The role of Urban topography when simulating mudflows: a case study of Quito, Ecuador

Abstract Urban areas are increasingly exposed to mudflow hazards as unplanned urban expansion intersects with steep drainage basins and extreme rainfall events intensify under climate change. Physics-based models offer powerful tools for hazard assessment, yet their performance in densely built environments remains insufficiently explored. In this study, we evaluate the applicability of the shallow-layer flow simulator Kestrel to simulate mudflows in complex urban settings, focusing on the catastrophic 31 January 2022 event in La Gasca, Quito (Ecuador), which resulted in 28 fatalities and significant property losses. High-resolution elevation models derived from UAV photogrammetry were used to test four topographic scenarios: a bare-earth Digital Terrain Model (DTM), a full Digital Surface Model (DSM), and two partial DSMs reintroducing permanent urban elements (buildings and fences) while excluding transient features and artefacts. Performance was assessed using flow depth and speed statistics, spatial inundation overlap, volume budgets, and field measurements. Urban topography exerts a first-order control on flow routing, depth, and velocity. The DTM substantially overestimates inundation extent, whereas the full DSM underestimates it because transient features obstruct the flow unrealistically. Notably, the mobilized volume is only weakly sensitive to topography, with all scenarios within 10–23% of field-based estimates, whereas the simulated inundation area varies twofold, showing that urban topography controls where the flow goes more than how much material is mobilized. The DSM–fences configuration performs best, with the highest spatial agreement and closest correspondence to measured deposits and velocities. These findings demonstrate that simplified, yet physically meaningful representations of urban topography can significantly improve simulations for urban hazard assessment.