Mechanics of a tectonized soil slope: influence of boundary conditions and rainfall

The Vadoncello landslide, mobilized in December 1993 and still active, is the reactivation of a landslide which took place, within the highly tectonized soils of a turbiditic formation, after the 1980 Irpinia earthquake (Southern Italy), when it was dragged by the movements of a larger landslide body at the toe of the slope, the Serra dell’Acquara landslide. The Vadoncello landslide has been studied by means of the results of comprehensive investigations and monitoring carried out within a EEC funded research project, as well as by means of successive data. Consequently the complex geological model of the slope has been defined, in which the chaotic successions of soil and rock strata are grouped into soil complexes, the location of different landslide bodies is identified within the slope (the 1993-‘95 Vadoncello landslide, the 1980 Vadoncello landslide and the 1980 Serra dell’Acquara landslide bodies) and two hydrogeological complexes are recognized. The soil mechanical properties are shown to be very poor, the deep soils being close to gross yield and therefore prone to large plastic straining due to even limited loading changes. The soil behaviour is consequently an important factor to the slope instability. The study of the soil displacements, both at the surface and at depth, shows that the landslide is composite, being formed of a shallow rotational slide at the top of the slope, a shallow earthflow downslope and an underlying mechanism of slow and long-lasting irrecoverable movements, which are also monitored on the Serra dell’Acquara landslide body, at the toe of the Vadoncello slope. These slow movements are considered to be consequent to the plastic flow of the weak clayey soils in the slopes, which may be activated by seasonal rainfall effects, by the frequent low-medium intensity seismic events occurring in the area, and also by the morphological changes resulting from the slow movements themselves. Rainfall intensity is not found to be the single direct cause of the shallow landslide reactivation, which instead is proposed to be due to the combination of the effects of the low return-period rainfalls in winter 1993 and the mechanism of slow movements active both at depth in the slope and at its toe.