Geostatistical Inversion of Seismic Reflection Data for the Characterization of Mining Deposits

Youssef Sultane Omar

Key information

Authors:

Youssef Sultane Omar (Youssef Sultane Omar)

Supervisors:

Leonardo Azevedo Guerra Raposo Pereira (Leonardo Azevedo Guerra Raposo Pereira)

Published in

November 28, 2024

Abstract

Seismic reflection methods find their primary application in the energy industry. However, advancements in seismic acquisition, processing and modelling have revealed the potential for its application across various areas. This method offers comparative advantages of affordability and reduced invasiveness when compared to alternative techniques based on direct measurements of the subsurface. As mineral exploration has progressed, significant deposits have been discovered at deep depths and complex geological settings creating a necessity for a cost-effective approach that would obviate the need for extensive drilling to characterize the ore deposits. Motivated by these factors, this thesis applies geostatistical seismic inversion to predict the spatial distribution and the occurrence probability of a given rock type within a mining site. The predicted seismic matches the observed one (real seismic) with a global correlation coefficient of 0.84. The results of the inversion are explored by comparing predictions against observed borehole data. Besides, the predicted acoustic models were classified to quantify the probability of occurrence of a given rock type: Massive Sulfides Unit (2.2%); STWK (2.0%); and Barren waste rock (95.8%). This thesis contributes to the characterization of mining deposits and promotes the application of this method in similar environments. It emphasizes the importance of meticulous data acquisition and pre-conditioning to ensure accurate and geologically plausible results.