Firefighters traditionally work with satellite data, but it is incredibly slow, often with a 12-hour delay, making the data ineffective. With a faster flow of information at their disposal, first responders' chances of effectively tackling a forest fire are greatly increased.
“Speed is one of the most important points. We want to be accurate, but also very fast in the predictions,” says Dr. Rossella Arcucci, Assistant Professor of Data Science and Machine Learning at Imperial College London.
Dr. Arcucci is another industry expert focused on harnessing the power of AI to improve emergency responses to wildfires.
But instead of using robotics, her focus is on using social media to identify the ignition point of an inferno and improve wildfire forecasting. She is working on this research together with the European Space Agency.
Their AI system scans social media – from TikTok and Instagram to Snapchat and X (formerly Twitter) – for videos and images of the wildfire. The content is then filtered and checked to ensure its reliability.
“If you post about a wildfire, you're not in the middle of it, hopefully you're running… so there's an error of a few miles,” Dr. Arcucci said.
“We combine the geolocation from the social media post with the risk map and then [the AI software] indicates the point with a higher probability of being the ignition point.
The margins of error are very low, says Dr. Arcucci, because the AI tool provides cross-validation. “When a wildfire happens, there's usually not one person talking about it,” she added.