Resource Classification Clause Samples

Resource Classification. A common method used in the classification of mineral resources involves geostatistical methods that define categories based on the confidence limits of the estimation. Measured resources are defined as material in which the predicted grades are within ±15% accuracy on a quarterly basis, at a 90% confidence limit. In other words, there is a 90% chance that the predicted grades for a quarter-year of production will be within ±15% of the actually achieved production grades. Similarly, Indicated resources include material in which the yearly production grades are estimated with ±15% accuracy at the 90% confidence limit. The method is based on the large sample normal theory that assumes that as the grade estimations from smaller blocks are combined into larger ones, the errors of the estimation become normally distributed (as described by B. ▇▇▇▇▇, 1997). The steps in generating the classification parameters for the CX and Range Front zones are described as follows. This exercise assumes a nominal daily production rate from either of the zones of 500 tons per day which equates to a monthly production rate of approximately 15,000 tons per month. At an average tonnage factor of 13 cubic feet per ton, a block measuring 60x60x60 feet represents approximately one month of production (16,600 tons). A block equal in size to the volume of one month’s production is created and the kriging variance is determined using a series of theoretical drill holes at intervals averaging 50, 100 and 200 foot spacing. The calculations are done over a series of drill-hole grids in order to evaluate the variation in the results with respect to the spacing of the drill data. The correlogram used to determine the kriging variance in the large block is derived from the actual gold sample data which has been composited to 5-foot intervals (Table 17-5). Because the correlogram was used, the normalized block kriging variance (a variable which is output from the OK run) was standardized to the underlying data by multiplying by the square of the coefficient of variation (CV=std dev/mean from the original 5 ft composite data). The relative standard error for a quarter year of production is determined by taking the square root of the standard block variance divided by three (i.e. divide by 3 for a quarter-yr of production or divide by 12 in order to determine the error for a full year of production). Finally, the 90% confidence limit is determined by multiplying the relative standard error by 1...