The IMIRILAND project focuses on aspects of management of risk in the case of large landslides, in its scientific, technical and land-planning aspects. Eight large landslide sites have been selected in the countries involved in the project, where geological, geomechanical and monitoring data already exist, with different levels of knowledge, in order to show how risk can be assessed in different situations.
These data, after a suitable organisation, are the basis for the following phases of the project. New methods or new applications of the existing methods are worked out in order to obtain the scenarios of the landslides evolution. Reference is made to geological, numerical. Then the developed methodologies are applied to the management of the endangered landslide zones. Actions for risk mitigation are proposed on the basis of risk assessment and the legal and economical framework. Finally, dissemination of risk management methodologies among the users is a major task of the project.
Large landslides affect many mountain valleys in Europe. They are characterised by a low probability of evolution as a catastrophic event but may have very large direct and indirect impact on man, infrastructures and environment. This impact is becoming more and more dangerous due to the increasing tourism development and the construction of new roads and railways in mountainous areas. Methodologies for the identification and mitigation of risk are therefore a major issue
As a matter of fact, many experiences during critical development of landslides have shown a lack of methodologies and above all a non-systematic approach of interpreted risks. Risk management is in practise accomplished by local and regional authorities only during the critical event in a necessarily improvised way. This "reactive" approach induces negative consequences on the identification procedure. For instance, very expensive monitoring systems have been installed on several large landslides without any well-established methodology linking the interpretation of the measures and the understanding of deformation mechanisms to the practical questions concerning the management of the risk.
Furthermore, risk can extend well beyond local damage (for instance, risk of river damming which may induce major hydrological hazards: floods, inundation of sewage plants, loss of drinking water resources), so that it must be considered in a wide perspective.
What do we know about landslide risks? A first theoretical definition was given by Varnes in 1984: risk is the probability of an event of a given magnitude multiplied by consequences.
Since then a lot of technical and scientific papers have been written on this topic (i.e. Einstein, 1988, Cherubini et al., 1993, Canuti & Casagli, 1994, CALAR, 1999). Recent works (IUGS Working Group on Landslides, 1997) propose the following multidisciplinary procedure for quantitative risk analysis of slopes:
However the evaluation of hazard is still made through different approaches leading to results which differ from the qualitative and quantitative point of view. Besides only a few works have been done (i.e. Leone et al., 1996) on vulnerability. Finally, it does not exist any well defined procedure to include the results of the risk analyses in land planning, taking the legal framework into account.
The present project intends thus to deepen the hazard analysis of different types of landslides, by the application of geological, geomechanical and statistical methods. Particular attention is given to numerical models where finite elements, boundary elements and finite difference methods will be applied in three dimensional conditions on the same slope stability problems. This multidisciplinary approach will thus allow to carry out a comparative analysis of different hazard prediction techniques.
Then, it is foreseen to develop the vulnerability and risk analysis for several landslides considering short and long term perspectives, direct and indirect consequences, as well as technical and social impacts, using tree event techniques. The combination of the results obtained in relation with hazard analysis and through the risk approach will allow the development of a new practical and quantified risk assessment programs which will be applied to several sites.
Cherubini C., Giasi C.I. & Guadagno F.M. (1993), Probabilistic approaches of slope stability in a typical geomorphological setting of Southern Italy, Risk and reliability in ground engineering, pp. 144-150, Thames Telford, London
Concerted action for forecasting, prevention and reduction of landslides and avalanche risks (CALAR) (1999). X-Calar’99 Proceedings, Innsbruck
Einstein H.H. (1988), Special lecture: Landslide risk assessment procedure, Proc. 5th Int. Symp. on Landslides, Lausanne, Switzerland, Vol. 2, pp 1075-1090
IUGS Working Group on Landslides, committee on risk assessment (1997), Quantitative risk assessment for slopes and landslides – The state of the art, Proc. of Int. Workshop on Landslide Risk Assessment, Honolulu
Leone F., Asté J.P. & Leroi E. (1996), Vulnerability assessment of elements exposed to mass-movement: working toward a better risk reception, Proc. of 7th Int. Symp. on Landslides, Trondheim, June 1996, p. 263-270
Varnes D.J. (1984), Landslide hazard zonation; a review of principles and practice, Natural Hazards 3, UNESCO, 63 pp.