The atlas in real tonnes

Absolute production: World Mining Data 6.4 (2026 ed.), production by country in metric tonnes, 2024. For each material we take the world total and the top producer’s tonnage share, and compare that share against the atlas’s existing USGS-derived top-producer share (in data.json). A third source, the British Geological Survey’s World Mineral Statistics (via its free OGC API, latest year per commodity), gives an independent third count for 18 of the materials. Same top country across sources + a small share gap = independently compiled compilations agreeing.

How independent is “independent”? We counted, because someone attacked this claim. The obvious objection to any second-source check is that the second source is just repackaging the first. World Mining Data tags every figure with where it came from, so the objection is answerable rather than arguable. Across 1,903 tagged figures in the 2024 edition: national statistics 60.0%, company reports 28.5%, questionnaire 6.7%, IEA 1.5%, ICG 1.1%, USGS 0.8% (16 figures), BP 0.7%, Kimberley 0.6%, WNA 0.2%. So WMD is not a repackaging of USGS — the circularity charge fails as usually put.

But the weaker version of the objection is right, and it bounds what this page can claim. Both compilations ultimately rest on the same upstream: national statistical returns and company reports. They are independent compilations, not independent measurements. If a country misreports its output, both inherit the error identically and agree perfectly — agreement would then be evidence of nothing. So what 26/28 demonstrates is compilation reliability: two teams, working separately from the same primary returns, made the same call. That is worth something and it is not nothing, but it is not measurement validation, and this page no longer says it is. No open source independently measures mine output. The closest thing the atlas owns is the satellite footprint layer — and it sees area, not tonnes. But area is measured from orbit, so it owes nothing to any national statistic — which makes it the one genuinely independent check available, if only a coarse one.

So we ran it. Does a country’s share of the world’s mapped mine footprint track its share of bulk open-pit output? This can only corroborate the aggregate mining geography — the polygons are undifferentiated coal/metal/aggregate, and area-per-tonne varies by orders of magnitude, so it cannot validate any per-metal share. But for the coarse claim “these are the world’s big mining nations,” it is the one check that does not rest on the same national returns as USGS and WMD — and it agrees.

How far the other agency sits from these figures. This page already shows the three compilations side by side; this is how far apart two of them sit, measured. For the commodities where the atlas has measured it, the USGS world total for the same commodity and stage differs from the BGS one by a median of: graphite 30.7%, antimony 13.1%, cobalt 12.7%, tungsten 6.9%, rare earths 5.7%, copper 0.6% (median absolute difference over 2002–2024, unsigned). That is a disagreement between two compilations, not an error bar and not a correction of either — it can be measurement, definition, coverage or vintage. The same measurement also shows how far the USGS moves its own current-year estimate, which is the figure to check before treating a how firm is a current-year production figure?

Caveats: the sources define commodities slightly differently (e.g. contained-metal vs concentrate, ore vs oxide), report different years’ vintages, and treat re-processing differently — so a few points of share difference is expected, and the genuine disagreements are flagged, not hidden. Notably, on bauxite and baryte the two independents (World Mining Data and BGS) agree with each other and put a different country on top than the atlas’s USGS figure (Guinea for bauxite, China for baryte) — a case where the third source adjudicates rather than rubber-stamps; titanium splits three ways on the ilmenite-vs-slag definition. Coverage: 28 of 32 materials have a WMD sheet; BGS holds a panel for 62 commodities; of the atlas’s 32, 25 have a recent (≥2020) BGS mine series in the cube. Silicon and helium have no BGS panel at all. Gallium and germanium do have BGS series — Production of primary gallium and Germanium metal — but they are refinery output, not mining, so they are absent from the mine-stage comparison above rather than from BGS; niobium shares a panel with tantalum. (Two earlier versions of this note were wrong in opposite directions: one said gallium, germanium and the PGMs had no BGS series, the other said niobium and tantalum had none.) Sources: world-mining-data.info · BGS World Mineral Statistics → production.json.