Datasets:

neet1797
/

burn2scar

For each wildfire recorded in EFFIS (European Forest Fire Information System), this dataset provides two Sentinel-2 acquisitions — one taken shortly after the fire and one taken several months later — each paired with a 7-class segmentation mask identifying clear land, fresh burn, old burn, cloud, cloud shadow, water, and nodata.

13,903 fires with at least one usable acquisition (19,443 queried; the remainder had no candidate scene passing the cloud/obscuration threshold in either window), 26,854 scene/mask pairs total
Countries: Italy, France (south of 46°N), Spain, Greece
Date range: 2020–2026
30.57 GB total

Classes

Mask value	Class	Definition
0	clear	No burn, cloud, shadow, or water detected
2	fresh burn	Burned area, ≤ 90 days since ignition
3	old burn	Burned area, > 90 days since ignition (up to 365 days)
4	cloud	Detected via OmniCloudMask
5	cloud shadow	Detected via OmniCloudMask
6	water	NDWI > 0.05, or Sentinel-2 SCL value 6
255	nodata	Outside the valid data mask

Fire population overview

Characteristics of the underlying EFFIS-sourced fire population, after filtering (before acquisition — this is about the input fire records, not the output imagery). Left to right: fire count by source country, distribution of burn area sizes within the 5–2,000 ha bounds, and seasonal distribution of fires by quarter across the 2020–2026 date range — wildfire activity concentrates heavily in the Mediterranean summer months, as expected.

Class distribution

Computed across all 26,854 masks (1.76 billion pixels total):

Class	% of valid pixels
clear	82.62%
old_burn	7.21%
fresh_burn	7.12%
water	2.32%
cloud	0.54%
shadow	0.19%

0.95% of all pixels are nodata (tile-edge artifacts from UTM reprojection), excluded from the percentages above. Fresh and old burn are closely balanced — a result of acquiring two independent time windows per fire rather than relying on incidental overlap with other fires' perimeters. See Limitations below for what clear does and doesn't represent.

Dataset structure

acquisitions/
  <fire_id>/
    <timestamp>/
      <fire_id>_<timestamp>_s2.tif     # 13-band image, uint16
      <fire_id>_<timestamp>_mask.tif   # 1-band label mask, uint8

Each fire_id (e.g. effis_52171) may contain one or two timestamp subfolders — one from the fresh-burn acquisition window (5–60 days post-fire) and, when a usable scene was found, one from the old-burn window (120–300 days post-fire). These are independent acquisitions of the same location, not a single paired before/after image — see Limitations below.

Image format

Bands (13, in order): B01 B02 B03 B04 B05 B06 B07 B08 B8A B09 B10 B11 B12
Resolution: 10 m (coarser native bands resampled up to this common grid)
Tile size: 256 × 256 pixels (2.56 km × 2.56 km), centered on the fire
Encoding: uint16, scale factor 10,000 → divide by 10,000 for TOA reflectance
Processing level: Sentinel-2 L1C (top-of-atmosphere) — no atmospheric correction, no sensor simulation, no synthetic degradation

How to load

This dataset is distributed as georeferenced GeoTIFF pairs rather than a flat image-classification layout, so the standard datasets.load_dataset() image loader does not apply directly. Load with rasterio:

import rasterio
import numpy as np
from huggingface_hub import hf_hub_download

repo_id = "neet1797/burn2scar"

s2_path = hf_hub_download(repo_id, repo_type="dataset",
                          filename="acquisitions/effis_52171/2021-09-19T.../effis_52171_..._s2.tif")
mask_path = hf_hub_download(repo_id, repo_type="dataset",
                            filename="acquisitions/effis_52171/2021-09-19T.../effis_52171_..._mask.tif")

with rasterio.open(s2_path) as src:
    bands = src.read()          # shape: (13, 256, 256), uint16
    reflectance = bands.astype(np.float32) / 10000.0

with rasterio.open(mask_path) as src:
    mask = src.read(1)          # shape: (256, 256), uint8, values in {0,2,3,4,5,6,255}

To browse all files programmatically, use huggingface_hub.HfApi().list_repo_files() or snapshot_download() for a full local mirror.

Fire selection criteria

Filter	Value
Countries	IT, FR (south of 46°N), ES, EL
Date range	2020-01-01 to 2026-08-04
Area	5–2,000 hectares

Area bounds exclude both marginal micro-detections and mega-fires large enough to make a single tile 100% burn scar with no useful class boundary.

Source data

Imagery: Copernicus Sentinel-2, via the Copernicus Data Space Ecosystem
Burn labels: EFFIS (European Forest Fire Information System), Copernicus Emergency Management Service
Cloud / shadow: OmniCloudMask
Water: NDWI + Sentinel-2 Scene Classification Layer (SCL)

Limitations

No pre-fire baseline. Both acquisition windows are strictly post-fire; this dataset does not provide a true "before" image for change-detection-style pairing.
Cloud cover permitted elsewhere in-frame. Scene selection requires the burn scar itself to be minimally obscured (≤ 50%), but up to 70% cloud cover is permitted in the wider scene outside the scar.
clear is a residual class, not a confirmed vegetation label. It covers anything not detected as burn, cloud, shadow, or water. Mediterranean terrain in particular can put bare soil, exposed rock, and volcanic terrain (present in both Italy and Greece) into clear, since there is no separate class for them. Users needing a purer vegetation mask should combine this dataset with a land-cover product (e.g. ESA WorldCover) and filter further based on their own criteria.
Burn labels are only as complete as EFFIS's own detection. EFFIS's documented minimum mapping unit for burnt-area products is 10 hectares — fires smaller than that are not labeled by EFFIS, and so appear as clear in this dataset rather than fresh_burn/old_burn, even where a real (sub-threshold) fire occurred.

Licensing

Sentinel-2 imagery: governed by the Copernicus Sentinel Data Legal Notice — free, full, and open, permitting reproduction, distribution, and adaptation, with attribution required ("Contains modified Copernicus Sentinel data").
EFFIS burn-perimeter labels: derived from Copernicus Emergency Management Service data. Redistribution terms for this specific product have not been independently verified beyond the general Copernicus data policy — consult EFFIS/CEMS directly before relying on this for redistribution outside your organization.

This is not legal advice; verify licensing terms independently for your use case.

Citation

If you use this dataset in your work, please cite:

@misc{thind2026burn2scar,
  author    = {Thind, Parampuneet Kaur},
  title     = {burn2scar: A Sentinel-2 Burn-Scar Segmentation Dataset},
  year      = {2026},
  publisher = {Hugging Face},
  url       = {https://huggingface.co/datasets/neet1797/burn2scar}
}

Please also retain attribution to the underlying source data, as required by its own license terms:

Contains modified Copernicus Sentinel data.
Burn labels derived from EFFIS (European Forest Fire Information System).

Author: Parampuneet Kaur Thind (Param)

Downloads last month: 3,080