Variability of geological interpretations of a mineral deposit: Quantitative study and implications

Assessing uncertainties is an essential step throughout the mineral resource development cycle, from exploration campaigns to development planning and remediation. These uncertainties reflect the scarcity of subsurface data, which need to be complemented by geological knowledge. This paper aims to assess the impact of conceptual variations on the geometry of uranium deposits. For this purpose, a case study was carried out on the interpretation of drilling data in a 2D section, in the context of an unconformity-related uranium deposit in the Athabasca Basin. Based on a reference section from this area, a blank cross-section with synthetic drillcores was produced and given to 32 interpreters. These 32 interpreters are from various backgrounds and are divided between experts in uranium geology (8) and non-experts (24). A set of fifty mathematical criteria is defined and correspond to five categories: mineralized zones, associated altered zones, structural network, geometric relationship between structures and mineralization, and annotations. Individual and group analyses of the defined criteria are performed. Results show that uranium experts and non-experts are difficult to discriminate by the metrics used in this study, but experts tend to filter out the most speculative conceptual models. A generalized parsimony of interpretations is observed for all survey respondents. What also emerges is that fault network connectivity is key to conceptual model discrimination in hydrothermal deposits. • Quantitative study of hydrothermal deposit dataset interpretation. • Comparative study of different groups of interpreters. • No clear differences between experts and non-experts in uranium deposits. • High-school students are easily distinguishable. • Experts tend to propose fewer outliers than the overall study group. • Interpretations tend to be parsimonious and it leads to conservative estimates. • Key role of characterizing fault network connectivity in hydrothermal deposits.