Venezuela Earthquakes, 2026-06-24

# Planet Crisis Response — Venezuela Earthquake, June 2026 (llms.txt)

STAC/Portolan catalog of Planet imagery documenting the 24 June 2026 earthquake on
Venezuela's coast. Imagery © Planet Labs PBC, licensed CC-BY-NC-4.0 (non-commercial).

Base URL: https://data.source.coop/planet/venezuela-earthquake-2026-06-24
Root catalog: {base}/catalog.json

## Structure
- pre-event/  → collection `quarterly-mosaic`: Planet Q1 2026 quarterly basemap, ~4.8 m
  RGB, 40 tiles (one Item per tile). Pre-earthquake, for before/after comparison.
- post-event/ → one collection per affected location, 50 cm SkySat/Pelican collects
  acquired 26–28 June 2026: caracas, la-guaira, catia-la-mar, puerto-cabello, valencia,
  independencia-ocumare, yumare.

## Items & assets
- One Item per scene (Planet scene id, e.g. 20260627_112621_ssc2_u0001).
- Assets (subset present per scene):
  - `visual`   — true-colour RGB COG (uint8)
  - `analytic` — pansharpened 4-band BGRN COG (uint16)
  - `udm2`     — Usable Data Mask 2 (per-pixel clear/cloud/shadow/haze + confidence)
  - `udm`      — Unusable Data Mask (single band)
- Asset hrefs are absolute public https URLs to COGs (open with /vsicurl, rio-tiler, etc.).
- Item properties include datetime, gsd (0.5), eo:cloud_cover, view:* and sun angles,
  constellation (skysat|pelican), platform, proj:epsg.

## Query (preferred): STAC-GeoParquet
Each sub-catalog folder has items.parquet aggregating its Items:
  {base}/pre-event/items.parquet
  {base}/post-event/items.parquet

DuckDB example (find post-event scenes over Caracas):
  INSTALL spatial; LOAD spatial;
  SELECT id, datetime, assets FROM
    read_parquet('{base}/post-event/items.parquet')
  WHERE collection = 'caracas';

## Notes
- Non-commercial license (CC BY-NC 4.0).
- Pelican assets have eo:bands by identity (blue/green/red/nir) but no center wavelengths.
- Source delivery (Planet zips + per-scene STAC) lives in the same source.coop repo;
  this catalog is a clean overlay built by scripts/catalog/build_catalog.py.