A different view on statistics.
With the disclaimer that experience is nearly always more definitive than projections based upon statistics, with regard to diamond mines in the making we have no other choice but ot go with the statistics.
A large number of small samples from various points in the resource may actually give us excellent indications of what the resource as a whole will provide.
Large diamonds in the small samples are not neccessary.
The actual size of the individual diamonds, the number of diamonds of each particular size, the difference between the sizes, and the difference between the size of the sample and the total weight of the diamonds recovered would tell us just about everything that we need to know.
It is reasonable to believe that the NW dyke is the result of a single emplacement event because of the narrow width of the dyke.
If point A is futherest from the source, it would be impossible to emplace B,C,D, or E after A was in place because the space occupied by B,C,D, and E would already be filled by the same material as A.
Comparing the samples from various points on the NW dyke would confirm this theory. We would then know for certain that we are dealing with a single emplacement.
A single emplacement Kimberlite which has cooled under similiar conditions (due to the narrow width of the dyke we can safely assume this) should be very uniform throughout with regard to diamond content. The size and quality of the diamond crystals is relative to the carbon content of the source and the heating and cooling conditions experienced in the emplacement process.
Kimberlite material with diamond content, emplaced and cooled under similar conditions will yeild a logarithmically predictable size distribution of diamonds. The grade wil be relative to the actual percentage of diamond content in the material.
The diamond gurus may tell you the actual formulas that they use for diamonds but we can all understand the mathematics.
Just as an example, we may find that for every 10 stones of 1 carat in size we may expect to find one stone of 10 carats and for every 10 ten carat stones we may find one of 100 carats.
We can expect this relationship to remain constant throughout the deposit. That would be the logarithmic relationship. The grade should be stable throughout the dyke as well.
By analyzing the logarithmic relationship of the microdiamonds we can extrapolate and predict the entire deposit. One could attempt to do this with one very small sample. It is not the actual weight of the sample that matters- it is more a matter of having enough diamonds in the sample to be accurate.
We could put all of our small samples together and process them as one larger sample or we could analyze each sample on it's own and then get russet to apply standard deviation analysis.
If a large percentage of the samples all gave similiar indications of grade and size distribution we may have more certainty than if we lumped them all together.
Basically the same priciples and mathematics will be applied to the macro distribution of a single large bulk sample-but the analysis of micro diamonds from a large number of small samples from various points in the dyke may actually have more mathematical reliability.
If the bulk sample projections match the projections from the micro analysis that would be even better.
russet will explain more fully. I hate statistics.
|
