AI Helps Spot Wildfires Faster
Wed, April 21, 2021

AI Helps Spot Wildfires Faster

Descartes Lab debuted its wildfire detector to spot blazes in California / Photo Credit: Krista Kennell (via Shutterstock)

 

A startup a few states away from California used AI to pinpoint the location of wildfires in California faster than firefighters or civilians, wrote Rachel Metz of business and financial news CNN Business. Santa Fe-based Descartes Lab debuted its wildfire detector in July. The firm’s AI software examines images coming in roughly every few minutes from two different US government weather satellites. The AI searches for any changes such as a shift in thermal infrared data and the presence of smoke. Descartes is testing its detector by alerting select forestry officials in New Mexico. Descartes told CNN Business that its detector has found about 6,200 blazes thus far. It said it can also detect fires when they are just about 10 acres in size. 

Clyde Wheeler, an applied scientist at Descartes who spearheaded the project, said, “Since wildfires are hot, they stand out pretty well.” It’s a big change from how fires are spotted in the country. Fires are spotted via planes or lookout towers, or having citizens report fires. Donald Griego, New Mexico's State Forestry Resource Protection Bureau Chief, informed the outlet that notifications from Descartes have “definitely” aided the state in finding wildfires more quickly and accurately “than they otherwise could.” 

The alerts also enable first responders to arrive at the scene faster. A motorist or an airplane pilot may report a cloud of smoke billowing from a “general area.” But Descartes’ text-based tool makes the location of the fire even more specific. Griego explained, “That's very beneficial," Griego added, "especially at night when it's hard to determine what mountain range this fire's actually on when you're on top of a peak 20 miles away.” 

However, Descartes is also looking to improve its services. For instance, Descartes’ applied science team lead Caitlin Kontgis stated that the firm is studying how additional data might help it track fires such as digital elevation models “showing the locations of steep slopes,” which would make it more challenging to put out the blaze. But there are limits. In New Mexico, a message from Descartes’ fire detector has to be forwarded to Griego and other forestry officials. Then, the text has to be relayed to an appropriate field office to verify the fire. Griego emphasized, “There are several steps you need to take before someone shows up and starts suppressing the fire.”