Abstract
Identification of possible relationship between the elements, in mineralization and geochemical modeling is very important. The Full Cubic Model (FCM) with 90 Predictors including the main effects, quadratic terms, and interaction effects of Ag, As, Cu, Hg, Mn, Nb, P, Pb, Sn, Sr, Zn and U elements was used to create an Optimal and Reduced Cubic Model (ORCM) at vein-style Au (Ag)-polymetallic mineralization of Glojeh. It was generated with multiple regression and variance analysis and enhanced by Backward Elimination Procedure (BEP) for insignificant predictors through 15 steps. All the predictors contributed in ORCM by lowest p-value of 0.07 at degree of freedom (DF) equaled to 36. F ratio and R2 (pred.) criteria showed increasing trend from 0% to 77.80% and 6.12–262.51, respectively in model optimization. The main and interacted elements at ORCM are Sn, U, Pb, Cu and Ag × Sn, As × Mn, As × Zn, Hg × P, P × Sn, Sn2, P2, respectively based on related t-values. Interaction occurred when a pair of elements produces no similar trend on the response at different levels (associated to populations) of another element. Therefore, interaction of two elements on Au concentration was considered, simultaneously. By investigation, a disordinal interaction effect occurs on deposit for Au-As–Mn, Au-Ag-As and Au-Hg-Mn when the third element changes across the background level to vein values. Whereas, ordinal effects was observed at both population for Au-Ag-Hg and Au-As-Hg, since the lines were parallel. It implies that As has similar trend on Au for each value of Hg. The Au-Ag line explains the general downward shift by increasing Hg concentrations from vein to background. It revealed that Hg is more concentrated in background area. The accuracy and precision of this approach have been studied extensively for sample analysis, parameter identification, and modeling to ensure about BEP.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have