boundary open data

National Boundary of the Germany, 2025. ADE=1. Raw layer holds 11 features; only GF=4 (ARS '000000000000', GEN 'Deutschland') is the land outline — the other 10 are coastal/marine water. Single-feature national outline; low join value but declared for completeness.

1 features

Super Data Zones (Northern Ireland) of the United Kingdom, 2021. NI only REPUBLISH 2026-07-20: first publication violated the attribute policy (population/household/area statistics retained as joinable — decision 0019 audit); declared keep-list now identifiers+names only.

849 features

Local councils (kunsilli lokali) of Malta — the LAU2 administrative level, all 68 local councils across Malta, Gozo and Comino. Official INSPIRE Administrative Units from the Planning Authority of Malta (INSPIRE Annex I, EU High Value Dataset). Join on nat_code (INSPIRE national code, e.g. MT01402 Mdina) or name (e.g. VALLETTA, ZEBBUG (Malta)). Names are uppercase.

68 features

Data Zones (Northern Ireland) of the United Kingdom, 2021. NI only; NISRA's own codes with published lookups REPUBLISH 2026-07-20: first publication violated the attribute policy (population/household/area statistics retained as joinable — decision 0019 audit); declared keep-list now identifiers+names only.

3,778 features

Territorial Collectivities of the France, 2025. Single territorial collectivities that merge dept+region powers (Corse, Martinique, Guyane, Mayotte, Métropole de Lyon, etc.). OVERLAPS region/departement by design — declared for completeness, not a distinct tiling. code_siren dropped from keep — null for ~90% of rows (only special/rattachées carry one); code_insee is the fully-populated key.

103 features

Cantons of the France, 2025. Electoral cantons (post-2015 redécoupage). code_insee is the canton code. Dropped: numero_du_canton, composition_du_canton, codes_insee_des_arrondissements (list-valued). REPUBLISH 2026-07-20: first publication shipped 2054 — dropEmptyFeatures discarded 2 cantons as GeometryCollections (RFC7946 winding artifacts, decision 0020); driver fix coerces to MultiPolygon, restoring 2056.

2,054 features

Municipal Arrondissements of the France, 2025. The 45 municipal districts of Paris (20), Lyon (9), Marseille (16). code_insee is the ARM code; code_insee_de_la_commune_de_rattach links to the parent commune.

45 features

Government Regions (Regierungsbezirke) of the Germany, 2025. ADE=3, ARS 3-digit. Raw 21 features, GF=4 gives 19. PARTIAL COVERAGE BY DESIGN: only some Bundesländer (BW, BY, HE, NRW) are subdivided into Regierungsbezirke — this level does not tile Germany. Not a coverage hole in the dataset; it reflects the real administrative structure.

19 features

Provinces and Territories of the Canada, 2021. 10 provinces + 3 territories. Fields: PRUID, DGUID, PRNAME, PRENAME, PRFNAME, PREABBR, PRFABBR, LANDAREA. Verified 206 application/x-zip-compressed, 133,730,024 bytes — huge for 13 features; full-resolution coastline.

13 features

Federal Electoral Districts of the Canada, 2023. PUBLISHER/SCHEMA EXCEPTION — read before processing. Publisher is ELECTIONS CANADA (hosted by NRCan), not StatCan; licence is Open Government Licence – Canada (open.canada.ca record 18bf3ea7-1940-46ec-af52-9ba3f77ed708, license_id ca-ogl-lgo): copy, modify, publish, distribute. Elections Canada specifies NO attribution string, so the OGL fallback applies. CHOSEN OVER StatCan's lfed000b21a_e.zip (which IS live, 206) because that file is the OBSOLETE 2013 Representation Order (338 seats); the 2023 Rep Order (343, proclaimed 2023-09-22) is current and was used for the 2025 general election, so real joins will be against 343. SCHEMA BREAK: fields are FED_NUM (LONG), ED_NAMEE(100), ED_NAMEF(100), REP_ORDER(4) — there is NO FEDUID, NO FEDNAME, and NO DGUID, so this type does NOT fit the CA-DGUID code_scheme and anything assuming StatCan field names will break. FED_NUM is an integer, not a zero-padded string. CRS CONFLICT UNRESOLVED: the ArcGIS REST layer reports EPSG:3978, but the technical guide's parameters (false easting 6200000, northing 3000000, central meridian -91°52', SP 49/77) match EPSG:3347 instead — read the .prj at convert, do not assume. Single geometry variant, no carto/digital split; 9,388,965 bytes (Last-Modified 2024-12-17), already generalised. Verified 206 application/zip.

