Precious Metals & Minerals Data Provider · Head-to-head
USGS Mineral Commodity Summaries vs BGS MineralsUK - UK and World Mineral Statistics
Which precious metals & minerals data provider data fits your job: USGS Mineral Commodity Summaries, or BGS MineralsUK. API, files, or your warehouse. Daily, weekly, or hourly.
USGS Mineral Commodity Summaries (annual mineral statistics incl. gold, silver, platinum-group metals)
BGS MineralsUK (UK and world mineral statistics)
Where the fields line up
1 shared field — join on these.
| Field | USGS Mineral Commodity Summaries | BGS MineralsUK |
|---|---|---|
Year | documented | Statistics reference year the figure belongs to, returned as an ISO datetime. |
Coverage, side by side
| USGS Mineral Commodity Summaries | BGS MineralsUK | |
|---|---|---|
| Granularity | One value per commodity x statistic x country x year; world totals rounded | One record per country x commodity x statistic type x year; nil and unavailable distinguished |
What each contains
Pick by fit, not by loyalty.
| USGS Mineral Commodity Summaries | BGS MineralsUK | |
|---|---|---|
| Documented fields | 11 | 27 |
| Field types | Integer year, string commodity and country, statistic family with optional detail breakdown, numeric value with unit, prose notes column, boolean critical-mineral flag | ISO date-time year, coded and translated commodity, translated country name plus ISO 3166 alpha-2/alpha-3 codes, statistic-type enum, numeric quantity with units, status and precision descriptors, footnote code/text pairs |
| Signature fields | `Commodity`; `Country`; `Statistics` + `Statistics_detail` (Mine production, Refinery: Secondary, Reserves, Price); `Unit`; `Year`; `Value`; `Notes`; `Is critical mineral 2025` | `year`; `country_trans` + `country_iso3_code`; `bgs_commodity_trans` ('silver, mine'); `bgs_statistic_type_trans`; `quantity` + `units`; `quantity_in_book_style`; `data_precision_description`; `sdmx_translation`; `erml_group` |
| Shared concepts | Period stamp, measured value with unit, commodity identity, statistic type, country attribution | Period stamp, measured value with unit, commodity identity, statistic type, country attribution |
| Entity granularity | One value per commodity x statistic x country x year; world totals rounded | One record per country x commodity x statistic type x year; nil and unavailable distinguished |
| Named operators or mines | None - country is the finest entity, with producing states called out in domestic narrative | None - country is the finest entity |
| Overlap verdict | Structural overlap only: place x material x measure x year on both sides. Current-year estimates, reserves and prices versus a half-century-plus annual panel. | — |
What each does better
USGS Mineral Commodity Summaries
Stock variables, not just flows. The world mine production and reserves tables pair each country's annual output with what remains in the ground - Australia at 13,000 tons of gold reserves against roughly 3,000 tons of world mine production - a forward-looking reserve base BGS's production-and-trade frame never carries.
Trade dependence made explicit. Import-source tables break reliance down by partner with percentages (Canada supplied 99 percent of US gold imports across 2021-24), net import reliance sits beside reported consumption, and the Is critical mineral 2025 flag says which commodities Washington currently screens as strategic. Add the Notes column's candor about proprietary rounding and you get a snapshot engineered for policy and market work, not just historical description.
BGS MineralsUK
A century of depth. Compiled since 1913, tabulated from 1970 through 2022, millions of rows across tens of thousands of distinct country-commodity-year combinations. USGS archives editions back to the 1990s, each holding only a five-year window; BGS hands you the whole panel in one shape.
Geography you can join on. Every record ships ISO 3166-1 alpha-2 and alpha-3 codes beside the country name, so 'Burkina Faso', 'BF' and 'BFA' never collide. USGS writes country names out in full and leaves normalization to you.
Honest zeros. data_precision_description separates 'Nil (nothing produced)' from genuinely unavailable, quantity_in_book_style keeps the yearbooks' own symbols, and SDMX status codes translate to plain language - see the sample row where Burkina Faso's 2006 silver output reads 0, renders as '-----' and is explained as nil. On the USGS side, absence tends to arrive as a blank cell or a footnote.
