Researchers are developing artificial intelligence to assess tipping points of climate change. The deep learning program could serve as an early warning system for climate change that has gotten out of hand.
Chris Bauch, professor of applied mathematics at the University of Waterloo, is co-author of a recent research paper detailing the results of the new deep learning algorithm. According to Bauch, research examines the points beyond which a system undergoes rapid or irreversible change.
“We found that the new algorithm could not only predict tipping points more correctly than previous techniques, but it could also provide information about the type of state that existed beyond the tipping point,” said Bauch. “A lot of these moments of change are unwelcome, and we would prefer to avoid them as much as possible.”
Melting arctic permafrost, which could release massive amounts of methane and stimulate even faster warming; the collapse of current ocean systems, which could lead to almost immediate changes in weather patterns; or the disintegration of the ice sheet, which could lead to a rapid change in sea level, are all examples of tipping points often associated with uncontrollable climate change.
According to the researchers, AI was trained to learn not only one type of tipping point, but also the characteristics of tipping points in general.
Related article: Several Factors for Environmental Tipping Point to Reach Critical Levels
Combine AI and mathematical theory
The power of the approach comes from the combination of artificial intelligence and mathematical tipping point theories, which achieves more than each method could do on its own. The researchers tested the AI at particular real-world tipping points in various systems, including historical climate core samples, after training it on a “universe of potential tipping points” that included around 500,000 models. .
“When we approach a critical tipping point, our improved technique can emit red signals,” said Timothy Lenton, director of the Global Systems Institute at the University of Exeter and one of the co-authors of the study. “Improving early warning of climate tipping points could help civilizations adapt and reduce their vulnerability to what is to come, even if they cannot prevent it.”
(Photo: Image by Comfreak from Pixabay)
Deep learning is making significant advances in pattern recognition and classification, with researchers for the first time converting tipping point detection into a pattern recognition problem. This is done in the hope of detecting patterns that occur before a tipping point and using a machine learning algorithm to predict whether or not a tipping point will occur.
“People know the tipping points in climate systems, but there are also tipping points in ecology, epidemiology and even the stock market,” said Thomas Bury, postdoctoral researcher at McGill University and one of the co -authors of the article. “What we found is that AI is effective at recognizing the failover traits common to a wide range of complex systems.”
The new deep learning system is a “game changer for the ability to predict significant changes, especially those related to climate change,” according to Madhur Anand, director of the Guelph Institute for Environmental Research and another of the project researchers.
Identify tipping points
Now that their AI understands how tipping points work, the team is working on the next step: providing them with data on current models of climate change. First, however, Anand warned of the dangers of such information.
She said: “It gives us a head start.” “However, it is for humanity to decide what we do with this information. I just hope that these new results lead to just and beneficial reform.”
Also read: Climate tipping points inevitably lead to dire environmental consequences
For more environmental news, be sure to follow Nature World News
© 2021 NatureWorldNews.com All rights reserved. Do not reproduce without permission.Source link