simulations in real time and integrate diverse datasets, Safer_RAIN, as part of SaferPlaces' digital twin platform, offers a transformative solution for flood risk intelligence, empowering cities to build preparedness and enhance climate resilience.

Keyword: Digital twin, Hierarchical Filling-&-Spilling Algorithms, 2D hydrodynamic models, pluvial flooding