Publication Date
5-2024
Advisor(s) - Committee Chair
Nahid Gani, Jerry Brotzge, M. Royhan Gani, Amin Beiranvand Pour, Mohammad Parsasadr
Degree Program
Department of Earth, Environmental, and Atmospheric Sciences
Degree Type
Master of Science
Abstract
Geological mapping is crucial for mineral exploration, especially in identifying lithological units, alterations, and critical minerals. Traditional methods, often limited by accessibility, cost, and environmental factors, rely heavily on fieldwork and geochemical sampling. This study demonstrates the effectiveness of leveraging multi-sensor remote sensing satellite imagery as a cost-efficient alternative for detailed mapping of mineral alteration zones, crucial as the demand for critical minerals intensifies amidst supply constraints. Focusing on the Mountain Pass District in California’s eastern Mojave Desert – known for its significant potential in rare earth element (REE) mining due to unique geodynamic conditions and extensive carbonate platforms – this research addresses a gap in comprehensive multi-satellite multispectral remote sensing analysis of the geologic framework related to carbonatite ore deposits and hydrothermal alteration zones. Employing data from Landsat-9, Sentinel-2, and the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), this study mapped hydrothermal alteration minerals across the visible, near-infrared, and shortwave-infrared spectrum. Advanced image processing techniques, including band ratios and principal component analysis (PCA), fuzzy logic modeling along with machine learning were utilized to integrate thematic layers from these satellites, creating precise mineral prospectivity maps.
The study successfully mapped iron-oxide and ferrous-bearing minerals, crucial for indicating ore deposits and surface weathering processes, alongside hydrothermally altered rocks such as clay minerals, using band ratio techniques. PCA and fuzzy logic modeling established a statistical foundation for creating a comprehensive mineral prospectivity map, identifying high prospective zones near active and abandoned mines. Advanced machine learning algorithms significantly improved the classification and mapping of hydrothermal alteration minerals, demonstrating the effectiveness of these technologies in geological exploration. The SVM algorithm showed better discrimination capability in detecting hydrothermal alterations, emphasizing its potential for detailed mineralogical analysis within complex geological settings like the Mountain Pass area. These outcomes not only enhance our understanding of employing multispectral satellite data in mineral exploration but also provide a robust framework for identifying and characterizing mineral zones. The findings have laid a solid foundation for future research in the eastern Mojave Desert region and advanced the field’s scientific innovation, offering new insights into carbonatite hosted REE deposits.
Disciplines
Earth Sciences | Environmental Sciences | Geography | Remote Sensing | Social and Behavioral Sciences
Recommended Citation
Esmaeilzadeh, Nima, "LEVERAGING MULTI-SENSOR REMOTE SENSING AND MACHINE LEARNING TO ENHANCE MINERAL EXPLORATION IN THE MOUNTAIN PASS DISTRICT, CALIFORNIA" (2024). Masters Theses & Specialist Projects. Paper 3719.
https://digitalcommons.wku.edu/theses/3719