Can artificial intelligence help predict Alaska wildfire growth?

Flames from the Funny River Wildfire flare up on May 24, 2016 in Soldotna, Alaska. The wildfire started unusually early in the season and burned nearly 200,000 acres on the Kenai Peninsula. (Photo by Rashah McChesney/Peninsula Clarion)

Using Alaska wildfires in their recent study, researchers at the University of California Irvine say they have a new way of predicting wildfire growth, right at the time the fire ignites.

The researchers developed an algorithm through what’s called machine learning that looks at information like climate and weather data to issue a prediction on how big a particular fire will grow. And knowing that, they say, could help fire managers figure out where to focus firefighting efforts.

Shane Coffield is a graduate student in the Earth System Science program at UC Irvine and lead author of the research paper. He spoke with Alaska Public Media’s Casey Grove.

Read more about Alaska’s 2019 wildfire season.