Table 5 Block model limits 2.11 Resource Estimation (OK)
Related Figures (24)
Fig. 1 A Geological Map of Burkina Faso showing the Study Area (Anon., 2007) ‘able 1 Summary Statistics of 1 m Au Composites Table 2 Directional Semi-variogram Parameters 2.9 Indicator Variography Downhole experimental semi-variograms were computed and modelled. A two-structure spherical scheme was found to be appropriate for modelling the semi-variograms. The downhole — semi- variogram was to provide an estimate of the nugget variance since it was calculated from a small distance along the paths of drill holes. The downhole semi-variogram is illustrated in Fig. 6. Horizontal experimental semi-variograms were also computed and modelled with a two- structure spherical model along strike and across strike as shown in Figs. 7 and 8 respectively. Lag spacing and angular tolerance were selected appropriately to capture more samples. From the semi-variogram models, the continuity of mineralisation was not the same in all the three orientations which means that the mineralisation is anisotropic. The direction of maximum continuity is along strike. Table 2 shows the directional semi-variogram model parameters where Cp, C,, Cz, a, and az represent the nugget variance, first sill, second sill, first range and second range respectively. Fig. 7 Directional Semi-variogram along Strike fable 4 Indicator Semi-variogram Parameters for 0.25 g/t Cut-off 2.10 Block Modelling Table 3 Interval Statistics of the Dataset Fig. 9 Indicator Semi-variogram Downhole for 0.25 g/t where Z(u) is a sample grade at u location and I(u; Zx) is the threshold grade defined at a location of u for a cut-off of Z,. Samples of such model for 0.25 g/t cut-off are shown in Figs. 9, 10, and 11, and the parameters are displayed in Table 4. ‘ig. 10 Indicator Semi-variogram along Strike for 0.25 g/t Table 6 Search Parameters for Ordinary Kriging Estimate Fig. 12 Cross-sectional View of Validated OK Model along N-S Section 21200 mE Fig. 13 Swath Plot of Drill Hole Grades and OK Model at 20 mE Table 7 Summary of OK and MIK Global Resource Estimates Fig. 15 Swath Plot of Drill Hole Grades and MIK Model at 20 m Fig. 16 compares the block model grades for OK and MIK at an elevation of 5 m. The ordinary kriging model grades are consistently lower than the drill hole grades except at 320 m, 335 m and 395 m where the block model grades are higher than the drill hole grades. At these locations, there is overestimation of grades by the ordinary kriging model. Underestimation of high Au grades is also common but it is so much at the extreme right end of the plots where high Au values are observed. Overestimation of low Au_ values and underestimation of high Au_ values indicate smoothing effect on the ordinary kriging grade estimates. Even though, the multiple indicator kriging grade estimates are also lower than the drill hole grades at some points, they are more close to the drill hole grades than the ordinary kriging grade estimates. It means that the multiple indicator kriging model gives estimates which are close to the actual grades. Comparatively, the multiple indicator kriging model is seen to give more accurate recoverable resource estimates than the ordinary kriging model. The study also established a_ grade-tonnage relationship, which relates the tonnes of material to its grade above cut-off grades. In Fig. 17, the multiple indicator kriging model produces higher tonnage of material with higher grade than the ordinary kriging model. The grade curve for multiple indicator kriging is somewhat above the grade curve for ordinary kriging except at cut-offs of 2.0 g/t and 2.5 g/t where the average grades are very close. Considering the four cut-offs used by the mine, the multiple indicator kriging model yields high ounces of gold more than the ordinary kriging model. Using the swath plots together with the grade tonnage curve, it can be said that the multiple indicator kriging model gives recoverable resource estimates of the deposit, where the distribution of grade is highly skewed. Mathematical Statistics. Prof. Sulemana Al-Hassan is an Associate Professor in the Mining Engineering Department at the University of Mines and Technology, Tarkwa. He holds BSc (Hons.) and Postgraduate Diploma in Mining Engineering degrees from the University of Mines and Technology, and PhD degree from the University of Wales, Cardiff. His areas of specialisation include Mineral Reserve Estimation, Mine Planning and Design, Mineral Economics. He is a member of GhIG and a consultant in Mineral Resource Management, Environmental Management and Small-scale wining Andrew Owusu Asante holds BSc in Geological Engineering from the University of Mines and Technology, Tarkwa, and MSc in Mineral Exploration from the Kwame Nkrumah University of Science and Technology, Kumasi. Currently, he is the Unit Manager, Mine Geology, Asanko Gold Mine, Ghana. He is experienced in mineral exploration, resource estimation and mine vgeolosv.
