Deepfire X Stanford Wildfire Spread Dataset: Per-Day Rasters for U.S. Fires

Multimodal per-fire and per-day raster data covering U.S. wildfire spread from 2016 to 2025. The dataset is hosted on Kaggle and appears to be a collaboration between Deepfire and Stanford. It provides daily snapshots of fire progression.

Use Cases

• Train machine learning models for wildfire spread forecasting based on daily raster progression.
• Analyze temporal patterns of wildfire growth across the United States based on per-day data.
• Develop multimodal fusion models for environmental monitoring based on the described raster data.

Strengths

• Covers a 10-year time range from 2016 to 2025.
• Provides data at a per-fire and per-day temporal resolution.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data may reflect geographic bias inherent to its U.S.-only coverage.

Provenance

Source

Kaggle, with implied collaboration between Deepfire and Stanford.

Time Range

2016-2025

Geography

United States