Internal structure and deformation of an unstable crystalline rock mass above Randa (Switzerland): Part II — Three-dimensional deformation patterns

The monitoring of slope displacements over time provides the basis for most rockslide early warning systems, yet the prediction of catastrophic failure from these records is highly problematic and tenuous, especially if the underlying kinematics and instability mechanism are poorly understood. An example is the moving slope above the town of Randa in the Swiss Alps. This slope is considered typical of those in crystalline rock that lacks a natural, highly persistent, weakness plane dipping towards the valley that can serve as a through-going detachment surface for kinematic release. In Part I (the companion paper to this), the findings from a comprehensive geological and geophysical investigation to image the internal structures of the unstable rock mass were presented. In this paper we develop a kinematic model that describes the pattern of displacement vectors for the rock mass along these structures, both on surface and at depth. The displacements were estimated from 5 years of data from an extensive monitoring system which included surface geodetic and crackmeter measurements, borehole inclinometer and extensometer measurements (up to 120 m depth), and microseismicity. The results showed that the displacement field is highly heterogeneous. Internal deformation is accommodated by both shear and opening-mode dislocation of faults and fracture zones which dip moderately to steeply into the slope. Microseismicity is most intense near the front of the scarp where surface translation in large part reflects the toppling of blocks accommodated by slip along the steeper dipping fractures. The small displacement rates at the study site of up to 2 cm/year coupled with a modest deviation of up to 15° from vertical of the boreholes posed severe problems for the estimation of horizontal displacements from the inclinometer data. We describe the error analysis of these data in some detail since it is relevant for similar installations elsewhere. Conclusions drawn from this work highlight the importance of integrating various types of data in order to better understand the complex block-kinematic processes whose evolution governs the long-term progressive failure of unstable rock slopes.