The process of estimating mineral resources is associated with a degree of uncertainty; however, recent research has suggested that overestimation of ore grades is a common problem in the mining industry. The research undertaken between 2020 and 2024 has suggested that geological modeling and estimation methodology are likely to overestimate ore grades by 20% or more.
The major reason for overestimation of ore grades is that structural geology is not properly considered in geological modeling. The research undertaken on structurally complex gold deposits in the Birimian terrane of West Africa has suggested that overestimation of ore volume is likely due to the lack of proper 3D modeling, and such overestimation of ore volume is likely to result in overestimation of contained gold ounces. The overestimation of ore volume is likely due to inappropriate connections between grade intervals in 3D models, as simple models are likely to improperly link intervals between drillholes (Stoch et al., 2022).
The accuracy of estimating ore grades is also compromised during the measurement process. The research undertaken on X-ray fluorescence (XRF) sensing in bulk ore sorting has suggested that three types of errors are associated with this process. These include instrumental errors, heterogeneity errors, and sampling errors. Out of these three types of errors, heterogeneity errors are considered more significant due to their unpredictable nature. The measurements are often taken on the surface of the ore; however, this is likely to result in a degree of discrepancy due to shedding of minerals during the crushing process (Li et al., 2021).
Human judgment also contributes to the variability. A study conducted in 2024 to evaluate the judgments made by different Competent Persons with the same datasets showed that subjective assumptions in resource classification parameters may result in considerable variability in the grades obtained. The subjectivity in classification parameters, coupled with the lack of universally accepted classification systems, may result in resources that are Indicated for one Competent Person being classified as Inferred for another (Owusu & Dagdelen, 2024).
Fortunately, some methodologies have demonstrated the possibility to overcome overestimation bias. For example, in the context of highly skewed grade distributions in gold deposits, it has been demonstrated that the application of machine learning algorithms such as Gaussian Process Regression with logarithmic normalization may be superior to the application of traditional algorithms such as kriging in estimating the grades while reducing the overestimation propensity (Zaki et al., 2022).
References
Li, G., Klein, B., Sun, C., & Kou, J. (2021). Lab-scale error analysis on X-ray fluorecence sensing for bulk ore sorting. Minerals Engineering, 164, 106812. https://doi.org/10.1016/j.mineng.2021.106812
Owusu, S. K. A., & Dagdelen, K. (2024). Impact of Competent Persons’ judgements in Mineral Resources classification. Journal of the Southern African Institute of Mining and Metallurgy, 124(7), 371–382. https://doi.org/10.17159/2411-9717/1538/2024
Stoch, B., Basson, I. J., Gloyn-Jones, J. N., & Lomberg, K. G. (2022). The influence of variable anisotropic search parameters on implicitly-modelled volumes and estimated contained metal in a structurally-complex gold deposit. Ore Geology Reviews, 142, 104719. https://doi.org/10.1016/j.oregeorev.2022.104719
Zaki, M. M., Chen, S., Zhang, J., Feng, F., Khoreshok, A. A., Mahdy, M. A., & Salim, K. M. (2022). A Novel Approach for Resource Estimation of Highly Skewed Gold Using Machine Learning Algorithms. Minerals, 12, 900. https://doi.org/10.3390/min12070900
