Three-dimensional Subsurface Geological Modeling of the Western Osaka Plane based on Borehole Data

Three-dimensional (3D) geological model of subsurface structure plays an important role in developing infrastructures. In particular, the 3D geological model in urban area is quite helpful to solve social problems such as underground utilization, environmental preservation, and disaster assessment. Over the past few years, many studies have been made on algorithms for 3D geological modeling. However, most of them have given little attention to objectivity of the model and traceability of modeling procedures. The purpose of this study is to develop an algorithm for constructing a 3D geological model objectively and for maintaining high-traceability of modeling procedures. For the purpose of our work, we proposed a new algorithm for 3D geological modeling using gridded geological boundary surfaces and the "logical model of geologic structure". The geological boundary surface is given by a form of Digital Elevation Model (DEM). The DEM is generated based on geological information such as elevation, strike and dip by using a unique spline-fitting method. The logical model of geological structure is a mathematical model that defines a positional relation between geological boundary surfaces and geological units. The model is objectively given by recurrence formula derived from a sequence of geological events arranged in chronological order. We applied the proposed algorithm into constructing a 3D subsurface geological model of the western Osaka Plane, southwest Japan. The data used for 3D geological modeling is a set of borehole data provided by Osaka City and Kansai Geoinformatics Agency. As a result, we constructed a 3D model consistent with the subjective model reported in other studies. In addition, all information necessary for modeling, such as the used geological information, the parameters of surface fitting, and the logical model, was stored in text files. In conclusion, we can not only construct 3D geological model objectively but also maintain high-traceability of modeling procedures. The current algorithm is unfit for constructing a 3D model including complicated geological structures such as overfold and lens. It is recommended that the surface fitting method be still improved.