S2 CONUS Hand Labeled Flood Data
Abstract
S2 multispectral dataset rasters and manual labels for the data/ folder of the flood mapping github repository. Data comes in N=146 tiles with size ~400 by ~400 pixels (4km x 4km). Run the code in the repository to patch the dataset into 64 x 64 patches for training. Rasters: rgb_*.tif: B04, B03, B02 b08_*.tif: B08 b11_*.tif: B11 b12_*.tif: B12 dem_*.tif: DEM waterbody_*.tif: waterbody roads_*.tif: roads flowlines_*.tif: flowlines tci_*.tif: TCI scl_*.tif: SCL nlcd_*.tif: NLCD
Fields of Study & Subjects
Flood Risk Assessment and Management
Remote-Sensing Image Classification
Hydrological Forecasting Using AI
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 8 February 2026
- Document type
- Dataset