Probability of Fire
Product Overview
A pure machine learning model trained on observed fire activity provides the probability of a fire to be detected given the predicted weather and fuel conditions.
Technical Details
Main hazard addressed:
Wildfires
Other related hazards:
Heatwaves
Type of Product:
Forecast
Data Type:
Raster Map
Licensing:
Free for use with European Centre for Medium-Range Weather Forecasts partners and member states
Probability of fire in Europe on 9 February 2026
Available Implementations
ARISTOTLE
Standard implementation from the ARISTOTLE project (All Risk Integrated System TOwards Trans-boundary hoListic Early-warning).
Horizon:
5 days
Spatial Resolution:
18 km
Time Resolution:
3 hours
Coverage:
Global
Update Every:
Day
Coverage: Array
Use Cases & Applications
Provides utility globally, therefore, available for all GOBEYOND’s pilot sites.
Associated Models
- Integrated Forecasting SystemView Model
Express interest
This project has received funding from the European Union’s Horizon Europe programme for research and innovation under grant agreement No. 101121135. Additional funding is provided by the Swiss Government.