Assessing Ore Mineralization and Alteration in the Copper-Gold Deposit Using Remote Sensing Data

Thesis Abstract

The effective assessment of ore mineralization and alteration zones within copper-gold deposits remains critical for optimizing mineral exploration strategies, yet traditional field-based methods are often limited by accessibility, high costs, and time constraints. This study aims to evaluate the potential of remote sensing data in delineating mineralized and altered zones in a prominent copper-gold deposit located in the Andes region. The specific objectives are to (1) analyze multispectral satellite imagery to identify surface mineralogical signatures associated with mineralization and alteration, (2) develop a spectral anomaly index tailored to copper-gold deposits, and (3) validate remote sensing findings through ground-truthing and geochemical sampling. The research adopts an exploratory mixed-methods approach, integrating qualitative spectral analysis with quantitative spatial statistical modeling, underpinned by the hydrothermal alteration theory and spectral mineralogy framework. The population comprises satellite data acquired from Sentinel-2 and ASTER sensors, with a sample size of 250 spectral signatures obtained from both remote sensing imagery and in-situ field points, collected through systematic sampling across identified geological features. Data collection instruments include high-resolution multispectral sensors and portable X-ray fluorescence (pXRF) analyzers for ground validation. The remote sensing datasets are processed through atmospheric correction, spectral unmixing, and Principal Component Analysis (PCA), followed by the application of machine learning algorithms such as Random Forest classification to delineate mineralization zones. The validation employs regression analysis and kappa statistics to assess the accuracy of remote sensing classifications against ground survey data. It is anticipated that the findings will reveal specific spectral signatures associated with alteration minerals such as clay, sericite, and sulfides, facilitating the identification of mineralized zones with an accuracy exceeding 85%. The spectral anomaly index devised may serve as a practical tool for rapid exploration and targeting in similar geological settings, while the integration of remote sensing and ground-truthing will demonstrate the efficacy and limitations of satellite-based mineral detection. This research contributes to the existing body of knowledge by advancing the application of multispectral remote sensing techniques in mineral exploration, particularly within the context of copper-gold deposits in tectonically complex terrains. It extends spectral mineralogy frameworks by incorporating machine learning tools for improved classification accuracy, thereby offering a scalable methodology adaptable to different mineral deposit types. The main conclusion affirms that remote sensing, when complemented by targeted ground validation, provides a reliable, cost-effective means of delineating mineralization and alteration zones, enhancing exploration efficiency. Recommendations include the adoption of satellite-based spectral analysis in regional exploration programs, the development of standardized spectral anomaly indices, and the integration of remote sensing data with other geophysical and geochemical datasets for comprehensive resource assessment. Further research should explore the use of hyperspectral data and three-dimensional modeling to refine mineral deposit estimation at depth, advancing the holistic application of remote sensing in mineral exploration practices.