A cloud-native geospatial data catalog of open Dutch government geodata, organized by
institution. Built on STAC metadata, GeoParquet for analytics, and PMTiles for visualization.
Published on Source Cooperative (https://source.coop/cholmes/portolan-nl).
Built with Portolan (https://portolan-sdi.org), a framework for cloud-native geodata
infrastructure. This catalog demonstrates how government geodata services like PDOK could
be extended with cloud-native formats — GeoParquet for scalable analytics, PMTiles for
instant map visualization from object storage, and STAC+AGENTS.md for AI/agent access.
How to Navigate
The catalog follows a nested STAC structure:
catalog.json → links to subcatalog per institution
Each subcatalog (kadaster/, rijkswaterstaat/, etc.) → links to collections
Each collection has collection.json (metadata), .parquet (data), .pmtiles (visualization)
Many collections have their own AGENTS.md with detailed field descriptions and queries
Institutions and Collections
Kadaster (Cadastre, Land Registry & Mapping Agency)
bag_light/ — 11.4 million buildings with construction year, usage, floor area
bestuurlijke_gebieden/ — 342 municipalities, 12 provinces, 1 national territory (2026)
inspire_buildings/ — 24.2 million INSPIRE-harmonized building records
bestand_bodemgebruik_2017/ — 171,543 land use polygons (~40 types) for all of Netherlands
existing_landuse_inspire/ — 171,543 INSPIRE-harmonized land use polygons (HILUCS classification)
bevolkingsspreiding/ — 31,033 population distribution features (1km grid, provinces, municipalities)
gezondheid/ — 24 provincial health indicators (doctors and deaths per 100K)
Getting Started
Use AI to explore the data. Point Claude Code
or Gemini CLI at any collection's AGENTS.md and
ask it to query the data, build interactive maps, or generate charts. Every collection includes
an AGENTS.md with field descriptions, query examples, and usage context.
Browse and download. Navigate to any collection directory to find GeoParquet files for
analytics and PMTiles for instant map visualization. All files are on Source Cooperative and can
be accessed directly via HTTP — no account or API key required.
Query with DuckDB. All GeoParquet files work with remote HTTP access:
Coordinate System
Most datasets use EPSG:28992 (Amersfoort / RD New) — the Dutch national coordinate system.
Coordinates are in METERS, not degrees. To convert to WGS84 (lon/lat):
CBS INSPIRE datasets (bevolkingsspreiding, existing_landuse_inspire, gezondheid) use
EPSG:3035 (ETRS89-LAEA), the pan-European standard — also in meters but on a European grid.
The one exception is inspire_buildings, which uses EPSG:4258 (ETRS89) — practically
identical to WGS84, so coordinates are in degrees and no transform is needed for most uses.
Highlights for Exploration
Building age analysis: BAG Light has construction years for 11.4M buildings (1600s–2030)
Flood protection: Combine waterkeringen (dikes) with bergingsgebieden (retention areas)
Nature vs development: Overlay Natura 2000 / National Parks with zero emission zones
Heritage mapping: 63K monuments categorized by type (residences, churches, castles, farms)
Administrative joins: Use bestuurlijke_gebieden as the reference for any municipality/province analysis
Fish connectivity tracking: rijksviswegen shows status changes from 2009–2033
Neighborhood demographics: Wijken en Buurten has population, income, and urbanity for every neighborhood