The Distribution Of Seismic Landslides And Route Project Disaster Prediction Based On SOC Theory

China is a both multi-seismically active and mountainous developing country, so it is one of heavy disaster areas of seismic landslides. Since the 21st century, the landslides triggered by 5 large earthquakes in the western of China have caused not only a large scale of deaths and injuries, but destructive damage to infrastructure also. So, there is important scientific significance and clear application background to study seismic landslides and disaster counter-measures. In view of the theory of self-organized criticality (SOC), it is the unique model to make the holistic theory applied to dynamic systems. Therefore, the SOC is as the theoretical framework of this paper, and many means are used to study the general dynamic characteristics of slope system under earthquakes, such as interpretation of satellite remote sensing data, investigations in field sites, the shaking table test of sand pile models, fast driven sand pile model test, variable fabric properties sand pile model test and numerical simulation of cellular automata.(1). A large number of landslides were triggered by the Ms 8.0 Wenchuan earthquake and the Ms 7.0 Lushan earthquake. Field data show that the distribution of landslides between the volume and the number was changed triggered by earthquakes in different intensity zones. This statistical result needs to be validated by physical experiments. Based on the theory of SOC, the shaking table test of sand pile models under excitations of seismic waves was conducted. The test results show that:?For PGA in the range of 0.075g?0.125g, the relation between the amount and cumulative frequency of sand follows negative power-law. ?When PGA is between 0.15g and 0.25g, the relation obeys lognormal distribution. ?For PGA between 0.35g and 0.45g, the relation turns to obey normal distribution. The shaking table test results recurs the evolution mechanism of seismic landslides from IX to XI intensity areas in physical perspective.(2). Aim at the shifting phenomenon of slope system dynamic characteristics with seismic loading increased, the mathematical and physical analysis are conducted. By mathematical analysis, the decrease of the coefficient of variation is the cause of probability distribution shifting between power-law, lognormal distribution, and normal distribution. Observed by experiment, the performance of grains on the surface of sand pile showed different models with increasing of seismic loading. Under weak vibration, the probability that the activity will die is overall balanced with the probability that the activity will branch. A large-scale collapse occurs when there is a chain interaction between neighboring grains. It has the greatest uncertainty. Under strong vibration, most of grains start independently. A large-scale collapse becomes a certain event. Hereby, the scale of collapse is transformed from chain reaction of unstable grains triggered by self-organized effect to the independent falling of unstable grains triggered by vibrating force effect, which is the physical mechanism of the reducing of the coefficient of variation. According to the above realizations, the phenomenon of the relation between the area and cumulative frequency changing from power-law distribution to lognonnal distribution with increasing of seismic intensity of Ms 8.0 Wenchuan earthquake may be universal. Above recognitions will provide an explanation for the mechanism that the distribution of seismic landslides will change in different intensity zones.(3) The results of Ms8.0 Wenchuan earthquake show that a large number of high-locality landslides are the major threat that causes seismic damage along route project. To identify whether seismic high-locality landslides could threaten the route project is of great important application value to location design at high-intensity mountainous areas. Therefore, potential disaster prediction of seismic high-locality landslide is proposed. According to hydrogeology condition, slope units are divided by GIS technology. Discriminate model of seismic high-locality slopes threatening route project is established. The model contains energy and path conditions. Both of two conditions are met, and it can ensured that seismic high-locality slope threaten the route project after losing stabilities. Zoning map of the seismic high-locality slopes threatening route project in study area can be obtained. This model approach not only provides an effective mean to identify which seismic high-locality landslides could threaten the route project, but also provides an important decision basis for location design in high-intensity mountainous areas and reconstruction of existing routes project. Finally, applied the distribution law of seismic landslides in IX intensity areas obtained by using SOC theory framework to seismic landslides disaster prediction project, and building an application framework about high-locality landslides disaster prevention engineering.(4) By statistics of seismic landslides in Ms 8.0 Wenchuan, Ms 7.0 Lushan ? intensity areas, it is found that the seismic landslides between the volume and the number obeys a power-law distributions, but there is a big gap between the values of exponent b. Therefore, applied the SOC theoretical framework to explore the reason of b values exists a big gap. Inspired by the physical property that b value of earthquake is positively correlated with fault fractal dimension; infer that there is a certain relationship between b value of seismic landslides and fractal dimension of system internal fabric properties. To verify this inference,7 groups of variable fabric properties sand pile model tests were conducted. The results showed that there are some positive correlations, but not linear relationship, between b values of sand amount and the fractal dimension D of sand pile fabric properties. The b values of sand amount will be increase with the fractal dimension D of sand pile fabric properties increasing, and vice versa. Thus, it is certain that the system fabric properties can affect the b values of seismic landslides. It can be inferred that the difference of geological conditions about two earthquakes may be one of reasons that the b values of seismic landslides is different in Wenchuan and Lushan.(5) Based on the SOC operating mechanism and the phenomenon of Ms 8.0Wenchuan earthquake, it can be inferred that slope system operating mechanism in meizoseismal area conforms to fast driven mechanism when ground uplifts rapidly in the earthquake. To study the distribution law of seismic landslides at rapid uplift zone, comparisons and statistics are carried out for landslide's area in Ms 8.0 Wenchuan earthquake within 1.5km of hanging wall and footwall. It shows that the hanging wall disaster obeys lognormal distribution, but the footwall disaster does not. Then the dynamic process of sandpile under fast driven condition is simulated through fast inclined floor. Four groups of fast driven sandpile model experiments were carried out. The result shows when the sand pile angle is near the critical angle, avalanche obeys power-law distribution, with sand pile angle gradually deviating from the critical angle, the statistics of avalanche obeys lognormal distribution, and does not obeys SOC features. Hereby, it can be inferred that the landslides in the hanging wall of fault triggered by Wenchuan earthquake obeys lognormal distribution, and this phenomenon may be the effect of system running mechanism shifting.In a summary, through the above study, this paper not only obtained the distribution of seismic landslides and the basic understanding of disaster countermeasures study, but also has a certain significance to study SOC theory.The process from phenomenological to exact science is expected to be promoted by the exemplary exploration.