| Abstract: | The ability to quantitatively predict Au lodes in Au mineralization series with data from large-scale mineralization information is greatly needed. This paper discusses how to (1) classify the oreforming information, (2) set up the mineralization information model, (3) divide the statistical units within the minimum area of the mineralization anomalies, (4) select the comprehensive ore-forming (controlling) information variables, and (5) carry out the quantitative prediction using some newly proposed statistical models. Finally, the quantitative prediction results for Au lodes in a Au deposit, the Aohan Banner in Mongolia, are provided. Among the three first-grade prediction targets, two were tested and have been found to have industrial significance. |
