I am involved in a diamond sedimentary deposit operations. The mine has been in operation for some years now. However when I joined the mine on the geology side, I realised that geo -statistical information was void and non existent at all. This is a given opportunity for me to do a thorough research, one of its kind on such a deposit. My challenge is all the variables( grade, litho thickness, SG etc are not consistent. That is expected of such a deposit. For a example to come up with a grade model is difficult because bulk samples can only be taken concurrently with mining operations due to the nature of the deposit. For example it is very common to get highest grades on a monolithic (5cm) conglomerate than a 1m thick conglomerate layer. What is the best way to go about considering that the number of samples do not even reach 100.