Exploration
Kamran Mostafaei; Mohammad Nabi Kianpour; Mahyar Yousefi; Meisam Saleki
Abstract
Discrimination of geochemical anomalies from background is a challenge in that elemental dispersion patterns are affected by a variety of geological factors, which vary from one to another area. While statistical and fractal methods are commonly employed for anomaly detection, they struggle with selecting ...
Read More
Discrimination of geochemical anomalies from background is a challenge in that elemental dispersion patterns are affected by a variety of geological factors, which vary from one to another area. While statistical and fractal methods are commonly employed for anomaly detection, they struggle with selecting optimal thresholds. This study proposes the Grey Wolf Optimizer (GWO) algorithm as a novel approach for identifying the optimal boundary between anomalies and background. Stream sediment geochemical data from a copper-mineralized area of the Sarduiyeh-Baft sheets in southeast Iran were utilized for analysis. The Geochemical Mineralization Probability Index (GMPI) was first calculated for Cu-Au, Mo-As, Pb-Zn, and porphyry distributions. Subsequently, fractal methods were used to identify anomalous populations within each GMPI. The GWO algorithm was then applied to these distributions to determine the optimal thresholds. Risk analysis, calculated as the ratio of covered copper occurrences to the covered area, revealed superior reliability for the GWO-derived limit compared to those obtained using fractal methods. For porphyry GMPI values, while the fractal reliability indices are 0.127, 0.44, and 0.5, the GWO limit achieved a value of 0.66. Risk analysis for Cu-Au distribution also caused more desired result for GWO limit rather that fractal ones. This demonstrates the enhanced performance and superior reliability of the GWO algorithm for optimizing anomaly detection thresholds in GMPI data.
Exploration
Kamran Mostafaei; Mohammad Nabi Kianpour; Mahyar Yousefi
Abstract
Mineral prospectivity mapping (MPM) is a multi-staged process aiming at delimiting exploration targets. Experts’ knowledge is an indispensable component of MPM, and might be required (i) while translating signature features of ore-forming processes into a suite of maps, namely evidence layers, ...
Read More
Mineral prospectivity mapping (MPM) is a multi-staged process aiming at delimiting exploration targets. Experts’ knowledge is an indispensable component of MPM, and might be required (i) while translating signature features of ore-forming processes into a suite of maps, namely evidence layers, (ii) while assigning weights to evidence layers, and (iii) while interpreting maps of mineral prospectivity. The latter is important as MPM integrates weighted evidence layers into a continuous map of mineral prospectivity. Although high values in prospectivity maps pertain to prospective zones, maps of mineral prospectivity are devoid of interpretation. One, therefore, should adopt a classification scheme to categorize or prioritize exploration targets from a map of mineral prospectivity. In addition to previous frameworks applied for interpreting maps of mineral prospectivity, this paper introduces an optimization-based framework, the Gray Wolf Optimizer (GWO) algorithm, for addressing this problem. In addition to GWO, we also used percentile maps of 85, 90, and 95% for interpreting the results of our prospectivity model. These methods were applied to a fuzzy-based map of mineral prospectivity derived for the Alut area, NW Iran. Overall, the map derived by the GWO has involved more Au occurrences, 66% of explored Au occurrences by GWO versus 33% by percentile maps; also introduces more targets as high-potential zones of Au mineralization that may be neglected by traditional methods like percentile maps.
Exploitation
K. Mostafaei; H. R. Ramazi
Abstract
Madan Bozorg is an active copper mine located in NE Iran, which is a part of the very wide copper mineralization zone named Miami-Sabzevar copper belt. The main goal of this research work is the 3D model construction of the induced polarization (IP) and resistivity (Rs) data with quantifying the uncertainties ...
Read More
Madan Bozorg is an active copper mine located in NE Iran, which is a part of the very wide copper mineralization zone named Miami-Sabzevar copper belt. The main goal of this research work is the 3D model construction of the induced polarization (IP) and resistivity (Rs) data with quantifying the uncertainties using geostatistical methods and drilling. Four profiles were designed and surveyed using the CRSP array based on the boreholes. The data obtained was processed, 2D sections of IP and Rs were prepared for each profile by inverting the data, and these sections were evaluated by some exploratory boreholes in the studied area. Based on the geostatistical methods, 3D block models were constructed for the 2D IP and Rs data, and the uncertainties in the prepared models were obtained. The mineralization location was determined according to the geophysical detected anomalies. In order to check the models, some locations were proposed for drilling in the cases that the borehole data was unavailable. The drilling results indicated a high correlation between the identified anomalies from the models and mineralization in the boreholes. The results obtained show that it is possible to construct 3D models from surveyed 2D IP & Rs data with an acceptable error level. In this way, the suggested omitted drilling locations were optimized so that more potentials could be obtained for copper exploration by the least number of boreholes.
