Threat of mining to African great apes

Abstract

The rapid growth of clean energy technologies is driving a rising demand for critical minerals. In 2022 at the 15th Conference of the Parties to the Convention on Biological Diversity (COP15), seven major economies formed an alliance to enhance the sustainability of mining these essential decarbonization minerals. However, there is a scarcity of studies assessing the threat of mining to global biodiversity. By integrating a global mining dataset with great ape density distribution, we estimated the number of African great apes that spatially coincided with industrial mining projects. We show that up to one-third of Africa’s great ape population faces mining-related risks. In West Africa in particular, numerous mining areas overlap with fragmented ape habitats, often in high-density ape regions. For 97% of mining areas, no ape survey data are available, underscoring the importance of increased accessibility to environmental data within the mining sector to facilitate research into the complex interactions between mining, climate, biodiversity, and sustainability.

Full text

Junker et al., Sci. Adv. 10, eadl0335 (2024) 3 April 2024
SCIENCE ADVANCES | RESEARCH ARTICLE
1 of 12
ECOLOGY
Threat of mining to African great apes
Jessica Junker1,2,3*†, Luise Quoss1,2†, Jose Valdez1,2†, Mimi Arandjelovic2,4, Abdulai Barrie5,
Geneviève Campbell3, Stefanie Heinicke6, Tatyana Humle3,7, Célestin Y. Kouakou8,9,
Hjalmar S. Kühl2,10,11, Isabel Ordaz- Németh3,10, Henrique M. Pereira1,2,12, Helga Rainer13,
Johannes Resch14, Laura Sonter15,16,17, Tenekwetche Sop3,10
The rapid growth of clean energy technologies is driving a rising demand for critical minerals. In 2022 at the 15th
Conference of the Parties to the Convention on Biological Diversity (COP15), seven major economies formed an
alliance to enhance the sustainability of mining these essential decarbonization minerals. However, there is a
scarcity of studies assessing the threat of mining to global biodiversity. By integrating a global mining dataset
with great ape density distribution, we estimated the number of African great apes that spatially coincided with
industrial mining projects. We show that up to one- third of Africa’s great ape population faces mining- related
risks. In West Africa in particular, numerous mining areas overlap with fragmented ape habitats, often in high-
density ape regions. For 97% of mining areas, no ape survey data are available, underscoring the importance of
increased accessibility to environmental data within the mining sector to facilitate research into the complex in-
teractions between mining, climate, biodiversity, and sustainability.
INTRODUCTION
Africa is experiencing an unprecedented mining boom (1) threaten-
ing wildlife populations and whole ecosystems. Mining activities are
growing in intensity and scale, and with increasing exploration and
production in previously unexploited areas. Africa contains around
30% of the world’s mineral resources, yet less than 5% of the global
mineral exploitation has occurred in Africa, highlighting the enor-
mous potential for growth in this sector (1). Substantial production
increases in the renewable energy sector are expected to cause a
boom in mineral exploitation (2). Africa, which is rich in ecological
diversity, harbors around one- sixth of the world’s remaining forests
and is home to one- quarter of the world’s mammal species (3).
Among these are primates, which are one of the most threatened
groups of species, with 67% of all primate species (Africa: 73.1%)
currently listed as threatened by the International Union for Con-
servation of Nature’s (IUCN) Red List of reatened Species and
42% with continuing declining population trends (4). Great apes
(hereaer “apes”) are particularly at risk, with all 14 taxa currently
listed as either Endangered or Critically Endangered (5).
Apes are our closest evolutionary relatives and are important in
many societies, contributing to livelihoods, cultures, and religions.
ey generate substantial income from tourism projects and serve
as powerful flagship species due to their anthropological signifi-
cance, helping to raise public awareness and millions in conserva-
tion spending (6). ey fulll the important role of umbrella species
implying that if conservation eorts focus on ape populations and
their habitats, this also increases the overlap with conservation pri-
orities identied for many other tropical plant and animal species
[e.g., (7)]. ey are essential for maintaining biodiversity and eco-
system services; they disperse seeds, consume and pollinate plants,
and create canopy gaps and trails (8). Last, habitats important to
apes, which mostly comprise tropical forests, play a crucial role for
global climate change mitigation due to their ability to extract car-
bon dioxide from the air, create clouds, humidify the air, and release
cooling chemicals (9).
e IUCN Red List recently estimated that only 2 to 13% of all
primate species were threatened by road and rail construction, oil
and gas drilling, and mining, whereas 76 and 60% were negatively
aected by agriculture and logging, and wood harvesting, respec-
tively (4). Similarly, mining currently ranks only fourth in the fre-
quency of reported threats across African ape sites documented in
the Ape Populations, Environments and Surveys (A.P.E.S.) Wiki
(10), 65 of 180 sites, i.e., 36% of all sites for which threats have been
documented (11); and is preceded by hunting (89% of sites), logging
(62%), and agricultural expansion (62%). However, given recent
ndings on the density of mining areas across Africa (2), these val-
ues might be a considerable underestimation of the real threat of
mining to apes. is discrepancy may be due to the lack of data from
mining locations (i.e., only 2 of the 180 African ape sites included in
the A.P.E.S. Wiki are mining areas as of March 2023). In addition,
mining companies that conduct Environmental Impact Assess-
ments typically practice data embargoes that prohibit use of the data
by second or third parties (see also 2022 Nature Benchmarks). As a
result, there are few published studies that scientically assess the
impacts of mining on wildlife populations (12).
1Institute of Biology, Martin Luther University Halle- Wittenberg, Am Kirchtor 1,
06108 Halle, Germany. 2German Centre for Integrative Biodiversity Research (iDiv)
Halle- Jena- Leipzig, Puschstrasse 4, 04103 Leipzig, Germany. 3Re:wild, 500 N Capital
of Texas Hwy Building 1, Suite 200, Austin, TX 78746, USA. 4Max- Planck Institute
for Evolutionary Anthropology, Department of Primate Behavior and Evolution,
Deutscher Platz 6, 04103 Leipzig, Germany. 5Ministry of Environment and Climate
Change, 55 Wilkinson Road, Freetown, Sierra Leone. 6Potsdam Institute for Climate
Impact Research (PIK), Member of the Leibniz Association, Potsdam, Germany.
7Durrell of Institute of Conservation and Ecology, School of Anthropology and Con-
servation, University of Kent, Canterbury CT2 7NR, UK. 8Université Jean Lorougnon
Guédé, BP 150 Daloa, Côte d'Ivoire. 9Centre Suisse de Recherches Scientiques
(CSRS), 17 Rte de Dabou, Abidjan, Côte d’Ivoire. 10Senckenberg Museum for Natural
History Görlitz, Am Museum 1, 02826 Görlitz, Germany. 11International Institute
Zittau, Technische Universität Dresden, Markt 23, 02763 Zittau, Germany. 12CIBIO,
Centro de Investigação em Biodiversidade e Recursos Genéticos, InBIO Laboratório
Associado, Campus de Vairão, Universidade do Porto, 4485- 661 Vairão, Portugal.
13Independent consultant, PO Box4107, 759125 Kampala, Uganda. 14Great Apes
Survival Partnership, United Nations Environment Programme, P.O. Box 30552,
00100 Nairobi, Kenya. 15School of the Environment, The University of Queensland,
St Lucia 4072, Australia. 16Centre for Biodiversity and Conservation Science, The
University of Queensland, St Lucia 4072, Australia. 17Sustainable Minerals Institute,
The University of Queensland, St Lucia 4072, Australia.
*Corresponding author. Email: jjunker@ rewild. org
†These authors contributed equally to this work.
Copyright © 2024 The
Authors, some rights
reserved; exclusive
licensee American
Association for the
Advancement of
Science. No claim to
original U.S.
Government Works.
Distributed under a
Creative Commons
Attribution
NonCommercial
License 4.0 (CC BY- NC).
Downloaded from https://www.science.org at Universitaet Leipzig on April 04, 2024

