SEN12VTS: Sentinel-1 and Sentinel-2 Vegetation Time Series over Europe (2022-2023)
- Provider
- LINKS Foundation (Oldani et al.), GUARDIANS project
- License
- MIT (source)
- Size
- 479 files
- Stored at
- Hugging Face (https://huggingface.co/datasets/links-ads/sen12vts)
- Access
- Free and public. No account with the data provider is needed.
- Category
- Earth Observation & Geospatial
- Source
- https://huggingface.co/datasets/links-ads/sen12vts
About this dataset
SEN12VTS is a multi-sensor time-series dataset for NDVI regression and vegetation monitoring across Europe, built by LINKS Foundation for the GUARDIANS project. Bounding boxes of 512 x 512 pixels at 10 m were sampled where at least 90% of pixels fall in vegetated ESA WorldCover 2021 classes (tree cover, shrubland, grassland, cropland, wetland, mangroves), giving 1,166 non-overlapping boxes over 2022 and 2023 (36 only in 2022, 454 only in 2023, 676 in both), cropped to 256 x 256 tiles. Each box folder holds Sentinel-1 radiometrically terrain corrected (RTC) VV/VH GeoTIFFs split into ascending and descending orbits, Sentinel-2 L2A GeoTIFFs with 12 bands resampled to 10 m plus the scene classification layer (SCL), and a WorldCover raster. The full set is about 825 GB, served as 476 split tar.gz parts under data/ (concatenate and extract, then read with rasterio or xarray). Uses include NDVI reconstruction under cloud, SAR-to-optical fusion and vegetation time-series forecasting. Released under the MIT license by LINKS Foundation; use is also subject to the Copernicus Sentinel and ESA WorldCover data policies.
How to load this dataset in Python
from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="links-ads/sen12vts", filename="data/sen12vts.tar.000.gz.part", repo_type="dataset")
print(path)Data structure
Files
Example file: data/sen12vts.tar.000.gz.part
Analyze this dataset without downloading it
On Scigantic, SEN12VTS: Sentinel-1 and Sentinel-2 Vegetation Time Series over Europe (2022-2023) 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
- California 30x30 data
- NASA TRACE-A Project
- Sentinel-1 Monthly Mosaic
- TMB-S3: Sentinel-3 OLCI Time Series for Burned Area Segmentation of 246 European Wildfires
- NREL WIND Toolkit: Bangladesh wind resource data (2014-2017) in HDF5 format
- maxar-opendata-v1.1
- Capella Space Synthetic Aperture Radar (SAR) Open Dataset
- Canopy Tree Height Map for the Amazon Forest (mean height composite 2020-2024) by CTrees.org