343 features

Dissemination Areas of the Canada, 2021. 57,932 CARTOGRAPHIC (not 57,936 — 4 DAs entirely in coastal water are excluded; the digital file has 57,936). This is the one level where StatCan explicitly documents the carto/digital delta. Fields: DAUID(8), DGUID, LANDAREA, PRUID — NO name, NO parent IDs. Verified 206, 197,042,003 bytes.

57,932 features

Census Subdivisions of the Canada, 2021. Municipalities and equivalents. Fields: CSDUID(7), DGUID, CSDNAME(100), CSDTYPE(3), LANDAREA, PRUID. No underscore after 'lcsd'. Verified 206, 155,981,521 bytes.

5,161 features

Census Agricultural Regions of the Canada, 2021. Fields: CARUID(4), DGUID, CARENAME(50), CARFNAME(50), LANDAREA, PRUID. NO single CARNAME column — English/French are separate. Verified 206, 136,732,193 bytes.

72 features

Significant Urban Areas of the Australia, 2021. count is a research estimate; the source is the truth

111 features

States and Territories of the Australia, 2026. Ed.4 re-seed (replaces the Ed.3 entry via newer-vintage rule). expected_units estimated from the ABS Ed.4 structure table minus Ed.3-pattern special-purpose codes; the source is the truth.

9 features

Statistical Areas Level 4 of the Australia, 2026. Ed.4 re-seed (replaces the Ed.3 entry via newer-vintage rule). expected_units estimated from the ABS Ed.4 structure table minus Ed.3-pattern special-purpose codes; the source is the truth.

89 features

Statistical Areas Level 3 of the Australia, 2026. Ed.4 re-seed (replaces the Ed.3 entry via newer-vintage rule). expected_units estimated from the ABS Ed.4 structure table minus Ed.3-pattern special-purpose codes; the source is the truth.

348 features

Statistical Areas Level 2 of the Australia, 2026. Ed.4 re-seed (replaces the Ed.3 entry via newer-vintage rule). expected_units estimated from the ABS Ed.4 structure table minus Ed.3-pattern special-purpose codes; the source is the truth.

2,552 features

Statistical Areas Level 1 of the Australia, 2026. Ed.4 re-seed (replaces the Ed.3 entry via newer-vintage rule). expected_units estimated from the ABS Ed.4 structure table minus Ed.3-pattern special-purpose codes; the source is the truth.

63,451 features

Postal Areas of the Australia, 2021. ABS approximation of postcodes — POA ≠ Australia Post postcodes; the caveat must survive into the description

2,641 features

Mesh Blocks of the Australia, 2026. Ed.4 re-seed. Special-purpose no-geometry blocks drop at convert (Ed.3: 31). Needs the #2137 DuckDB streaming path — GeoJSON is ~700MB.

374,803 features

Greater Capital City Statistical Areas of the Australia, 2026. Ed.4 re-seed (replaces the Ed.3 entry via newer-vintage rule). expected_units estimated from the ABS Ed.4 structure table minus Ed.3-pattern special-purpose codes; the source is the truth.

