The Importance of Grade Control and Mining Dilution What went wrong in the 2015 Resource estimate ?
Very thin veins require closely placed drills ( grade control ) to ensure that the Kriging model does not show mineralization where it isn't .
The 2015 Resource was based on 2.5 m on either side of the narrow veins.
The updated grade control showed that this level of mining dilution interpolated gold where it did no exist.
The appropriate dilution factor for spatial interpolation at San Albino was just 0.5 m on either side of the veins.
Hence substantial collapse of the size of San Albino gold reserves when updated in 2020.
I know what the mining dilution factor is at Goldboro after more research.
Goldboro is a thin vein resource .
Perhaps the mining dilution factor is appropriate for Goldboro.
In my view, the consequences of dilution error are more on the side of over estimation than under estimation .
So, what happened at Point Rousse when grades turned out to be just half of what was expected?
Lack of adequate grade control....ie dilution factor was too high.
This is why I sold all of my shares ....risk aversion.
It may have been an incorrect decision but I sleep better after doing so.
Check out the Goldboro mining dilution factor in the technical report on Sedar.
It will show you the direction of errors on grade and estimation results..
Dilution
Mining dilution is an important factor at San Albino in part because the vein is, in some places, so thin that the dilution can render it uneconomic, but also because the operation requires very detailed grade control. A helpful feature of the vein is that in many places it has distinct hanging wall and footwall contacts, which aid with grade control. This mineral resource includes an approximately 0.5 m rind of dilution both above and below the vein. The dilution grade varies from zero to averaging up to 0.5 g/t Au. In all cases, the dilution grade is taken from the mineral resource estimate and is not a singularly applied grade. It is expected that underground dilution will be greater because of ground conditions, shallow dip, minimum mining height, and less control on locating hanging wall and footwall. MDA is reporting full-block-diluted grades for the underground for all material within the optimized stopes at a mining cutoff of 2.5 g/t Au (blocks are 1.0 m high by 1.0 m across strike and 2.0 m long).
Potential for Increased Grade
Clustering of high-grade data necessitated use of a quadrant search to minimize spreading out the high grade. This clustering is clearly demonstrated by the difference between Measured versus Indicated mineral resource grades. While higher grade in higher-classification material is not unusual, at San Albino the effect was exaggerated by normal exploration and follow-up drilling to better define the higher grades. Given that history, both the Indicated and Inferred mineral resource grades may increase as well.
Post Model Performance
After the effective date of the current mineral resource estimate, there were seven new holes with assays in the mineralized zone or projection of the mineralized zone. Four of those holes drilled areas where the model predicted no mineralization, which was confirmed by this later drilling. The other three holes were drilled into areas where mineralization was expected. One intersected good grade precisely where mineralization was expected, although the model had interpolated that area as low-grade halo mineralization. The second hole intersected mineralization with good grade where adjacent holes on either side of a fault were lower grade, however the location of the mineralization was substantially different than expected as the mineralized zone had been offset by faulting. The area around this fault is classified as Inferred mineral resources. Indeed, all mineralization around these faults is classified as Inferred mineral resources because of this known lack of certainty of location. The third hole intersected the vein where predicted and encountered better than expected grade.
Importantly, San Albino is open along strike and at depth, and with additional drilling, MDA believes that mineral resources are likely to be upgraded and expanded.
2015 Resource Estimate
The 2015 PEA estimated mineral resources totaling 152,000 Indicated ounces of gold and 787,000 Inferred ounces of gold. This updated mineral resource estimate totals 177,800 Measured and Indicated ounces and 100,900 Inferred ounces of gold. Comparing the 2015 and 2020 models, it was evident that the general location of the 2015 mineralized zone was modeled appropriately with respect to locations of the principal domains albeit generous in thickness and extrapolation. Several 2015-modeled solids depicting vein mineralization have since been drilled showing that mineralization does not exist where modeled in 2015. Another reason for the difference is that MDA took a more conservative approach to modeling extrapolations and projections from drill data, where no drilling was done since 2013. The 2020 volume of modeled veins and veins-plus-halo were 89% and 46% lower, respectively, compared to the 2015 single-domain volumes.
An important difference between the 2015 and 2020 models is the interpretations of the mineralization. The vein and vein zones (both averaging >15 g/t Au) have distinct, sharp-bounded contacts usually within low-grade halos. By modeling these veins and halos as one in 2015, the extreme high-grades in the vein were spread out into the larger volume of the halo in the estimate, thereby over-estimating metal, albeit at lower grades than exist in the vein and vein zones.