Twelve thousand pins on a map do not settle the mining industry's nature problem, but they do make it much harder to pretend the problem is evenly spread. ICMM's new Global Mining and Metals Biodiversity Dataset overlays 12,000 large-scale formal mining, smelting, refining and processing facilities with ecosystem-condition and conservation indicators. For investors watching the scramble for transition minerals, that is useful. Not comforting, necessarily, but useful.
The central finding is uneven exposure. Twenty-two percent of facilities sit in landscapes classified as high or very high ecosystem condition, places where original biodiversity is largely intact and habitat connections remain. Five percent intersect Key Biodiversity Areas, while 4% intersect Protected or Conserved Areas. Those percentages can look modest at first glance. They are not a permission slip. A relatively small slice of facilities can matter enormously when the locations are ecologically irreplaceable or politically sensitive.
Location Is Becoming a Harder Business Variable
Mining has always been dictated by geology. Ore is where it is, and executives cannot relocate a lithium deposit the way a retailer relocates a warehouse. Still, the old industry habit of treating environmental constraints as a late-stage permitting nuisance is getting worn out. The Kunming-Montreal Global Biodiversity Framework gives the policy backdrop real teeth: 196 countries agreed to bring loss of high-integrity landscapes close to zero by 2030 and to conserve 30% of land, water and sea.
That does not mean every facility near a sensitive landscape becomes unviable. The dataset explicitly does not measure an individual facility's biodiversity impact, operating quality or compliance. It also does not assess company performance. But it identifies where scrutiny, permitting friction, conservation expectations and stakeholder pressure may concentrate. That is plenty relevant to capital allocation.
Commodity Labels Do Not Tell the Whole Story
The uncomfortable wrinkle is that biodiversity exposure does not line up neatly with the market's preferred clean-energy storyline. Facilities producing antimony, diamond and lithium are overrepresented in high-condition ecosystems. Manganese, tin and diamond facilities are overrepresented in Key Biodiversity Areas relative to the global average. Meanwhile, platinum group metals, barium and chromite are underrepresented in high-condition areas; steel, lithium and coal are proportionally underrepresented in KBAs.
Read that carefully. Lithium's profile differs depending on which indicator is being considered. Anyone trying to collapse this into a lazy “green metals good, coal bad” scorecard is going to get the wrong answer. Geography, asset type, local protections and surrounding land pressures all matter. Indonesia is a sharp example: 84% of its mining and metals facilities in the dataset, mostly thermal coal facilities, are in high-ecosystem-condition areas also under pressure from other drivers.
- For governments: the map can help target protection, enforcement and permitting decisions where the trade-offs are most acute.
- For operators: voluntary responsible-mining standards may carry more weight where formal protection is absent.
- For investors: asset-level location deserves more attention than broad commodity narratives alone.
The KBA Numbers Deserve a Longer Look
ICMM identifies 607 facilities intersecting KBAs across 359 sites. Twelve of those sites contain six or more facilities. Chrissie Pans in South Africa is the sort of detail that makes an abstract dataset feel very real: the wetland is described as 100% irreplaceable for the global survival of one species and contains 18 coal facilities, none within a Protected or Conserved Area.
Globally, 73% of KBAs containing facilities are not fully formally protected. That leaves conservation outcomes dependent, at least in part, on company commitments and voluntary standards. I would not wave that away, nor would I blindly trust it. Voluntary action can be meaningful, but it is uneven by nature. Where government recognition and enforcement are weak, reputational and operational risk can pile up quickly after a local dispute, a permit challenge or a change in political mood.
Useful Evidence, Not a Company Scorecard
There is a big methodological warning label here, and ICMM deserves credit for putting it in plain view. Facilities are represented as points, not operational footprints. The figures show the share of facilities, not the land area involved. The data excludes artisanal and small-scale mining, illegal operations and closed sites. It also cannot isolate a facility's contribution to the condition of the surrounding landscape.
In other words, do not use this to declare a particular mine destructive or a particular company responsible. That would be sloppy. The dataset also does not distinguish ICMM members from the wider sector. Its value is as a sector-wide triage tool: a better first map of where questions need asking.
The next improvement is obvious and difficult: better facility-boundary data. Turning points into operational footprints would make the analysis more decision-ready. Until then, the sensible takeaway is neither celebration nor panic. The energy transition needs minerals and metals, full stop. It also cannot afford to bulldoze the ecological systems that climate policy is supposed to help protect. This dataset does not resolve that collision, but it finally shows where the collisions are most likely to happen.