Flows as well as production. Imports and exports ride in the same schema as production - for all countries up to 2002, then selected European countries through 2018 - and the UK Minerals Yearbook layer adds a domestic breakdown into energy minerals, crushed rock, construction minerals and industrial minerals that no world table provides.
Where they're equivalent
More than the field counts suggest. Both reduce the mineral economy to the same four-part key: a place, a material, a measure, a year. Both treat gold, silver and the platinum-group metals as ordinary members of a much wider commodity list rather than as specialist feeds. Both resolve to annual grain and nothing finer - neither carries a monthly or daily cut. Both preserve provenance (USGS in Notes, BGS in its footnote-code columns), both expose their units explicitly rather than burying them, and both publish verified field dictionaries.
Neither names mines, companies or operators: the finest geographic entity on either side is the country - with the single exception of USGS calling out producing states inside its domestic narrative. If your question needs asset-level resolution, both records stop at the same border.
The verdict
Verdict: sample both, pick by fit - near-tied on quality, split by tense.
Accept a five-year rolling window, rounded world totals and no named operators. That shape suits the investors quants use cases.
Take BGS MineralsUK if your question lives in the past tense or needs joins: multi-decade annual panels per country and commodity, ISO-standardized geography, nil-versus-unavailable precision, trade flows in the same schema as production, plus the UK domestic layer. Accept that the tabulated panel closes at 2022. That shape fits the data scientists use cases.
Rule of thumb: if the question begins "how much gold did the US produce last year and what is left in the ground", take USGS; if it begins "build me a fifty-year panel of silver mine production across every reporting country", take BGS. Both live in the precious metals & minerals data hub.
Sample both, pick by fit. See USGS Mineral Commodity Summaries · See BGS MineralsUK
Or take both in one feed
Yes - they interlock almost perfectly because each one's weakness is the other's strength.
Two cautions straight from the records. First, there is no join key - normalize country spelling yourself, which is easier working from BGS's ISO codes toward USGS's written-out names than the reverse. Second, keep units apart before merging: metric tons of metal on the USGS side, kilograms of metal content on the BGS side, with BGS's nil convention marking true zeros rather than gaps.
Datadory ships either record alone or merged onto one delivery calendar, normalized so units stop colliding, delivered daily, weekly, or hourly - your call. Or take both in one feed.
API, files, or your warehouse. Daily, weekly, or hourly.
Fair questions
Is USGS Mineral Commodity Summaries better than BGS MineralsUK?
Better at different jobs, scored 10/10 versus 9/10. BGS wins the past tense: an annual country-by-commodity panel compiled since 1913, tabulated from 1970 through 2022, with ISO-coded geography ready to join.
Do the two datasets cover the same ground anywhere?
Structurally, yes. Both collapse to country x commodity x statistic x year, both carry gold, silver and the platinum-group metals inside much wider commodity lists, and both resolve to annual grain only.
Which dataset is more current?
USGS, by construction. Its 2026 edition carries estimated 2025 figures across the whole commodity list - gold mine production put at roughly 160 tons against a $3,300 per-troy-ounce mark - inside a five-year rolling window. BGS's tabulated panel closes at 2022, with the printed archive holding the longer lineage. When the question is this year's number, USGS is the record that has one.
Which dataset reaches further back in time?
BGS, decisively. World Mineral Statistics has been compiled since 1913, a century of lineage with the tabulated panel readable from 1970 through 2022. USGS archives prior editions back to the 1990s, and each edition holds only a five-year salient-statistics window, so a long USGS series means stitching editions together yourself.
Can Datadory deliver both datasets together?
Yes. Either record arrives alone or merged onto one delivery calendar, with country keys mapped, units reconciled between metric tons and kilograms of metal content, and BGS's nil conventions carried through intact. Name the commodities, countries and year range when you request the sample and it lands pre-cut, with field definitions attached.