Junker et al., Sci. Adv. 10, eadl0335 (2024) 3 April 2024
SCIENCE ADVANCES | RESEARCH ARTICLE
2 of 12
e direct and indirect impacts of industrial mining (hereaer
“mining”) are manifold (Fig.1). Mining areas are highly dynamic and
impactful activities already start during the exploration phase. During
this phase, high noise production, caused by extensive drilling and
blasting, can disturb the communication of species, such as primates
(13), and result in functional loss of otherwise intact habitat (14).
Physiological responses to noise pollution have also been docu-
mented in various other wildlife species and include among others,
increased heart rate, damage to the auditory system, and ultimately,
a decrease in survival probability (15). Removal of vegetation may
already be initiated during this phase where very distinct drilling
lines can oen be visible from satellite imagery (16). During the ex-
ploitation phase, digging, blasting, and the use of heavy machinery
typically result in direct impact within the project’s development
area in the form of habitat destruction, fragmentation, and degrada-
tion (Fig.1). e release of pollutants, such as heavy metals and
toxic chemicals, can contaminate air, water sources, and soil, poten-
tially causing health issues (17–19) and disrupting food chains.
While studies on the eect of light pollution are still scarce and non-
existent for apes, a recent meta- analysis found that exposure to arti-
cial light at night induces strong responses for physiological measures,
(e.g., reduced melatonin levels), longer daily activity, and life history
traits (e.g., reduced reproductive success), also in diurnal species (20).
Indirect mining impact beyond the mining lease boundary is much
more dicult to quantify and only a few studies on this topic have
been published to date [e.g., (21–23), Fig.1]. In 2017, Sonter etal.
(24) demonstrated that large- scale industrial mining operations
caused signicant deforestation over time and up to 70 km from
mining lease boundaries in Brazil’s Amazon Forest. Furthermore, a
recent global pan- tropical assessment found that in two- thirds of
the 26 investigated countries, deforestation rates were higher close
to the actual mining areas than in areas farther away, even when
controlling for other known determinants of tropical deforestation
(24). In some of these countries, the authors found high statistical
signicance for mining driving deforestation in the surrounding ar-
eas up to 50 km outside the mining areas. is is largely ascribed to
in- migration of people and increased access resulting in an in-
creased demand for land, charcoal, fuelwood, and roads (23).
Once extracted, many minerals are typically transported to the
nearest port from where they are shipped to destinations around the
world. Associated infrastructures, such as road and rail develop-
ment, therefore go hand- in- hand with activities in and around the
concession site. e threat to wildlife posed by linear infrastructure
is mostly indirect as demonstrated by numerous studies [e.g., (25–28)];
however, collisions of vehicles with apes trying to cross the
road have been reported previously (29, 30). Recently, Andrasi etal.
(31) estimated that western chimpanzee density is negatively aect-
ed within a distance of about 16 to 19 km away from major roads
and 5 to 6 km from minor roads. Various underlying threats nega-
tively inuence wildlife along roads: ey include induced access,
increased re incidence, soil erosion, landslides, biological inva-
sions, increased hunting pressure, and proliferation of agriculture
(32). Last, apes in mining areas are likely to have an increased risk of
contracting disease from humans due to increased frequency in
contact (33). is is aggravated by the fact that people and goods are
moving more rapidly and further into remote locations potentially
introducing diseases that were not known to those areas (34). How-
ever, an additional complex issue is the link between large- scale
Fig. 1. Schematic overview of the approximate potential direct (10 km) and indirect threats (50 km) on apes linked to mining activities. Expected high and mod-
erate to lower risk of impact is indicated by red and yellow pointers, respectively.
Downloaded from https://www.science.org at Universitaet Leipzig on April 04, 2024

Junker et al., Sci. Adv. 10, eadl0335 (2024) 3 April 2024
SCIENCE ADVANCES | RESEARCH ARTICLE
3 of 12
development projects and the resulting habitat change and emer-
gence and spread of diseases. Deforestation in tropical regions has
oen been associated with increased outbreaks of infectious diseases
such as dengue fever, malaria, and yellow fever; some of these
diseases aect great apes as well (35). e underlying mechanisms
are oen complicated: A study of zoonotic malaria, transmitted by
long- and pig- tailed macaques (Macaca fascicularis and Macaca
nemestrina) in Malaysian Borneo, conrmed the link between
zoonotic spillovers and deforestation but showed complex and dif-
ferent eects of forest degradation at dierent scales (36).
To quantify the potential impact of industrial mining on wildlife
population abundance, we used African great apes as a case study.
ey are particularly important in this context, because they are the
only taxon specically mentioned in the International Finance Cor-
poration’s (IFC) Performance Standard 6 Guidance Note 73 as a
taxon that is likely to trigger so- called “Critical Habitat” (CH),
which imposes strict environmental regulations on mining compa-
nies that are seeking IFC funding (or loans from other lenders align-
ing with these standards) and that want to operate in these areas. It
requires companies to reach out to the IUCN/Species Survival
Commission (SSC) Primate Specialist Group, Section on Great Apes
for consultation (37). Specically, mining projects operating in CH
must implement mitigation measures to eectively counteract their
ecological impact, ultimately resulting in a net increase in the over-
all population of great apes.
Using data spanning 17 African nations (tableS2) over an area of
1,507,811 km2, we estimated the extent of the potential direct and
indirect negative impact from mining activities on ape abundance
in and around operational and preoperational mining areas. To do
this, we integrated a global mining dataset with range- wide esti-
mates of ape density distribution. We investigated (i) how many
African apes could potentially be negatively aected by mining
activities across their range, (ii) whether mining areas oen over-
lapped with high ape density areas, and (iii) to what extent great ape
survey data were available across these mining areas. Furthermore,
we (iv) quantied the spatial overlap of mining areas with likely CH
triggered by biodiversity features unrelated to apes and (v) identied
hotspots of spatial overlap of high mining and ape densities.
RESULTS
Geographical distribution of mining density in relation to
ape density
High ape densities broadly coincided with operational and preop-
erational mining areas (mining locations and their 50- km buers)
throughout most of the ape range in West Africa, in Gabon, south-
ern and western Republic of Congo (from here on “Congo”) and
southern Cameroon in Central Africa, and in Uganda along the bor-
der of the Democratic Republic of Congo (DRC) (Fig.2). Here, it is
important to note that although artisanal mining poses a serious
Central Africa
C
Mining density
Ape density
A
East Africa
West Africa B
Fig. 2. Spatial distribution of mining and ape density. Bivariate choropleth showing the relationship between mining density (using 50- km buers around mining loca-
tions) and ape density in (A) West Africa (operational=18.4%; preoperational=81.6%), in (B) Central Africa (operational=8.3%; preoperational=91.7%), and in
(C) East Africa (operational=12.2%; preoperational=87.8%). Each color change indicates a 20% quintile change in mining and ape density. Lower bounds for both min-
ing and ape density are indicated in the color matrix.
Downloaded from https://www.science.org at Universitaet Leipzig on April 04, 2024

Junker et al., Sci. Adv. 10, eadl0335 (2024) 3 April 2024
SCIENCE ADVANCES | RESEARCH ARTICLE
4 of 12
threat to apes and other wildlife in and around protected areas [e.g.,
(38)], it was not included in this analysis for reasons described in the
methods. Central Africa included the largest percentage of areas
with high ape densities outside mining areas (63%), followed by East
(20%), and West Africa (14%), i.e., areas potentially not threatened
by mining (g.S1). e most critical areas, i.e., those with relatively
high ape densities (0.16 to 6.07 apes/km2, median=0.3) and mod-
erate to high mining densities (3 to 42 mining areas/km2; medi-
an=3.8) are currently not protected (g.S2).
Mining overlap with high versus low ape density areas
Preoperational, and to a lesser degree, operational mining locations
and their 10- km buers in Liberia, Senegal, and Sierra Leone in
West Africa more oen overlapped with high– than with low–ape
density areas (Fig.3). In these countries, chimpanzee range is either
very restricted (i.e., Senegal) or chimpanzees are widely distributed
but their range is highly fragmented (i.e., Liberia, Sierra Leone)
and competition for dierent land uses is high. In countries with
relatively large and/or less fragmented ape populations, such as the
Republic of Guinea (from here on “Guinea”) in West Africa and in
Cameroon, Congo, Equatorial Guinea and Gabon in Central Africa,
mining areas consistently had lower ape densities than nonmining
areas. In Burundi and in Côte d’Ivoire, most of the apes occur in a
few protected areas, where industrial mining is less of a threat be-
cause industrial- scale natural resource extraction activities are usu-
ally prohibited in these.
Positive spatial correlations between mining and ape density
(i.e., more mining areas located in high– than low–ape density ar-
eas) were observed more frequently when analyzed for mining areas
with 50- km buers (g.S3). When using 50- km buers to approxi-
mate potential negative indirect impact of mining activities [see e.g.,
(24, 39, 40)], mining areas in ve of eight West African range coun-
tries (Guinea, Guinea- Bissau, Liberia, Mali, Senegal) overlapped
more oen with high– than low–ape density areas within each of
those countries. Mining areas in Tanzania and Uganda in East
Africa, and in Gabon and Cameroon in Central Africa, also more oen
overlapped with high than low ape densities. Some relatively small
countries (Burundi, Rwanda, and Equatorial Guinea) and those with
Fig. 3. Box plots comparing the average dierence in randomly selected samples of ape densities between areas within a 10- km buer of preoperational and
operational mining areas and randomly selected nonmining areas across countries in West Africa, Central Africa, and East Africa. The dotted line indicates no
dierence between these areas. Values above the dotted line indicate that mining areas are located more often in areas with high than low ape densities and vice versa.
Nigeria and Rwanda are excluded as they do not include pixels that occur inside the ape range. Signicant dierences are marked with an asterisk (*P<0.01, **P<0.001).
WA, West Africa; CA, Central Africa; EA, East Africa.
Downloaded from https://www.science.org at Universitaet Leipzig on April 04, 2024

