Initial testing of the MAWA-1 automated wildfire detection system was successfully completed in Barrie, Ontario.During this phase, the system was flown over multiple fire hotspots of varying sizes at different altitudes and airspeeds. MAWA-1 consistently detected all hotspots from a wide range of altitudes, angles, and velocities—demonstrating its robustness across diverse flight conditions.
For advance testing, Next1 partnered with Yukon Wildland Fire Management to simulate real-world wildfire scenarios. All testing took place in and around Whitehorse, Yukon.
Each detection mission involved a designated target grid ranging from 1,000 to 2,500 hectares, located approximately 50 to 75 nautical miles from the base of operations. These grids presented complex terrain, including mountain peaks, dense vegetation, steep valleys, small towns, and bodies of water.
Within each target area, a simulated lightning strike was created using a small bush fire (2–4 square feet) along with smaller hotspots—represented by hot charcoal and camping stoves—dispersed approximately 100 feet apart. The goal was to assess MAWA-1’s ability to detect smaller heat sources in the presence of a more intense fire.
During Phase II testing, the Next1 team collected extensive data to analyze the effectiveness of MAWA-1 in real world conditions. All flights were conducted around midday, which is the most difficult time of day to detect heat sources due to the heat of the sun; testing for a worst-case scenario.
Comparison of simulated hotspot detected by MAWA-1(top image) vs human observer(bottom image)
Extensive analysis of the data collected during Phase II suggests that MAWA-1 was able to automatically detect over 85% of the hotspots in various target areas. This includes small hotspots in the vicinity of larger fires that are not visible to the human eye from the air.
The successful completion of Phase I and II testing marks a major milestone in the development of the MAWA-1 automated wildfire detection system. Demonstrating high accuracy in both controlled and real-world environments, MAWA-1 consistently detected a wide range of fire signatures under challenging conditions, including midday operations and complex terrain. The system's ability to identify small, otherwise undetectable hotspots near larger fires highlights its potential to significantly enhance early wildfire detection.
With further optimization and deployment during ideal detection windows, MAWA-1 is projected to achieve over 95% detection accuracy—offering a powerful, scalable solution for wildfire prevention and environmental monitoring.
Socials
Send us your thoughts