Shallow failure, which often occurs on mountain slopes during heavy rainfall, was investigated. A three dimensional slope stability analysis method, called the three-dimensional multi-planar sliding surface method reported elsewhere, was used to determine the site and volume of the critical sliding mass, as defined by the minimum safety factor. The site and volume of the critical sliding mass were compared to those of the actual failed mass. The site appeared to be at nearly the same place as the actual failure, and the volume was nearly the same; nevertheless the mesh space of the digital elevation model differed. The sliding direction of the assumed sliding mass that includes the most unstable cell was assumed in these reports to be the same as that of the most unstable cell. The sliding direction of an assumed sliding mass, however, may change according to the shape of the assumed sliding mass that includes the most unstable cell. The sliding direction of the assumed sliding mass was calculated from the equilibrium of the forces to be at a right angle to the sliding direction. The calculation method then was applied to an actual failed slope, and results show that the sliding direction obtained is nearly the same as that of the failure that took place on a mountain slope in 1986. The sliding direction having been obtained and a setting method giving a non-rectangular shape having been used, the safety factor was calculated using three analyses methods : the 3-D multi-planar sliding surface method, the 3-D Hovland method, and the 3-D simplified Janbu method. The most probable unstable mass found with the 3-D multi-planar sliding surface method is very nearly at the same site and has very nearly the shape of the actual failed mass.

View full abstract

[in Japanese]

View full abstract

Forest damage which was caused by eruptions of Mt. Fugen, Unzen volcano, and the transition until 1992 was investigated with remote sensing and ground survey. Five images by SPOT satellite showed that the course of pyroclastic flow changed from Mizunashi river to Oshiga valley and to Akamatsu valley, and that the forest damage which was caused by pyroclastic flow and volcanic gas spread out gradually. Ground survey showed that most pine trees and many fir trees were dead and deciduous broad-leaved trees were declined in Nita mountain pass area where typical damage was observed. The area damaged by volcanic gas reached down to Mt. Nodaka and Mt. Iwatoko which were about 3 km south-east from the crater. Damaged Hinoki arificial forest especially stood out there. A forest damage transition map and a temporary forest damage map derived from a SPOT image were drawn based on the results described above. Causes of forest damage were classified as pyroclastic flow, volcanic gas and ash. The damage types and the mechanism were discussed. The damage caused by volcanic gas was characteristic of the wide area and the mechanism. It was estimated that forest damage by volcanic gas was generated by not only SO₂, but sulfuric acid mist which SO₂ and moisture of fog were compounded into. Chronic damage and decline of forest by volcanic ash on forest soil and leaves will spread on the continuous volcano activity.

View full abstract

[in Japanese]

View full abstract

The major factors contributing to slope failures debris run-out limit using probability linear regression were evaluated. It was concluded that the major factors are, firstly, the average gradient of the source area, secondly, the average gradient of the deposit area and its catchment, and, lastly, the failure depth and the nature of the soil. Of these factors, 60% is contributed by the size and quality of debris, and 40% is contributed by the condition of the slope. Therefore, both of these factors are necessary to evaluate the debris run-out for failures. The evaluation indicates that the failure depth is also a factor that increases the width of the possible limits of the debris run-out limit. These wider possibilities may have been caused by the fact that a shallower failure is more easily affected by any objection, or may have been caused by the fact that it is difficult to measure the failure depth.
Using the analysis method can make it easy to estimate the debris run-out limit. Applying the method to other cased under research, and increasing the precision of the models, can make it possible to evaluate the danger of debris run-out, particularly for longer flows.

View full abstract