REMOTE SENSING APPLICATION FOR RADIOACTIVE MINERALIZATION EXPLORATION IN THE MAMUJU PROSPECT AREA, WEST SULAWESI

• Main Author: Rafli Irham, Mohammad
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/84725
• Institution: Institut Teknologi Bandung

Indonesia possesses significant radioactive mineral in several islands, including in the Mamuju region, West Sulawesi, known for its high natural radiation levels. This study aims to map the distribution of radioactive mineral occurrences in the Mamuju prospect area by leveraging remote sensing technology and machine learning methods. the study acquired reflectance spectroscopy data from radioactive mineral samples. Machine learning algorithms were then employed to classify the mineral groups based on their spectral characteristics. High-resolution PlanetScope PSB.SD satellite imagery was processed using supervised classification techniques to map land cover. Prospect areas were identified through band math analysis, vegetation indices, directed principal component analysis, and linear spectral unmixing to delineate indicator minerals associated with radioactive mineralization. The results show that the study area exhibits radiation dose rate anomalies ranging from 982.8 to 26,769 nSv/h, with uranium equivalent anomalies from 34.4 to 984.2 ppm and thorium from 222.6 to 6,066 ppm. The distribution of radioactive-bearing minerals, such as davidite and thorianite, has been identified in the Adang, Ampalas, Tapalang, and Malunda volcanic complexes. Indicator minerals like iron oxides and clay minerals were also identified, suggesting potential for radioactive mineralization in the region. This study demonstrates the significant potential of integrating remote sensing and machine learning for efficient, cost-effective, and accurate radioactive mineral exploration and mapping.