Junker et al., Sci. Adv. 10, eadl0335 (2024) 3 April 2024
SCIENCE ADVANCES | RESEARCH ARTICLE
5 of 12
very small and spatially restricted ape populations (Côte d’Ivoire
and Nigeria), showed the reverse pattern (i.e., mining areas over-
lapped more oen with low than high ape densities) and in Congo,
a country with a very large and widely distributed ape population,
mining areas consistently had lower ape densities than nonmining
areas. In Ghana, there was no dierence between operational and
preoperational mining and nonmining areas, neither for mining lo-
cations with 10- nor 50- km buers, probably because of the extreme-
ly small population size [≅25 chimpanzees; (41)] and restricted area
of this ape population resulting in low chimpanzee densities both
inside and outside of mining areas. For detailed statistics of the
t tests, refer to tableS1.
Overlap of ape populations with mining areas
Mining areas and their 10- and 50- km buers overlapped with 3 and
34% of the total ape population in Africa, respectively (Table1). e
spatial overlap of preoperational and operational mining areas with
habitat important to apes was highest in West Africa, followed by
East and Central Africa (Fig.4). However, it is important to note
that most of these areas (84.6%) represent mineral exploration areas
(i.e., preoperational mining areas), which may or may not become
operational in the future. Countries with the largest overall overlaps
in ape population abundance and mining areas (in terms of num-
bers of apes potentially aected) included Gabon, Congo, and Cam-
eroon in Central Africa and Guinea in West Africa (table S2).
Although our dataset included fewer mining areas in Central
(12% of total mining areas) than in East (27%) and West African range
countries (61%), more individual apes would potentially be threat-
ened by mining in this region, because of higher overall ape densi-
ties in this region (42). Countries that had the largest proportional
overlaps between ape population abundance and mining areas (in
terms of proportion of population potentially aected) included
Liberia, Sierra Leone, Mali, and Guinea, all of which are located in
West Africa (Fig.4). erefore, Guinea had one of the largest pro-
portional and overall overlaps of mining- and chimpanzee density,
where >23,000 individuals or up to 83% of Guinea’s population
could be directly or indirectly inuenced by mining activities soon.
All country- specic overlap statistics are available in tableS2.
Overlap with critical habitat triggered by biodiversity
features other than apes
We found that 20% of mining locations and their 10- km buers
overlapped with potentially additional CH triggered by biodiversity
features other than apes (g.S4). is suggests that many areas con-
taining critical habitat features not specically related to great apes
may face potential threats from mining activities. When we com-
pared CH to ape density distribution, we found large areas that did
not overlap between these two layers (g.S5). is indicates that the
Global Critical Habitat Map (43) omits extensive areas of ape habitat
that, according to international standards like the IFC Performance
Standard 6, should actually qualify as CH. is discrepancy is most
profound in Guinea and Sierra Leone in West Africa, and in Congo
and Gabon in Central Africa, suggesting that in these countries, CH
is particularly maldened and needs to be more inclusive of areas
important to apes.
Availability of ape data for mining areas
At the time of analysis, only 3% of pixels included in mining areas had
survey data stored in the IUCN SSC A.P.E.S. Database (11), and only
1% of the total area surveyed and archived in the A.P.E.S. Database
overlapped with operational or preoperational mining areas (g.S6).
DISCUSSION
Corporations and their operations are the most important contribu-
tors to worldwide biodiversity loss and ecosystem destruction (44).
Mining is one of the top drivers of deforestation globally with tropical
rainforests standing out as mining- induced deforestation hotspots
(24). Moreover, deforestation within current mining leases suggests
that the rate of mining- related forest loss has increased signicantly
over the past 10 years (24). ese patterns, which are driven by a
rapidly growing global demand for critical metals vital to energy
transitions, are expected to exacerbate deforestation over the com-
ing years if companies continue business as usual. Until now, private
sector contributions to a more sustainable and nature- positive fu-
ture have remained low. In a recent ranking published by the World
Benchmarking Alliance (45), only 5% of the 400 assessed companies
had carried out science- based nature and biodiversity impact as-
sessments of their operations and business models.
To address these issues, the Sustainable Critical Minerals Alli-
ance (SCMA) was announced at the 15th Conference of the Parties
to the Convention on Biological Diversity (COP15). Its work plan,
funded by member countries and private sector partners, focuses on
four key areas: (i) promoting responsible mining practices, (ii) de-
veloping new low- impact technologies, (iii) creating circular econo-
mies for critical minerals, and (iv) sharing benets equitably. Related
to key area 1, this study provides species- level data on the potential
threat of mining on population abundance across the entire range of
African great apes, a taxon threatened by extinction and of high eco-
logical, economical and anthropological signicance. Our results
Table 1. Total and proportional overlap between ape density distribution and mining areas with 10- and 50- km buers in West, Central, and East
Africa.
Region
No. of apes potentially
threatened by mining (10- km
buers)
Proportional overlap (10- km
buers)
No. of apes potentially
threatened by mining (50- km
buers)
Proportional overlap (50- km
buers)
West Africa 5,691 12% 39,599 82%
Central Africa 10,711 2% 135,042 29%
East Africa 292 4% 4,175 62%
Total 16,694 3% 178,816 34%
Downloaded from https://www.science.org at Universitaet Leipzig on April 04, 2024

Junker et al., Sci. Adv. 10, eadl0335 (2024) 3 April 2024
SCIENCE ADVANCES | RESEARCH ARTICLE
6 of 12
indicate that the extent of the potential threats of mining on apes in
Africa has been grossly underestimated. In many instances and
throughout their range in Africa, preoperational and operational
mining areas coincide with areas of high importance to apes, where
many of these overlapping areas currently lack adequate protection
measures (Fig.2). Although DRC was not included in our analyses,
there is evidence that mining has had signicant impacts on the East-
ern chimpanzee (Pan troglodytes schweinfurthii) and Grauer’s gorilla
(Gorilla beringei graueri) populations inside and outside protected
areas, supporting our results. In particular, Plumptre etal. (46) un-
covered a marked decline in Grauer’s gorilla densities of more than
80% over a 20- year period. e authors ascribe this to widespread
insecurity, along with evidence of armed militias and rebel groups
engaging in poaching of apes in and around artisanal mining sites in
the study area.
e overlap of mining and ape habitat was particularly profound
in West Africa, which was also the region with the largest number of
mining areas. Here, ape range is highly fragmented and spatially re-
stricted and areas with large mineral deposits that are not yet devel-
oped, are directly competing with areas that are important to apes.
Furthermore, great ape densities were signicantly higher inside
than outside mining locations and their 10- km buers in three of
eight West African range countries (Fig.3) and in ve of eight coun-
tries when using 50- km buers (g.S3). We estimated that more
than one- third of the entire great ape population in Africa—nearly
180,000 individuals—could be directly or indirectly threatened by
mining now and in the near future. Apes in West Africa could be
most severely aected, where up to 82% of the population currently
overlaps with operational and preoperational mining locations and
their 50- km buers (Fig.4).
Given the increase in overlap between areas developed by min-
ing projects and areas preserved in their natural state to protect apes
and other threatened wildlife species, we have to substantially step
up our eorts to integrate conservation goals with economic devel-
opment targets. e “mitigation hierarchy” (37, 47), as articulated
by the Business and Biodiversity Osets Programme and the IFC, is
a best practice approach to managing potential impact on biodiver-
sity by development projects that receive funding from IFC or other
lenders that align with their standards. is approach advocates ap-
plying eorts early in the development process to avoid adverse im-
pacts to biodiversity wherever possible, then reduce impacts that
cannot be avoided, rehabilitate aected areas, and then compensate
for any residual impacts (48, 49). However, mining companies fre-
quently only apply measures to mitigate (i) direct impact (ii) during
exploitation and (iii) within the mining lease boundaries. ey fail
to consider that their impacts, whether direct or indirect, occur dur-
ing all project development stages and spill over to a wider geo-
graphic area. To allow ape populations to disperse and relocate,
mitigation of both direct and indirect impact should extend beyond
the administrative boundaries of the mining project. At the same
time, companies should make a greater eort to identify and antici-
pate indirect impacts induced by e.g., mining- related human in-
migration and zoonotic disease transmission. e time frame over
which a net gain in ape population abundance is achieved is also all
Fig. 4. Overlap between ape density distribution and mining areas in Africa. (A) Proportion of ape population threatened by mining (operational and preoperational
mining areas) with a 10- km buer (dark shades) and with a 50- km buer (light shades) for range countries in the dierent regions. Total regional estimates of the propor-
tion of ape populations threatened by mining in (B) West Africa, in (C) Central Africa, and in (D) East Africa.
Downloaded from https://www.science.org at Universitaet Leipzig on April 04, 2024

