Physikalisch basierte Gefährdungsmodellierung flachgründiger Translationsrutschungen in einem geologisch heterogenen Gebiet

Author(s)

Ekrem Canli

Abstract

Although there have been major advances in the analysis of landslide hazard
within the last decade, landslides still pose a substantial threat to infrastructure
and lives on a global scale. Within the field of risk management it is crucial to obtain information on landslide susceptibility for a specific area beforehand. This thesis contributes to that related topic. Based on a deterministic approach, the slope stability for the whole district of Waidhofen/Ybbs is being calculated with the SINMAP model. SINMAP has its theoretical roots in the infinite slope stability model combined with a hydrological model that obtains wetness from a high resolution Digital Terrain Model (DTM). The stability index is calculated for different calibration areas by the parametrization of different geotechnical parameters. The study area consists of very heterogeneous geological conditions in which particularly the northern part exhibits more than 800 landslides resulting in a very high landslide density. Due to these different lithologies it was necessary to split the study area into a northern part (consisting mainly of Flysch and Klippen zone) and a southern part (Limestone and Dolomite). Modeling results revealed that at least 77% of the mapped landslides are within the three lowest stability classes. The modeling was carried out in different spatial resolutions to estimate the difference caused by a changing data basis. A comparison with a previously realized statistical modeling for the same study area revealed substantial differences. The unified susceptibility classes displayed a good match for the northern part where more than 70% of the pixels were classified with the same susceptibility class. Whereas the southern part revealed a high mismatch, as only 29% of the pixels are within the same susceptibility class. Results of the sensitivity analysis proved that changing values of the friction angle bear the greatest impact on the model. Only slight changes to the model outcome occur when changes in the hydrological parameter take place. Although the results of the northern part are considered to be adequate, there are still a lot of questions to be answered. This is due to the comparison with the statistical model, as differences, particularly in the southern part, are enormous. A combination of both modeling approaches statistical and deterministic, is advised since this adds a better physical interpretation to empirical models.

Organisation(s)

Department of Geography and Regional Research

No. of pages

159

Publication date

2013

Austrian Fields of Science 2012

105404 Geomorphology, 105408 Physical geography, 105101 General geology

Portal url

https://ucrisportal.univie.ac.at/en/publications/4cfff426-5d40-425b-94ff-3fd76bc6ef0b