Orebody Solid Modelling Accuracy--A Comparison of Explicit and Implicit Modelling Techniques Using a Practical Example from the Hope Bay District, Nunavut, Canada

摘要

Accurate geometric models of orebodies establish the framework for reliable mineral resource estimates that provide a basis for further exploration and mine planning, as well as for financial models that ultimately determine the economic viability of a deposit. Traditional, explicit geometric modelling techniques that are utilised for resource modelling by the general mining software packages, such as MineSight~R , involve manually digitising interpreted ore outlines that must be carefully linked to create valid geometric solid models. An alternative, implicit modelling method, employed by Leapfrog? software relies on a fast method of global interpolation, which offers several advantages over traditional modelling techniques. Although previous case studies have documented the relative efficiencies of the two approaches, the accuracy of the two methods has not been previously quantified and compared. This study quantifies the accuracy of solid models created using explicit and implicit methods, and compares the differences between the models. Solid models of the Doris Hinge vein gold deposit (located in the Hope Bay district, Nunavut, Canada) are constructed from preliminary and follow-up drilling data using both modelling techniques. The accuracies of the solid models generated using preliminary drilling data are determined by measuring the 3D distance between the surface of the models and the actual contact location defined by the follow-up drilling data. Additional comparisons based on volume, cross-sectional area and surface differences were made between both phase 1 solid models and solid models constructed from phase 2 drilling information. The overall results of this study find that the MineSight~R and Leapfrog~(TM) solid model accuracies are virtually the same, and that their volume, cross-sectional area, and surface differences are comparable. This study concludes that the implicit modelling technique generates accurate solid models that can be used in place of explicit models as ore shells for resource estimation in certain circumstances.