[en] As one task of the research project DEUFI (DEtailing Urban Flood Impact), laboratory experiments were performed representing the flow intrusion from a street into a building through openings representing open doors or windows. Various obstacles (representing cars, trees, etc) were inserted in the street in order to deviate the flow and to measure their effect on the flow discharge through the opening. In this paper, various models solving 2-D shallow water equations are used to simulate these experiments. Main conclusion is that the intrusion discharge is always underestimated if the water elevation in the street in front of the opening is well predicted. This latter flow discrepancy does not appear to be linked with a discrepancy in the representation of the detailed flow pattern. The value of the intrusion flow depends on the turbulence model that changes the velocity field close to the opening and only slightly on the other parameters of the numerical model (topography, mesh, boundary conditions, friction). The percentage of change in the intrusion flow due to the obstacles is not accurately modelled although the trend is generally satisfactory. This latter result means that the representation of the obstacles in the model is only necessary if the expected effect is high, and that the calculated intrusion discharge should be used carefully.
Research Center/Unit :
UEE - Urban and Environmental Engineering - ULiège
Mignot E, Camusson L, Riviere N (2020) Measuring the flow intrusion towards building areas during urban floods: Impact of the obstacles located in the streets and on the facade. J Hydrol 583:124607. https://doi.org/10.1016/j.jhydrol.2020.124607
El Kadi Abderrezzak K, Paquier A, Mignot E (2009) Modelling flash flood propagation in urban areas using a two-dimensional numerical model. Nat Hazards 50:433–460
Erpicum S, Dewals BJ, Archambeau P, Pirotton M (2010) Dam break flow computation based on an efficient flux vector splitting. J Comput Appl Math 234(7):2143–2151