Junker et al., Sci. Adv. 10, eadl0335 (2024) 3 April 2024
SCIENCE ADVANCES | RESEARCH ARTICLE
7 of 12
too oen underestimated. If the time frame is too short, the popula-
tions may not have enough time to increase suciently.
Considering the complex social organization and dynamics of
African great apes and associated elevated risks of mortality, as well
as the paucity of suitable release sites, translocation of groups from
highly aected areas is not a feasible option (50). In addition, trans-
location and relocation of wildlife potentially raises several ethical
and legal issues due to the stress inicted on the animal and risks
associated with starvation and predation by other species (51). Last,
restoring habitat simply takes too long for resident apes to benet
from this intervention. erefore, unless great ape habitat is avoided
entirely, mitigation is unlikely to prevent ape population declines.
Companies should therefore reconsider the long- term feasibility of
exploration sites in areas important for apes, due to their environ-
mental responsibilities and the costs associated with achieving no
net loss/net gain in ape abundance. Also, lending banks should re-
frain from funding projects in these areas. To illustrate this, if cor-
porations ceased their exploratory activities in areas likely to contain
a minimum of 20 apes, this would result in 38% (22 of 58) of puta-
tive mining projects situated within the African ape habitat to re-
main undeveloped. Notably, nine of these areas exhibit the potential
to accommodate populations exceeding 50 apes.
To compensate for any residual impact that could not be avoided,
reduced, or restored, mining companies can implement compensa-
tion measures by creating biodiversity osets to ensure that an equal
or greater area of identical habitat or ape population is protected or
improved (52). However, osets are controversial and their eec-
tiveness for apes has yet to be demonstrated (53–55). Oset design
and implementation is frequently guided by company internal stan-
dards, lending banks, or international best practices and few African
ape- range countries have national policies guiding or requiring o-
sets (53). A recent independent assessment by the ARRC Task Force
of the Section on Great Apes and Section on Small Apes of the
IUCN SSC Primate Specialist Group (56) has shown that even the
most ambitious and cutting- edge eorts by the private sector to o-
set residual impacts on apes and their habitat are not sucient to
eectively mitigate the total loss they incur to great ape populations.
One key factor is the duration of osets, which is oen set equal to
the length of exploitation activities (generally c.20 years). is time
period is too short to achieve any signicant gains for apes. ese
temporary actions do not ensure long- term conservation of apes,
while most impacts at the mining sites are permanent. Osets also
do not consider impacts from mining exploration activities, and
legacy impacts when projects are sold to dierent companies.
Where compensation schemes are considered, osets must be
designed in such a way as to take into account the cumulative threats
across the landscape or region, ideally forming part of existing na-
tional or regional conservation strategies. e estimates provided in
this study could serve as an approximation based on which an initial
screening for suitable aggregated oset schemes could be conducted.
Our study also provides some guidance with regards to where to
compensate for residual impact. Investing in increased protection
might be more feasible where high ape densities exist outside of
mining areas. Alternatively, aggregated oset strategies could focus
on contributing to existing protected areas to improve their eec-
tiveness (e.g., by nancially investing in management activities and
sta) (g.S2).
We also found that 20% of mining areas overlapped with areas
that likely qualify as CH triggered by biodiversity features other
than apes (g. S4), which, according to international regulatory
frameworks, would hinder projects from receiving nancial support
[i.e., (37)]. Similarly, another study found that 32% of all mammal
species worldwide with more than 30% of habitat within mining ar-
eas are currently listed as reatened with extinction on the IUCN
Red list (57). Because species of conservation concern would likely
trigger CH status, companies operating in these areas should have
adequate mitigation and compensation schemes in place to mini-
mize their impact, which seems unlikely, given that most companies
lack robust species baseline data (45). What is of even greater con-
cern is the spatial overlap between areas set aside for conservation
and those potentially inuenced by mining. For example, it is esti-
mated that 8% of the global area potentially inuenced by mining
overlaps with protected areas, 16% with Remaining Wilderness and
7% with Key Biodiversity Areas (2). Another study that examined the
intersection of mines with protected areas identied 2558 boundary
violations totaling about 6232 km2, or 9.5% of all areas identied as
mining projects (58). is is supported by the information on world-
wide downgrading, downsizing, and degazettement of protected ar-
eas (PADDD), providing evidence for more than 3000 enacted cases
of PADDD in nearly 70 countries, covering about 1,300,000 km2
[updated from (59)].
Our results conrmed the lack of data sharing by mining projects,
where only 1% of the ape survey data from Africa that is currently
stored in the IUCN SSC A.P.E.S. Database—the public repository for
data from surveys of apes and their habitats—was collected in and
around mining areas (g.S6). is lack of transparent data sharing
hampers science- based quantication of impacts of mining on apes
and their habitat and the development of eective mitigation strate-
gies. is was reected in the results of the rst global synopsis of the
eects of primate conservation interventions examining approxi-
mately 13,000 publications, which found a marked absence of studies
on the eectiveness of conservation strategies specically designed to
reduce the impact of mining on apes (60). We therefore stress the
need for mining companies to make their biodiversity data publicly
available in a central database, such as A.P.E.S. or the Global Bio-
diversity Information Facility and call on the IFC and other regulatory
frameworks to urge companies to provide access to their data.
e large overlap between mining areas and areas important to
apes is partly because many of the minerals needed for the energy
transition are in places that have not yet been industrialized, which
typically include rural or remote parts of the world. is means that
current climate solutions could lead to more industrialization in
these places, which could worsen the climate crisis (61). e pro-
duction of biofuels from food and feed crops exemplies this para-
dox, where increases in bioenergy cropland to meet global demands
in biofuel are expected to cause severe impacts on biodiversity that
are not compensated by lower climate change impacts (62). In addi-
tion, the injustices inicted by the expansion of industrial develop-
ment are already immense (63) and may worsen with an increase in
unsustainable economic development in previously undeveloped
areas (64). To illustrate this, 69% of energy transition minerals and
metals projects worldwide are on or near land that belongs to Indig-
enous people or small holder farmers and pastoralists, with an even
higher proportion (77%) of overlap in Africa (61).
e SCMA is a signicant step forward in the global eort to en-
sure that the transition to a low- carbon economy is sustainable and
equitable. However, Africa’s great apes and many other threatened
wildlife species are at high risk from industrial mining activities,
Downloaded from https://www.science.org at Universitaet Leipzig on April 04, 2024

