LandCoverNet South America: Annual Land Cover Labels on Sentinel-1, Sentinel-2 and Landsat-8 Time Series (2018)
- Provider
- Radiant Earth Foundation
- License
- CC-BY-4.0 (source)
- Format
- TIFF, CSV (in sampled files)
- Stored at
- Azure Blob Storage, https://radiantearth.blob.core.windows.net/mlhub/landcovernet_sa
- Access
- Free and public. No account with the data provider is needed.
- DOI
- 10.34911/rdnt.6a27yv
- Category
- Earth Observation & Geospatial
- Source
- https://doi.org/10.34911/rdnt.6a27yv
About this dataset
The South America part of LandCoverNet, a global training dataset for annual land cover classification from multi-sensor satellite time series, built by Radiant Earth Foundation with support from Schmidt Futures, NASA ACCESS and Microsoft AI for Earth. It holds 1,200 image chips over 40 tiles. Each 256 x 256 pixel chip at 10 m resolution comes with every 2018 observation from Sentinel-1 GRD (radiometrically calibrated and orthorectified backscatter), Sentinel-2 L2A surface reflectance and Landsat-8 Collection 2 Level-2 surface reflectance, and one annual label per pixel in seven classes: water, natural bare ground, artificial bare ground, woody vegetation, cultivated vegetation, (semi) natural vegetation and permanent snow/ice. Labels were drawn by annotators from the full annual time series, with per-chip consensus scores. Files are GeoTIFFs under data/v1.0/2018/ / / with a CSV of per-chip metadata and STAC JSON; measured 2026-11-15 the listing exceeds 40,000 files (at least 6.4 GB). Read the rasters with rasterio or rioxarray, or stream them with torchgeo. Released under CC BY 4.0; cite Radiant Earth Foundation (2022), LandCoverNet South America, Radiant MLHub, doi:10.34911/rdnt.6a27yv, and Alemohammad and Booth, arXiv:2012.03111.
How to access this dataset in Python
import adlfs
fs = adlfs.AzureBlobFileSystem(account_name='radiantearth', anon=True)
fs.ls('mlhub/landcovernet_sa/')[:10]Lists the files with anonymous access. Runs on any machine, with no cloud account.
Data structure
File types in sampled files: TIFF, CSV
Files
Example file: data/v1.0/2018/17NRB/00/17NRB_00_2018_LC_10m.tif
Path pattern: {id}/{seg1}/{YYYY}/{seg3}/{seg4}/{filename}
Analyze this dataset without downloading it
On Scigantic, LandCoverNet South America: Annual Land Cover Labels on Sentinel-1, Sentinel-2 and Landsat-8 Time Series (2018) 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
- PALSAR-2 ScanSAR Flooding in Rwanda (L2.1)
- State of Colorado Imagery
- NASA Flood Extent Detection Dataset: Sentinel-1 SAR VV/VH Scenes with Water Labels (Bangladesh and US floods, 2017-2019)
- Sentinel-5P Level 2
- CanElevation - Canada Digital Elevation Models
- Smallholder Cashew Plantations in Benin (Pleiades-derived labels with Sentinel-2)
- SEN2NEON: Sentinel-2 to NEON Hyperspectral Multispectral Super-Resolution Benchmark
- NUVIEW - Multi-State Geospatial Data