Abstract
The aerial photograph interpretation of 964 slope failure cases proved the following facts. (1) The angle of elevation formed by failed slope height and length, θ, shows normal distribution when the failed slopes were divided into some group by an appropriate combination of failure trigger, area/geology, topography, and vegetation. (2) In case of rainfallinduced failures, the factors of area/geology and/or topography influence the distribution of θ. (3) On the other hand, in case of earthquake-induced failures, the influence of the above factors on the θ distribution is very small. Using the above θ distributions, we developed a stochastic simulation system for predicting hazard areas of shallow slope failure debris.
The algorithm of this simulation is based on the following steps. (1) Get the distance of hazard areas of “Random walk” was calculated from Monte Carlo method using probability distribution of θ. (2) Random walk was performed until the distance of hazard areas. (3) And these processes were repeated.
The result of this simulation roughly corresponded with past cases of slope failures. This simulation is an effective method of hazard mapping on shallow slope failure.
