SenForFlood Global Flood Detection Dataset
Comprehensive ML training dataset for flood mapping with 353 flood events from all continents using Sentinel-1 SAR and Sentinel-2 optical imagery with terrain data and analyst-verified flood masks
github.com
At a glance
- Region
- Global
- Coverage
- 353 flood events across all continents (2025)
- Access
- Free
- Category
- Research
What's inside
- Flood Maps
- SAR Flood Detection
- Training Data
- Flood Masks
What you can do with it
Typical work that data of this type supports.
- 01
Hazard maps
Map flood-prone areas and who and what is exposed.
Because it has: Flood Maps
- 02
Model validation
Check flood models against observed events.
- 03
Design storms
Build rainfall intensity-duration-frequency curves for drainage design.
Because it has: Training Data
Where AI adds value
Ideas for machine learning and AI agents built on this kind of data.
- 01
Flood extent from SAR
Segment flooded areas in radar imagery, even through cloud.
- 02
Depth estimates
Learn flood depth from terrain, rainfall and past events.
- 03
Downscaling
Turn coarse rainfall grids into field or ward scale estimates.
Related water data
Find more water data
Search and compare 1,900+ water data sources by region, type and access.