While there is an instinct to blame humans for the carnage of wildfires, lightning is increasingly responsible for some of the greatest devastation and damage the blazes create, per researchers.
In the wake of the tragic Los Angeles wildfires that took dozens of lives, burned 1.5 million acres, and caused hundreds of billions in damages, a group of Israeli researchers developed a remarkable tool that can predict lightning strikes with over 90% accuracy.
The Bar-Ilan University-led team's machine learning model was unveiled in the Scientific Reports journal, and they outlined their findings in a news release.
The researchers revealed that lightning-caused fires now are responsible for the majority of the carnage in regions like Canada and the western U.S. Making the picture even bleaker, they also say that the warming planet is making the extreme lightning strikes worse and more frequent due to factors like record-setting heat and drought.
The team described lightning strikes as "sneaky" because they can occur far away from population centers, build up underground, and then emerge in uncontrollable fashion. Preventive actions are also hard to manage.
"Firebreaks in which trees and undergrowth are cleared between sections of trees can reduce the danger, but these are very difficult and expensive to maintain because they grow back quickly," researcher Dr. Assaf Shmuel told The Jerusalem Post.
Watch now: Giant snails invading New York City?
Thus, being able to predict where lightning strikes will occur could make a huge difference in fire preparedness. The researchers trained a machine-learning model on high-resolution, global satellite data from 2014 to 2020.
The researchers said the AI-powered predictive tool also incorporated other factors that could exacerbate a fire, like vegetation, local weather, and topography. Given all the inputs, the model was tested on wildfire data from 2021. The model's 90% or more accuracy is unprecedented and game-changing, per the scientists.
"We hope this research empowers countries around the world to better anticipate and mitigate these fires," researcher Dr. Oren Glickman commented to The Jerusalem Post. "Meteorological services, fire departments, and emergency planners can respond earlier and more effectively."
Glickman revealed that models used for human-started fires "struggle to predict lightning-induced fires that behave very differently and often start in hard-to-reach areas." Given that their model indicated that lightning strikes will only become more frequent, the researchers say different tools are needed for each fire type.
TCD Picks » Upway Spotlight
Upway makes it easy to find discounts of up to 60% on premium e-bike brands
|
What would it take for you to get rid of your gas stove?
Click your choice to see results and speak your mind. |
I'd pay for it myself 
The researchers are just one of many teams using machine learning to help ward off some of the extreme weather's greatest challenges, including separate efforts taking on wildfires and droughts.
The Israeli team's model has a key step before it can be deployed to predict lightning-based fires. It will have to become operational on real-time inputs and forecasting systems to offer live predictions for lightning strikes. The team is bullish about its potential and said the model "lays critical groundwork" to meet the increasing challenges.
"Machine learning offers the potential to revolutionize how we predict and respond to lightning-ignited wildfires, providing insights that could save lives and preserve ecosystems," Glickman concluded to The Jerusalem Post.
Join our free newsletter for weekly updates on the latest innovations improving our lives and shaping our future, and don't miss this cool list of easy ways to help yourself while helping the planet.
Cool Picks
