Artificial intelligence (AI) could help scientists predict when\nclimate tipping points will occur, potentially more than two years\nbefore they happen. This capability could open up opportunities to\nanticipate environmental changes whose effects are difficult or\nimpossible to reverse.<\/p>\n
The term \u201ctipping point\u201d was popularised by Canadian journalist and\nauthor Malcolm Gladwell in a book published in 2000. In a social\ncontext, it describes the moment when a trend or behaviour reaches a\ncertain threshold and spreads rapidly.<\/p>\n
In science, a tipping point occurs when a system undergoes a\nfundamental change that makes it difficult or impossible to return to\nits original state. In the climate system, this can happen when global\nwarming pushes an environmental component beyond a certain threshold,\nsuch as when an ice sheet begins melting in a way that is difficult to\nreverse.<\/p>\n
AI has been widely used to predict daily weather and extreme events.\nPredicting when tipping points in the climate system will occur,\nhowever, remains a challenge because climate change can alter the Earth\nsystem\u2019s baseline conditions.<\/p>\n
A recent study published in the journal Proceedings of the National\nAcademy of Sciences (PNAS) found that a new AI approach could help\nestimate when such tipping points will occur.<\/p>\n
Researchers from the California Institute of Technology (Caltech),\nthe University of Colorado Boulder (CU Boulder) and technology company\nNVIDIA developed an approach based on recurrent neural operators (RNO)\nto predict tipping points using simplified models.<\/p>\n
Caltech professor of computer science and mathematics Anima\nAnandkumar explained that a tipping point is a condition in which a\nsystem loses its ability to return to its original state as instability\nincreases.<\/p>\n
\u201cA tipping point is when you have no chance of going back because the\nsystem is already unstable. A change in part of the system means it\ncannot return to its original state. That is why a tipping point can be\nseen as a disaster,\u201d Anandkumar said, as quoted by Phys.org on Saturday\n(17 October 2026).<\/p>\n
Anandkumar and her colleagues\u2019 research was also cited in a recent\nUnited Nations report on climate tipping points. The researchers said\nthe AI approach could potentially be applied more broadly, including in\nclimate science, ecology and epidemiology, as well as aircraft\ndesign.<\/p>\n
\u201cThis is a stepping stone towards characterising tipping points and\npredicting when they will occur in the future. We demonstrated it on\nsimplified climate models and models of air turbulence over aircraft\nwings\u2014systems that could not previously be predicted,\u201d Anandkumar\nsaid.<\/p>\n
One scenario examined was the loss of low cloud cover over the oceans\nas atmospheric carbon dioxide concentrations rise. This could occur when\nglobal warming pushes the cloud system beyond a certain threshold,\nalthough the process has not yet been widely observed directly.<\/p>\n
The loss of low cloud cover could exacerbate global warming. Low\nclouds generally reflect some solar radiation back into space. When they\ndisappear, more of the Sun\u2019s energy can be absorbed by the Earth\u2019s\nsurface, raising temperatures.<\/p>\n
To test their approach, the researchers used simulation data to train\nand evaluate an RNO-based AI programme for estimating when the clouds\nwould disappear. The model produced a prediction more than two years\nbefore the simulated tipping point, with an estimated timing error of\nabout 11 days.<\/p>\n
Despite the promising results, further testing using real-world\nobservations is needed to establish how well the approach can be applied\nto actual climate systems.<\/p>\n
Anandkumar said earlier AI approaches were generally developed on the\nassumption that past weather patterns could serve as a guide to\npredicting future conditions.<\/p>", "url": "https:\/\/jawawa.id\/newsitem\/ai-could-predict-critical-climate-tipping-points-1792291827", "image": "" }, "sponsor": "Okusi Associates", "sponsor_url": "https:\/\/okusiassociates.com" }