Assessment of the magnitude-frequency relationship of landslides and rockfalls: application to hazard mapping
- Autores: Guillem Domènech Surinyach
- Directores de la Tesis: Jordi Corominas i Dulcet (dir. tes.)
- Lectura: En la Universitat Politècnica de Catalunya (UPC) ( España ) en 2015
- Idioma: inglés
- Tribunal Calificador de la Tesis: Joan Manuel Vilaplana Fernández (presid.), Marcel Hürlimann Ziegler (secret.), Rosa María Mateos Ruiz (voc.)
- Programa de doctorado: Programa Oficial de Doctorado en Ingeniería del Terreno
- Materias:
- Enlaces
- Tesis en acceso abierto en: TDX
- Resumen
Due to the exponential grown of the population within the last decades, the landslide hazard assessment of earthflows and rockfalls and their hazard mapping have become an essential tool for the territory management, mostly in mountainous areas. The landslide hazard was defined as the probability of occurrence of a potentially damaging phenomenon in a certain area and within a given period of time. Thus, the probability of occurrence (or frequency) for each type of landslide and magnitude must be known. The aim of this research is to develop an objective, quantitative and reproducible methodology to obtain the magnitude-frequency relations for medium and large size earthflows and rockfalls. This procedure has been set up in the Barcedana Valley and Montsec Range for large earthflows and rockfalls, respectively. Both study areas are located within the Tremp Basin (Eastern Pyrenees). Concerning the earthflows, they have been split between 1) reactivations and 2) intact slopes. The frequency of the reactivations has been obtained by analysing 11 sets of orthophotos covering a period from 1956 to 2013. The magnitude has been calculated as the area of the landslides obtained from the landslide mapping and from the orthophotos. The resulting magnitude-frequency relation for reactivations has been used to derive the probability of landslide reactivation for a given volume. It has been compared with probability of landslide reactivation obtained from the rainfall threshold responsible for the reactivation of 4-large landslides located within the Tremp Basin as well. The reactivation date has been estimated by means of dendrogeomorphology and the rainfall threshold has been determined by means of ROC analysis. The susceptibility of first-time slope failures have been obtained using a deterministic model named SINMAP. The frequency for each susceptibility class has been calculated using the inventory of first-time failures identified in the field and by means of orthophotos. An algorithm to obtain the area of the earthflows larger than the pixel size has been developed through an automatic aggregation of pixels located within the same slope and having the same susceptibility class. The obtained magnitude-frequency relation of first-time failures has been compared with the one obtained from the mapped first-time failures. Finally, the magnitude-frequency matrix for hazard mapping of intact slopes has been defined. Concerning the rockfalls, a methodology to obtain the rockfall scar size distribution of a cliff has been defined. It has been assumed the rockfall scar volumes as proxy for the rockfall volumes. In that case, the distribution of rockfall scars has been calculated using a high resolution point cloud of the rockwall obtained by a terrestrial laser scanner and following. Several volume distributions have been calculated to take into account the different detachment mechanisms and the consequent range of detached volumes. Finally, a procedure has been developed to convert form statistical frequency (% of scar volumes), calculated in the previous step, to temporal frequency (annual number of the scar volumes). To this, the total volume of material lost has been computed using the afore-mentioned point cloud. The elapsed time within the total volume has been removed has been estimated by dating the initial surface, from which the current rockfall activity started, by means of terrestrial cosmogenic nuclide, 36Cl.
