A three-dimensional model for the Leeuwpoort tin mine and its application to exploration prediction

Abstract

Due to tighter financial margins, the need for better knowledge of grade data is required. Geological models allow the user to have a better understanding of the geological environment in which mining is taking place. Three dimensional (3D) geological, ore deposit, and mining models are therefore becoming increasingly important in the mining industry. 3D models are being used more frequently for mineral potential targeting, as well as resource assessment, because good quality geological models allow the user to determine grade domains within mineralized environments. The aim of this thesis is to create a geological model for Leeuwpoort Tin Mine (C-Mine) and test the applicability of this 3D model for exploration purposes. The 3D geological and interpolant models created of Leeuwpoort Mine was used to determine the probability of intersecting a lode of economic interest, if 500 drillholes were randomly drilled in a specific boundary. Resampling was conducted using the Bootstrap method, in order to determine how the probability will change as different borehole sample sizes are used. When conducting a quantitative resource assessment of mineral deposits, grade-tonnage models form a fundamental part in the estimation and prediction process. Grade and tonnage models are used during quantitative resource assessment to predict the values of the known deposits for a specific type, and can also be used to determine the potential value of undiscovered deposits in a specific area. If sufficient geological data is available, the tonnage of mineralized bodies can be calculated and a grade-tonnage model can be created from the 3D geological model. 3D models can be used as a summary and visualization tool for geological environments. The 3D visualization of deposit give a much better representation of the orebody than 2D cross-sections from a few drillholes. The effectiveness of 3D model, as a possible tool for mining, is limited by the quality and quantity of the data. Poor quality data will result in poor quality models, whereas limited data will result in a higher level of uncertainty of the estimates based from these 3D models. However, even limited data can be used to visualize geological environments. In the case of Leeuwpoort Mine, limited structural and lithological information was available to create the 3D models. However, the provided peg index noted the positions of the mine pegs, and consequently indicate where the lodes were mined. In this instance, with limited mine peg data, the lode “volumes” could be reconstructed. The peg index was used to delineate the mining area in order to model the lodes, as well as interpret geological features. Once the lode “volumes” were created in the 3D model, the specific orezone with the geometric relationship between individual lodes of Leeuwpoort Mine could be defined. In addition, the probability of intersecting a certain number of lodes was derived from these modelled surfaces, resulting in the estimation for the predicted probability of success. The 3D geological and interpolant models created of Leeuwpoort Mine were used to determine the probability of intersecting a lode of economic interest, if exploration drilling were to be done. Resampling was conducted using the Bootstrap method, in order to determine how reliable this prediction is as a function of number of boreholes. The results obtained from the Bootstrap analysis indicates that the average probability of intersecting a lode of economic significance, for each of the different sample sizes stays the same. A higher level of confidence in the probability of intersecting a lode can be assumed for areas that have large quantities of drilling. If a geological environment similar to that of Leeuwpoort Mine is considered for an exploration project, the chances of intersecting an economical orebody or lode during exploration is very low. When considering Greenfields exploration (virgin exploration), the chances of intersecting a mineralized body is extremely low, and a lot of money has to be invested to obtain meaningful results. However, if Brownfields exploration (exploration on a known area) takes place, the chances of intersecting a mineralized body is higher, because prior knowledge of the area exists and can be used to make informed decisions on where to drill. Less money needs to be invested for the drilling than for Greenfields exploration. In the case of Leeuwpoort Mine, unfortunately even Brownfields exploration will bear a very limited chance of identifying a mineralized body, which means that in such a scenario, the information about potential success rate of a drilling program is even more important for drilling and financial planning.