Modélisation et visualisation infographiques tridimensionnelles de structures et propriétés géologiques

This project, the computer-aided three-dimensional modeling and visualization of geological structures and properties, was conceived in the framework of the INFOGEOL project, "Modeling of Underground Civil Engineering Works Intersecting Geologic Structures". This project is carried out by the Laboratory of Geology of the Swiss Federal lnstitute of Technology in Lausanne (GEOLEP-EPFL). The work presented here is oriented to solving problems connected with modeling geologic structures rather than civil works, which already benefit from sophisticated 3-D modeling CAD (Computer Aided Design) packages. The research had three primary goals: to determine the 3-D graphic computer situation available in the modeling and visualizing of geological structures and properties; to present a method and software that solve the major geological modeling problems; and to illustrate the benefits of 3-D modeling in solving problems in the Earth Sciences. Statement of the 3-D representation of geological objects The main characteristics of the application of the 3-D computer modeling in geology are discussed in this section. The visualization principle is based on the representation of block diagrams similar to those created by geologists since the first geological studies. The evolution of computer technology has made the construction of such models easier. In addition, computer technology has optimized the use of such models by creating tools for viewing these blocks from any angle, for separating the blocks into the objects that compose them, for slicing through the blocks in any direction, and so on. To obtain this goal of visualizing and manipulating these block diagrams, some additional constraints must be discussed : they concern the available hardware and software techniques. Because of their design, workstations are the only computers that allow efficient work in the 3-D computing domain. As such, these workstations were used in the development of our methods. From the standpoint of the methodology, the techniques used in this study are different from those of a CAD system: the aim is not to create objects as an architect does (an ab initio process), but to build objects from existing data (an interpolation process). The main problems encountered in developing the new method are : the modeling of overturned structures, such as recumbent folds; the computation of intersections between surfaces; and the volume representation of each object. In addition, the modeling process must be done with the maximum amount of data precision, which ultimately determines the quality and the validity of the model. Presentation of the method and software In the framework of this research project, a partnership has been established with Dynamic Graphics, Inc., a Californian software Company specializing in mapping and modeling of geologic data. From that arrangement, developments have allowed the production of a system that solves all the problems discussed above. In addition, t