Researchers at the University of Liège and Purdue University have developed a computer model to measure the influence of urbanisation on flood damage. Coordinated by Benjamin Dewals (EEU Research Unit - Faculty of Applied Sciences) the ARC FloodLand project suggests the implementation of urban planning recommendations that could limit flood height in cities. This study was published in Science of the Total Environment (1).
F
loods are responsible for almost a third of the economic damage caused by natural disasters. However, we know that the risk of flooding in urban areas will increase over the course of the century as a result of climate change and growing urbanization. If flood risk cannot be totally avoided, its impact on cities could be better controlled. This is in any case what Benjamin Dewals and his colleagues at the University of Liège and Purdue University (USA) tried to model within the framework of the FloodLand project (*).
“The objective of the FloodLand scientific project was to measure the influence of urban pattern on potential flood damage," explains Benjamin Dewals, hydraulic engineer at the HECE Lab (Hydraulics in environmental and civil engineering). This project, based on an interdisciplinary approach, brought together various groups of the Urban and Environmental Engineering Research Unit (UEE) of the Faculty of Applied Sciences, including researchers in urban planning, hydrology, hydraulics and transport modelling.
Some examples of urban centre simulations studied during the project.
“Most of the studies carried out so far have analysed many aspects of the influence of urbanisation on floods, but generally ignored the impact of the geometry of the urban model on the severity of floods," explains Professor Jacques Teller, Director of the LEMA and partner of the project. Urban features - such as street width, orientation or curvature - can have a strong influence on flood flow because they affect the distribution of water between streets, as well as the depth and velocity of flow."
As part of the FloodLand project, the researchers looked at the entire risk chain from climate impact, hydrological and hydraulic modelling to damage and risk estimation. The analysis was carried out on two spatial levels: the catchment level (mesoscale analysis) and the floodplain level (microscale analysis).
“The originality of this study lies in the parametric analysis of the impact of urban patterns on floods," says Ahmed Mustafa, a doctoral student at LEMA." It was within this framework that the researchers decided to study the influence of small-scale urban configurations on the severity of floods caused by overflowing rivers in floodplains. To do this, the researchers used a system of automatic generation of urban configurations developed at Purdue University to generate nearly 2,000 "synthetic" urban districts based on several parameters such as street width, degree of urbanization or distance between buildings.
Hydraulic analysis of 2,000 synthetic urban fabrics to determine the urban characteristics that maximize flood resilience.
Once these 2,000 synthetic urban districts were generated, the researchers used the hydraulic model WOLF2D - developed at the University of Liege - to calculate the flood characteristics for each of them." Our hydraulic model has been enhanced to simulate flooding in this set of configurations," says Martin Bruwier, postodoctoral researcher at the HECE Laboratory and first author of the publication (2). The model developed has calculated hydraulic flows in a particularly efficient way by representing the urban fabric as a kind of "porous medium" whose characteristics depend on topographic information and urbanistic parameters.
Analysis of the relationships between urban and flood characteristics revealed that the increase in water levels as a result of new urban developments could be compensated for by an appropriate layout of the buildings. For the configurations analysed, increasing fragmentation of the urban fabric and increasing distances between buildings would contribute to a significant reduction in flood heights. The FloodLand project therefore paves the way for recommendations to promote more flood-resistant urban development to limit damage.
(*) The Floodland project is part of the ARC - Concerted Research Actions - programme of the University of Liège. This programme aims to develop centres of excellence in research, which are a priority for the University of Liège, by means of coherent projects introduced by confirmed teams. These extend over 4 years and preferably integrate teams of complementary disciplines. The projects are submitted to the evaluation of international experts, before being examined by the Research and Development Councils.
(1) M. Bruwier, A. Mustafa, D. G. Aliaga, P. Archambeau, S. Erpicum, G. Nishida, X. Zhang, M. Pirotton, J. Teller, and B. Dewals, “Influence of urban pattern on inundation flow in floodplains of lowland rivers,” Science of The Total Environment, vol. 622–623, no. Supplement C, pp. 446–458, 2018. http://orbi.ulg.ac.be/handle/2268/216580
(2) M. Bruwier, P. Archambeau, S. Erpicum, M. Pirotton, and B. Dewals, “Shallow-water models with anisotropic porosity and merging for flood modelling on Cartesian grids,” Journal of Hydrology, vol. 554, no. C, pp. 693–709, 2017. http://orbi.ulg.ac.be/handle/2268/207585
UEE Research Unit I HECE Lab (Hydraulics in Environmental and Civil Engineering)
Martin BRUWIER
UEE Research Unit I LEMA (Local Environment Management and Analysis)
Ahmed MUSTAFA
A new study at ULiège and a call to established entrepreneurs and young entrepreneurs in student incubators
Prof. David Lyden (Weill Cornell Medical College) inaugurated at ULiège the series of lectures and meetings in Belgium on The pre-metastatic nice concept.
A look back at the honor bestowed on Lisette Lombé on Thursday April 9 at ULiège.
