THE OZ MACHINE AND SMALL COUNTY: ONLINE RESOURCES FOR EDUCATION IN SUBSURFACE CHARACTERIZATION

We present two applications for interactive instruction in the geological aspects of petroleum reservoir characterization, both freely available through the Kansas Geological Survey’s web site. The first application, the Oz Machine, is a Java applet that provides a simple exercise in the interpretation of lithologies from petrophysical well logs. Each time the applet is run, a Markov chain simulation procedure generates a new sequence of lithologies from which a corresponding suite of logs is computed. The student’s task is then to identify the lithological sequence based on the logs, with the applet optionally providing feedback on the accuracy of the student’s interpretation. Despite the simplicity of the applet, it provides a compellingly realistic exercise in geological log interpretation. The second application, Small County, expands the basic concepts Oz Machine to three dimensions and adds identification of fluid contents (oil and water) to the interpretive problem, employing a virtual subsurface closely reflecting the geology of the US mid-continent, in the fictional setting of Small County, Kansas. Stochastic simulation techniques are used to generate the subsurface characteristics, including the overall geological structure, distributions of facies, porosity, and fluid saturations, and petrophysical logs. The student then explores this subsurface by siting exploratory wells and examining drilling and petrophysical log records obtained from those wells. A version of the application aimed at students in introductory geology courses implements a simpler exercise in which students search for the peak of an anticlinal structure, providing experience in the interpretation of contour maps and the improvement in information as more data points (wells) are added.