Geochemistry-based machine learning approach applied to an archaeological provenance study: the obsidian blades of Tulūl al-Baqarat (Iraq)

A groundbreaking study identifies Nemrut Dağ stratovolcano (Southeastern Turkey) as the likely source of obsidian blades found at the 4th millennium BCE archaeological site of Tulūl al-Baqarat in Iraq. Using a geochemistry-based machine learning approach, researchers analyzed nine non-destructive obsidian tools and compared their compositional data with volcanic sources across the Near East. Key elements like Zr and Rb, combined with principal component analysis (PCA) modeling, revealed that these blades match volcanic glasses from Nemrut Dağ, excluding most Turkish and Armenian outcrops. This finding provides compelling evidence for ancient trade networks linking Turkey to the Persian Gulf region during the 4th millennium BCE.

Vaggelli, G., Cossio, R., Borghi, A., Lippolis, C., & Ghignone, S. (2025). Geochemistry-based machine learning approach applied to an archaeological provenance study: the obsidian blades of Tulūl al-Baqarat (Iraq). Frontiers in Earth Science, 13. https://doi.org/10.3389/feart.2025.1675908