copernicus-ems¶

copernicus-ems is a small, typed, async Python library for Copernicus EMS Rapid Mapping, the service that maps floods, fires, landslides and earthquakes from satellite images within hours of a disaster. It reads the activations, their products, and the observed event layers (flooded areas, burnt areas, landslides…) as GeoJSON, together with the observation masks that make them usable as ground truth.

async with RapidMapping() as ems:
    events = await ems.observed_events("EMSR664")  # Emilia-Romagna, May 2023
# 6,334 flood polygons in 11 products, saved in ~/.cache/copernicus-ems/EMSR664/

No account, no API key: the dashboard API is public. One dependency: httpx.

Get started See the roadmap

What it does¶

	Feature	In plain words
	Activations	Every public activation, filtered by country and category.
	Products and observation masks	Each product with its label, the time the satellite saw the event, and the AOI that was looked at.
	Observed events as GeoJSON	The flooded polygons of each product, filtered to real floods (no landslides, no AOI).
	Disk cache	Layers are saved once; a second run reads them from disk. One environment variable moves the folder to a Docker volume.
	Self-contained	`RapidMapping()` opens its own HTTP client, retries 502s and network errors, closes it at the end.
	Pure functions too	Every parsing step is a plain function, if you have your own HTTP stack.

Why not the product package¶

The obvious way	What this does instead
download `{code}_products.zip` (~100 MB per activation, zips of zips) and read shapefiles	read each product's vector layer as GeoJSON, the same features
some product packages answer 403 (EMSR664)	their GeoJSON layers are open: EMSR664 alone is 6,334 flood polygons
take the area of interest as the event	take `observedEventA`: on EMSR762 AOI01 the AOI is ~1,850 km², the flood 1.19 km²
filter activations on category `Flood`	many Italian floods are filed as `Storm`: filter features on `event_type`
keep every feature of a flood activation	the same rain's landslides are in it too (`6-Mass Movement`)

Per activation the product package is about 100 MB; the GeoJSON flood layer of EMSR762 AOI01 is 62 KB

Who is it for¶

Hydrologists and flood modellers who need observed extents to validate a forecast or a hazard map.
GIS and data engineers who want every flood of a country, repeatably, without clicking through the viewer.
ML practitioners who need labelled flood / not-flood areas, with the mask that says where "not flooded" was actually observed.

It started inside Limen, where the observed flood extents are the ground truth of a flood backtest.

Data licence

Rapid Mapping data is © European Union, Copernicus Emergency Management Service, free to use with attribution. Anything you publish from these layers must credit the source, for example: Contains modified Copernicus Emergency Management Service information, © European Union.