I will "simplify" the whole geological geostat thing for you. No bull. OK. Complicate it truthfully.
It's from McTaggart page 134. "No assay or group of assays can be relied upon without comparison of that ore block to a bulk sample of the ore in question processed by a mill of a reliable method."
This paraphrase is slightly circular, if you think about it. As how do you know the mill got the gold unless you assay, and how do you know the assays unless you mill. hint you assay the mill stuff too. ALWAYS RUN GRAVITY CONCENTRATION, STUPID, AS OTHERWISE YOU WILL NEVER KNOW WHAT YOU MIGHT HAVE MISSED.
Kreiging: "When sampling becomes very sparse the kreiging method degenerates to arithmetic averageing."
SAMPLING BY DEFINITION IS 'ERRATIC' AND SPARSE. USE ARITHMETIC AVERAGING, STUPID. AND DRILL ENOUGH HOLES SO YOU CAN STOP SCRATCHING YOUR HEAD AND MINE THE STUFF.
Geological structure is what defines the areas where gold precipitates. If the structure is not there, either the gold values increase as it narrows, or the increase as it blows out (boiling releases gold, asp releases gold, finer structure are better grade, wider ones are better..). You have to know that. KNOW YOUR STRUCTURE AND YOUR DEPOSITIONAL HABITS OR STOP COUNTING. THE ORE IS THERE FOR A REASON. IF IN DOUBT HIRE A REAL GEOLOGIST.
Where drill holes are sparse and grade varies widely, and width does too, it is true that 1./10 of an ounce over 1 foot, AT 100 FOOT HOLE SPACINGS where the average width is 4 feet and the average grade is really 0.50 ounces IS NOT 800 TONS OF 1/10 OF AN OUNCE, IT'S 3000 TONS OF HALF AN OUNCE, YOU IDIOT.
How do you know this if using the figures is what you are doing to establish the grade..?
Glad you asked that. You first establish what the average width is. Then you look to see if the closely related info can quickly change spatially the width or the grade or that width changes do not carry for a distance. If it can, then the high variability of width and grade does not allow one to multiply width times grade to get grade-width figures over large sectional blocks, unless the width and grade ARE PROVEN TO STAY CONSTANT OVER THE ENTIRE SECTION AS A RULE. What are saying here is that pinch outs may not be pervasive structurally for volume calculation purposes, and grade drops may not carry over the distance either. In this case the grade and width are independant for calc purposes, and have no such power to carry over a section. In this case the width and the grade MUST be AVERAGED SEPARATELY. This is an important part of Kreiging too.
If grade and width are dependent, then grade time width becomes important. It is then necessary to establish the relation of grade to width in order to calculate ounces. Width is still averaged separately and the appropriate factor for grade applied. In this case it may be that the wider the vein, the better the grade. The true volume of the blow out veins must be calculated versus the pinch outs. This is very hard stuff to calculate as there may be no arithmetical rule that is easy to apply.
A rule is the average assay in the ore zone, which is arrived at by inference not by calculation, is the average assay you will mine, REGARDLESS OF THE WIDTH OF THE ASSAY. This is *not* operative in cases where the nugget effect is pronounced. As no matter how many drill holes or channel samples you assay you CANNOT COUNT NUGGETS.
At Dome mines they used to do 1200 assays underground a day. The also assayed the belt going into the mill. That is called the heads. They averaged assay of all these samples was 0.12 ounce per ton. The mill recovery was 0.24 ounces per ton. The fudge factor was assay the crap out of the ORE and DOUBLE THAT.
At the Salmita Mine they used to to assay mostly below 1/10 of a ounce and every once in while the would assay 3.0 ounces per ton, If you arithmetically averaged all the samples it came to a mineable grade. If you cut the sample back to one ounce that were over one ounce, the mine did not exist. It was a mine. The lesson here is THERE IS ALMOST NEVER ANY JUSTIFICATION TO USE CUTBACKS TO ONE OUNCE. THERE IS NO PRINCIPLE THAT OCCASIONALLY HIGH ASSAYS ARE UNREPRESENTATIVE, OR ARE SPIKES. IF PEOPLE USE CUTBACKS OR IGNORE HIGH ASSAYS OR TALK ABOUT NUGGETS CAUSING UNREPRESENTATIVE ASSAYS, THEY ARE JUST STUPID. IT IS NOT GEOLOGY, AND NOT A PRINCIPLE OF GOLD MINING. THAT IS, IF THE SAMPLING IS CAREFUL AND THE ASSAYING GOOD.
Bad milling methods that don't get the gold out, and going off the vein, or not knowing what the hell your tails samples are are 60% of the problem with missing ore.
Assuming some kind of block like structure when the veins are in some kind of complex fold or drilling at one angle and the veins run at another are common causes of misplaced ore. If a company says the got XYZ tons of stuff, and they try to mine it bulk and the grade goes for a crap, it is I will bet a structural problem. They assumed one structure and the stuff is very differntly placed. Geostatistics cannot elucidate structure, at least at our incarnation of it. (IE. flat veining when they thought it was vertical. Any drill program that does not take into account the shape AND ORIENTATION of the ore fabric has no right to hang numbers on a sheet of paper. Know how many drill programs orient their drills in the holes with marking devices so they can tell flat veins from vertical ones? About 0.25 % Maybe.
EC<:-} |
