Kuro Siwo: Global Multi-Temporal Sentinel-1 SAR Flood Mapping Dataset (GeoTIFF)

Provider
Orion Lab, National Observatory of Athens / NTUA (Bountos et al.)
License
CC-BY-4.0 (source)
Size
53 files
Stored at
Hugging Face (https://huggingface.co/datasets/orion-ai-lab/Kuro-Siwo-GeoTIFFs)
Access
Free and public. No account with the data provider is needed.
Category
Earth Observation & Geospatial
Source
https://proceedings.neurips.cc/paper_files/paper/2024/hash/43612b0662cb6a4986edf859fd6ebafe-Abstract-Datasets_and_Benchmarks_Track.html

About Kuro Siwo: Global Multi-Temporal Sentinel-1 SAR Flood Mapping Dataset (GeoTIFF)

Kuro Siwo is a manually annotated, multi-temporal Sentinel-1 SAR dataset for rapid flood mapping, presented at NeurIPS 2024 (Datasets and Benchmarks). It covers 43 flood events on six continents and three climate zones between 2015 and 2022, about 33 billion square metres of annotated area at 10 m. For every event there is one post-flood acquisition and two pre-flood acquisitions, a DEM layer, and an expert photo-interpreted mask with three classes (no water, permanent water, flood) plus a valid-pixel mask. Two components are provided: GRD (VV and VH backscatter) and SLC (four channels: VV phase, VH amplitude, VV amplitude, VH phase), each under its own folder; a sample's info.json carries event and acquisition metadata. This repository holds the GeoTIFFs (about 1 TB) for building embeddings or extending the dataset; a WebDataset copy sits in orion-ai-lab/Kuro-Siwo-Webdataset. The authors release Kuro Siwo under a CC BY licence; cite Bountos et al. 2024.

How to load this dataset in Python

from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="orion-ai-lab/Kuro-Siwo-GeoTIFFs", filename="GRD/00000.tar", repo_type="dataset")
print(path)

Data structure

Files

Example file: GRD/00000.tar

Analyze this dataset without downloading it

On Scigantic, Kuro Siwo: Global Multi-Temporal Sentinel-1 SAR Flood Mapping Dataset (GeoTIFF) 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