Towards Cross-Border Landslide Hazard and Risk Assessment in Central Asia

Thursday, 18 December 2014: 4:45 PM
Annamaria Saponaro, Marco Pilz, Marc Wieland, Massimiliano Pittore, Dino Bindi and Stefano Parolai, Deutsches GeoForschungsZentrum GFZ, Potsdam, Germany
The countries of Central Asia are known to be among the most exposed in the world to landslide hazard and risk. In the past, several devastating slope failures have caused intense economic and human losses across the entire region. The large variability of local geological materials, difficulties in forecasting heavy precipitation locally, and problems in quantifying the level of ground shaking, call for harmonized procedures to better quantify landslide hazard. Moreover, due to uncontrolled urban expansion in mountainous areas, a growth in vulnerability of exposed population as well as overall risk has to be expected. In order to mitigate landslide risk, novel and strategic approaches are required mainly for enhanced understanding of causal factors, for reducing exposure to hazards, and for controlling land-use practices in a harmonized transnational way.

We have already presented a regional landslide susceptibility assessment for Kyrgyzstan. First results allow for the identification of most potential landslide areas all over the country, with sufficient degree of accuracy. Based on this, we hereby propose a procedure for obtaining cross-border risk map of earthquake-induced landslides among central Asian countries, by employing statistical tools and updated input information in such remote and data-scarce regions. The method is initially applied to Kyrgyzstan where the majority of input parameters is available, and subsequently extended to Tajikistan and Uzbekistan. At first, the influence of diverse potential parameters (topography, geology, tectonic lineaments) as well as seismic triggering to landslide activation is evaluated. Elements at risk are then analyzed in relation to landslide hazard, and their vulnerability is hence established. A sensitivity analysis is carried out, and results are validated to an independent dataset.