The aim is to improve landslide hazard assessment using two different approaches and scales: (a) landslide occurrence analysis over large areas by means of both detailed field work and the use of computer based maps implemented on Geographical Information Systems (GIS); and (b) predicting the behaviour of individual slopes by means of physically based models of slope stability conditions.
Approach (a) will include the improvement of data capture and mapping techniques for landslide hazard assessment and standardization procedures. New techniques of systematic description of field observations will be used to generate databases at each field station utilising geo-referenced cartographic symbols.
The data base will be interfaced with a Computer Aided Drafting (CAD) and GIS for data processing. The accuracy of computerized data capture will be checked in the field. GIS based landslide hazard maps at three study areas (Corvara-IT, Lisbon-PT and Cantabrian Range-ES) will be prepared by means of an statistical analysis (multivariate statistical techniques and favourability functions) of both automatic data captured and field work data;
Approach (b) will consider several steps. First, new low-cost equipment (wire extensometers and tilt and deformation meters) and techniques for the monitoring of ground movement (GPS) and for landslide geometry definition (ground penetrating radar) will be tested at four landslide sites in Vallcebre (Eastern Pyrenees), Alver` (Dolomites), Roughs (British Coast) and Super-Sauze (French Alps). The second step includes the analysis of the performance of different hydrological and slope stability models in predicting groundwater fluctuations and landslide behaviour at each landslide site.
Existing hydrological models will be improved by investigating the effect of vegetation and the significance of unsaturated and fast flow by-passing conditions (macropores and vertical fractures) in triggering or reactivating landslides. Coupled hydrology-stability models will be prepared and validated against real-time monitored data.
The last step will include downscaling of General Circulation Models (GCM's) by means of a statistical treatment which interrelates the characteristic patterns of observed simultaneous variations of regional climate parameters and the large-scale atmospheric flow. Rainfall occurrence obtained by this procedure will be used as a data input of hydrological-slope stability models in a climatic scale. Future climatic scenarios will be coupled with the hydrological models used in this project at established monitoring stations. In this way, future landslide hazard will be predicted.
Funding SchemeCSC - Cost-sharing contracts
7511 AL Enschede
WC2R 2LS London
3508 TC Utrecht