Junker et al., Sci. Adv. 10, eadl0335 (2024) 3 April 2024
SCIENCE ADVANCES | RESEARCH ARTICLE
8 of 12
which are likely to increase as the world transitions to a low- carbon
economy. e inclusion of great apes in IFC’s Performance Standard
6 Guidance Note 73 and the creation of the ARRC Task Force, com-
posed of foremost experts in ape conservation tasked with oering
independent guidance on how to mitigate the adverse eects of en-
ergy, extractive, and related infrastructure projects on apes, instills
optimism that eorts to integrate conservation goals with economic
development targets are increasingly being taken up by environmen-
tal policy and private investors. Our ndings highlight the need for
the mining sector to increase transparency and make their environ-
mental data more accessible. We therefore call upon lending banks,
such as the World Bank to ensure that World Bank–supported infra-
structure and other development projects make their ape survey data
accessible in a central database like the A.P.E.S. Database. is would
allow for better independent assessments of the risks posed by min-
ing activities to endangered ora and fauna. We also call upon com-
panies, lenders, and nations to reevaluate investments in exploration
activities in areas of high biodiversity and importance to great apes
and recognize the greater value of leaving some regions untouched
by industrial activity, as these actions are vital for preserving ecosys-
tem services, preventing disease spillovers addressing future epidem-
ics or pandemics, and mitigating climate change.
Limitations of the approach
We opted to use point locations for mining properties, because this
is the only dataset currently available that includes preoperational
and operational mining locations. Including only those sites that are
operational (i.e., areas where direct impact caused by mining activi-
ties is visible on satellite images) would considerably underestimate
potential negative impact of mining on African great apes as the
majority of mining areas are still in the exploration phase (proportion
preoperational mines: West Africa=81.6%, Central Africa =91.7%,
East Africa=87.8%).
One limitation of this dataset is that it does not include artisanal
mining areas. Although small- scale, informal, and artisanal mining
areas constitute only 1.63% of the total mine area globally, the propor-
tional magnitude of the artisanal mining footprint is likely substantial,
because these areas are oen associated with severe environmental
risks and no ecological protection measures (58). erefore, our esti-
mate of the impact of mining activities on apes is probably an under-
estimate of the true impact. Adding to this, mining activities have
been observed to cause indirect impacts that expand across space and
persist over time (58).
On the other hand, because the majority of mining areas included
in this study are still in the exploration phase, it can be expected
that not all of the preoperational mining areas will become opera-
tional in the future. A number of studies estimated the success rate
of mining exploration (i.e., the proportion of exploration sites that
become extraction sites) and calculated that the likelihood of dis-
covery of a major deposit in areas where little to no previous mining
activity has occurred, ranges from 0.3 to 0.5%, and is 5% in areas
where mining activities have taken place previously (65). However,
the geological potential of a site is not the only factor determining
the success of a mine and other aspects, such as economic viability,
market demand, social acceptance, global economic conditions, and
regulatory and environmental factors, among others, inuence re-
turn on investments in mining. While the return on investments is
less than 1% globally, for Australia and Africa returns on invest-
ments in mining are considerably higher and estimated at 12 and
38%, respectively (65). Also, while a mine might be regarded as eco-
nomically unfeasible at one point in time, it may become feasible at
another point in time (e.g., as demand or the price for the mineral
increases).
Likewise, operational mines may be implementing eective miti-
gation measures, thereby not aecting all great apes within 10- or
50- km buers. Because the eects of these processes are dicult to
quantify with the limited data at hand, they were also not considered
in this analysis. Furthermore, because robust data on the extent of
the direct and indirect impact of mining activities on apes in Africa
are lacking, the buers used in this study are mere approximations
of true impact and will vary greatly from mine to mine. In some
instances, they may be an overestimate of the actual impact, e.g., in
the case of relatively recent mines, or mines that have implemented
appropriate avoidance and minimization measures, and in other
cases they may underestimate the true impact of the mine, e.g., well-
established and relatively large- scale operational mines. We exclud-
ed road development from our impact assessment because of the
challenge of determining whether a road was built as a result of min-
ing activities or for other reasons. Last, another source of uncer-
tainty is the highly dynamic nature of impact from mines. A mine
life cycle may involve periods of expansion followed by periods of
reclamation or revegetation, further complicating the interpretation
of results. Despite these limitations, we think that the results pre-
sented in this study provide a useful global assessment of the poten-
tial threats of mining on apes in Africa. To be able to address these
limitations in the future, we stress the need for conducting scienti-
cally robust impact assessments inside and around mining projects
in dierent range countries with dierent species of apes at varying
densities and over suciently long periods of time.
MATERIALS AND METHODS
Study design
We used various data sources for analysis related to mining density
and ape density in dierent geographical locations (Table2). We
used Mollweide equal area projection to analyze all data listed in
Table1 and matched all spatial layers at a 1 × 1 km pixel resolution.
We combined great ape density distributions modeled by (41) and
(7, 42) and mapped this for each range country in Africa (referred to
as “range country” throughout the text). We excluded the DRC and
the Central African Republic from analysis because of a lack of ape
density information (42). However, in DRC there is extensive min-
ing occurring within the Eastern chimpanzee and Grauer’s gorilla’s
range, inside and outside protected areas, and thus, the impact on
their population is likely signicant (46). e metric for ape density
distribution is the number of apes per pixel.
We had two mining datasets: a dataset that included industrial (i)
preoperational (i.e., exploration) and (ii) operational (i.e., exploita-
tion) mining locations both with a 10- cell and a 50- cell radius, col-
lectively referred to as “mining areas” throughout the text (2). e
point layer distinguishes neither between open pit or underground
mines nor between dierent mining materials. Values in these spa-
tial layers estimate mining density (i.e., number of overlapping min-
ing areas per pixel). Because none of the preoperational sites are
currently being mined, we use these as a proxy for potential future
mining areas, recognizing that some of these sites may never be de-
veloped. We converted mining densities to binary values to indicate
mining inuenced areas where mining density was >0.
Downloaded from https://www.science.org at Universitaet Leipzig on April 04, 2024

Junker et al., Sci. Adv. 10, eadl0335 (2024) 3 April 2024
SCIENCE ADVANCES | RESEARCH ARTICLE
9 of 12
Buer areas
e global dataset on mining locations used in this study includes
point locations only, and as such, the boundaries of the mining con-
cession were not known. We therefore dened buers to reect the
approximate extent of direct and indirect impacts from mines. To do
this, we considered the results of previous studies that estimated av-
erage mining area, which is the area likely to be included within
mining lease boundaries and which ranges from 0.36 to 12.3 km2.
e study that estimated average mining area sizes at >12.3 km2 fo-
cused on larger- scale operations (66), whereas studies that reported
average mining area sizes at <2 km2 included artisanal mining areas
(16, 58, 67–69). Because the dataset used in this study provides coor-
dinates of larger- scale mines, and because mining- related threats
like light and noise pollution or hunting cannot be visualized from
satellite images, we believe that using a 10- km buer to approximate
the direct impact from mines is justied. Indirect impact of mining, on
the other hand, has commonly been assumed to extend 50 to 70 km
beyond the boundaries of mining areas (24, 39, 40) and we therefore
decided to use a 50- km buer to assess potential indirect impact of
mining on African great apes. We would like to emphasize that
these boundaries do not serve as a universally precise distinction
between what is considered direct and indirect impact. Instead, in
the context of this study, they function as guidelines to broadly char-
acterize impact patterns and simplify spatial analyses.
Statistical analyses
Geographical distribution of mining density in relation to
ape density
We used mining density, i.e., the number of mining areas (i.e., point
locations and their 10- and 50- km buers) overlapping with each
Table 2. Name, description, spatial resolution, spatial extent, and source of datasets used in this analysis.
Name Description Spatial resolution Spatial extent Source
Global mining areas Global map of operational
and preoperational mining
locations using 10- cell and
50- cell radii based on the
mining properties database
1 × 1 km Global (2)
Range- wide African great ape
density distribution
Model continent- wide great
ape density distribution
based on site- level estimates
of African great ape abun-
dance
10 × 10 km African great ape range, ex-
cluding DRC, Central African
Republic, Liberia
(42)
Range- wide western
chimpanzee density
distribution
Range- wide density
distribution model based
on reconnaissance and line
transect data in the IUCN SSC
A.P.E.S. Database
30 arc sec Western chimpanzee range (41)
Liberia chimpanzee density
distribution
Nationwide density distri-
bution model based on line
transect data
1 × 1 km Liberia (7)
Global Critical Habitat map Global screening layer of
Critical Habitat in the terres-
trial realm based on global
spatial datasets covering the
distributions of 12 biodiver-
sity features aligned with
guidance provided by the IFC
1 × 1 km Global (43)
This study dened CH on the
basis of biodiversity features
grouped into ve broad cat-
egories: (i) protected areas,
(ii) Key Biodiversity Areas,
(iii) threatened ecosystems,
(iv) critical sites for selected
species (tigers and sea
turtles), and (v) the distribu-
tions of threatened species
qualifying under IUCN Red
List criterion D
IUCN SSC A.P.E.S. Database Archive of existing ape popu-
lation survey data
Site- specic Global ape range (11)
IUCN African apes range layer Merged boundaries of distri-
butional ranges of all African
great ape ranges
Species- level African great ape range (5)
Downloaded from https://www.science.org at Universitaet Leipzig on April 04, 2024

