A collection of datasets from various Dutch institutions to demonstrate a Spatial Data Infrastructure built on Portolan.
High-resolution aerial orthophotos from the 2024 Beeldmateriaal Nederland summer campaign, covering 356 tiles across 12 kaartblad sheets in the central Netherlands — from the North Sea coast (Haarlem, IJmuiden) to the Veluwe (Apeldoorn, Harderwijk) and from Alkmaar in the north to Gorinchem and Tiel in the south. Served as Cloud-Optimized GeoTIFFs (COGs).
Each tile covers one "kaartblad" (Dutch national map sheet) of approximately 5 km × 6.25 km at 25 cm ground sampling distance, yielding 20,002 × 25,002 pixels per tile.
Two asset variants:
All COGs include JPEG compression (quality 75), 512×512 internal tiling, 6 overview levels (AVERAGE resampling), and baked-in per-band statistics (min, max, mean, stddev).
Source: Beeldmateriaal Nederland via GeoTiles (TU Delft) Provider: Beeldmateriaal Nederland — a partnership of Dutch municipalities, provinces, water boards, Rijkswaterstaat, and Het Kadaster for joint aerial photography since 2011. License: CC BY 4.0 (attribution to beeldmateriaal.nl requested)
The data lives on Source Cooperative as Cloud-Optimized GeoTIFF in EPSG:28992 (Amersfoort / RD New). COGs support HTTP range requests — GDAL, rasterio, and other tools can read arbitrary windows without downloading entire files.
Base URL: https://data.source.coop/cholmes/portolan-nl/beeldmateriaal/luchtfoto_2024/
Tiles are organized into sub-catalogs by kaartblad sheet number:
{base}/kb{sheet}/luchtfoto-2024-{tileid}-{variant}.tif
sheet: Kaartblad sheet number (19, 20, 24, 25, 26, 27, 30, 31, 32, 33, 38, 39)tileid: Lowercase kaartblad tile code (e.g., 32dn1, 25az2)variant: rgb (3-band, ~80-97 MB) or rgbnir (4-band, ~293-365 MB, KB32 only)356 tiles across 12 kaartblad sheets in the central Netherlands:
KB 32 (Amersfoort) contains the origin of the RD New coordinate system (X=155000, Y=463000 — the Onze Lieve Vrouwetoren church tower).
The items.parquet file contains all 356 STAC items as a GeoParquet table, queryable
with DuckDB for spatial search and filtering:
All bands are uint8 (0–255). Per-band min/max/mean/stddev statistics are embedded in each COG.
COGs can be rendered as web map tiles using TiTiler, which reads HTTP range requests from the COG and serves XYZ tiles on the fly.
https://titiler.xyz/cog/tiles/WebMercatorQuad/{z}/{x}/{y}?url={encoded_cog_url}
RGBNIR files are larger because 4-band JPEG cannot use YCbCr colour space optimisation.
to-import/beeldmateriaal-orthophotos.md for how to add
more tiles from the full nationwide archive.valid = (red + green + blue) > 0.Beeldmateriaal Nederland (literally "Imagery Netherlands") is the Dutch national collaboration for aerial photography, established in 2011. It brings together municipalities, provinces, water boards (waterschappen), Rijkswaterstaat (public works), and Het Kadaster (land registry) to jointly commission and share aerial surveys.
Two campaigns fly each year:
All imagery is published as open data under CC BY 4.0.
Website: https://www.beeldmateriaal.nl/ Open data: https://opendata.beeldmateriaal.nl/ PDOK services: https://service.pdok.nl/hwh/luchtfotorgb/wmts/v1_0
Hand-maintained agent guide (ported from this collection's llms.txt in August 2026). Update it alongside collection.json.
| 32 |
| Amersfoort |
| 32 |
| Amersfoort, Zeist, Leusden, Woudenberg, Veenendaal |
| 33 | Apeldoorn / Veluwe East | 31 | Apeldoorn, Ede, Wageningen, Barneveld |
| 38 | Gorinchem / Rivierenland | 32 | Gorinchem, Vianen, Leerdam, Culemborg |
| 39 | Tiel / Betuwe | 32 | Tiel, Rhenen, Wageningen (south), Betuwe |