Direct deforestation within mining areas

The investigated mining areas covered 11,467 km² of land that included 7,019 km² of tropical forest in 2000. By 2019, 3,264 km² (46.5%) of these forest areas were directly lost to industrial mine expansion. With 1,901 km² of deforested area, Indonesia was by far the most affected country, accounting for almost 60% of direct forest loss by mining across all 26 investigated countries (Figure 1). Mine expansion in East Kalimantan on the island of Borneo for coal production was the main factor behind this development in Indonesia. Deforestation within Brazil’s mining areas located in tropical forest biomes extended over 327 km² since 2000. Ghana (213 km²), Suriname (203 km²), and Côte d’Ivoire (99 km²) follow as the countries with the highest direct forest loss. All other countries together made up 16% of tropical forest loss by mining observed across the 26 countries.

Figure 1: Direct tropical forest loss due to industrial mining from 2000 to 2019 in the top 15 countries with the highest absolute deforestation by industrial mining, together accounting for 98% of direct deforestation across all 26 investigated countries.

While for most countries the direct deforestation effects of mining (on-site) are rather small compared to other land-intensive activities, such as the production of soybeans and palm oil or cattle farming, some countries show considerably higher shares in total forest loss, e.g. Suriname with 11% and Guyana with 4%.

Indirect deforestation induced by mining

We find strong evidence that mining also induces indirect deforestation outside areas designated for mining activities. In 18 of the 26 investigated countries, deforestation rates are higher close to the actual mining areas than areas farther away than 50 km, even when controlling for other known determinants of tropical deforestation, such as proximity to agriculture or roads. These indirect impacts are estimated using the distance to the closest mine in a spatial statistical model. Thus, if deforestation increases with higher proximity to the mines, we expect this to indicate indirect deforestation effects of the corresponding mines.

In Brazil and Indonesia, we find high statistical significance for higher deforestation 50 km around the actual mining sites. On average, and holding all other parameters fixed, reducing the distance to a mine in Brazil by 10% (e.g. from 10 to 9 km) leads to a 3% increase in deforestation. For Indonesia, this value is 2.3%. The absolute effects of these results can be illustrated with a scenario simulation: For instance, if all mines in Indonesia were to expand their borders by 100 m, this would induce an additional deforestation between 194 and 215 km². In Brazil, the same 100-m expansion would lead to 147 to 154 km² of additional forest loss. Apart from Indonesia and Brazil, the statistical relation of mines causing indirect deforestation can also be observed for many other tropical countries, including Guyana, Colombia, the Philippines, Papua New Guinea and the Democratic Republic of the Congo (DRC).

Figure 2 provides a visualisation of the statistical results. Map (a) shows the coefficients for mining-induced deforestation in all 26 investigated mining countries with tropical forests. A negative coefficient indicates that mining drives off-site deforestation. The impacts of industrial mining on forest loss can also be illustrated on a spatially explicit level by considering the distance of each grid cell to the nearest mine (b). These three maps for mining regions in Brazil, the DRC, and Indonesia thus illustrate the importance of mining-induced deforestation in total forest loss in each grid cell. White areas indicate that no forest loss has been observed since 2000. The importance of forest loss due to industrial mining is clearly visible in the state of Minas Gerais in Brazil, where iron ore and gold are particularly mined in the “Iron Quadrangle” in the south. The indirect deforestation induced by mining in the Central African Copperbelt, stretching between Zambia and the DRC, is illustrated in the second example. As a third case, we show the Indonesian mining regions on the island of Borneo, where coal, nickel, and tin mining significantly expanded since the year 2000.

Figure 2: Visual representation of indirect deforestation induced by industrial mining outside mining areas. (a) The national coefficient of distance to mine across 26 investigated countries. (b) Granular representation of the national coefficient in three selected mining areas. The closer a grid cell is located near a mining site, the higher is the share of mining-induced deforestation.

Discussion

Our study has provided an assessment of deforestation caused by industrial mining in countries across the tropical biome. The results show that the direct impacts of mining on tropical forests have been concentrated in only a few countries, with Indonesia, Brazil, and Ghana being the most heavily affected. In Indonesia, the highest deforestation rates were observed for the years 2010 to 2014, a period that was marked by a doubling of coal production volumes, particularly driven by demand from China and India. A fragmented and opaque governance system for issuing new coal extraction licenses facilitated this development. Institutional reforms after 2014 implemented caps on coal extraction growth rates, which also slowed down direct deforestation. Also in Brazil, forest loss within mining areas decreased after 2014. Declining global commodity prices and an economic crisis in Brazil after 2014 are among the explanatory factors for that temporal pattern. We also found that mining played a large role in the deforestation trajectories in other countries with relatively less absolute forest loss (e.g., 11% of deforestation in Suriname between 2000 and 2019 was directly caused by mining). Further, we revealed that industrial mining indirectly drives deforestation in the surroundings outside mining areas. Indirect deforestation effects caused by industrial mining are most considerable for Brazil and Indonesia, but these effects can be observed in more than two-thirds of the investigated countries. Against the background of rapidly increasing global demand for mineral resources, e.g., for housing and transport infrastructure or green energy technologies, our results emphasise important yet unevenly distributed and largely unmanaged future threats to tropical forests.

Implications for policy and industry

Our results suggest that Environmental Impact Assessments and licensing procedures for new mining projects should also consider potential impacts outside the actual mining extents. Applications for new mining concessions should also not be considered in isolation – in particular, if they include additional economic sectors, such as agriculture – but should take their potential cumulative impact on forest loss into account. Policy initiatives that aim at raising transparency and due diligence along global supply chains could also play an increasingly significant role in managing environmental impacts of mining.

Apart from policy measures, other actors, including conservation organisations, multilateral organisations, and industry groups, will play a key role in setting stricter environmental standards. Although companies in the extractive sectors have started to manage direct deforestation impacts of mining activities, only a few examples consider indirect impacts on deforestation.

Best-practice approaches to mitigate mining impacts on forests are promoted by private sector organisations, such as the International Council on Mining and Metals (ICMM). However, off-site impacts are not addressed appropriately yet. ICMM also has a requirement in place for its member companies not to explore or mine in World Heritage areas listed by the International Union for Conservation of Nature (IUCN). Opportunity exists to expand these commitments to avoid impacts on other biodiverse places.

One key requirement to limit losses of tropical forests due to mining is the development and implementation of monitoring programs. Satellite-based data systems on mining activities and deforestation, as employed in our study, could form the starting point for developing monitoring systems that allow for a regular and consistent identification of deforestation linked to the expansion of specific mines. Such information is also a precondition to set up supply chain initiatives in the private sector to reduce deforestation, following examples from the agriculture and food sectors. Such initiatives would increase the traceability and transparency of mining products and could thereby help slowing down mining-induced deforestation as part of national and global forest conservation efforts.

Citation

Giljum, S., Maus, V., Kuschnig, N., Luckeneder, S., Tost, M., Sonter, L.J., Bebbington, A.J., 2023. Mining causes direct and indirect loss of tropical forests. FINEPRINT Brief No. 17. Vienna University of Economics and Business (WU). Austria.