Junker et al., Sci. Adv. 10, eadl0335 (2024) 3 April 2024
SCIENCE ADVANCES | RESEARCH ARTICLE
10 of 12
pixel, and mapped this in relation to ape density distribution across
range countries. Here, we merged operational and preoperational
mining areas. We then grouped the values for mining density and
ape density into quintiles and classied each pixel depending on the
product of these two factors, resulting in a total of 25 classes. We
mapped these 25 classes over geographical space to visualize pris-
tine ape habitat (low mining density; high ape density) versus ape
habitat threatened by mining (high mining density; high ape densi-
ty) versus areas where mining does not threaten ape populations
(high mining; low ape densities) and where neither mining densities
nor ape densities are high. We excluded areas with values of zero for
both mining and ape density. For each region, we also plotted the
percentage area across quintiles of varying ape density and mining
density, where we restricted this analysis to areas with high ape den-
sities (i.e., including only ape density values that fell into the h
quintile) and across all ve quintiles of varying mining density.
Mining overlap with high– versus low–ape density areas
To compare ape density dierences between mining and nonmining
areas across the ape range and within each country, we rst overlaid
mining areas with ape densities. We then compared the distribution
of ape densities from pixels that overlapped with mining areas to
those outside of mining areas, but within the ape range. To account
for the large variation in pixel numbers between countries, as well as
mining and nonmining areas (mining areas always had much fewer
pixels than nonmining areas), we selected the total number of pixels
from within mining areas and randomly selected the same number
of pixels without replacement from nonmining areas separately for
each country. We then performed a t test, repeating the process for
1000 iterations, to determine whether there were density dierences
between mining and nonmining areas. e large number of itera-
tions and a random selection approach minimized the likelihood of
biased results stemming from specic pixel selection and resulted in
more representative samples. is process was done separately for
preoperational and operational mines and for mining locations with
10- and 50- km buers.
Overlap of ape populations with mining areas
We overlaid the mining areas with ape density distribution and
summed the number of apes estimated for each pixel at 1 × 1 km
resolution to estimate the proportion of total ape population poten-
tially threatened by current and future mining activities in each re-
gion and ape range country. Each pixel in the ape density distribution
layer was weighted by the amount of overlap with mining areas. If,
for example, 30% of the pixel area fell into a mining area, then only
30% of the number of apes in that pixel was included in the overall
estimate of threatened apes per region and range country.
Overlap with critical habitat triggered by biodiversity features
other than apes
We followed the procedure described in the previous section and
summed the number pixels at 1 × 1 km resolution to estimate the
proportion of area identied as likely and potential CH triggered by
biodiversity features other than apes (Global Critical Habitat map;
Table1), that overlapped with operational and preoperational min-
ing areas in each region (West Africa, East Africa, Central Africa)
and range country. Each pixel in the Global Critical Habitat map
was then weighted by the amount of overlap with mining areas. To
investigate how likely CH triggered by the occurrence of apes com-
plemented (or not) the areas identied as likely or potential CH trig-
gered by biodiversity features other than apes, we compared the
Global Critical Habitat map (clipped to range countries) with ape
density distribution. is allowed us to identify additional areas
of likely CH not yet included in the output maps produced by
Brauneder etal. (43).
Availability of ape data for mining areas
We consulted the data in the IUCN SSC A.P.E.S. Database (11) to
determine whether survey data existed for sites that overlapped with
mining locations and their 10- km buers. Here, we only included
mining areas within the distributional range of great apes [(5);
Table1]. To know whether an ape survey was conducted in the area
or not (which also included surveys that did not report the presence
of apes in the area), we mapped all observations recorded during
surveys over the global mining areas layer (Table1) and calculated
the proportion of pixels included in mining areas for which survey
data were available (i.e., via request to the A.P.E.S. Database). Here,
we also included in the analysis the DRC and the Central African
Republic because we assessed the spatial overlap of survey data from
the A.P.E.S. Database (and not ape densities as in the previous anal-
ysis) with mining areas.
Data processing
All analyses were performed in R (Version 4.2.0) using the following
R packages: “raster” (70), “terra” (71), “sp” (72), “sf” (73), “rgdal” (74),
“ggplot2” (75), and “dplyr” (76), “tidyr” (75), and “reshape2” (77). In
addition, we used QGIS (V 3.26.2) and ArcMap (V 10.7.1) to spa-
tially visualize our data on maps.
Supplementary Materials
This PDF le includes:
Supplementary Text
Figs.S1 to S6
TablesS1 and S2
REFERENCES AND NOTES
1. D. P. Edwards, S. Sloan, L. Weng, P. Dirks, J. Sayer, W. F. Laurance, Mining and the African
Environment. Conserv. Lett. 7, 302–311 (2014).
2. L. J. Sonter, M. C. Dade, J. E. M. Watson, R. K. Valenta, Renewable energy production will
exacerbate mining threats to biodiversity. Nat. Commun. 11, 4174 (2020).
3. IPBES, The IPBES regional assessment report on biodiversity and ecosystem services for
Africa. doi: 10.5281/ZENODO.3236178 (2018).
4. A. Estrada, P. A. Garber, A. B. Rylands, C. Roos, E. Fernandez- Duque, A. Di Fiore,
K. A.- I. Nekaris, V. Nijman, E. W. Heymann, J. E. Lambert, F. R overo, C. Barelli, J. M. Setchell,
T. R. Gillespie, R. A. Mittermeier, L. V. Arregoitia, M. de Guinea, S. Gouveia, R. Dobrovolski,
S. Shanee, N. Shanee, S. A. Boyle, A. Fuentes, K. C. MacKinnon, K. R. Amato, A. L. S. Meyer,
S. Wich, R. W. Sussman, R. Pan, I. Kone, B. Li, Impending extinction crisis of the world’s
primates: Why primates matter. Sci. Adv. 3, e1600946 (2017).
5. IUCN, The IUCN Red List of Threatened Species, Version 2015.1; www.iucnredlist.org.
6. E. J. Mace, E. A. Williamson, Best Practice Guidelines for Great Ape Tourism. (IUCN, 2010).
7. J. Junker, C. Boesch, T. Freeman, R. Mundry, C. Stephens, H. S. Kühl, Integrating wildlife
conservation with conicting economic land- use goals in a West African biodiversity
hotspot. Basic Appl. Ecol. 16, 690–702 (2015).
8. E. Tarszisz, S. Tomlinson, M. E. Harrison, H. C. Morrogh- Bernard, A. J. Munn, Gardeners of
the forest: Eects of seed handling and ingestion by orangutans on germination success
of peat forest plants. Biol. J. Linn. Soc. 123, 125–134 (2017).
9. D. Lawrence, M. Coe, W. Walker, L. Verchot, K. Vandecar, The Unseen Eects of
Deforestation: Biophysical Eects on Climate. Front. Forests Global Change 5, 1 (2022).
10. IUCN, IUCN SSC A.P.E.S. Database.
11. S. Heinicke, I. Ordaz- Németh, J. Junker, M. E. Bachmann, S. Marrocoli, E. G. Wessling,
D. Byler, S. M. Cheyne, J. Desmond, D. Dowd, M. Fitzgerald, M. Fourrier, A. Goedmakers,
R. A. Hernandez- Aguilar, A. Hillers, K. J. Hockings, S. Jones, M. Kaiser, K. Koops,
J. M. Lapuente, F. Maisels, J. Riedel, E. Terrade, C. G. Tweh, V. Vergnes, T. Vogt,
E. A. Williamson, H. S. Kühl, Open- access platform to synthesize knowledge of ape
conservation across sites. Am. J. Primatol. 83, e23213 (2021).
12. A. T. Martins- Oliveira, M. Zanin, G. R. Canale, C. A. da Costa, P. V. Eisenlohr,
F. C. S. A. de Melo, F. R. de Melo, A global review of the threats of mining on mid- sized and
large mammals. J. Nat. Conserv. 62, 126025 (2021).
Downloaded from https://www.science.org at Universitaet Leipzig on April 04, 2024

