A grade-3 method of zonation for seismic slope stability: An Italian case study

This paper presents a seismic microzonation study performed in an area of the Northern Apennines (Italy) in order to reduce the level of risk from landslide hazard, to support decisions in urban planning and to provide useful information for building design. The area in question measures about 2.6 km(2) and is characterised by considerable landslide susceptibility also in non seismic conditions as evidenced by a number of surveys carried out over the past years. The slope deposit, lying over a marley-arenaceous formation, shows a variable thickness from a few metres to 30 m and consists of fine silty-clay, clay-sandy and silty-sandy soil in a variable amount. Quiescent and active landslide areas with ground movements involving the shallowest layers of the soil deposit have been detected. Given the relatively high seismicity of the region, the aim of this study is to describe the principles and criteria adopted in a seismic risk analysis conducted over a limited area in order to carry out a very high and detailed level of zonation for ground motion and slope stability (i.e., Grade 3 according to the definition in the Manual for Zonation on Seismic Geotechnical Hazards). For this purpose, reliable expected earthquakes were assumed and in-depth geologic, geomorphologic, geotechnical and geophysical analyses were carried out to define a subsoil model for performing numerical analyses of local seismic response and slope stability in both non-seismic and seismic conditions. The proposed methodology consists of the following three steps: (i) identification of a number of seismic inputs for the numerical analyses; (ii) implementation of numerical analyses to evaluate the amplification of ground shaking and to assess potential for slope failure; (iii) mapping of homogeneous areas with respect to expected site amplifications and landslide risk. The study allowed the mapping of four zones which are expected to be homogeneous in terms of both amplification factor of seismic motion (FA) and safety factor against landslide (FS). Quiescent and active landslide zones were thus identified with greater accuracy together with the most probable sliding surfaces. For this purpose, a 1-D numerical code was adopted for seismic response analyses and limit equilibrium methods were implemented and numerically solved in static and pseudo-static slope stability analyses. (C) 2012 Elsevier Ltd. All rights reserved.