Accurate modeling of geology, rock fracture, and geologic property distributions is an important problem for natural resource explorations, rock engineering, and underground storages of CO2 and high-level radioactive wastes. This paper presents 3D spatial modeling techniques that are suitable for clarifying these distributions for a wide study area. The techniques are composed of two methods, OPTSIM for discontinuous geologic layers and physical properties and GEOFRAC for fractures including joints and faults. A mechanical minimization criterion and a stochastic simulation were combined for OPTSIM. GEOFRAC is a geostatistics-based method which incorporates the orientations (strikes and dips) of sampled fracture data into the simulation. Fracture locations are generated randomly following the fracture densities assigned by a sequential Gaussian simulation. Fracture orientations are transformed into an indicator set consisting of eight binary variables and the variables are compressed using the principal component analysis. Ordinary kriging is then employed to estimate the distributions of these principal values and the results are back-transformed into the original coordinate system. Fracture orientations are generated randomly using their histograms within the defined directional sector. Finally, fracture facets are determined from the simulated locations and orientations, and the fractures within the angle and distance tolerances are connected to form a fracture plane. The Tono Area, chiefly underlain by Cretaceous granite in central Japan, was selected as a study area, because this area includes the 19 deep boreholes raging from 500 to 1000 m depth. The main target area size is 12 km (E-W) , 8 km (N-S) , and 1.5 km depth range. GEOFRAC was shown to be able to draw a plausible fracture system, because the simulated orientations reproduced those measured: in addition, the locations and directions of the continuous fractures corresponded to the zones of abrupt changes of resistivity and P-wave velocity and the known main faults.
