Next Day Wildfire Spread: 1 km Fire Mask Transition Dataset for the Contiguous US (TACO)
- Provider
- Huot, Hu, Goyal, Sankar, Ihme, Chen (Google Research); TACO repackaging by ISP, Universitat de Valencia
- License
- CC-BY-4.0 (source)
- Size
- 9 files
- Format
- PNG (in sampled files)
- Stored at
- Hugging Face (https://huggingface.co/datasets/isp-uv-es/next-day-wildfire-spread-taco)
- Access
- Free and public. No account with the data provider is needed.
- Category
- Earth Observation & Geospatial
- Source
- https://arxiv.org/abs/2112.02447
About this dataset
Next Day Wildfire Spread (IEEE TGRS, 2022) is a curated benchmark for predicting where an active wildfire will be tomorrow, aggregated in Google Earth Engine from nearly a decade of fires across the contiguous United States. It has 18,545 chips of 64x64 pixels at 1 km, each pairing yesterday's MODIS fire mask and twelve covariates with the next day's fire mask: SRTM elevation, VIIRS NDVI, gridMET wind speed and direction, minimum and maximum temperature, humidity and precipitation, drought index (PDSI), energy release component, and population density. Both fire masks use -1 for uncertain pixels, kept as an ignore index. The release carries no coordinates or dates, so none are given here. Splits are 14,979 train, 1,877 validation and 1,689 test. About 1.7 GB in TACO format, read with tacoreader. Published by the authors on Kaggle under CC BY 4.0; cite Huot et al. 2022.
How to load this dataset in Python
from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="isp-uv-es/next-day-wildfire-spread-taco", filename="assets/logos.png", repo_type="dataset")
print(path)Data structure
File types in sampled files: PNG
Files
Example file: assets/logos.png
Analyze this dataset without downloading it
On Scigantic, Next Day Wildfire Spread: 1 km Fire Mask Transition Dataset for the Contiguous US (TACO) 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
- Global Human Settlement Layer (GHSL) on AWS: population, built-up surface, volume, height, land and settlement model grids
- Sentinel-5P Level 2
- geovibes-datasets
- Satlas Tree Cover: Global 10 m Five-Level Tree Canopy Cover Rasters from Sentinel-2, January and July 2016-2023 (Ai2)
- DEM-Terrain
- SSL4EO-L Benchmark: Landsat Land Cover (NLCD) and Crop Type (CDL) Segmentation
- New Zealand National Coastal LiDAR 1 m Digital Elevation Model (Topo-Bathymetric DEM), LINZ
- Digital Earth Africa - Copernicus Global Land Service - Lake Water Quality