Automated Identification of Hazards, Vulnerabilities, and Risks Associated with Large-Scale Wildfires Using Multispectral Images and AI Algorithms
Application for the Automated Identification of Large-Scale Wildfire Hazards
Vexiza, LLC
Castile and León Regional Government
Identification of the Risk of Large-Scale Wildfires
What is it and what is it used for?
Geospatial representation of data calculated by an algorithm based on artificial intelligence techniques capable of predicting with high accuracy the type and potential behavior of large-scale wildfires, thereby improving the accuracy of identifying areas that require vegetation management and fire prevention measures.
The problem it solves
The inability to continuously monitor the current state of vegetation cover across large tracts of land, especially those located on privately owned property. In this scenario, the ability to identify in advance the areas where aggressive fire behavior is expected is a critical factor not only for the effectiveness of the operational response but also for the safety of the firefighting resources deployed on the ground.
"Artificial intelligencemakes it possible to process massive amounts of data on the hydrological and structural conditions of vegetation obtained from satellite data, enabling the automatic assessment of wildfire risk and severity across all types of landscapes."
What makes it innovative?
Availability of satellite data with high spectral and temporal resolution
Such as those provided by the Copernicus program through Sentinel-2
The Potential of Machine Learning Algorithms
By extracting complex patterns from large volumes of heterogeneous data, this opens up the possibility of developing prediction systems capable of continuously characterizing the condition of forest fuel and its potential for ignition and spread. The resulting prediction maps are presented in standard geospatial formats and are updated periodically based on the latest available satellite data, without requiring user intervention.
Expected impact
The algorithm is capable of assessing the risk and severity of large-scale wildfires based on the current state of vegetation, taking into account various environmental and structural conditions of the landscape—especially in forested areas where the vertical structure of the fuel plays a critical role in the potential behavior of the fire. All of this helps improve the accuracy of identifying areas that require vegetation management and fire prevention measures.
