AI developed by NCAR will identify forests at high fire risk

BOULDER, Colo. — The threat of wildfires has been a particular focus in Colorado for months, and projections show the worst could still be to come as drought conditions worsen. However, firefighters will have a new tool at their fingertips – developed right here in Colorado – that will help them know in advance where fires could spread quickly.

The artificial intelligence developed in National Center for Atmospheric Research (NCAR) in Boulder allows scientists to identify the locations and amounts of dead or dying trees and vegetation in Colorado’s forests, which are burning rapidly and, therefore, causing wildfires to spread faster.

The NCAR scientists behind its development illustrated the power of the technology by retroactively mapping the East Troublesome Fire of 2020, which grew much more explosively than expected.

NCAR

Projections of the East Troublesome Fire of 2020, using NCAR artificial intelligence to map downed trees and beetle destruction.

“We can run a simulation and basically use it to understand how the fire spreads, what areas are burning, and how much smoke is produced by the fire,” said project scientist Timothy Juliano of the East Troublesome Fire projection. “Before this, the fuel map inside our model didn’t even account for beetle mortality. It assumed the forest was healthy and had live trees that were still standing. In reality, we know that this area was devastated by beetle destruction and many trees felled and died… If we had this product available [then]we could have been much closer to observations of how fast the fire was going to spread.

Currently, researchers are keeping tabs on forest health through a combination of human cataloging and various ecological datasets. This leads to a laborious process and, at times, unreliable results.

“You can imagine it takes a lot of time and effort to put this data together, so it’s only released about every five years,” said Amy DeCastro, the study’s lead researcher. “Our AI-based method is able to update vegetation maps during these periods between release and accurately capture any vegetation disturbances that may have occurred in the meantime, such as drought and infestation. beetles.”

As Colorado firefighters look to the future — and into the prospects of longer, more severe fire seasons — technology that gives a more robust understanding of forest health will better equip them to show up in the right places. at the right time. This will provide a head start that could literally save homes and lives.

“We intentionally designed this project around freely available satellite imagery and an online modeling platform,” DeCastro said. “So any firefighter or field manager who wants to update their vegetation data before running the wildfire forecast has access to this method.”

“They can issue their evacuation warnings faster and have a better idea of ​​where the fire will be and at what time,” Juliano said.

The NCAR team has been working on this artificial intelligence for more than a year and knew its potential power very early on. It was in December, however, that its impact was truly felt for Juliano, as he was forced to evacuate his Louisville home during the Marshall fire.

“I think for me it’s like an extra motivation, honestly, just to be really passionate about research and trying to solve some of these problems,” he said. “It hits home when you see it first hand.”