Metaheuristic-optimized machine learning framework for remote sensing-based alteration mapping of porphyry copper
Mahnaz Mahboobi1, Homayoon Katibeh2, Yousef Bahrami1
1Department of Mining Engineering, Amirkabir University of Technology, Tehran, Iran.
Scientific Reports
|June 7, 2026
Summary
This study introduces an optimized machine learning framework for critical mineral exploration, significantly improving the mapping of hydrothermal alteration zones for porphyry copper deposits. The advanced method enhances discrimination accuracy using satellite imagery and metaheuristic optimization.
Area of Science:
- Geoscience
- Remote Sensing
- Machine Learning
Background:
- Remote sensing is crucial for critical mineral exploration, especially for porphyry copper deposits (PCDs).
- Conventional methods face limitations due to linear assumptions, spectral mixing, and parameter selection issues, hindering accurate alteration mapping.
- Advanced techniques are needed to overcome these limitations and improve the identification of complex alteration assemblages.
Purpose of the Study:
- To develop and evaluate a metaheuristic-optimized machine learning framework for enhanced hydrothermal alteration mapping.
- To improve the discrimination of alteration zones associated with porphyry copper deposits using multispectral satellite data.
- To integrate Boosted Trees (BT) and Quadratic Support Vector Machines (QSVM) with the Shuffled Frog Leaping Algorithm (SFLA) for superior classification performance.
Main Methods:
- Integration of ASTER SWIR bands with Sentinel-2 VNIR and red-edge features.
- Application of a metaheuristic optimization framework combining BT and QSVM with SFLA.
- Validation using field mapping, petrographic analysis, and X-ray diffraction.
Main Results:
- Metaheuristic optimization significantly improved classification performance, with AUC values increasing from 0.89 to 0.94 for BT and 0.88 to 0.93 for QSVM.
- The optimized BT model demonstrated superior performance over QSVM in spectrally heterogeneous environments.
- Independent validation confirmed strong geological consistency with over 78% spatial agreement between predicted and observed alteration zones.
Conclusions:
- The proposed framework effectively learns complex, non-linear spectral-mineralogical relationships from multispectral data.
- Metaheuristic optimization enhances the reliability and accuracy of alteration zone mapping for PCD exploration.
- The workflow is computationally efficient, scalable, and transferable, serving as a generalizable decision-support tool for global exploration in arid/semi-arid regions.
Related Concept Videos
01:28Methods of Medium Optimization
Optimizing growth media enhances microbial proliferation and maximizes product yield. Statistical experimental design methodologies provide structured and reproducible approaches, offering progressively higher levels of robustness and efficiency.The One-Factor-at-a-Time (OFAT) MethodThe One-Factor-at-a-Time (OFAT) method involves adjusting a single variable while keeping all others constant. However, it cannot detect interactions between variables, often leading to suboptimal outcomes when...
01:21Manipulation and Analysis
GIS manipulation and analysis functions are vital for decision-making and planning. These activities range from data retrieval tasks, such as selecting information based on specific criteria, to advanced analytical techniques that address complex spatial problems.One critical GIS analysis method is overlaying, which combines multiple data layers to examine impacts. For example, overlaying a river-dammed lake boundary with road networks can identify affected infrastructure. Another common...