LIGHTNING STRIKESAI Model Predicts Lightning Wildfires with 90% Accuracy
Israeli researchers use seven years of weather and satellite data to predict future wildfires caused by lightning strikes.
They say lightning never strikes twice, but it might be possible to find out where it’s going to strike next, thanks to a new AI model developed by Bar-Ilan University (BIU) researchers.
Designed with wildfire prevention in mind, the AI can predict the timing and location of wildfire-inducing lightning strikes with over 90 percent accuracy, according to a test performed using wildfire data from 2021.
“We are at a critical moment in understanding the complexities of wildfire ignitions,” said project co-lead Oren Glickman, from BIU’s department of computer science.
“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 and fellow BIU researcher Assaf Shmuel trained their model using seven years’ worth of global satellite data and detailed information on vegetation, weather patterns and topography.
According to the study, published in Scientific Reports, the BIU model is much more accurate than traditional wildfire prediction methods, thanks to its unique utilization of satellite and weather data. They say this model could help emergency services thwart wildfires in a much safer, more effective manner.
Lightning strikes are a significant natural cause of wildfires globally. In the United States, lightning accounts for 16% of wildfire ignitions but is responsible for 56% of the total acreage burned due to remote locations and delayed detection. Global impacts vary by region.
Climate models predict future increases in wildfire ignitions due to more frequent lightning strikes under warming conditions.
The model is not yet ready for real-time implementation, but its developers say it could shift the balance of power between man and nature in the struggle to keep forests and vegetation safe from fire damage.
As Shmuel notes: “With the growing implications of climate change, new modeling tools are required to better understand and predict its impacts; machine learning holds significant potential to enhance these efforts.”
Zachy Hennessey is a writer and editor at ISRAEL21c. This article is published courtesy of Israel21c.
