Mediterranean Sea and anthropogenic influences on ambient vibration amplitudes in the low-frequency and high-frequency domains in the Algiers region

Ambient vibrations have been continuously recorded at Dar El Beida, about 20 km from Algiers (Algeria). This data set allows determining that, in the low-frequency domain (<1 Hz), ambient vibration sources are mainly linked to Mediterranean Sea effects, while in the high-frequency domain, they are closely related to anthropogenic activity. Climatic conditions have an influence on the ambient vibration spectral amplitudes in the low-frequency domain, which is not the case in the high-frequency domain. The limit between the low-frequency and high-frequency domain, based on natural versus anthropogenic activity, is not clear cut and lies between 1.25 and 1.50 Hz. Variations of H/V peak amplitudes in the low-frequency domain are clearly linked to the climatic conditions. In the high-frequency domain, H/V peaks are not related to climatic conditions and cannot be clearly related to anthropogenic source changes.     

Numerical modeling to predict the spread of landslide of schist’s area under climatic event of Ain El Hammam (Algeria)

The several reactivations of the landslide of Ain El Hammam (AEH) after each important weather event compel us to look closely at its triggering factors and predict its mechanisms and its evolution at the longer term. In this sense, the prediction of the slope behavior becomes necessary. This paper presents a numerical model of the AEH landslide using Plaxis® software. This model considers hydraulic effects such as precipitation and pore pressure even in the unsaturated parts. Soil and rock behaviors are described with proper elasto-plastic models named Hardening Soils and Jointed Rocks. The first model takes into account hardening on isotropic and deviatoric mechanism as well as a non-associated flow rule. The second model considers a non-isotropic elasticity with perfect plasticity along with given sliding directions. The hydraulic and mechanical models are coupled with an effective stress concept. To detect unstable areas in the landslide, we developed a Matlab® program to take into account the Hill’s bifurcation criterion, which is based on sign of the second-order work. It has been proved that this criterion allows detecting all failure modes that can appear in rate-independent materials and especially the ones that develop before the plasticity limit criterion. From such computations, we can predict the shape and position of slip surface responsible of the actual ground movement of the slope. To validate the numerical results, analysis of field measurement is included. We use high resolution of electrical tomography to delineate the geometry and position of failure surface and approve our results.