Most recent developments in resource estimation have taken place in the following fields:
- Geological modelling: 3D modelling has become much faster and more accurate, by implementing fast interpolation algorithms within specialised software like Leapfrog, GOCAD, Geomodeller. More traditional mining packages have added implicit modelling to their capabilities (Minesight, Micromine, etc.). The advantages of these methods are well known: a true 3D approach (as opposed to the traditional wireframing technique), a very fast creation of models, eliminating the tedious tasks of working section by section and allowing for tests of alternative geologically plausible solutions.
- Domaining: the definition of “homogeneous” (stationary) 3D zones is made easier by the availability of good statistical methods and software (clustering techniques, contact analysis, border effects).
- Geostatistical techniques: while no radically new techniques have appeared in the last few years, progress in computing power has led to a significant increase in the use of non-linear methods like uniform conditioning and conditional simulations for geological variables, such as lithology and grades.
The combination of the modern 3D geological modelling approach and powerful geostatistical resource estimating techniques is illustrated by SRK’s recent evaluation of the Miriam Uranium deposit in Niger for GoviEx.
The Miriam deposit was discovered in 2012, at 400m drill spacing, within the Madaouela 1 exploration license of the GoviEx Uranium Madaouela Project. Since then, the deposit has been drilled intensively (from 200m to 50m grid). The latest drilling phase, at 50m center grid, delivered the current resources used in preparing the preliminary feasibility study of the complete Madaouela Project.
The presence of mineralisation linked with a ‘vertical’ redox front was confirmed and, at the 100m center grid, it became clear that there was more than one vertical, structurally-oriented ‘redox front’.
After the 50m center grid was completed with controls by oblique core drilling, the interpretations showed that the high-grade zones are related to the intersection of discrete expansion faulting with favorable sedimentological units, resulting in the ‘Christmas tree’ aspect of these high-grade zones.
The ore envelope was constructed three-dimensionally, using the interpreted sedimentological controls and the vertical interpreted structures. The model was built within Leapfrog software with two cut-off grades defined shells.
The estimation within the Leapfrog envelope was carried out using a uniform conditioning method, reflecting both the data density and the potential mine selectivity in an open pit. The method is well adapted to estimating the resource for open pit projects at this PFS or FS stage, particularly in cases of relatively complex geometry.
The Miriam deposit appears to be the only one of this type in the region and it opens new ideas for further discoveries in a similar geological context.
