Spatial continuity of geological models must be controlled rigorously because they have a direct impact on the mineral resource estimate. These models typically integrate drillhole assay and geological information—such as lithology, alteration, and mineralization along with some structural data; however, the spatial trends used are often scarce or inaccurate, especially between different structural domains. In addition, modelling databases may have suboptimal codification, and drillhole and blasthole data may not be integrated when available. These issues can lead to simplified models that tend to show under- or over-projection of geological units.
To generate geological models and evaluate their spatial continuity, SRK uses implicit modelling software combined with integrated Python workflows. Leveraging its established geological modelling expertise, SRK has developed a methodology that addresses most of the issues described above.
Modelling databases can benefit from post-processing scripts that apply multielement thresholds or ratios to refine geological codification. For example, clay spectrometry or arsenic grades may help improve alteration coding. For validation, Python workflows can evaluate geological codification by generating clusters using algorithms such as k-means, hierarchical clustering or DBSCAN, and by assessing spatial continuity when coordinates are included as features.
Geological units may exhibit different spatial continuity across structural domains, so these domains must be modelled beforehand to identify the boundaries where local trends control continuity and orientation. Also, implicit modelling relies on parameters such as ranges and ellipsoid shapes, which are often defined based on geological knowledge rather than quantitative methods. These parameters should be validated using indicator variograms to assess continuity ranges and directional anisotropy, thereby determining appropriate ellipsoid shapes.
For post-processing, sample support should be evaluated for each orebody to assess the spatial continuity of geological units. Using Python workflows, mesh diameters along their major, intermediate, and minor axes can be measured while accounting for anisotropy. Sample support can then be flagged per mesh and expressed as a diameter ratio to quantitatively evaluate whether the geological units exhibit under- or over-projection.
Implementing this methodology across different mineral deposits has allowed SRK to assess the spatial continuity of structurally controlled geological models more accurately. Whether the objective is geological modelling or due diligence, this approach provides mining companies with robust feedback and recommendations, helping reduce risk within the mineral resource estimate.
