Assessment of Landslides Triggered by Earthquakes Based on the Combination of Peak Ground Motion and Critical Acceleration Analysis

There is a need for landslide susceptibility models that can be used to quickly predict the locations of earthquake-triggered landslides after large seismic events. As a triggering factor, peak ground acceleration (PGA), which is a measurement of the magnitude of seismic ground motion, has a close relationship with the landslides occurrences and usually is used as an indicator in the assessment of landslides hazards. However, the landslides triggered by the 2014 Ludian earthquake, Yunnan, China show an exception. Different from other events, the landslides exhibit a particular pattern of spatial distribution. They did not occur along a fault or structural zone linearly, instead being relatively concentrated in several locations southeast and west of the epicenter. The usually used factors for landslides assessment such as earthquake magnitude, the distance to epicenter or faults as well as PGA cannot give a reasonable explanation to this phenomenon. Considering the physical mechanism of earthquake triggered landslides, a slope performance during a shaking event mainly depends on two parts: one is the stability of itself, which can be represented by critical acceleration obtained by Newmark method model analysis, and the other is the trigger intensity, which can be measured by PGA. Thus, for a given PGA, whether or not a landslide occurs depends on not only the PGA, but also the stability of the slope itself. Based on these, we use the Newmark’s method model to analysis critical acceleration in the landslides affected area during the Ludian earthquake, and find that the results can make it explicable for the particular distribution patter.