Researchers at the University of Waterloo are developing a new artificial intelligence (AI) that could act as an early warning system against climate change tipping points. The new research is focused on thresholds beyond which rapid or irreversible change happens in a system.
Chris Bauch is a professor of applied mathematics at the University of Waterloo, and he is co-author of the research paper.
“We found that the new algorithm was able to not only predict the tipping points more accurately than existing approaches but also provide information about what type of state lies beyond the tipping point,” Bauch said. “Many of these tipping points are undesirable, and we’d like to prevent them if we can.”
Climate-Change Tipping Points
These various climate-change tipping points can include melting Arctic permafrost, which could release mass amounts of methane that leads to further rapid heating. It also includes the breakdown of oceanic current systems, which can result in immediate changes in weather patterns. Another possibility is ice sheet disintegration, which could cause rapid sea-level change.
According to the researchers, this new approach is innovative given that it was programmed to learn about more than one type of tipping point. Instead, it learns the characteristics of tipping points in general.
The new algorithm is based on hybridizing AI and mathematical theories of tipping points, which results in better results than just one method on its own. The AI is trained on a “universe of possible tipping points,” which includes around 500,000 models. It is then tested on specific real-world tipping points in various systems, such as historical climate core samples.
Timothy Lenton is director of the Global Systems Institute at the University of Exeter and one of the other co-authors of the study.
“Our improved method could raise red flags when we’re close to a dangerous tipping point,” said Lenton. “Providing improved early warning of climate tipping points could help societies adapt and reduce their vulnerability to what is coming, even if they cannot avoid it.”
Deep Learning Algorithm
The researchers relied on deep learning, which is increasingly impacting pattern recognition and classification in a positive way. The researchers have converted tipping-point detection into a pattern-recognition problem for the first time, and this helps detect the patterns that are present before a tipping point. This in turn helps a machine-learning algorithm be able to say whether a tipping point is coming.
Thomas Bury is a postdoctoral researcher at McGill University and another one of the co-authors of the paper.
“People are familiar with tipping points in climate systems, but there are tipping points in ecology and epidemiology and even in the stock markets,” said Bury. “What we’ve learned is that AI is very good at detecting features of tipping points that are common to a wide variety of complex systems.”
Madhur Anand is another one of the researchers and director of the Guelph Institute for Environmental Research.
According to Anand, the newly developed deep learning algorithm is a “game-changer for the ability to anticipate big shifts, including those associated with climate change.”
The team will now work to give the AI the data for contemporary trends in climate change. However, Anand warns that the outcome is based on how these findings are used.
“It definitely gives us a leg up,” she said. “But of course, it’s up to humanity in terms of what we do with this knowledge. I just hope that these new findings will lead to equitable, positive change.”