Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
The 1987 Hirschigrabe landslide was triggered by a rain intensive period and formed a natural dam to the Zulg river. The natural dam did not break and as such, only the uncertainties concerning the security of some houses within the surrounding settlement cone persist. Ongoingmovements and small spontaneous events indicate shallow and deep displacements. This work establishes a geological model of the site by literature review, site investigation and merging of existing data. With the geological model a inverse problem on the 1987 landslide is established. The complete rigid-plastic problem is solved using upper and lower bound theory of limit analysis with a commercial geotechnic software. The resolution of the inverse problem delivers the material parameters of the formations, whereas the location and shape of the strata is assumed fixed (e.g. emerging from the geologicalmodel). After a verification of the back analysis results, a 2D stability analysis is used to establish predictions on the security of the most exposed residential building. The solution of the inverse problem is particularly sensitive to the friction angle of the gravel formation incorporated in the problem. As to obtain coherence with the 1987 event, it’s value needs to be assumed especially low (Á0 ¼ 10°). Nonetheless, verification with the retained parameter combination shows satisfying results. For the stability analysis a stochastic variance of the important properties is assumed. The risk assessment for the building with the highest danger, emphasizes security factors ranging from 0.3 to 0.5 As the corresponding movements do not affect the close environment of the housings, such lowvalues can nonetheless be accepted. The landslides remain localized on the rupture edge even for worst case scenarios. Despite the limited extend of the stability analysis, no critical tendencies guiding towards an elevated risk for the population are encountered.
Quoc Viet Hung Nguyen, Thành Tâm Nguyên
,