Catálogo de publicaciones - libros

Landslides from massive rock slope failure (MRSF) are a major geological hazard in many parts of the world. Hazard assessment is made difficult by a variety of complex initial failure processes and unpredictable post-failure behaviour, which includes transformation of movement mechanism, substantial changes in volume, and changes in the characteristics of the moving mass. Initial failure mechanisms are strongly influenced by geology and topography. Massive rock slope failure includes rockslides, rock avalanches, catastrophic spreads and rockfalls. Catastrophic debris flows can also be triggered by massive rock slope failure. Volcanoes are particularly prone to massive rock slope failure and can experience very large scale sector collapse or much smaller partial collapse. Both these types of failures may be transformed into lahars which can travel over 100 km from their source. MRSF deposits give insight into fragmentation and emplacement processes. Slow mountain slope deformation presents problems in interpretation of origin and movement mechanism. The identification of thresholds for the catastrophic failure of a slow moving rock slope is a key question in hazard assessment. Advances have been made in the analysis and modeling of initial failure and post-failure behaviour. However, these studies have been retrodictive in nature and their true predictive potential for hazard assessment remains uncertain yet promising. These processes, which can be instantaneous or delayed, include the formation and failure of landslide dams and the generation of landslide tsunamis. Both these processes extend potential damage beyond the limits of landslide debris. The occurrence of MRSF forms orderly magnitude and frequency relations which can be characterized by robust power law relationships. MRSF is increasingly recognized as being an important process in landscape evolution which provides an essential context for enhanced hazard assessment. Secondary processes associated with MRSF are an important component of hazard.

Landslides resulting from massive rock slope failure (rockslides, rock avalanches and the failure of volcano slopes (including edifice collapse)) are an important geological hazard in many regions of the world and have been responsible for some of the most destructive natural disasters in world history. Massive rock slope failure occurs with measurable frequency in the mountains of the world; based on twentieth century data massive rock slope failure involving volumes equal or greater than 20 M m3 occur every 2.7 years. Historical data indicates that the frequency in the Alps is one every century. The magnitude and frequency of massive rock slope failure from three datasets were analysed and robust power law scaling relations were obtained which had similar negative values of the exponent. A database of 38 landslide disasters involving 1000 (or more) fatalities in the period A.D. 1000-1999 indicates that 75% of the total fatalities were due to massive rock slope failure. 68% of these fatalities were due to the indirect effects of such failure, viz. bursting of landslide dams and the occurrence of landslidegenerated waves. The Landslide Destructiveness Index (LDI) expresses losses as loss (in this case life loss) per unit volume of a landslide. LDI varies inversely landslide volume. The magnitude and frequency of fatalities resulting from landslide disasters shows power law scaling with the value of the exponent ~ -1. Whilst the frequency of landslide disasters (in the period 1000-1999) with a given number of fatalities is comparable to disasters related to volcanic eruptions, it is an order of magnitude less frequent for comparable magnitude disasters involving destructive earthquakes.

The 250 million m³ Vaiont landslide of 1963 is the largest failure for which monitoring data has been collected in the period in which the failure was initiated. Despite this, and the availability of good geological and geotechnical datasets, considerable uncertainty remains regards the processes occurring within the landslide prior to and during the failure. In this paper, the monitoring data are re-examined to attempt to understand the triggering mechanisms and controls on movement. by plotting 1/velocity against time for the 1960 and 1962 movement events, and the 1963 final failure, it is demonstrated that the landslide appears to have undergone a change in deformation mechanism. The 1963 data suggest that the final failure was brittle in nature, but the lack of seismicity effectively rules out large-scale cracking in the limestone. The 1960 and 1962 failures suggest a ductile mechanism. This is used to support the previously proposed initiation mechanism in which deformation of the basal clays in the brittle-ductile transition regime allows movement that is characterised by micro-cracking on the small scale but apparent ductility on the large scale to be superseded by brittle failure when the microcracks coalesce.

