Structural Damages to Residential Buildings Induced by the July 2021 Flood in Belgium

Abstract

The floods of July 2021 in the Belgian part of the Meuse basin were disastrous. This exceptionally significant event is classified as extreme in the literature. The consequences were numerous, both material and psychological, particularly for the people affected but also on a regional scale. Numerous studies have been carried out in order to prevent the consequences, but also to change the development of the affected valleys in the face of potential new and unusual floods. The aim of these studies is to understand what causes these floods, but also what can lead to such damage. For this, models are needed. In order for these models to develop and function properly, a lot of data must be collected. In this respect, the exceptional floods of July 2021 are a very important source of information. In this master thesis, the objective was to create a model for predicting structural damage to buildings caused by flooding. For this purpose, a sample of about 200 buildings was used. Some of the data collected after the July 2021 event has been used to create an interesting database. In particular, it contains information on the damage caused to buildings. The analysis of this data has made it possible to assign a degree of damage to each of the buildings studied in this work. This classification was established according to what can be found in the literature. Next, a statistical model using logistic regressions was developed in order to predict the degrees of damage associated with the buildings on the basis of predictor variables. The latter are building or flow characteristics listed in the database. The reader of this final study should be aware throughout that the results presented are representative of the sample studied and not of a general behavior observable at the scale of the whole valley.