Probabilistic Seismic Landslide Hazard Analysis

Seismic landslide is a significant seismic hazard that induced by a mountain slope under the action of a strong earthquake.It may be accompanied by the occurrence of the earthquake and slide,or after the earthquake,in a certain period of time,the mountain slope is subjected to rain,gravity instability and other factors appear instability slide.Seismic landslide is a system involving many factors and is full of complexity and uncertainty.As the basis of regional seismic landslide zoning,regional seismic landslide hazard assessment is of great significance to reduce the hazard of earthquake secondary disasters.Regional seismic landslide probability assessment is chosen as the research and exploration direction in this paper.This paper focuses on the influence of uncertainty of Newmark model on seismic landslide hazard,the influence of threedimensional fault source on seismic landslide hazard in near-fault area,and the influence of coupled rainfall in seismic landslide hazard assessment on seismic landslide hazard.On the basis of data collection,interpretation of indoor remote sensing image and field investigation,probabilistic model and probabilistic evaluation method are established by using mathematical statistics and other methods.This paper has made innovative progress in the following aspects:1.Proposed a set of technical methods to establish the probability distribution representing the uncertainty of Newmark model.The probability distribution model of rock mass physical and mechanical parameters(φ’,C,γ)uncertainty in Newmark model is established by using Beta-PERT distribution with definite physical meaning.The Newmark model set was formed by Monte-Carlo random sampling technique,and the probability distribution model of critical displacement ac which represented the uncertainty of Newmark model was established.2.Newmark displacement Dn evaluation method with hazard level(quantile)is proposed and applied to regional seismic landslide hazard evaluation.Based on the probability distribution of critical displacement ac and the empirical relationship between ground motion parameters,ac and Dn,the probability distribution of Dn uncertainty was obtained,so as to determine Dn values under different quantiles.Quantile hazard expression is introduced for the first time in regional seismic landslide hazard zoning based on Dn scale.Also,in the statistical prediction of earthquake landslide probability,the quantitative relationship between Dn and landslide probability under different quantile was established for the first time,and the earthquake landslide probability with given quantile was introduced.Based on the coseismic landslide catalogue of Jiuzhaigou Mw6.5 earthquake in 2017,an example is calculated.3.A probabilistic evaluation method of earthquake landslide hazard is proposed,which combines the uncertainty of Newmark model,the uncertainty of Dn empirical relation and the probability distribution of ground motion.Considering the probability distribution of critical displacement ac,lognormal distribution of Dn empirical relation and probability distribution of ground motion occurrence,the hazard probability of landslide represented by the critical Newmark displacement threshold λis calculated by using the total probability method.4.The Newmark model considering rainfall influence is discussed,and the probabilistic evaluation method of earthquake landslide hazard coupled with rainfall influence is established for the first time,and the application of this method is realized by taking Zhouqu area as an example.5.In the probabilistic evaluation of seismic landslide hazard,threedimensional fault source is used to improve the prediction of ground motion in the near-fault area of large faults,so that the near-fault effect can be considered in the probabilistic prediction of seismic landslide hazard in the near-fault area,and the rationality of the probabilistic evaluation of seismic landslide hazard in the near-fault area can be improved.Taking the northwest section of Xianshuihe fault as an example,the application has achieved obvious results.