No abstract

This paper presents the results of numerical analyses of ground motion in the Red Zone sector of Amatrice hill, violently struck by the 2016-2017 Central Italy seismic sequence. The methodologies used in processing the data to define the numerical model are firstly described. The results obtained from the computational analyses are then presented and discussed by comparing them with experimental data set of weak motion recordings. Computational analyses were performed via both a 2D-numerical FEM model and a pseudothree-dimensional hybrid model (SiSeRHMap) which develops multispectral maps taking into account topographic effects. Starting from available geological data and geophysical measurements, an original and specific subsoil GIS model was developed and utilised to perform the computational analyses. The preliminary map for fundamental periods computed from the subsoil model is in good agreement with the experimental data. A restricted set of weak ground motions acquired from an accelerometric station located in a reference site was used as input for the numerical analyses, while the signals of the corresponding events recorded at the top of the hill were used as targets in the reliability evaluation analysis of the outputs. In the area of Amatrice hill, which is characterized by a complex geological and topographical context, the reliability analysis shows a good performance of the hybrid model compared to the 2D-FEM model in the prediction of seismic response. Agreement generally was also good with regards to the experimental and computational results, both in relation to the amplitude and to the shape of the spectral amplification that change depending on the hill sector. Considering the predictive reliability of the models, a high amplification, due to topographic effects, was observed for the Red Zone by performing a back-simulation of the 24th August 2016 main shock. The analysis results highlight also that the maximum amplification factors, based on the definition of the Housner intensity, occur in the interval of periods 0-0.5 s covering the fundamental period range of the buildings in this area.

No abstract

No abstract

The investigation of soil response to dynamic loads is necessary to predict site-specific seismic hazard. This paper presents the results of cyclic and dynamic laboratory tests carried out after the 2016-2017 Central Italy Earthquake sequence, within the framework of the seismic microzonation studies of the most damaged municipalities in the area. The database consists of 79 samples investigated by means of dynamic resonant column tests, cyclic torsional shear tests or cyclic direct simple shear tests. Results are firstly analysed in terms of field and laboratory values of small-strain shear wave velocity, highlighting the influence of the sample disturbance and of the mean effective consolidation pressure. The cyclic threshold shear strains as a function of plasticity index are then compared with findings from the published literature and the outliers are analysed. Subsequently, the dynamic soil behaviour is investigated with reference to the small-strain damping ratio. Differences between results from different tests are analysed in the light of the loading frequencies. Finally, the database is used to develop a predictive model for soil nonlinear curves according to plasticity index, mean effective confining stress, and loading frequency. The model represents a useful tool to predict the nonlinear stress-strain behaviour of Central Italy soils, necessary to perform site-specific ground response analyses. Keywords Soil dynamics • Laboratory tests • Shear modulus and damping ratio • Smallstrain material damping • Shear wave velocity Communicated by S.I.: Seismic Microzonation of Central Italy.