B.C. Company Develops AI 'Scout' to Forecast Wildfire Risk Months Before First Sparks Ignite
A groundbreaking artificial intelligence system developed by a British Columbia company could predict wildfire risk months before the first flames appear, potentially preventing catastrophic events like the devastating 2021 Lytton wildfire.
Early Warning System Analyzes Multiple Risk Factors
SkyScoutAi, a B.C.-based technology company, has created an AI model that draws from various data sources including drone footage, weather patterns, terrain analysis, and historical fire data to generate comprehensive threat assessments. The system constantly calculates and updates wildfire risk based on five critical factors: fuel moisture, flammability, weather conditions, terrain characteristics, and historical fire patterns.
Michal Aibin, program head in applied computing at the B.C. Institute of Technology and chief technology officer for SkyScoutAi, developed the fire prevention tool after recognizing a significant gap in wildfire management technology.
"While we have excellent technology for detecting and fighting fires, we lacked sophisticated tools for prevention," Aibin explained. "You cannot prevent random natural events like lightning strikes, but you can prepare areas to respond more effectively when fires do occur."
Real-Time Monitoring Across British Columbia
A demonstration on SkyScout's website shows live data for hundreds of locations across British Columbia. The system provides detailed risk assessments that can guide fire prevention work and resource allocation long before traditional fire season begins.
In one example from Wednesday's data, an area on Alpine Way in Whistler showed an overall low threat level despite terrain scoring 39 percent risk. Other factors including fuel moisture and flammability registered in the single digits, contributing to the favorable assessment.
Conversely, a location near Field Road in Kelowna displayed moderate threat levels with individual risk factors showing elevated scores: terrain at 67 percent, fuel moisture at 43 percent, and weather at 28 percent. A bar graph tracking trends over the previous 30 days revealed the threat level alternating between low and moderate as daily weather changes influenced fuel moisture conditions.
Retrospective Analysis Validates System Effectiveness
A recently published white paper from SkyScoutAi presents compelling evidence of the system's capabilities through retrospective analysis of the 2021 Lytton wildfire. Using only data that would have been available at that time, the AI model identified extreme risk conditions present and measurable long before the fire started.
The analysis revealed high fuel loads including dead trees, beetle-damaged timber, and dry brush observed through drone imaging, combined with other risk factors like weather patterns, terrain characteristics, and fire history. These conditions created a perfect storm that the system would have flagged months in advance.
The technology represents a significant advancement in proactive wildfire management, moving beyond reactive firefighting to strategic prevention. By identifying vulnerable areas before conditions become critical, communities can implement targeted mitigation measures including fuel reduction, strategic planning, and resource allocation.
As climate change increases wildfire frequency and intensity across British Columbia and beyond, such predictive technologies could become essential tools for protecting communities, infrastructure, and natural resources from devastating wildfires.
