Laboratoire d'accueil : Bureau d'évaluation des risques sismiques pour la sûreté des installations (BERSSIN)
Date de début de thèse : octobre 2017
Nom du doctorant : Flomin TCHAWE NZIAHA
Descriptif du sujet
Surface ground motions are a result of seismic waves emitted during faults rupture processes; these waves travel through the earth's crust and then through superficial geological layers. The influence of the latter on the incident ground motion is known as "site effects" in seismo-logical literature. In regions with high contrasts in soil properties (between sedimentary layers and the surrounding bedrock) and complex layers geometries as the case of sedimentary basins, the superficial layers tend to trap the seismic waves hence resulting in longer and more amplified motions. As a result, buildings and installations present in such areas are more likely to be damaged with the occurrence of and earthquake as it was the case during the Michoacán earthquake of 1985 in Mexico. It is therefore important to correctly assess site effects in order to better prevent such risks.
To this aim, numerical simulations provide a precious tool to understand the propagation of seismic waves in complex media. Recent studies (e.g. Oral et al. 2017) highlight the strong influence of the soil mechanical behavior on the wave propagation. Indeed, under strong motion, the soil behavior may become non-linear (soil properties depend on the level of strain). In Oral et al. (2017), it was shown that considering excess pore pressure and nonlinear effects in numerical simulations allowed to better reproduce observed data.
The goal of this thesis is to study from a physical point of view: 1) the impact of the spatial variability of the distribution of heterogeneities in media with linear and non-linear behavior on ground motion, 2) the impact of the source complexity on surface motions, 3) the structures response to the simulated wave field can be studied as well.
Our first case study is the sedimentary basin of Nice (a relatively small size basin) on which we study wave propagation considering different material rheologies. The results obtained from these numerical simulations are going to be used to quantify the basin response under different conditions. These results will allow the code verification with already published studies (e.g., Gandomzadeh, 2011; Peyrusse et al., 2014). This is the first step before studying the Rhone valley's response with the same approach. Finally, in order to see the effect of lateral heterogeneities, we are going to perturb the wave propagation model by adding random spatial variability of soil properties, and analyze their effect on the computed ground motion.