Skip to main content

CBS - On The Dot with David Schechter: Wildfires & climate change

Link to Article

By CBS Baltimore Staff

All week, we're exploring how climate change is leading to more extreme wildfires.  

And, we're learning about solutions that can protect communities and save lives., including data, technology and collaboration.  

Together they can help locate a wildfire sooner and predict its movement, to stop it faster.   

One of the things that would be interesting to look at is how big data and massive amounts of information are being used to solve this problem of fire.   

"You know, we need to respond best to using the best of the technology, best of the data," said Dr. Ilkay Altintas, the director of the Wildfire Lab at the University of California, San Diego.

Pinpointing a fire from the moment it starts is critical to containing it fast.

Altintas is trying to address that challenge across California using big data.

A timelapse from a huge fire in 2017 shows the fire continues to flare up for more than a month.   

"If you managed to stop this in the first couple of hours, it's a lot easier to stop," Altintas said.   

The first pieces of data she uses come from 911 callers.  

"That location has a level of accuracy, but it's not as accurate as we'd like it to be," Altintas said. 

If a 911 caller says, 'Hey, I see a fire, we don't know exactly where that was."  

"Yeah, you know, the proximity and where the person is calling from," Altintas said. "There's a little bit of lag in between that location that they call from and where the smoke is to pinpoint the fire."

Altintas said the next data source is mountaintop cameras.   

"And these cameras then could triangulate and say, 'Yeah, it's pretty much right here,'" Altintas said.  

With that location, a scientist at UC San Diego pulls in data from nearby weather stations to project where the fire's headed and how big it will get.  

So, you can see air temperature, humidity, fuel, moisture, wind and gusts.   

"Yeah, and these are measured in real-time," Altintas said. "Instead of a day like we are in today. Let's say this was a Santa Ana day. Within a matter of hours, it's hitting more population and more housing."   

So, it could go from 1081 people to 3702 people.   

"And many more homes. The acreage grows quite a bit," Altintas said. "So these are every half an hour."   

Finally, an aircraft sends back real-time video so scientists can map the perimeter of the fire.   

"It gives us data about the fire front as it's as the perimeter is growing," Altintas said.   

So what happened? Why did it just change?   

"What we've just seen is the change to IR," Altintas said. "It's a different type of sensing that could see through the smoke and gives us a better understanding of where the fire is burning. And once you have that, you can use that as information to see what can happen in the next couple of hours."   

Are we moving in a direction, do you think that fire will be more predictable and more containable and more manageable?  

"I think this is the most important thing that the recognition that the wildland fire problem is solvable if you  do some things right collaboratively," Altintas said.

Researchers at UC San Diego are working to develop artificial intelligence tools they hope can help with challenges that we are facing now because of climate change.