Site effects modelling applied to the slope affected by the Suusamyr earthquake (Kyrgyzstan, 1992)

In 1992, a large magnitude earthquake (Ms=7.3) hit the northern part of the Kyrgyz Tien Shan range where it triggered rockslides and many debris slides or flows. One of these mass movements occurred on the Chet–Korumdy ridge located in the Suusamyr Basin. It consists of a multi-rotational debris slump in its upper part that turned into a debris flow in its lower part. Involving arenitic material overlying silty clays, it has a volume of about 0.5 to 1.106 m3, a maximum thickness of 40mand a run-out of 200 m. The field observations and measurements carried out on this slope suggest that local amplification effects could have contributed to the initiation of the seismic failure. To test this hypothesis in the lack of instrumental evidence of
local ground-motion recordings, we conducted a sensitivity study of site effects based on a numerical analysis in the visco-elastic domain with a twodimensional finite difference code. Varying the topography and the geology of the investigated slope, topographic site effects are found to be less important than geological site effects which are controlled by the contrast of impedance between the surface materials and the bedrock. The geometry of the low-velocity surface layer has also an influence on site effects, which is often difficult to be distinguished from pure topographic effects.
Considering all modelling results, we conclude that site amplifications alone cannot have triggered the Suusamyr landslide during the 1992 earthquake. The static slope stability analyses done in previous studies revealed that the Suusamyr failure neither can have a purely static origin. Even if the water table is very high within the arenite layer, only a minor failure develops in the lower part of the slope. Therefore, we believe that the triggering of the Suusamyr landslide is a consequence of pore pressure build up in areas characterized by significant ground-motion amplifications.