Orebody Knowledge From Calculated Mineralogy, GRE46 Gold Deposit, Australia

The Cowal Gold Operation in central New South Wales (Australia) has recently commenced production from the underground GRE46 gold deposit. GRE46 is expected to contribute a significant portion of the site's future gold production, which was previously derived exclusively from the E42 open pit. This new, more variable ore-type introduces additional complexity to the processing circuit. Deposit-scale quantification of the mineralogical variability that underpins rock physio-chemical behaviour during the extraction cycle is therefore a critical aspect of orebody knowledge that reduces uncertainty and technical risk. In this study we demonstrate a workflow to develop, refine and subsequently verify a spatially constrained three-dimensional calculated mineralogy model for GRE46 that can be used to correlate with comminution characteristics that may influence the processing performance. Here we present the results of two element-to-mineral conversion models using common multi-element geochemical assays with increasing levels of geological input, and compare the calculated mineralogy to mineralogy data determined by X-ray diffraction. The initial model adopts nominal end member and stoichiometric mineral compositions, while the refined model incorporates domain-specific mineral solid solution compositions for chlorite and white mica measured by laser ablation inductively-coupled plasma mass spectrometry. The accuracy of predicted white mica and chlorite content is improved substantially between the initial and refined models relative to the measured X-ray diffraction data (white mica R 2 0.61 initial vs 0.90 refined ; chlorite R 2 0.3 initial vs 0.67 refined ). This work highlights the importance of appropriate geochemical domaining and mineral solid solution chemistry on the predictive accuracy for several bulk minerals at GRE46.