Résumé
Although the impact of floodwaters on buildings is determined by a series of physical mechanisms that, when triggered, result in a wide range of physical damage to buildings and monetary losses, existing research on the topic appears polarized. The majority of work focuses either on the physical mechanisms that transform contact with floodwaters into damage, omitting the estimation of economic consequences, or on improving the estimation of direct damage, but neglecting the mechanisms mentioned above.In the work we present here, we seek to reconcile these two approaches to investigate the potential implications of both. We begin by considering the exchange of floodwater through the exterior and interior openings of a building to assess direct material damage and estimate the monetary value of the loss. To do this, we build a building-level simulation model by coupling a hydraulic model and an economic model. Our hydraulic model simulates the flow exchanges and the flood water height between the interior and exterior of the building and between the rooms inside the building. The economic model is based on a decomposition of the building into elementary components (wall material, cladding, insulation, floor, furniture, etc.), identifying the room and the height at which they are located. For each elementary component, the damage is modeled for all combinations of water depth and duration of flooding. We use our coupled model to simulate four different scenarios on exemplary buildings. The first scenario corresponds to the classical methodology, with homogeneous water depths outside and inside the building. The other three represent: i) street-to-building exchanges in an open plan building; ii) street-to-building/interior exchanges with closed openings; and iii) street-to-building/interior exchanges with open openings. Designating the conventional method as the baseline, we adopt a comparative approach based on monetary valuation of flood damage costs to determine the magnitude of bias of each alternative scenario. Given the high elasticity that flood damage functions exhibit for shallow water depths, the magnitude of the bias in the monetary flood cost assessment for properties may be non-negligible.The estimates of monetary costs show heterogeneous results. The magnitude of the bias with respect to the reference situation varies from rather modest to non-negligible. This heterogeneity is related to multiple sources: duration and depth of flooding, condition of openings, location of rooms in relation to the entry of floodwater into the building. These results encourage us to carry out additional analyses with buildings having more varied room layouts, but also to take into account the resistance of materials (partitions and openings in particular). They also seem promising to develop a coupled approach to refine the modeling for the estimation of damages, but also to qualify the danger of buildings in case of structural failure.