Integrated interpretation of geology and geophysics, using inversions, to predict geology under cover

Abstract

Predictive 3D geological models of the subsurface can be developed using a range of available tools. Each tool is suited to slightly different problems and datasets. The method described here, using the UBC-GIF inversions algorithms, allows rapid development of models using an objective, automated procedure. It has flexibility to include as little or as much geological information as is available, making it ideal for greenfields exploration or mapping programs. The steps involved are: 1) develop a solid understanding of the expected physical properties; 2) convert geological observations into physical property constraints; 3) perform geologically-constrained inversions; 4) apply geological classifier to recovered 3D physical property models. The procedure is demonstrated for the southern Agnew- Wiluna greenstone belt in WA, a highly mineralised region with a high proportion of surface cover. The predictive 3D lithology models developed for the area are particular effective at mapping the extent of dense mafic and magnetic ultramafic rocks, and provide new insights about their distribution at depth.