The deceleration processes of debris flows and their depositions due to the change of channel width, at the flume which has constant bed slope, are experimentally studied.
Change of the front velocity of debris flow is explained as a hypothetic model in conjunction with decrement of the concentration in the front of debris flow. The deceleration along the channel are different according to the incremental ratio of channel width represented by T. As the T increases, the possibility to abate the impactive force and the velocity is improved to some extent.
Also, characteristics of residual deposits in relation to T or C_d are discussed.

View full abstract

Mountainslope failures usually occur during heavy rainfalls. On granite slopes, these failures occasionally appeare in a thin layered form. In order to predict the most dangerous site of these failures, the failure potential layer had been defined by employing the results of field investigations, examinations and laboratory tests. These results have already been reported elsewhere (Okimura and Tanaka, 1980). Using a distribution of the said layer and the longitudinal cross section of this slope, the most dangerous site and the length of the failure was analyzed by a slope stability analysis method. This method is a multi-planar sliding surface method which is developed from a tri-planar sliding surface method (Chowdhury, 1978).
This method is applied to the data obtained from a granite test field in which slope failures had occurred in 1972. The analysis results showed that the assumed sliding block, which presented the minimum safety factor, appeared at the same place where failure had occurred. Failured slopes and non-failured slopes could be grouped by using the modified safety factors which were calculated by multiplying the shape factor of the catchment area by original safety factor. Therefore, if we obtain the distribution of the failure potential layer, we can point out the most dangerous site of failure on a slope by using the multi-planar sliding surface method.

View full abstract

T and N Eq. (2.1) are the total shear stress along and total pressure normal to circular arc failure plane in a semi-infinite simple earth slope (cf: Fig. 1), derived by Dr. S. Ogihara by means of integration. Substituting T, N and L into Fellenius's formula of safety factor F, and letting dF/dθ=0, where θ=half the central angle of a circular arc, we obtain the equation of condition for a critical arc.
A circular arc is determined with θ and r given; the radius of arc r can be reduced to a function of θ plus a constant λ where λ stands for a given specific substratum geometry. r/λ are given in the form r₀ in Eqs. (2.3) - (a) through- (d), where, as well as in Fig.2,
(a) =homogeneous earth slope
(b) =substratum-bounded slope where α=β
(c) =substratum-bounded slope where α≠β or δ=|α-β|>0
(d) =substratum-bounded slope where β₁<α<β₂
When both effective cohesion c′ and effective angle of internal friction φ′ are positive, for example, the condition for critical arc is expressed as in Eq. (2.11), where κ₀ is a known as defined in Eq.
(2.6)′. The value θ which satisfies Eq. (2.11) is the half of the central angle of the critical arc. Substituted into Eq.(2.6)′ or (2.4), this brings values of F₀ and F or T₀, N₀, L₀ and so on. Steady state seepage is considered here by means of a weighted average of pore pressure ratio r_u, which defind where u=pore water pressure; ω=unit weight of soil; and h is the depth of the point in the soil mass below the ground surface.
Expressions which appear with suffix 0 or 1 are all dimensionless and are so arranged to facilitate making relevant nomographs. Nomographs will help a great deal find critical arc or its factor of safety. Where α, β, r_u and λ are given, a single critical arc corresponds to a pair of c′ and φ′, and vice versa. This helps in estimating values of c′ and φ′ or susceptibility to landsliding in a given tract of earth slope.

View full abstract

The application of gamma distribution model is tried to forecast landslide area ratio in Shizuoka prefecture. The procedure to forecast is as follows. To begin with, the resistance index C of stratum belong Cenozoic group is presumed from the mean monthly precipitation J in July by next formula, where the coefficient k₀ is decided from data of past landslides caused by heavy rain.
C=k₀J, k₀=0.0284mm^-1
Secondly, the resistance index of objective stratum in absolute year T_i is revised from the resistance index C of Cenozoic group in year T by the next formula. C_i=C(T_i/T)^0.32
The maximum daily precipitation in 20 years and 100 years probability are taken as planning rainfall r. Then the landslide area ratio P is computed by substitution above C and r into the next gamma distribution model.
P%=(1-∑^c-1_j=0(0.01r)^j/j!exp^-0.01r)×100 As a result of above mentioned operation, the local distribution maps on landslide area ratio are drawn like geomorphological map.The feature as a whole and regional differences are investigated from the stand point of zoning work. Consequently, it is presumed that the lanslide area ratio will reach high value at Izu peninsula and outskirts of plain from Mikkabi, Sagara, Shizuoka to Atami. This tendency coincides with the feature of another. distribution map on the maximum landslide density from disaster statistics.
If the magnitude, period and location of heavy rainfall in future will be forecasted, the estimation of landslide area ratio will develop in concrete shape. The accuracy will rise up by collecting of fundamental data and improvement of forecast method on resistance index.

View full abstract