Junker et al., Sci. Adv. 10, eadl0335 (2024) 3 April 2024
SCIENCE ADVANCES | RESEARCH ARTICLE
11 of 12
13. M. H. L. Duarte, M. C. Kaizer, R. J. Young, M. Rodrigues, R. S. Sousa- Lima, Mining noise
aects loud call structures and emission patterns of wild black- fronted titi monkeys.
Primates 59, 89–97 (2018).
14. L. I. Rabanal, H. S. Kuehl, R. Mundry, M. M. Robbins, C. Boesch, Oil prospecting and its
impact on large rainforest mammals in Loango National Park, Gabon. Biol. Conserv. 143,
1017–1024 (2010).
15. G. Shannon, M. F. McKenna, L. M. Angeloni, K. R. Crooks, K. M. Fristrup, E. Brown,
K. A. Warner, M. D. Nelson, C. White, J. Briggs, S. McFarland, G. Wittemyer, A synthesis of
two decades of research documenting the eects of noise on wildlife. Biol. Rev. 91,
982–1005 (2016).
16. V. Maus, S. Giljum, J. Gutschlhofer, D. M. da Silva, M. Probst, S. L. B. Gass, S. Luckeneder,
M. Lieber, I. McCallum, A global- scale data set of mining areas. Sci. Data 7, 289 (2020).
17. M. J. Patiño Ropero, N. Rodríguez Fariñas, R. Mateo, J. J. Berzas Nevado,
R. C. Rodríguez Martín- Doimeadios, Mercur y species accumulation and trophic transfer
in biological systems using the Almadén mining district (Ciudad Real, Spain) as a case of
study. Environ. Sci. Pollut. Res. 23, 6074–6081 (2016).
18. R. E. Junge, C. V. Williams, H. Rakotondrainibe, K. L. Mahefarisoa, T. Rajaonarivelo,
C. Faulkner, V. Mass, Baseline health and nutrition evaluation of two sympatric nocturnal
lemur species (avahi laniger and lepilemur mustelinus) residing near an active mine site
at ambatovy, madagascar. J. Zoo Wildl. Med. 48, 794–803 (2017).
19. Y. Zhao, Y. Chen, A. M. Ellison, W. Liu, D. Chen, Establish an environmentally sustainable
Giant Panda National Park in the Qinling Mountains. Sci. Total Environ. 668, 979–987 (2019).
20. M. J. Sanders, L. Miller, S. A. Bhagwat, A. Rogers, Conservation conversations: A typology
of barriers to conservation success. Oryx 55, 245–254 (2021).
21. N. L. Alvarez- Berríos, T. M. Aide, Global demand for gold is another threat for tropical
forests. Environ. Res. Lett. 10, 14006 (2015).
22. A. González- González, N. Clerici, B. Quesada, Growing mining contribution to Colombian
deforestation. Environ. Res. Lett. 16, 64046 (2021).
23. H. A. Seki, J. P. R. Thorn, P. J. Platts, D. D. Shirima, R. A. Marchant, Y. Abeid, N. Baker,
M. Annandale, A. R. Marshall, Indirect impacts of commercial gold mining on adjacent
ecosystems. Biol. Conserv. 275, 109782 (2022).
24. S. Giljum, V. Maus, N. Kuschnig, S. Luckeneder, M. Tost, L. J. Sonter, A. J. Bebbington, A
pantropical assessment of deforestation caused by industrial mining. Proc. Natl. Acad. Sci.
U.S.A. 119, e2118273119 (2022).
25. S. A. Lahm, R. F. W. Barnes, K. Beardsley, P. Cer vinka, A method for censusing the greater
white- nosed monkey in nor theastern Gabon using the population density gradient in
relation to roads. J. Trop. Ecol. 14, 629–643 (1998).
26. W. F. Laurance, B. M. Croes, L. Tchignoumba, S. A. Lahm, A. Alonso, M. E. Lee, P. Campbell,
C. Ondzeano, Impacts of roads and hunting on central African rainforest mammals.
Conserv. Biol. 20, 1251–1261 (2006).
27. C. B. Yackulic, E. W. Sanderson, M. Uriarte, Anthropogenic and environmental drivers of
modern range loss in large mammals. Proc. Natl. Acad. Sci. U.S.A. 108, 4024–4029 (2011).
28. B. Galea, T. Humle, Identifying and mitigating the impacts on primates of transportation
and service corridors. Conserv. Biol. 36, e13836 (2022).
29. M. Cibot, S. Bortolamiol, A. Seguya, S. Krief, Chimpanzees facing a dangerous situation: A
high- trac asphalted road in the Sebitoli area of Kibale National Park, Uganda. Am.
J. Primatol. 77, 890–900 (2015).
30. S. Krief, A. Iglesias- González, B. M. R. Appenzeller, J. P. Okimat, J. B. Fini, B. Demeneix,
S. Vaslin- Reimann, S. Lardy- Fontan, N. Guma, P. Spirhanzlova, Road impact in a protected
area with rich biodiversity: The case of the Sebitoli road in Kibale National Park, Uganda.
Environ. Sci. Poll. Res. 27, 27914–27925 (2020).
31. B. Andrasi, J. A. G. Jaeger, S. Heinicke, K. Metcalfe, K. J. Hockings, Quantifying the
road- eect zone for a critically endangered primate. Conserv. Lett. 14, e12839 (2021).
32. F. Kleinschroth, J. R. Healey, Impacts of logging roads on tropical forests. Biotropica 49,
620–635 (2017).
33. C. A. Devaux, O. Mediannikov, H. Medkour, D. Raoult, Infectious Disease Risk Across the
Growing Human- Non Human Primate Interface: A Review of the Evidence. Front. Public
Health 7, (2019).
34. D. G. Randolph, J. Resch, S. MacMillan, C. Y. Wright, B. Bett, D. Robinson, B. Wernecke,
H. S. L. Lee, W. B. Karesh, C. Machalaba, A. Fraenkel, M. Barbieri, M. Kappelle, “Zoonotic
diseases and how to break the chain of transmission: A scientic assessment with key
messages for policy- makers” [United Nations Environment Programme (UNEP), 2020].
35. B. Wilcox, B. Ellis, B. A. Wilcox, B. Ellis, Forests and emerging infectious diseases of humans,
Unasylva 57, 11–18 (2006).
36. K. Zimmer, Deforestation is leading to more infectious diseases in humans (National
Geographic, 2019); www.nationalgeographic.com/science/article/deforestation- leading-
to- more- infectious- diseases- in- humans.
37. IFC, Performance standards on environmental and social sustainability. International
Finance Corporation (IFC) 4.0, 942–944 (2012).
38. C. Spira, A. Kirkby, D. Kujirakwinja, A. J. Plumptre, The socio- economics of artisanal mining
and bushmeat hunting around protected areas: Kahuzi–Biega National Park and
Itombwe Nature Reserve, eastern Democratic Republic of Congo. Oryx 53, 136–144 (2019).
39. L. J. Sonter, D. Herrera, D. J. Barrett, G. L. Galford, C. J. Moran, B. S. Soares- Filho,
Mining drives extensive deforestation in the Brazilian Amazon. Nat. Commun. 8,
1013 (2017).
40. M. Kramer, T. Kind- Rieper, R. Munayer, S. Giljum, R. Masselink, P. van Ackern, V. Maus,
S. Luckeneder, N. Kuschnig, F. Costa, L. Rüttinger, Extracted forests: Unhearting the role of
mining- related deforestation as a driver of global deforestation. (2023).
41. S. Heinicke, R. Mundry, C. Boesch, B. Amarasekaran, A. Barrie, T. Brncic, D. Brugière,
G. Campbell, J. Carvalho, E. Danquah, D. Dowd, H. Eshuis, M.- C. Fleury- Brugière, J. Gamys,
J. Ganas, S. Gatti, L. Ginn, A. Goedmakers, N. Granier, I. Herbinger, A. Hillers, S. Jones,
J. Junker, C. Y. Kouakou, V. Lapeyre, V. Leinert, F. Maisels, S. Marrocoli, M. Molokwu- Odozi,
P. K. N’Goran, L. Pacheco, S. Regnaut, T. Sop, E. Ton, J. van Schijndel, V. Vergnes, M. Voigt,
A. Welsh, E. G. Wessling, E. A. Williamson, H. S. Kühl, Advancing conservation planning for
western chimpanzees using IUCN SSC A.P.E.S.—the case of a taxon- specic database.
Environ. Res. Lett. 14, 64001 (2019).
42. I. Ordaz- Németh, T. Sop, B. Amarasekaran, M. Bachmann, C. Boesch, T. Brncic, D. Caillaud,
G. Campbell, J. Carvalho, R. Chancellor, T. R. B. Davenport, D. Dowd, M. Eno- Nku,
J. Ganas- Swaray, N. Granier, E. Greengrass, S. Heinicke, I. Herbinger, C. Inkamba- Nkulu,
F. Iyenguet, J. Junker, K. S. Bobo, A. Lushimba, F. Maisels, G. A. F. Malanda, M. S. McCarthy,
P. Motsaba, J. Moustgaard, M. Murai, B. Ndokoue, S. Nixon, R. A. Nseme, Z. Nzooh,
L. Pintea, A. J. Plumptre, J. Roy, A. Rundus, J. Sanderson, A. Serckx, S. Strindberg, C. Tweh,
H. Vanleeuwe, A. Vosper, M. Waltert, E. A. Williamson, M. Wilson, R. Mundry, H. S. Kühl,
Range- wide indicators of African great ape density distribution. Am. J. Primatol. 83,
e23338 (2021).
43. K. M. Brauneder, C. Montes, S. Blyth, L. Bennun, S. H. M. Butchart, M. Homann,
N. D. Burgess, A. Cuttelod, M. I. Jones, V. Kapos, J. Pilgrim, M. J. Tolley, E. C. Underwood,
L. V. Weatherdon, S. E. Brooks, Global screening for Critical Habitat in the terrestrial realm.
PLOS ONE 13, 1–16 (2018).
44. IPBES, Summary for policymakers of the global assessment report on biodiversity and
ecosystem services. Zenodo [Preprint] (2019); https://doi.org/10.5281/zenodo.3553579.
45. World Benchmarking Alliance, World benchmarking alliance, Wba (2018); www.
worldbenchmarkingalliance.org/nature- benchmark/.
46. A. J. Plumptre, S. Nixon, D. K. Kujirakwinja, G. Vieilledent, R. Critchlow, E. A. Williamson,
R. Nishuli, A. E. Kirkby, J. S. Hall, Catastrophic Decline of World’s Largest Primate: 80% Loss
of Grauer’s Gorilla (Gorilla beringei graueri) Population Justies Critically Endangered
Status. PLOS ONE 11, 1–13 (2016).
47. BBOP, Standard on Biodiversity Osets Business and Biodiversity Osets Programm (BBOP)
3.0, 102 (2012).
48. W. N. S. Arlidge, J. W. Bull, P. F. E. Addison, M. J. Burgass, D. Gianuca, T. M. Gorham,
C. D. S. Jacob, N. Shumway, S. P. Sinclair, J. E. M. Watson, C. Wilcox, E. J. Milner- Gulland, A
Global Mitigation Hierarchy for Nature Conservation. Bioscience 68, 336–347 (2018).
49. B. Phalan, G. Hayes, S. Brooks, D. Marsh, P. Howard, B. Costelloe, B. Vira, A. Kowalska,
S. Whitaker, Avoiding impacts on biodiversity through strengthening the rst stage of
the mitigation hierarchy. Oryx 52, 316–324 (2018).
50. T. Humle, “The dimensions of ape- human interactions in industrial agricultural
landscapes” (Cambridge Univ. Press, 2015), pp. 1–12.
51. C. G. Thulin, H. Röcklinsberg, Ethical Considerations for Wildlife Reintroductions and
Rewilding. Front Vet Sci 7, (2020).
52. A. Lanjouw, “Mining / oil extraction and ape populations and habitats Introduction” in
State of the Apes - Ex tractive Industries and Ape Conservation, A. Lanjouw, H. Rainer,
A. White, Eds. (Arcus Foundation, 2013), pp. 126–161.
53. R. Kormos, C. F. Kormos, T. Humle, A. Lanjouw, H. Rainer, R. Victurine, R. A. Mittermeier,
M. S. Diallo, A. B. Rylands, E. A. Williamson, Great Apes and Biodiversity Oset Projects in
Africa: The Case for National Oset Strategies. PLOS ONE 9, 1–14 (2014).
54. S. O. S. E. Ermgassen, J. Baker, R. A. Griths, N. Strange, M. J. Struebig, J. W. Bull, The
ecological outcomes of biodiversity osets under “no net loss” policies: A global review.
Conserv. Lett. 12, e12664 (2019).
55. L. J. Sonter, J. S. Simmonds, J. E. M. Watson, J. P. G. Jones, J. M. Kiesecker, H. M. Costa,
L. Bennun, S. Edwards, H. S. Grantham, V. F. Griths, K. Jones, K. Sochi, P. Puydarrieux,
F. Quétier, H. Rainer, H. Rainey, D. Roe, M. Satar, B. S. Soares- Filho, M. Starkey, K. ten Kate,
R. Victurine, A. von Hase, J. A. Wells, M. Maron, Local conditions and policy design
determine whether ecological compensation can achieve No Net Loss goals.
Nat. Commun. 11, 2072 (2020).
56. ARRC Task Force, IUCN SSC Primate Specialist Group Section on Great Apes (2020); www.
arrctaskforce.org/.
57. L. J. Sonter, T. J. Lloyd, S. G. Kearney, M. Di Marco, C. J. O’Bryan, R. K. Valenta,
J. E. M. Watson, Conservation implications and opportunities of mining activities for
terrestrial mammal habitat. Conserv. Sci. Pract. 4, e12806 (2022).
58. L. Tang, T. T. Werner, Global mining footprint mapped from high- resolution satellite
imagery. Commun Earth Environ 4, 134 (2023).
59. M. B. Mascia, S. Pailler, R. Krithivasan, V. Roshchanka, D. Burns, M. J. Mlotha, D. R. Murray,
N. Peng, Protected area downgrading, downsizing, and degazettement (PADDD) in Africa,
Asia, and Latin America and the Caribbean, 1900–2010. Biol. Conserv. 169, 355–361 (2014).
Downloaded from https://www.science.org at Universitaet Leipzig on April 04, 2024

