This study presents methods of deciding input data to model jointed rock masses for numerical analysis. First, a laboratory shear test method is presented in which shear testing is performed applying normal stress level of field condition to establish the value of mechanical properties including normal and shear stiffness of natural rock joints. In addition, a field investigation method is also proposed to identify the joint geometry in a rock mass. Finally, these methods are applied to determine the input data for DEM(distinct element method) analysis while carrying out the stability analysis of a jointed rock slope situated in Omaru river dam site, Kyushu island, Japan. The results of the stability analysis convincingly reveal the positive significance of the proposed methods predicting the plane failure along a geological contact as a response to the excavation of the slope for a dam foundation.
NTRODUCTION
While analyzing, designing and predicting the performance of engineered structures built in and around jointed rock slopes, it is essential to have a good understanding of the mechanical behavior(strength and deformability) of the rock mass under imposed boundary conditions. A rock mass of a site is not continuum and its behavior is dominated by discontinuities such as bedding, joints, faults and fractures. Presence or absence of these discontinuities has a very important influence upon the stability of rock slopes and therefore, these geological features have to be taken into consideration properly for practical designs.
Stability analysis of the rock slope consisted of a single or a few blocks associated with a single persistent discontinuity or small number of discontinuities can be carried out by the limit equilibrium method as has been representatively described by Hoek and Bray (1981). In case of jointed rock masses comprised of a finite number of discrete and interacting blocks, the finite element method (FEM) and the distinct element method (DEM) have often been used. The FEM, however, uses continuous formulation and is therefore not suitable to simulate the progressive failure accompanied with large deformation in rock masses. On the other hand, the DEM is based on discontinuum approach in analyzing a blocky rock masses and it utilizes a force displacement law which specifies forces between blocks, and a motion law which specifies the motion of each block due to unbalanced forces acting on the blocks. In addition, the rock mass is modeled as an assemblage of rigid or deformable blocks and discontinuities are regarded as dis.tinct boundary interaction between these blocks. Cundall (1971) has described that the DEM is useful in analyzing the progressive, large-scale deformation of discontinua. But Hoek et al. (1995) has mentioned that further experience in the application of this method to practical design situation is still required so as to understand where, when and how the DEM could be applied properly. In addition, while modeling a rock mass of a site with numerical methods one has to determine the joint geometry and its mechanical properties as accurately as resembling of their field situation. The former can not be predicted convincingly except at a few observation points and the latter is usually determined using some empirical methods. Consequently, simplification has to be introduced extrapolating available data which in turn adds uncertainties in the results.
This paper presents a laboratory shear test method which gives full attention to sample selection, sample preparation, shear testing with normal stress level of field condition and data processing to establish the mechanical properties including normal and shear stiffness of natural rock joints. Base
