Evaluation of the Hydrodynamic Forces on an Isolated Building Induced by a Dam-Break Flow

Abstract Flood-induced forces on buildings are a key concern for designing resilient infrastructure in flood-prone areas. Standard guidelines in several countries provide simplified approaches to account for different types of loads, such as hydrostatic pressure, impact forces from flood waves or tsunamis, and drag forces. Each type of load depends on specific parameters usually related to the water depth and flow velocity. This study assesses the reliability of various strategies for estimating the forces exerted by floodwaters on buildings, using output from two-dimensional numerical simulations. A review of existing formulations for impact and drag forces was first conducted. In parallel, two computational approaches were investigated: the drag-based method, which relates the force to local momentum flux through empirical coefficients, and the physically based method, which integrates hydrostatic and hydrodynamic pressure contributions obtained from numerical simulations along the building walls. To achieve this, dam-break flow experiments impacting an isolated obstacle were conducted and pressure sensors were used to measure the resulting force. The same configurations were reproduced numerically with a finite-volume shallow-water solver. Simulation outputs were then used to compute forces through the two computational approaches and relevant formulations from the literature, with all results being compared to experimental measurements. Particular attention was paid to mesh resolution: while the physically based method requires a detailed grid explicitly representing the building geometry, the drag-based method can be applied on coarse meshes without geometric representation of the obstacle. The findings highlight that drag-based formulations, when properly calibrated, can provide accurate predictions at low computational cost. Conversely, physically based methods offer a detailed representation of the impact and quasi-steady phases but at the expense of significantly higher computational effort. These results provide new insights into the applicability of existing empirical formulas and guide the choice of modeling strategies for flood risk assessment and structural design.