You can enable subtitles (captions) in the video player
The world is facing an alarming surge in forest fires. Between 2001 and 2023, the amount of tree cover the world lost each year due to fires rose dramatically. But new AI-supported technology could help slow some blazes.
In March, FireSat - a non-profit collaboration of corporations, wildfire authorities and others - launched the first of some 50-planned firefighting satellites. They'll make up an AI-enhanced network, designed to cover the globe and detect flames before they spread beyond a '5 by 5 metre square' area in near real time.
In the US, the University of California, San Diego's Alert California Programme has now deployed more than a thousand cameras and sensor arrays that use AI to analyse footage in real time, and alert firefighting authorities if necessary. In Turkey, officials said an interactive wildfire risk map, launched in 2022, that harnesses AI and machine learning, has led to an 80 per cent accuracy rate in predicting wildfires 24 hours before their outbreak.
But technology is certainly no silver bullet. And costs can be a major drawback. For example, a single camera station being maintained and monitored by a commercial operation can cost around $50,000 a year. Another hurdle, according to a 2022 UN report, typically, close to 60 per cent of authorities' expenditure on wildfires goes on response, while prevention and preparation receives significantly less funding. The report strongly suggests that this ratio needs to change, and warns that by the end of the century, the number of wildfires the world faces is projected to increase by some 50 per cent.
=
