Long-term monitoring of a large deep-seated landslide (La Clapiere, South-East French Alps): initial study

The large-scale deformation of high mountain slopes finds its origin in many phenomena (inherent parameters, external stresses) with very different time constants (instantaneous to geological scale). Gravitational effect, tectonic forces and water infiltration are generally the principal causes of slope instability. However, it can be very difficult to distinguish which cause is dominant and which are their respective effects. To gain a better understanding of the complex processes taking place during the evolution of an unstable slope and separate the causes responsible of the landslide dynamic, an observational study based on geodetic, meteorological, seismological and electrical data has been performed on the La ClapiSre rockslide (Southern French Alps). This deep-seated landslide (DSL) is known for many years as one of the largest and fastest rock slide in Europe (60 million m(3) of highly weathered metamorphic material, moving at 1 to 3 m year(-1)). The set-up of the "Observatoire Multidisciplinaire des Instabilit,s de Versants" (OMIV, in 2011 has allowed the production and availability of an important and original data set over several years of accurate monitoring. Thus, for the first time, the long-term study of geodetic data permitted us to highlight acceleration phases in the general movement of the landslide that affect its dynamic. These modifications are associated with variations of the velocity by a factor 3 to 6. The characterization of the origin of these variations was possible due to the comparison with meteorological, electrical and seismological data. Based on these various signals, we were able to establish correlations and contributions of meteorological water infiltration in the dynamic evolution of the La ClapiSre slope. We determine several response times to the meteorological stress for seismic endogenous events (mainly rockfalls), the resistivity of the ground (quasi-instantaneous) and the kinematics of the slope (from 2 weeks to 2.5 months). Moreover, our results strongly suggest the existence of rainfall threshold of 3.5 +/- 1 mm day(-1) from which the number of seismic endogenous events is highly increased.