Reliable prediction of rockfalls is a major need in mountain areas, both for hazard assessment and countermeasure design. In this paper, a 3D kinematic modelling approach is presented, which is able to solve site-specific rockfall problems using an original simulation code with high resolution input data. The code allows detailed rockfall simulations on a 3D topography computed from a Digital Elevation Model. The case study of the Mt. S. Martino - Mt. Coltignone cliffs, impending on the Lecco urban area (Lombardia Region, Northern Italy) is discussed. The example is particularly interesting, due to the occurrence of frequent rockfall events, valuable elements at risk (urban areas, corridors) and countermeasures (barriers and retaining walls). Model calibration issues are discussed by comparing model results to available experimental data. The scale dependency of model results is also discussed.
Rockfalls threaten lives, settlements and transportation corridors in mountain areas. Despite Usually limited in volume (Rochet 1987), rockfalls are characterised by high energy and mobility. Rochet (1987) defined four categories of rockfalls, namely: single block falls (involved volume in the range 10-2-102m3), mass falls (102-105 m3), very large mass falls (105- 107 m3) and mass displacement (more than 107 m3). We deal with rockfalls involving less than 105 m3, i.e. "fragmental rock falls" (Evans & Hungr 1993), characterised by negligible interaction among the falling blocks. Rockfall modelling aims to assess the potential trajectories of "design" blocks, the maximum run out distance, the spatial distribution of kinematic parameters and the probability to stop at specific locations. This information results in the design of slope benching, retaining fences, fills and rock sheds, or in hazard assessment for land-planning purpose. Rockfall dynamics is a complex function of the starting point and the geometry and mechanical properties of both the block and the slope. Knowing the initial conditions, the slope geometry, and the relationships describing the energy loss at impact or by rolling, it should be possible to compute the position and velocity of a block at any time. Nevertheless, the relevant parameters are hardly ascertained both in space and time, even for an observed event. Usually, the geometrical and geomechanical properties of the blocks (size, shape, strength, degree of fracturing) and of the slope (gradient, length and roughness, longitudinal and transversal concavities and convexities, grain size distribution, elastic moduli), and the exact location of the source areas are unknown and characterized by extreme spatial variability. In addition, the energy lost at each impact or during rolling depends on a variety of factors including the velocity of the block and the impact angle, the block to slope contact type, the density of vegetation and even the time of the year (Broili 1977, Bozzolo & Pamini 1986, Azzoni et al. 1995, Jones et al. 2000). These parameters are difficult to determine at any spatial scale. Thus, "contact functions" relating the kinematics of the block (in terms of velocity) or its dynamics (in terms of energy) before and after each impact, are introduced to model the energy loss.
