Risk and reliability of exploration methods used to define a heavy mineral sand deposit in Kwa-Zulu Natal, South Africa

Abstract

The objective of the treatise is to review the sample methodology and the exploration methods at a heavy mineral sands deposit. The level of confidence and the risk associated with the resource calculations will be established based on the sample methodology. Two drilling methods used at Hillendale are compared statistically and visually. It is clear from the comparison that the Wallis Aircore (WAC) method is significantly better than the Reverse Circulation (RCN) method. This is evident also when comparing results of the two methods with the recovery from the mined out areas. The WAC drilling method reduces the risk of downhole contamination due to the change in the drilling method allowing air to be forced back in an inner tube after the sample is collected compared to the air blowout of the RCN method. The WAC method reduces the variability introduced in the sample collection as compared to the RCN method. The average calculated value (for total heavy minerals (THM) and Ilmenite) of the global population for the two methods is similar. The reduced variability in the sample population of the WAC method results in a lower population being required for the confidence interval about the mean at a 95% confidence level. The lower number of required samples results in a lowering of project cost. Blockmodels created from the two drilling methods are compared statistically (mode, median, standard deviation). The global mean of the input data compares well with the global mean of the estimated blockmodel. In reconciling the resource blockmodels with actual production data, the variability as introduced by the RCN drilling method is clear. The WAC drilling information produces a blockmodel with an acceptable level of variance (deviation of the geological model from the actual values obtained during mining) as defined by Hillendale mine management of less than 10%. This is determined by reconciling the estimated geological blockmodel for a specific mined out area against the actual tonnages and plant recovery values achieved in the same area. Ultimately, because the drill spacing is standard in the mining operation under study, the required confidence in the mean might not be achieved by either of the two drilling methods in certain portions of the mine, but the results from the WAC method will be closer to the reality than the RCN method with the same number of drillholes. Three main risks are identified in the sample methodology; they have a high probability to occur or have a high cost impact. The risk inherent in sampling and analysis is the most important of the risks identified in this study. The risk can be effectively reduced by implementing a quality assurance and quality control programme (QAQC). The other high risks are introduced by the drilling method and drillhole spacing. The risk can be reduced by continuous improvement and keeping up to date with new developments in the industry such as improved drilling techniques and by improved knowledge of the ore-body. The knowledge will help in understanding the risk - It may become clear in the text how this can reduce the risk. In conclusion, the improvement and implementation of systems such as improved drilling technique and quality assurance and quality control programmes enables one to establish an acceptable confidence level in the resource calculation as well as reducing the inherent risk to an acceptable level for future decisions. Copyright