Abstract
In the case of complex geological models, these tools do not give satisfactory results and new approaches are necessary. In this talk, we explore the possibilities offered by a new approach based on the use of geostatistical tools to create non-stationary models of the attributes. The approach offers significant flexibility to create a variety of features in the resulting models. These models are the first step into the creation of digital models of the Earth, capturing complex connectivity and locally changing features, and generalize the current tools available in the geostatistical toolbox. They are based on the intersection of random fields, that is, probabilistic models of the spatial distribution of the attributes. Depending on the specific geometries of these fields, results show locally varying anisotropies and non-stationarities that are closer to what we see in real deposits. We discuss the potential and challenges of this approach and provide an outlook of their use in building digital models of ore deposits.