Application of three-dimensional weights of evidence in modeling concealed ore deposits: Case study of a porphyry Cu deposit in the Urmia-Dokhtar magmatic belt of Iran
share
Given the challenges in data acquisition and modeling at the stage of detailed exploration, developing a prospectivity model particularly for disseminated ore deposits is difficult. Recent work has shown that the weights of evidence-based modeling has good potential for discovering of such deposits. In our approach, the qualitative geological and quantitative geochemical data obtained from boreholes are used to create a three-dimensional prospectivity model of a porphyry Cu deposit within the Urmia-Dokhtar magmatic arc, Iran. This prospectivity model is created using the weights of evidence method which is further extended for a three-dimensional (3D) space. We demonstrate that this method has the ability of integrating qualitative and quantitative exploration criteria in a 3D space based on the metallogenic model of the study area through prospecting for a concealed ore body. The geological data used in this study, include lithology, alteration and rock origin data. The results indicate a high correlation between monzodiorite units, silicific alteration and as expected volcanic rocks and the Cu mineralization. The input evidential models are integrated using the weights of evidence and three models including posterior probability, uncertainty and studentized posterior probability are created. The anomalous voxels in probability models are determined using concentration-volume fractal models and validated by prediction-area plots. The results show that the posterior probability model is more efficient through discovering Cu mineralization-bearing voxels, but the studentized posterior probability model is more reliable because of the effect of uncertainty in determining the voxel values. Using the Python scripts enclosed to this study, the same procedure can be implemented for exploring concealed ore bodies in other regions and locating potential zones in depth.
READ FULL TEXT