This is going to be a short treatment because I don’t believe that anyone really understands the topic of geostatistics. Geostatistics is the art of applying specialized statistical methodologies to predict the value of something in three dimensional space based on the values of that something located more or less close to the point of interest. For example, if you are walking along the sidewalk and you notice a dime sitting a crack and at the next crack you notice two dimes and the next crack you notice three dimes pretty soon you say, “Wow, there is a pattern here and the next crack will have four dimes” If it does then you will conclude that you are the smartest person in the world for being a bit richer and you will start speculating on the number of dimes in the hundredth crack. We like to create models remember. 

Now replace the sidewalk with a drill hole and the number of dimes with assay values of gold along the drill hole and you can see why we use statistics in creating models. Geostatistics is an elegant way of predicting the grade of the rock immediately east, north, west and south of the drill hole. It is a way of looking for patterns in grade distribution (or some other value; porosity, hardness etc.) and then making predictions of unsampled points. But consider this; geostatistics was created by the French who also gave us Post-modernism. It is widely rumoured that the French father of Post-modernism did it as a joke and everyone thought he was serious. Could this be the case for geostatistics as well? “Oh those French!” ... if you know what I mean. I am not saying this is the case but it does warrant some thought.

Geostatisticians use terminology like semi-variograms, range, Kriging etc. These are weirdnesses used to create a sense of the arcane in the field as a way of keeping the numchucks out. It is a tactic used extensively by shamans of all primitive religions and it works like a charm. “If you use the semi-variogram incorrectly not only will your model be incorrect but you hair will fall out.” Works for me and I always have an expert evaluate my models for geostatistical purity. Why take a chance on something as important as my hair? And I might add that, in my observation, most successful geostatisticians have full heads of hair. Laugh if you will but it is true.

So all the data that is carefully gathered and put into databases, crosschecked and fed into computers is now converted into mathematical models that attempt to replicate the location of assay values in space. It is a pretty tall order as you can imagine and one of the most important functions of the model is to tell you how much more data you need to collect so that the model is more accurate. It always means more drilling but that is how economies are created. Once the model is created and the geostatisticians think the error in the predictions can be lived with there are so many really cool things to do with it. It becomes the basis for everything that follows in terms of designing the mine, the metallurgical plant and the retirement of the company executives. Geostatistics is a very useful tool - regardless of the foregoing.

Having a model of the mineral occurrence leads inevitably to a calculation of the tonnes and grade of the occurrence. This is called a resource estimate which then becomes a reserve estimate. I’ll try to explain that process now.