16 features

Australia of the Australia, 2026. national outline, Ed.4 re-seed — ABS ships no generalised product; simplification tolerance carried over from Ed.3 (1.66M-vertex single feature exceeded per-row limits), disclosed in evidence

1 features

Australian Drainage Divisions of the Australia, 2021. count is a research estimate; the source is the truth

14 features

Electoral Wards of the United Kingdom, 2025. December 2025 boundary set; UK-wide

8,405 features

Intermediate Zones (Scotland) of the United Kingdom, 2022. Scotland only — the complement to E&W MSOAs — zip ships EoR and MHW clipping variants; MHW (Mean High Water, shoreline-clipped) matches the generalisation policy

1,318 features

Data Zones (Scotland) of the United Kingdom, 2022. Scotland only — the small-area complement to E&W LSOAs REPUBLISH 2026-07-20: first publication violated the attribute policy (population/household/area statistics retained as joinable — decision 0019 audit); declared keep-list now identifiers+names only.

7,366 features

Westminster Parliamentary Constituencies of the United Kingdom, 2024. July 2024 boundaries — all four nations re-coded for the 2024 review; UK-wide

650 features

Parishes and Non Civil Parished Areas of the United Kingdom, 2024. England and Wales only — Scotland/NI have no civil parishes

11,541 features

Output Areas of the United Kingdom, 2021. England and Wales census 2021 output areas — ceiling removed 2026-07-16; attempted, bounded by machine limits only

188,875 features

Middle layer Super Output Areas of the United Kingdom, 2021. England and Wales only, census 2021

7,261 features

Lower layer Super Output Areas of the United Kingdom, 2021. England and Wales only, census 2021 (near the 50k cap); Scotland/NI equivalents are Data Zones, next pass

35,672 features

Local Authority Districts of the United Kingdom, 2025. December 2025 boundary set (newer than the researched May 2025); UK-wide

361 features

Integrated Care Boards of the United Kingdom, 2026. England only — April 2026 boundaries, post the 2025-26 NHS reorganisation (the re-verify gate from the research is satisfied: this IS the reorganised set)

36 features

Counties and Unitary Authorities of the United Kingdom, 2025. December 2025 boundary set; UK-wide

218 features

ZIP Code Tabulation Areas of the United States, 2020. DECENNIAL — 2020 IS the latest vintage. ZCTA ≠ ZIP: label it 'ZIP Code Tabulation Areas' and surface the caveat

33,791 features

Urban areas of the United States, 2020. DECENNIAL — use the _corrected file

2,644 features

Tribal Census Tracts of the United States, 2025. count is a research estimate; the source is the truth

493 features

Census Tracts of the United States, 2025. ceiling removed 2026-07-16; attempted, bounded by machine limits only

85,184 features

States of the United States, 2025. incl. DC and territories — the 500k layer, NOT 20m: 20m drops four island territories (52 vs 56)

56 features

Census Regions of the United States, 2025.

4 features

PUMAs of the United States, 2020. DECENNIAL

2,487 features

Metropolitan Divisions of the United States, 2025. count is a research estimate; the source is the truth

37 features

Census Divisions of the United States, 2025.

9 features

Combined Statistical Areas of the United States, 2025.

184 features

County subdivisions of the United States, 2025. REVIEWED: seed the full national layer. Near the cap. Real governing towns in ~29 states; 'unorganized territory' placeholders elsewhere

36,360 features

Statistical Sectors (Statistische sectoren / Secteurs statistiques) of the Belgium, 2025. Finest statistical unit, the smallest join level. 20,812 sectors. niscode = 9-char sector code ('11001A00-') = 5-digit municipality NIS + sub-sector suffix, so it rolls up to municipality. Code-only (no names — join on code). Full-res from the 218MB gpkg → GeoParquet + 0.0001°.

20,812 features

Full-time students as a proportion of total population at LSOA level in Bristol.

268 features

Number of persons in NS-SeC higher managerial, administrative and professional occupations, along with total population, by electoral ward.

34 features

Higher managerial, administrative and professional occupations (NS-SEC L1, L2, L3) as a share of all usual residents aged 16+ by LSOA.

268 features

A collection of community facilities in the Grafenwoehr area, including medical, educational, religious, and recreational services. Note: This dataset covers the entire specified boundary and includes off-post facilities.

26 features

Population density (persons per square kilometre) by electoral ward in London.

679 features

Bus stations within the Guwahati boundary for route planning.

12 features

Gas station locations within the Guwahati boundary.

28 features