ON THE SLOPE DEVELOPMENT CAUSED BY THE SURFACE LANDSLIDES

1979 
In order to examine the relationship between the landslides and the subsurface structure of the slope, the cone penetration test were carried out on the weathered granite hillslope in Aichi Prefecture, where a heavy rainfall caused many landslides in 1972. The results are shown in Fig. 2, where N10 represents the number of the impacts necessary to drive a cone resister at each incremental depth of 10cm. Broken lines and chained lines in Fig. 2 show the depths where N10 exceeds 10 and 50, respectively. At the middle and the lower segments of the slopes, where landslides frequently occur, the soft layer (N10<10) is much thicker and the transient zone (10_??_N10<50) is extremely thin. So the subsurface structure can be regarded as the two-layered, which means that the bedrock underlies the soft layer without the transient zone. when the landslide occurs, the soft layer slides down, and the bedrock is exposed. The net rate of wasting _??_ is represented by _??_ ' where _??_l is the average wasting rate by landslides, and the wasting rate by slow and steady process of erosion (soil creep, rain wash, etc.). Each of the wasting rates is defined in the normal direction of the slope. It seems that the landslide can not occur until the thickness L of the soft layer reaches the critical thickness Lcr. In the case of the straight slope with slope angle θ, Lcr is approximated by _??_ (2)' where C is the cohesion, φ the angle of internal friction, γs the saturated unit weight, and γb the submerged unit weight of the soil. Because the heavy rainfall which can induce the landslides is frequent in Japan, it can be said that the landslide will occur immediately after L becomes equal to Lcr. Therefore, the return period T of the landslides nearly equals to the time for the soil thickness to grow up from zero to Lcr. Thus, Eq. (1)' becomes _??_ (3) The time change in the thickness L of the soft layer is given by _??_ (4) where υw is the descending velocity of the weathering front. Rearranging Eq. (4) and integrating it with the conditions L=0 at t=0 and L=Lcr at t=T, we obtain _??_ (5) Under the assumptions that υw is expressed by _??_ with A and L1 being constants and that _??_ is constant, we can calculate _??_ from Eq. (5). Therefore, Eq. (1)' becomes _??_ (8) Denoting Lcr/Li as a and _??_/A as β, Eq. (8) is rewritten as _??_ (8)' The relationship between β and V/A is shown in Fig. 7. The line R-S in this figure represents the condition that the landslide can not occur, for L is always less than Lcr. The interval P-R represents the condition that the landslides take place on the slope. It is notable at this range that the increase in β brings about less preferable conditions for the occurrence of the landslide and leads to smaller value of V/A. The decrease of _??_/A is remarkable near the point R. When _??_ is equal to zero, we can obtain the relative rate of slope retreat _??_ as a function of slope angle from Eqs. (2)' and (8). The relationship between θ and Vr/A is shown in Fig. 8. If L1 is large enough, Vr/A decreases with an increase of θ after it reaches a maximum at a certain slope angle.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    2
    References
    5
    Citations
    NaN
    KQI
    []