By analyzing the observations of mass movement and underground-water level for various landslides, the creep characteristics of soil mass are estimated. Moreover, the correlation between the creep characteristics and the geological properties of the landslide areas are examined. Besides, the plastic and the creep characteristics are investigated through laboratory tests on model soils. Based upon these results, the influence of the geological factors on the creep characteristics of landslides is discussed by paying attention to the soil properties of a sliding layer.

View full abstract

Run-out distances of rockfall mass had been investigated on granitic rock blocks by means of a large-scale rockfall apparatus. Rockfall specimens had been prepared under dry and water-saturated conditions. Longitudinal run-out distances between gravity centres before experiments and at depositions were 10 % longer in water-saturated specimens than dry specimens when the specimens were with the same number of rock blocks in our rockfall experiments. Lateral expansions at deposition were larger in experiments with larger number of rock blocks. However, the standard deviations in lateral direction at deposition normalised by those at initial arrangements before experiments became smaller in experiments with larger number of blocks. Equivalent coefficients of friction between gravity centres were in positive correlation with number of rock blocks, which were contradicted to the fact of a negative correlation between landslide mass and equivalent coefficient friction in the classic studies on the landslide run-out distances. These were likely attributable to the facts that pore-structures were not formed in the rock-mass and no excess of pore-pressure was built-up during the downslope motion regardless of the rock-mass volume in our experiments, and rock blocks made inelastic collisions much more when the specimens consisted of larger number of rock blocks, in which much kinetic energy was consumed.

View full abstract

Ninety percent of all tsunamis are generated by earthquakes; however, the remaining 10% are caused by meteorite collisions, volcanic activity, and submarine landslides. Submarine landslides occur due to slope instability, and large-scale submarine landslides can generate very large tsunamis. Therefore, from the viewpoint of disaster prevention, it is important to clarify the generation mechanism and features of submarine landslides. In this research, we compared the features of submarine landslides generated in two different regions, and we calculated their volumes. The first environment is a debris avalanche that occurred by the collapse of a volcanic body; the debris flowed into the sea and generated the Kanpo Tsunami along the Sea of Japan in 1741 in the case of Oshima-Oshima Volcano southwest off Hokkaido in the Japan Sea. A hummocky topography was found from the sea floor at the northern part of Oshima-Oshima, it is assumed to be the debris avalanche in 1741. In the second place, a submarine landslide exists from the continental shelf margin southwest of the Satsuma Peninsula in the southern part of Kyushu up to a depth of 320 m. We calculated the volume of the submarine landslide, and the volumes of the two landslides were similar (-2km³) . However, those distributions and morphology are significantly different. We clarified that these differences are caused by the difference in the environment (topography, width of head wall, collapsed materials, etc.) .

View full abstract

Influence of the geometric elements (length“L”, thickness“D cos α ”and slope“α ”) and groundwater level of landslides on the c′ - tan φ′ diagram of stability analysis and the thrust force of landslide were studied by means of a model landslide, where the shift of average strength prameters at slip surface was also investigated.
Influences of L, D cos α , α and groundwater level on the c′ - tan φ′ diagram were discussed in relation to the alteration of c′ -tan φ′ line (c′ -intercept“c′ ”, tan φ′ -intercept“tan φ′ ”and inclination“β ”) . Obtained results were as outlined below. The influence of L is very minimal. D cos α has an effect on c′ and β , but has little impact on tan φ′ where as α has a significant impact on c′ and tan φ′ , but does not seem to have any effect on β . Influence of the geometric elements on the c′ - tan φ′ diagram was measured quantitatively and the importance of determining the slip surface was clarified futher. Influence of groundwater level appears as a large rightward shift of tan φ′ .
In the analysis of the alteration of average strength parameters at slip surface (c′_ave , φ′_ave ) and the thrust force of landslide corresponding to differences in relative positions of slip surface and groundwater table, two important observations were noted. A difference of one meter in the groundwater level has a significant impact on φ′_ave , especially in the case of landslides having a small L, a steep α and a high groundwater level. A difference of one meter in the position of the slip surface has a significant impact on the percentage of change of thrust force, especially in the case of landslides having a small L and D cos α . Given these conditions, there are possibilities of change of thrust force by nearly 15 percent.

View full abstract

This paper examined typical examples of slope failures that occurred during the past 15 years in Mie prefecture, clarifying the features of slope failures controlled by geology. Sedimentary rocks and schists have been subjected to landslide with planar sliding surfaces parallel to their foliations. On their anaclinal slopes, toppling has occurred. Igneous rocks have been subjected to destructive slope failures, if they are severely weathered or fractured with adversely oriented joints. Decomposed granite in the Ryoke Belt has been subjected to compound slide. Large rockfalls and slope failures may occur in granite porphyry of the Kumano acidic rocks, in which columnar joints are developed. Weakly consolidated lake deposits have been subjected to landslide with low angle sliding surfaces parallel to the bedding plane.

View full abstract

Because the traditional limit equilibrium method of 3D stability analysis assumes moving body of a landslide as rigid mass, it ignores horizontal ground pressure exerted on a slip surface. However, frictional resistance exerted by horizontal earth pressure on the flanks of a slide is believed to be greatly influential to 3D stability analysis, hence the authors examined to include horizontal earth pressure in the limit equilibrium method of the analysis. Furthermore, the authors built equations reflecting this alternative method applied in the Janbu methods of the two-dimensional stability analysis as well as of the three-dimensional simplified analysis. In this paper, calculation results of the equations applied to data samples from actual landslide areas are presented in detail. Safety factor was 1 to 4% higher with the two-dimensional analysis and 3 to 16% higher with the three-dimensional analysis both compared to the traditional methods.

View full abstract