Wildfires-CEMS: Burned Area Delineation and Severity Dataset from Copernicus EMS and Sentinel-2

Provider
LINKS Foundation (Arnaudo, Barco, Merlo, Rossi)
License
CC-BY-4.0 (source)
Size
21 files
Format
PNG (in sampled files)
Stored at
Hugging Face (https://huggingface.co/datasets/links-ads/wildfires-cems)
Access
Free and public. No account with the data provider is needed.
Category
Earth Observation & Geospatial
Source
https://github.com/links-ads/burned-area-seg

About this dataset

Burned area segmentation dataset focused on Europe, built from Copernicus Emergency Management Service (EMS) rapid mapping activations and Sentinel-2 imagery, introduced with the paper Robust Burned Area Delineation through Multitask Learning. Each tile pairs a 12-channel Sentinel-2 L2A image (reflectance x 10,000) with a binary burned area delineation map, a damage grading map (0 to 4) where EMS provides one, ESA WorldCover 2020 land cover, a CloudSEN12-derived cloud mask, and two extra land cover variants (ESRI annual 9-class and a static 10-class map). All layers are GeoTIFF. The listing (measured 2026-11-14) shows data/train, data/val and data/test folders of split multi-part tarballs (tar, gzip, split) that are concatenated and extracted before use; the repository uses about 18.4 GB. Read tiles with rasterio; training code is in the links-ads/burned-area-seg GitHub repository. Supports burned area mapping, fire severity grading and multitask land cover segmentation. Released under CC-BY-4.0, with Copernicus EMS and Sentinel attribution.

How to load this dataset in Python

from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="links-ads/wildfires-cems", filename="assets/sample.png", repo_type="dataset")
print(path)

Data structure

File types in sampled files: PNG

Files

Example file: assets/sample.png

Analyze this dataset without downloading it

On Scigantic, Wildfires-CEMS: Burned Area Delineation and Severity Dataset from Copernicus EMS and Sentinel-2 opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Earth Observation & Geospatial datasets