Abstract

Geological modeling is an essential step in mineral resource assessment. Geologists typically use deterministic methods to create interpreted models, which ignore the uncertainty in the geological domain layout. This study explores a hierarchical approach for simulating a categorical variable (lithological domain) and a continuous variable (iron grade) in a deposit located in Cameroon, enhancing confidence and accounting for geological uncertainty in resource evaluation. To achieve this, lithological domains are simulated using a non-stationary variant of the plurigaussian model, followed by the simulation of iron grade while accounting for its spatial correlation within and across domains. Cross-validation is used to validate the model, and the simulated block models are processed to assess in-situ and recoverable resources and their uncertainties.