JERUSALEM, March 31 (Xinhua) -- Israeli researchers have developed an AI model to predict lightning-induced wildfires with over 90 percent accuracy, offering a new tool in the fight against climate change, Bar Ilan University said in a statement on Monday.
The study, published in Nature Communications, could enable meteorological services, fire departments, and emergency planners to respond sooner and smarter, and potentially save lives and ecosystems.
While human activity is a leading cause of wildfires, lightning ignites some of the most destructive and unpredictable blazes, particularly in remote areas where firefighting resources are limited. These fires can smolder underground for days before erupting, making them especially dangerous. Climate change has worsened the situation, with more frequent lightning storms and hotter, drier conditions providing more fuel for fires.
The new AI model, trained on seven years of high-resolution global satellite data, assesses fire risk by analyzing multiple factors, including lightning strikes, vegetation density, weather patterns, and topography.
A key finding is that lightning-induced wildfires behave differently from human-caused ones, highlighting the need for specialized approaches tailored to each fire type.
Moreover, the study reveals that climate change is accelerating the risk of lightning-driven fires. The model forecasts a rising risk of lightning fires due to increased lightning activity and more frequent "fire weather" conditions.
By leveraging Big Data and AI, the model represents a significant advancement in wildfire prediction, enhancing global climate resilience and emergency preparedness. ■
