Interpolation in the context of mining and mineral resource estimation is a mathematical and geostatistical process used to estimate the value of a variable — such as ore grade, density, hardness, or geotechnical parameter — at an unsampled location within a deposit, based on the known values measured at surrounding sample points. Because it is physically and economically impractical to sample every cubic meter of an ore deposit, interpolation methods are used to construct continuous three-dimensional models of grade distribution and geological attributes from discrete sample data collected through drill holes, channel samples, and blast hole assays. Interpolation is a foundational technique in resource and reserve estimation for bauxite, gold, iron ore, and diamond deposits.
Multiple interpolation methods are applied in modern mining geology, each with different underlying assumptions, computational approaches, and applicability characteristics. Nearest-neighbor interpolation assigns the value of the closest sample to each estimation point, producing blocky, discontinuous models useful for checking purposes. Inverse Distance Weighting methods weight sample contributions inversely proportional to their distance from the estimation point, providing smooth and easily computed estimates but without statistical optimality guarantees. Kriging — the industry-standard geostatistical interpolation technique — uses a variogram model of spatial correlation structure to compute minimum-variance, unbiased linear estimates with quantifiable estimation uncertainty, expressed as kriging variance.
In bauxite deposits, interpolation of reactive silica and available alumina grades is critical for processing plant feed quality management. In gold deposits, high nugget effect in gold grade distribution makes variogram modeling and kriging parameter selection particularly challenging, requiring careful attention to outlier treatment and compositing strategies. In iron ore, interpolation of Fe, SiO2, Al2O3, and moisture across large stratiform deposits supports blending optimization. In diamond deposits, interpolation of diamond frequency and value models determines project economics, though the extremely heterogeneous distribution of diamonds requires specialized simulation methods. Interpolation results underpin resource classification, mining dilution estimation, and mine planning decisions.