GNGTS 2013 - Atti del 32° Convegno Nazionale

points (on the surface and along depth), the location of existing wells/boreholes (in a radius of 250 m), and the geologic material constituting the different strata are represented on the resulting cross-sections. In addition, the faults present in the region have been reported by a procedure developed ad hoc for the present study. This procedure takes into account the exact location of the fault on the surface and its strike and dip. In such a way it is possible to project in depth the fault and then to analyze how it can modify the behaviour of the different strata. Fig. 2 shows a 3D representation of the geotechnical model: the superficial geology is characterized in a manner similarly to Fig. 1 and the profiles illustrate the geologic interpretation in depth. Furthermore, data about the bathymetry of the bay have been added to define the depth of the sea. Each profile is geo-referenced and the contact between the different layers in depth was modified by hand when it was required. The soil properties, such as type of material, density, and VS, were assigned to each layer: these parameters are the input data in local response modelling. After having defined all profiles, the geotechnical model of the study area has been developed. An E-W oriented geological cross-section is represented in Fig. 3, the acronyms for geology are the same as in Tab.1. It is the framework from which it is possible to identify the sectors to be modelled for deriving the expected soil amplification. Both 1D and 2D modelling techniques are expected to be applied. In sectors where the geotechnical model shows a homogeneous soil behaviour in depth (almost flat parallel layers), an equivalent linear 1D analysis is planned, according to the computer code PSHAKE (Sanò and Pugliese, 1991). Conversely, in areas of complex stratification with strong lateral variations, a linear 2D analysis is planned with the BESOIL code (Sanò, 1996). This code applies a fast numerical calculation of the seismic wave propagation in space, based on the boundary element technique. Conclusions. For seismic response modelling it is worth to have a good geotechnical model of the investigated area. Good geological, geotechnical, topographical, geophysical, tectonic and seismic data contribute in a fundamental manner to the construction of a geotechnical model. A geotechnical model for the Santiago de Cuba broader area has been defined on the basis of the analysis of all the available geological and geophysical information. Moreover, 61 stratigraphic profiles, which cover the entire study area, have been constructed being the profiles calibrated on data coming from a huge number of boreholes. The obtained geotechnical model is intended to be used for 1D and 2D modelling of the local soil amplification. Fig. 3 – E-W oriented geological cross-section, the acronyms for the geology are the same as in Tab. 1 and the red lines represent the deep geometry of the faults. 333 GNGTS 2013 S essione 2.2