Subsurface mapping: selection of best interpolation method for borehole data analysis

Abstract

Selection of the best interpolation method as well as the appropriate software parameters that give the closest approximation to known geology is usually a difficult task. This is because geologic factors such as lithologic and structural feature occurrences are not easily predictable, especially where outcrop occurrences are fortuitous. Borehole data provides an alternative to sparse outcrops especially in areas where the borehole occurrence is well-distributed. However, most boreholes were drilled for exploration purposes and may not be well-distributed over all the study areas. Getting information in-between boreholes becomes a big challenge, especially in areas where we have very large volume of borehole data that are randomly distributed. The use of interpolation technique to estimate in-between borehole points becomes imperative. Comparison of interpolation techniques was carried out with borehole log data so as to determine the best interpolation method output that is closest to known geology. The test area is located in the southwestern part of the Western Bushveld Complex and the northeastern part of the Eastern Bushveld Complex in South-Africa. Eight interpolation tools available in Rockworks® 15 were evaluated and utilized in contouring the major stratigraphic peaks within the area. The best results were achieved with Kriging and trend surface analysis interpolation methods. The resultant maps were interpreted and compared with geology and structures inferred from existing geological and geophysical records and a good correlation is present.