Ore Reserve Estimation at the Feasibility Study Stage - The Period of Greatest Risk

Gold mineralization like other low concentration high value commodities has gained the notoriety of being difficult to sample. This adversely affects the confidence that can be placed in resource estimates and ore reserves. As an example, it is not uncommon to find differences of the order of 30% between duplicates of a single sample sent to two different commercial laborartories. Two halves of drill core generally show marked differences in grade. Again differences of 50% are not uncommon in halved core. In the context of this type of "error" within the base data used for the resource estimate, many of the subtleties of computation methods and expert judgement of resource categories, are almost a case of shutting the gate after the horse has bolted. Managers, financiers and consultants must question the quality of the base data upon which resource estimates are made. They must insist that standard and duplicate samples are used to measure the accuracy and precision of commercial analytical laboratories, that sampling practices have been checked for possible bias and give tolerable repeatability. Further they must ensure there can be a realistic confidence in the estimates by correctly using these measures of data quality to quantify risk.