The Use Of Conditional Lognormal Probability Distribution For The Estimation Of Open-Pit Ore Reserves In Stratabound Uranium Deposits - A Case Study

Stratabound uranium deposits consist generally of two comingled geologic and statistical populations. Barren zones, usually having radiometric assays too low for recording on digitized logs, comprise a large portion of a mineralized stratum. In between are mineralized zones; where the grades are distributed essentially lognormally. Estimation of open-pit ore reserves requires the prediction of the tonnages of selective mining units (the smallest mined volume which may be segregated as ore or waste) and their grades within panels of convenient size for open pit planning and economic analysis. The proportionate tonnage of mineralized selective mining units within each panel can be predicted by kriging an an indicator function assigned to each composite (1 if mineralized, 0 if barren). Next, using logarithms of mineralized assays, the conditional lognormal distribution of grades of mineralized selective mining units can be found. Using this distribution and the krigedestimate of the probability a selective mining unit is mineralized, the grade and tonnage of selective- mining units above any cutoff can be predicted. A case study illustrating this two- step procedure is presented for the Imouraren uranium deposit located in the Republic of Niger. Using cross validation, the estimation procedures were checked and were shown to be adequately accurate for mine planning purposes.