Three Decades of Artisanal Gold Mining Reveal Contrasting Outcomes for Conservation Areas and Indigenous Lands

Elmontaserbellah Ammar , Sean J Glynn, Kerry Anne Kansinally, William Clemens, Emilius Richter, Matthew J Struebig, Jake E Bicknell

Artisanal gold mining (AGM) is a critical driver of tropical forest disturbance, yet is challenging to quantify. Using deep learning and three decades of satellite imagery (1995-2024), we mapped AGM expansion across three contrasting countries in northern Amazonia. We evaluate mining trends, and extent across differing land designations and associated impacts on carbon stocks. AGM area increased >2000% over the study period, evolving from dispersed sites, into larger, consolidated mining landscapes, cumulatively releasing ~27,000 Gg of aboveground carbon from the region. Crucially, we show that protected areas slowed new mine establishment, while Indigenous lands faced rapid, sustained mining intensification, particularly since 2010. These findings show how long-term, high-resolution monitoring and machine learning can support policy responses across governance contexts, including providing key data needed in regulating legal and illegal mining, identifying emerging frontiers, and prioritising areas for restoration or protection where mining pressures intersect biodiversity and carbon priorities.