Geostatistical Modeling of Ore Deposits with Transitional Boundaries

In mineral resources and ore reserves estimation, a crucial point is the geological domains that will be used for the modelling, as well as the type of boundaries between these domains. The most common geostatistical techniques are based on the assumptions of stationarity of the variable within estimation domains considered hard boundaries (sharp contact between geological units). However, in most cases, the geological mechanisms that generate a deposit are transitional -overlapping in geological units- in nature. In transitional boundary deposits, each geological zone has its own mineral grade distributions and spatial variability, but with an overlapping between geological zones. Hence, any method for estimation models, affects the mine planning with a significant sensitivity particularly in transition areas. Due to this point, the identification of the exact boundaries of mineralization is essential for an accurate estimate of resources. The objective of this dissertation is to develop a methodological framework to be used in presence of transitional boundaries. The methodological framework is introduced and then explained through a case study with transitional boundaries. Moreover, through a mining case study, it will be shown how choosing appropriate methodology for modelling variables and for interpreting the deposit geology will help to optimize parameters identification. The methodological framework in general allows decreasing the uncertainty in resources estimation and reserves selection. The method is general and can be used in other field of geoscience that incorporate numerical modelling, such as environmental modelling, petroleum or mining industry where complex geology deposits should be characterized