The treatment of faults is one of the greatest problems in the application of computers to petroleum exploration. Typically, too little information and too many alternative interpretations make analytic solutions economically unfeasible or otherwise unsatisfactory. Most computer techniques treat XYZ data points as though they are samples from a continuous surface. As a result, data points on opposite sides of faults are treated together with consequent distortion of analytically derived surfaces in the fault vicinity. Interactive computer graphics offers a solution. The geologist uses a light pen to interact with a series of graphic displays which enables him to (1) collect data on the position of fault surfaces from well logs, contour maps, and cross sections; (2) build umerical models of fault surfaces; (3) use the fault surfaces to separate data points on opposite sides of a fault; (4) build separate numerical surfaces from data points thus separated; (5) composite these surfaces with the fault surface to obtain a final model in which the surface is not distorted by the influence of points on opposite sides of the fault; (6) display the resultant model in the form of contour maps, cross sections, or perspective displays; and (7) iterate on the above process by altering

End_Page 556------------------------------

points on the fault surfaces, defining additional faults, and exchanging points between fault surfaces, etc., until a final interpretation is reached.

The computer performs the tedious tasks of information retrieval, numerical computation, and display generation, whereas the geologist uses his specific knowledge of the problem at hand to evaluate results and propose alternatives until a satisfactory interpretation is reached. The interpretations made in this interactive environment can be geologically and economically superior to those reached by entirely manual or entirely analytic methods.

End_of_Article - Last_Page 557------------