aceMetaheuristic-optimized machine learning framework for remote sensing-based alteration mapping of porphyry copper systems
Remote sensing-based hydrothermal alteration mapping is a pivotal technique in critical mineral exploration, particularly for identifying porphyry copper deposits (PCDs). However, conventional multispectral approaches are often constrained by linear assumptions, spectral mixing, and suboptimal parameter selection, limiting their ability to resolve complex alteration assemblages. This study presents a metaheuristic-optimized machine learning framework that integrates Boosted Trees (BT) and Quadratic Support Vecto…
"Army scientists energize battery research" by U.S. Army CCDC, CC BY-SA 2.0.
Researchers developed a metaheuristic-optimized machine learning workflow that integrates Boosted Trees and Quadratic Support Vector Machines with the Shuffled Frog Leaping Algorithm to map hydrothermal alteration from multispectral satellite imagery. Tested in the Shahr-e-Babak district of Iran's Urumieh-Dokhtar Magmatic Belt, the method used ASTER and Sentinel-2 bands to discriminate argillic, phyllic, propylitic and iron oxide zones.
By the June 2026 publication date, the optimized models showed higher classification performance than baselines, with BT AUC improving from 0.89 to 0.94 and QSVM from 0.88 to 0.93, and field checks indicated more than 78% spatial agreement. The authors present the workflow as computationally efficient and transferable for porphyry copper exploration in arid and semi-arid regions, though broader transferability beyond this case study remains to be demonstrated.
- Framework combined Boosted Trees and Quadratic Support Vector Machines with Shuffled Frog Leaping Algorithm optimization using ASTER shortwave infrared and Sentinel-2 visible, near-infrared and red-edge bands.
- Applied to Shahr-e-Babak district within the Urumieh-Dokhtar Magmatic Belt, Iran, to map argillic, phyllic, propylitic and iron oxide/hydroxide zones.
- Optimized BT outperformed QSVM, with AUC rising from 0.89 to 0.94 for BT and 0.88 to 0.93 for QSVM, and independent field, petrographic and XRD validation showed over 78% agreement.
Integrating Shuffled Frog Leaping Algorithm optimization with Boosted Trees and QSVM classifiers improved discrimination of argillic, phyllic, propylitic and iron oxide alteration zones from ASTER and Sentinel-2 data, raising AUC and achieving strong field validation.
The rundown
The study fused ASTER shortwave infrared bands with Sentinel-2 visible, near-infrared and red-edge features to resolve four alteration types linked to porphyry copper systems. Conventional multispectral approaches were described as constrained by linear assumptions and spectral mixing.
Optimization was performed with the Shuffled Frog Leaping Algorithm applied to Boosted Trees and Quadratic Support Vector Machines. Validation relied on field mapping, petrographic thin-section analysis and X-ray diffraction in the Shahr-e-Babak district, a well-exposed mineralized province in Iran.
Sources
- Peer-reviewedScientific Reports2026-06-07
How should this claim be treated?
The debate