Despite improvements in recognition, prediction and mitigation, rock slope instabilities still exact a heavy social, economic and environmental toll in mountainous regions. This is largely due to the complexity of the processes driving slope failure and our inadequate knowledge of the underlying mechanisms. Ever increasingly, experts are called upon to analyse and predict the stability of a given slope - assessing its risk, potential mode of failure and possible preventive/remedial measures. To do so, it has become essential for the practitioner to be cognisant of the slope analysis tools that are available and to fully understand their strengths and limitations. This paper examines the use of numerical modelling and its role in aiding rock slope stability analyses by providing key insights into potential stability problems, failure mechanisms and mitigative solutions. Several examples will be presented to demonstrate the cause and effect relationships shaped by geological conditions (e.g. rock mass structure, strength degradation through weathering), coupled hydro-mechanical processes, interactions with engineered structures, and aspects of progressive failure as they apply to massive natural rock slopes.

Pre-collapse creep is a widespread component of complex slope movements. Its study requires a preliminary kinematic classification, based on the relations between style and rate of gravitational dislocations and geological structure of rock massifs. Complexity and variability of slope deformations determine the shortcoming of the traditional deterministic approach. Thus, probabilistic analysis considering different scenarios and ‘event trees’ seems to be on the mainstream of rock slopes stability assessment. General aspects of geomechanical modelling and risk analysis of slope processes are discussed.

We examine the choices model-users have to face when applying models to better understand slope failure and prefailure movements of large rock slopes and illustrate how modelling can help understand rock slope movements. At first, the questions that the models can help to answer are listed. The different types of models available to understand the hydromechanical behaviour of rock slope movements are then specified. Different types of models have been applied to the La Clapière case. This movement is described and three different modelling analyses of this landslide are outlined. Each model allows an understanding of a part of the problem raised by rock slope movements, but none of them can actually represent totally the phenomena involved in the movement.

This paper illustrates the use of a combined finite-discrete element code with fracture propagation capabilities in the simulation of rock slope failure. Using this approach it is possible to investigate failure from initiation through transportation to deposition; in effect from a “total slope failure process” perspective. Selected failure cases and mechanisms are simulated, not as definitive analyses, but in order to show the future potential of this techniques in rock slope analysis.

Earthquake-triggered landslides are an important secondary effect of strong shaking. Such phenomena have occurred during a large number of earthquakes and have resulted in significant loss of life. Seismically induced slope failures have occurred in both rock and soil masses, and had sizes ranging from a few individual blocks up to several million cubic meters in volume. Since the mid 1960s numerous methods of analysis have been developed to assess the risk posed by these hazards, however, such investigations normally use a simplified form of ground motion input such as peak ground acceleration, or Arias Intensity (Ia). These take no account of potential amplification effects associated with steep topography. In this paper the mechanisms of topographic amplification are considered, along with the potential impacts on the stability of slopes during large earthquakes. The key uncertainties have been outlined along with some of the major research questions for the future.

A rockslide of about 10¶ m³ was reactivated in April 2002 on Monte Beni near the town of Firenzuola in Northern Tuscany (Italy). The rockslide caused the evacuation of 3 private houses and the interruption of the Regional Road n.65 “della Futa”, one of the main transportation routes connecting Firenze and Bologna. During the emergency, in order to rapidly acquire data on the state, distribution and style of activity of the moving rockslide, a monitoring campaign was carried out, using an innovative radar device capable of a remote sensing assessment of ground displacement fields with a high resolution and accuracy. The system is a ground-based radar interferometer, known as LISA (Linear Synthetic Aperture radar), which has been successfully tested in past experiences for landslide monitoring. The production of multi-temporal maps of ground displacements, over a time span of 5 days, has provided a clear picture of the rockslide mechanism and activity. These data were utilised by public authorities and decision makers to define temporary measures for risk reduction and risk management.

The evaluation of the state of activity is an essential step for landslide hazard assessment. To attribute a state of activity class to a landslide a suitable set of monitoring data is needed. Different types of landslides present different characteristics and are subjected to different spatial and temporal evolution; these become relevant when working at a regional scale or when a catastrophic evolution of the movement is expected. The Permanent Scatterers (PS) technique is used to determine state of activity and long term behavior of rock slope instabilities. The PS technique, overcomes several limitations of conventional differential SAR interferometry (DInSAR) applications, and proves to be effective for high accuracy monitoring of gradual very slow slope deformation which may eventually transform into extremely rapid failures. The success depends on various factors including: available data, location and morphology of the area, PS density; and motion of the targets. We present the results of the application of the technique at a regional scale in the Central Italian Alps involving different types of instabilities in different materials. The utility of the technique in landslide hazard assessment is demonstrated by the quality of the results and by the integration of different datasets (inventory maps, site investigations, remote sensing).