Critical Materials Atlas
Method · the mechanism

How the atlas works

Not a summary of what each page shows — the actual machinery. How two countries reporting the same shipment differently become one number. How a physical limit catches a false origin. What every analysis takes in and puts out. And the five checks a figure must survive before it is published.

Everything here is built from public data by script, and rebuilt from scratch each time. There is no hand-entered spreadsheet anywhere in the chain: if a number appears on this site, a named source and a named calculation produced it, and both are written down below.

The engine, step by step

Five operations turn raw public files into the figures on the site. Each is stated with its actual rule, because “we reconcile the data” is not a method.

1

Decide who to believe

Every shipment is reported twice — once by the exporter, once by the importer — and the two rarely agree. Import values also include freight and insurance that export values do not, so they are not even measuring the same thing. The engine first removes that wedge by estimating the CIF/FOB markup, then combines the two sides. Version 1 uses an equal-weighted geometric mean of the adjusted pair.

Reporters are also scored, so that combination can improve. Each country’s systematic bias is estimated by robust two-way median centering (removing partner and product effects), and its reliability = 1 / residual variance, shrunk toward the average so a country with few flows cannot look artificially precise.

Input
Both sides of each bilateral flow (CEPII BACI, UN Comtrade mirrors)
Method
CIF/FOB adjustment → geometric mean of the pair; reporter bias and reliability estimated separately
Output
flows_reconciled — one agreed value per flow, plus a quality score per reporter
2

Count each shipment once

Seven sources overlap. The same tonne can appear in a national release, a mirror record and a compilation. Records are matched on flow identity and de-duplicated to a single best row per flow, with the losing rows kept rather than deleted so the choice can be inspected.

Input
All source rows for the same flow
Method
Match on (period, exporter, importer, code); keep the best record, retain the rest
Output
flows_best
3

Apply the physical limit

Trade data says where a shipment came from, which is not the same as where the material was produced — a hub can export what it never mined. So the atlas imposes a physical constraint: a country cannot ship out more than it produced and did not itself consume.

exportable(c) = max(0, production(c) − consumption(c)) exportable_share(c) = exportable(c) / Σ exportable · the physical prior guardrailed(c) = min(observed_share(c), exportable_share(c)) · HARD limit corrected(c) ≈ w·guardrailed(c) + (1−w)·exportable_share(c) · w = 0.5

The weight w is how much we trust the trade record against the physical prior. It is set at 0.5 and documented as a knob, not a discovery — which means a corrected share is the midpoint of a range, never a point finding, and the pages say so.

Input
Reconciled flows + production + the consumption estimate from step 4
Method
Hard cap at what is physically exportable, then a documented blend
Output
Corrected origin shares with the size of each correction shown
4

Estimate what a country actually uses

Consumption is mostly unpublished, so it is estimated from what a country visibly does — how much steel it makes, how many cars, how much construction — times how much of each material that activity needs. The intensities are not guessed. Each is back-solved from a known world total, so the country cells can be uncertain but the world sum cannot drift:

intensity(m, driver) = KNOWN_WORLD(m) × end_use_share(m, driver) / WORLD_TOTAL(driver) demand(c, m) = Σ over drivers activity(c, driver) × intensity(m, driver)

This is deliberately trade-independent: it never looks at imports, which is what lets it be used as an independent check on the trade data in step 3 rather than a restatement of it.

Input
National activity series (steel, vehicles, construction, semiconductors…) + published end-use splits
Method
Activity × self-calibrating intensity, anchored to a known world total
Output
Demand by country and material — an order-of-magnitude estimate, labelled as one
5

Turn it into indices

Shares become comparable measures: concentration (HHI of national shares), a weighted supply-risk score, resilience under supplier removal. Where an input is an interval rather than a measurement, the index carries a score band and a rank band instead of pretending to a point.

Input
Production, refining and trade shares; recycling rates; origin gaps
Method
Explicit weights, published and adjustable; uncertainty propagated, not hidden

What each analysis takes in, and puts out

Six analyses, each attacking supply risk from a different side. Same question for each: what goes in, what is done to it, what comes out.

1 · Value chains

Where does this material actually come from, stage by stage?
In
Production by stage, reconciled trade, published chain literature per material
Method
Trace mine → concentrate → refine → component; attach the leading producers to each stage
Out
57 chain pages + 59 material and country profiles, each with a research library

2 · Chokepoint map

Is this a rock problem or a factory problem?
In
Every chain stage, plus feedstock signatures showing whether a leader mines or imports its input
Method
Classify each chokepoint by mechanism: geology, processing capability, policy, or furnace physics
Out
58 chokepoints — only 4 geological, so 54 are in principle rebuildable

3 · Risk & scenarios

Could the world cope without the dominant supplier?
In
Mine share, refining share, export concentration, origin gap, recycling rate
Method
Weighted index with a recyclability discount; N−1 removal; optimal-transport reallocation
Out

4 · Trade engine + anchor

Which trade numbers do we actually believe?
In
Both sides of every flow, from seven overlapping sources
Method
Steps 1–3 above: reconcile, de-duplicate, then constrain origin by what is physically exportable
Out
The anchor — corrected origin shares, each with the size of its correction

5 · Concentration

Did supply actually get more concentrated?
In
Production tonnages only — no trade, no model, no forecast
Method
HHI of national shares per decade, then leave-one-out and a frozen pre-window definition to test whether the result is an artefact
Out
The finding: the story runs two ways, and the cited average is two materials

6 · Cross-source comparison

Does a second compilation report the same numbers we do?
In
The cube: BGS national sums against USGS world estimates, paired form by form
Method
Explicitly declared pairings — never matched on material name — then the ratio of the two sums
Out
34 of 53 within 10%, and a recorded reason for every figure that differs

The five checks a number passes before it appears

This is the part worth knowing before you use anything here. Most publishing on critical minerals gives you a number. The atlas gives you the number and what it is worth — and takes numbers down in public when they fail. Every figure runs this gauntlet.

1 · Mechanical

Nothing is published if a link is broken, a dataset a page depends on is missing, or a retracted figure has reappeared. This runs before every single deployment and has never been bypassed.

2 · Provenance

Every estimate carries a clickable basis note naming the source, the quoted row and the computation. Reported, computed and estimated are never silently mixed — and a figure with no recorded source says so on its face rather than borrowing the authority of the ones that have it.

3 · Adversarial

Findings are attacked before publication, not after. The concentration finding went through five review passes and two complete framings were discarded, because a claim that only survives friendly reading is not a finding.

4 · Drift

A corrected source does not correct its copies. Every derived figure is compared against the source it was copied from, so a page cannot keep publishing a number the rest of the site has already retracted.

5 · Census

A sum over a handful of reporting countries is not a world total, however plausible it looks. 13 of 53 materials fail that test — including lithium and rare earths — and the pages that use them say so rather than quietly rounding the limitation away.

What this cannot do. The atlas measures public data, so it inherits what public data cannot see: artisanal and informal output that never enters a national return, company-confidential cells, and stockpiles nobody publishes. Consumption is an estimate and is labelled as one. Corrected origin shares are midpoints of a documented range, not measurements. Where two sources disagree we show both and explain the difference rather than picking a winner. The full limits page states the rest, and the challenge page invites you to break any of it.