BYU student helps enhance wind tracking tool for wildfires

PROVO, Utah (ABC4) — A Brigham Young University (BYU) student has used machine learning and math to improve a key tool that firefighters rely on while they are out in the field battling wildfires.

“I think it’s really cool when you study math, you end up working on problems that you never would have really guessed, like I have done things in so many different fields,” Jane Housley is a BYU mathematics graduate student and

a wildfire modeling researcher.

Housley recently wrapped up her master thesis in partnership with the U.S. Forest Service’s Missoula Fire Sciences Lab and focused on improving WindNinja.

WindNinja is used by fire crews and analysts to predict how wind will move through terrain during a fire. It is a simulation tool created by the Missoula Fire Sciences lab.

According to the U.S. Forest Service, the behavior of wildland fires and the dispersion of smoke from these fires depends, in part, on ambient and fire-induced winds that work to spread fires across the landscape and mix fire emissions into the atmosphere.

Machine learning and wildfire modules

Housley’s study focused on improving the device to model what’s called a cavity zone. That’s the area directly behind a mountain or ridge where wind tends to swirl backward and create a whirlpool-like motion.

This movement can dramatically shift how and where a fire spreads.

Housley helped improve two key areas of the WindNinja, the mass-conserving solver and the computational fluid dynamics (CFD) solver. The first one is fast but less accurate but on the other hand, CFD is more precise but much slower.

“This is the first time that we’ve taken machine learning and AI and applied it to the field of wildfire modeling. The real way to study this was to try and understand the physics and mechanics of how things work,” Housley told ABC4.com.

Through her research, Housley trained the neural network to learn patterns of error in the mass-conserving solver, using the CFD solver accuracy as the goal. 

The device could now be able to recognize wind patterns the way facial recognition spots a familiar face. Housley said she still remembers the feeling of excitement when she saw how accurate and efficient her new model was.

“Once I had the network built and plugged in the data and ran the simulation, the results were really good. I thought, ‘I must be doing something wrong,’” Housley said in a press release. “I combed through every single line of code and found that it was working correctly. I was really excited.”

Collaborating with firefighters and scientists at the Missoula Fire Sciences Lab will be an experience that Housley says she will cherish forever.

Alexa Mcfadden contributed to this report.