Junker et al., Sci. Adv. 10, eadl0335 (2024) 3 April 2024
SCIENCE ADVANCES | RESEARCH ARTICLE
12 of 12
60. J. Junker, H. Kühl, L. Orth, R. Smith, S. Petrovan, W. Sutherland, Primate Conservation:
Global Evidence for the Eects of Interventions (University of Cambridge, 2017).
61. J. R. Owen, D. Kemp, A. M. Lechner, J. Harris, R. Zhang, É. Lèbre, Energy transition
minerals and their intersection with land- connected peoples. Nat Sustain 6, 203–211
(2023).
62. C. Hof, A. Voskamp, M. F. Biber, K. Böhning- Gaese, E. K. Engelhardt, A. Niamir, S. G. Willis,
T. Hickler, Bioenergy cropland expansion may oset positive eects of climate change
mitigation for global vertebrate diversity. Proc. Natl. Acad. Sci. U.S.A. 115, 13294–13299
(2018).
63. EJAtlas - Global Atlas of Environmental Justice, The Global Atlas of Environmental Justice;
https://ejatlas.org/.
64. C. M. Kennedy, B. Fariss, J. R. Oakleaf, S. T. Garnett, Á. Fernández- Llamazares, J. E. Fa,
S. Baruch- Mordo, J. Kiesecker, Indigenous Peoples’ lands are threatened by industrial
development; conversion risk assessment reveals need to support Indigenous
stewardship. One Earth 6, 1032–1049 (2023).
65. M. R. González- Barros, J. A. Espí, The returns on mining exploration investments.
Boletin. Geologico y Minero. 130, 161–180 (2019).
66. T. T. Werner, G. M. Mudd, A. M. Schipper, M. A. J. Huijbregts, L. Taneja, S. A. Northey,
Global- scale remote sensing of mine areas and analysis of factors explaining their extent.
Glob. Environ. Chang. 60, 102007 (2020).
67. L. Tang, X. Liu, X. Wang, S. Liu, H. Deng, Statistical analysis of tailings ponds in China.
J. Geochem. Explor. 216, 106579 (2020).
68. T. Liang, T. T. Werner, X. Heping, Y. Jingsong, S. Zeming, A global- scale spatial assessment
and geodatabase of mine areas. Glob Planet Change 204, 103578 (2021).
69. V. Maus, S. Giljum, D. M. da Silva, J. Gutschlhofer, R. P. da Rosa, S. Luckeneder, S. L. B. Gass,
M. Lieber, I. McCallum, An update on global mining land use. Sci Data 9, 433 (2022).
70. R. Hijmans, raster: Geographic data analysis and modeling. R package version 2.3- 12., R: A
Language and Environment for Statistical Computing (2023); http://cran.r- project.org/
package=raster.
71. R. J. Hijmans, R. Bivand, E. Pebesma, M. D. Sumner, Terra: Spatial Data Analysis, 26.03.2023.
Spatial Data Science with R and “terra.”.
72. E. Pebesma, R. Bivand, Spatial Data Science: With Applications in R (Chapman and Hall/
CRC, 2023).
73. E. Pebesma, Simple features for R: Standardized support for spatial vector data. R J. 10,
439–446 (2018).
74. R. Bivand, T. Keitt, B. Rowlingson, rgdal: Bindings for the “Geospatial” Data Abstraction
Library (2023); https://r-forge.r-project.org/projects/rgdal/.
75. H. Wickham, D. Vaughan, M. Girlich, tidyr: Tidy Messy Data (2023); https://github.com/
tidyverse/tidyr.
76. H. Wickman, R. François, L. Henry, K. Muller, dlpyr: A Grammar of Data Manipulation
(2021); https://github.com/tidyverse/dplyr.
77. H. Wickham, Reshaping data with the reshape package. J. Stat. Softw. 21, 1–20 (2007).
Acknowledgments: We would like to thank E. Wendt for helping with formatting the
manuscript and G. Rada/iDiv for helping with graphic design. We thank E. Abwe for valuable
comments on an earlier version of the manuscript. Funding: This work was nancially
supported by the European Union’s Horizon 2020 research and innovation program under
grant agreement no. 101003553. J.J., L.Q., J.V., and H.M.P. acknowledge the suppor t of iDiv
funded by the German Research Foundation (DFG- FZT 118, 202548818). Author contributions:
Conceptualization: J.J. Data curation: J.J. Methodology: J.J., L.Q., and J.V. Investigation: J.J., L.Q.,
J.V., and T.S. Visualization: J.J., L.Q., and J.V. Writing—original draft: J.J. Writing—review and
editing: J.J., L.Q., J.V., M.A., A.B., G.C., S.H., T.H., C.Y.K., H.S.K., I.O.- N., H.M.P., H.R., J.R., L.S., and T.S.
Competing interests: J.J. and M.A. have acted as paid consultants for various mining
companies over the past 3 years, and L.S. works part- time as a consultant for The Biodiversity
Consultancy. All coauthors have seen and agree with the contents of the manuscript, and there
is no nancial interest to report. Data and materials availability: All data needed to evaluate
the conclusions in the paper are present in the paper and/or the Supplementary Materials.
Submitted 28 September 2023
Accepted 29 February 2024
Published 3 April 2024
10.1126/sciadv.adl0335
Downloaded from https://www.science.org at Universitaet Leipzig on April 04, 2024