Application of 3D models and numerical simulations as a predictive exploration tool in western Victoria

Abstract

3D models and computer-based numerical simulations have been used in the exploration industry for some time to visualise the geometry and mechanisms resulting in the formation of orebodies. However, due in part to computational limitations, few numerical simulations have been run on complex (real) geometries in order to predict the location of new ore systems. Presented here are the results of an exploration program developed by the Predictive Mineral Discovery Cooperative Research Centre (pmd*CRC) and MPI (now Leviathan Resources) in the orogenic-gold system of western Victoria that utilised 3D modelling and numerical finite-element simulations to successfully target several new orebodies and predict their geometries and extent. Existing drillcore databases were utilised to constrain the geometries of known deposits and associated mafic domes, the effects of known post-mineralisation faulting was systematically removed and syndeformation fluid flow was then modelled within the system. The results of these simulations were compared with the known geometry of the mineralised systems about these deposits in order to test the simulation parameters and accuracy. 3D models were also developed of poorly constrained target domes in regions with no outcrop utilising potential-field datasets and limited drilling data. Simulations were then run on these model geometries using the tested parameters in order to predict the likelihood of mineralisation in these systems, its geometry and (most importantly) its location. These targets were then drilled resulting in the discovery of previously unknown gold deposits associated with the Kewell Dome northwest of Stawell.

Key words:

• 3D modelling
• FLAC
• fluid flow
• gold
• numerical modelling

Acknowledgements

We gratefully acknowledge the support of the pmd*CRC in funding and promoting the project. MPI (now Leviathan Resources) are thanked for providing access to all available datasets and locations, and for having the confidence in the results of this work to drill the proposed targets, thus testing the validity of the process. David Giles and John McLellan are also thanked for their constructive reviews which greatly improved the manuscript.