DEVELOPMENT OF PYTHON BASED VERTICAL ELECTRICAL SOUNDING AND AZIMUTHAL RESISTIVITY SURVEY DATA PROCESSING PROGRAM EQUIPPED WITH GRAPHICAL USER INTERFACE

• Main Author: Atmaodi, Berith
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/61807
• Institution: Institut Teknologi Bandung

The physical properties of underground material are generally anisotropic, which are quantified by the anisotropic coefficient. The presence of fractures is one of the possible causes of these anisotropic properties. By studying the variation of the anisotropic coefficient with depth, one might learn about the fracture orientation beneath the earth. One of the many geophysical methods developed to learn about anisotropic properties under the earth is the azimuthal resistivity survey (ARS) which is the improvement of the vertical electrical sounding (VES) method. Measurement is done in one dimension with various electrode spacing and azimuth of electrode lines. The horizontal anisotropic coefficient will be illustrated by the difference of measurement results from the same point of measurement and electrode spacing, but for the different azimuth of the electrode line. Measurement data is plotted with a polar plot to express the anisotropic property under the surface. The anisotropic coefficient calculation is done by carrying out the one-dimension inversion of the measurement data from every azimuth, for a user-defined number and constant layer thickness. The algorithm used to create the polar plot, ellipse fitting, and to conduct the Schlumberger configuration VES data inversion is combined in one program based on Python programming language. This program is equipped with a graphical user interface (GUI). The test result with synthetic VES data shows that the program can model noise-less data well, but it has its limitations with data containing noise, and models consisting of a thin layer. The test result with synthetic ARS data shows that the program is capable of calculating the anisotropic coefficient from noiseless data (the largest error is 0.1), with the largest error in fracture orientation prediction is 11.76o. The test also shows that the use of a solution model with a large number of layers and thinner layer thickness will result in a more continuous anisotropic coefficient variation calculation with depth