Mining triggers extensive additional deforestation in sub-Saharan Africa

Article Mining triggers extensive additional deforestation in sub-Saharan Africa https://doi.org/10.1038/s41586-026-10551-2 Received: 7 October 2025 Accepted: 15 April 2026 Published online: xx xx xxxx Open access Check for updates Oscar Morton1,10 ✉, Christopher G. Bousfield1,10 ✉, Prince Dégny Valé2,3, Ieuan Lamb1, Victor Maus4,5, Robert G. Bryant6 & David P. Edwards7,8,9 ✉ Demand for minerals sourced from sub-Saharan Africa is expanding rapidly1–5. If poorly managed, mining expansion poses a key threat to tropical forests across the continent6,7. Here we present a spatiotemporal assessment of mining-driven deforestation of dense forests across Africa, using continent-wide data on postdeforestation land uses and a robust difference-in-differences framework to assess 16,627 mines between 2001 and 2020. In total, we find 187,000 hectares of direct mining-driven deforestation, that is, deforestation due to features directly associated with mining operations, such as pits, tailing ponds and spoil heaps. We estimate that mining also triggers an additional 8.0 percentage points (pp; 95% confidence interval (CI): 7.2–8.9 pp) increase in deforestation within 1 km of a mine compared with unmined areas. Increased levels of deforestation (1.1 pp, 95% CI: 0.7–1.5) persist up to 20 km from mines even after ten years. For every hectare of direct deforestation due to the mine footprint, mining triggers, on average, 34 hectares of additional offsite loss within five years through ancillary activities, including agriculture and settlements. Mines extracting cobalt and copper—key energy transition minerals— caused the highest amount of additional deforestation. Embedding offsite deforestation levels into environmental impact assessments for new mining projects will be key to ensuring zero-deforestation or no-net-loss supply chains for critical minerals and reduce future mining-driven forest losses in sub-Saharan Africa. Global demand for metal ores and minerals is rising markedly1,2. The extraction of metal ores has quadrupled3 since 1970, fuelling rapid economic growth. The total, or embodied, volume of material stock used in buildings, infrastructure and machinery increased4 23-fold between 1990 and 2010. Moreover, the conversion of currency into gold during global periods of financial upheaval has increased gold exploitation5, and the global shift towards renewable energy has further heightened demand for key energy transition minerals (ETMs), including cobalt, lithium and copper8. Globally, mining directly threatens 8% of vertebrate species and drives widespread habitat loss9,10. This is concerning given that more than 20% of the remaining intact tropical forest lies in mineral, gas or oil concessions, many of which are still in the exploration stage11. Mine expansion affects forests directly when land is cleared for mining pits or shafts, on-site processing, spoil heaps, or tailing storage facilities. More than 325,000 hectares (ha) of tropical forest was lost directly to industrial mining6 between 2000 and 2019. However, focusing solely on mine footprints ignores ancillary offsite anthropogenic activities triggered by mining indirectly, including the construction of ports, roads and railways, settlements and agricultural expansion12–14. Each of these activities drives further deforestation15,16 and increases forest accessibility, facilitating hunting and trapping17,18. National-scale attempts to estimate total offsite deforestation caused by mining vary substantially from no-detectable-net impact in Zambia and Madagascar19,20, to significant net-negative impacts in the Democratic Republic of the Congo (DRC) and Brazil21,22. Such variability emphasizes the need for continental-scale assessments of the spatial and temporal impact of mining on direct and offsite (indirect) deforestation. Here we uniquely quantify direct and offsite mining-induced deforestation of dense forest across sub-Saharan Africa by combining 16,627 distinct clusters of mines mapped in forested areas23 with 20-years of forest-cover data24 and by using recent advances in robust differences-in-differences (DID) inferential approaches25,26. We account, at the continental scale, for both direct mining deforestation driven by the mines themselves (such as pits and tailing ponds), and offsite deforestation through other processes (including road construction and agricultural and urban expansion) that may be triggered by mine establishment. Sub-Saharan Africa has the largest known global reserves of many strategically important minerals7, including 48% of cobalt and manganese reserves27, and has seen surges in foreign investment fuelling a mineral boom and rapid mining expansion28. Its forests contain globally vital carbon stocks29 and are hotspots of biodiversity30; thus, it is imperative that mining expansion be sustainable and well managed across the continent. Specifically, we address Ecology and Evolutionary Biology, School of Biosciences, The University of Sheffield, Sheffield, UK. 2Université Jean Lorougnon Guédé, Daloa, Côte d’Ivoire. 3Centre Suisse de Recherches Scientifiques en Côte d’Ivoire, Abidjan, Côte d’Ivoire. 4Institute for Ecological Economics, Vienna University of Economics and Business, Vienna, Austria. 5Advancing Systems Analysis Program, International Institute for Applied Systems Analysis, Laxenburg, Austria. 6School of Geography and Planning, The University of Sheffield, Sheffield, UK. 7Department of Plant Sciences, University of Cambridge, Cambridge, UK. 8Centre for Global Wood Security, University of Cambridge, Cambridge, UK. 9Conservation Research Institute, University of Cambridge, Cambridge, UK. 10 These authors contributed equally: Oscar Morton, Christopher G. Bousfield. ✉e-mail: ; ; 1 Nature | www.nature.com | 1 Article a b 5 Direct mining-driven deforestation (%) GNQ 4 3 2 GHA SWZ ZAF RWA 1 BDI ZWE ZMB CMR 0 GIN 0 GNB 10,000 AGO MDG 20,000 COD 30,000 40,000 Direct mining-driven deforestation (ha) Direct mining-driven deforestation (ha) c Direct mining-driven deforestation (ha) COD 3,000 GHA 2,000 MDG 1,000 ZAF AGO 0 0 1 10 50 100 500 1,000 5,000 2000 2005 2010 2015 2020 Year Fig. 1 | Direct mining-induced deforestation across sub-Saharan Africa between 2001 and 2020. a, Area (ha) of direct mining-induced deforestation of dense forest per 30 km 2 cell. b, Direct mining-induced deforestation compared with the percentage of national deforestation accounted for by direct mining-induced deforestation between 2001 and 2020. c, Three-year average annual direct mining-induced deforestation per country between 2001 and 2020. Each line represents a single country. The five largest countries in terms of summed direct mining-induced deforestation during the analysed period are labelled. Only countries with more than 100 ha of mining-induced deforestation are shown. A full list of three-letter country codes and the corresponding country name is pr (...truncated)