Numerical Insights Into the Effects of Model Geometric Distortion in Laboratory Experiments of Urban Flooding
Xuefang Li, Sébastien Erpicum, Emmanuel Mignot, Pierre Archambeau, Nicolas Rivière, Michel Pirotton, Benjamin Dewals
Abstract
Abstract Geometrically distorted scale models have been a valuable tool for physical modeling of urban flooding in a network of streets. However, little is known so far about the bias induced in such cases by the model geometric distortion. Here, we use 2‐D computational modeling to provide a first systematic quantification of this bias in the case of a synthetic urban layout. The bias is found to be generally small, with the maximum deviations of the upscaled flow depth and discharge partition from the corresponding values of the undistorted model being around 10% in the case of relatively rapid and shallow flow conditions. When the geometric distortion is increased, the computations reveal a nonmonotonous pattern of the flow variables (depth, discharge partition, and size of flow separation zones), which results from a competition between declining frictional losses and growing local losses in the model. These findings may guide the design of distorted scale models of urban flooding and assist the interpretation of laboratory observations for assessing flood protection measures, for process understanding or for validating computational modeling.