Assessing a physical-based approach for landslide susceptibility in the Río la Carbonera, SE flank of Pico de Orizaba volcano, Mexico

In volcanic environments of Mexico, landslides and debris flows occur continually along stream systems of large stratovolcanoes. This is the case on Pico de Orizaba volcano, the highest mountain in Mexico. The volcano has a great potential to affect and damage human settlements and economic activity by landslides. People living along its lower valleys are under continuous threat from the coalescence of upstream landslide sediments that increase the destructive power of debris flows. These debris flows not only produce floods, but also cause the loss of life and property. Despite the importance of assessing these processes, there are few landslide inventory maps and scarce landslide susceptibility modelling. As a result, in Mexico, landslide susceptibility models have not been assessed in order to evaluate their advantages and disadvantages. In this study, a comprehensive assessment of landslide susceptibility models using GIS technology is carried out for the SE flank of Pico de Orizaba volcano. A detailed multi-Temporal landslide inventory map in the La Carbonera watershed is used as framework for the quantitative assessment of a physical-based model for mapping landslide susceptibility. Two landslide susceptibility maps based on the Stability Index MAPping (SINMAP) model are created: 1) by using default geotechnical parameters, and 2) by using field observations of the geotechnical properties of volcanic soils. SINMAP combines the factor of safety derived from the infinite slope stability model with the theory of a hydrologic model to produce the susceptibility map. It has been claimed that the model analysis is reasonably successful in defining areas that intuitively appear to be susceptible to landsliding in regions with sparse information. The resulting susceptibility maps are validated by comparison with the inventory map under the LOGISNET system and SPSS Statistics Software. This indicates the advantages and limitations of the individual models in prediction of landslide locations. Although the model tends to improve with the use of calibrated field data, the landslide susceptibility map tends to overestimate the occurrence of regions prone to landsliding; from the point of view of human safety, this is not necessarily a disadvantage.