AI-enhanced research shows regional warming will exceed critical thresholds faster than expected, with most regions surpassing 1.5°C by 2040. Vulnerable areas like South Asia face heightened risks, urging swift adaptation actions.
Three leading climate scientists have analyzed data from 10 global climate models, utilizing artificial intelligence (AI) to enhance accuracy. Their findings indicate that regional warming thresholds are likely to be reached sooner than previously estimated.
Published in Environmental Research Letters by IOP Publishing, the study projects that most land regions, as defined by the Intergovernmental Panel on Climate Change (IPCC), are likely to surpass the critical 1.5°C warming threshold by 2040 or earlier. Additionally, several regions are expected to exceed the 3.0°C threshold by 2060—significantly earlier than previous estimates suggested.
Regions including South Asia, the Mediterranean, Central Europe, and parts of sub-Saharan Africa are expected to reach these thresholds faster, compounding risks for vulnerable ecosystems and communities.
The research, conducted by Elizabeth Barnes, professor at Colorado State University, Noah Diffenbaugh, professor at Stanford University, and Sonia Seneviratne, professor at the ETH-Zurich, used a cutting-edge AI transfer-learning approach, which integrates knowledge from multiple climate models and observations to refine previous estimates and deliver more accurate regional predictions.
Key Findings
Using AI-based transfer learning, the researchers analyzed data from 10 different climate models to predict temperature increases and found:
- 34 regions are likely to exceed 1.5°C of warming by 2040.
- 31 of these 34 regions are expected to reach 2°C of warming by 2040.
- 26 of these 34 regions are projected to surpass 3°C of warming by 2060.
Elizabeth Barnes says: “Our research underscores the importance of incorporating innovative AI techniques like transfer learning into climate modeling to potentially improve and constrain regional forecasts and provide actionable insights for policymakers, scientists, and communities worldwide.”
Noah Diffenbaugh, co-author and professor at Stanford University, added: “It is important to focus not only on global temperature increases but also on specific changes happening in local and regional areas. By constraining when regional warming thresholds will be reached, we can more clearly anticipate the timing of specific impacts on society and ecosystems. The challenge is that regional climate change can be more uncertain, both because the climate system is inherently more noisy at smaller spatial scales and because processes in the atmosphere, ocean, and land surface create uncertainty about exactly how a given region will respond to global-scale warming.”
Reference: “Combining climate models and observations to predict the time remaining until regional warming thresholds are reached” by Elizabeth A Barnes, Noah S Diffenbaugh and Sonia I Seneviratne, 10 December 2024, Environmental Research Letters.
DOI: 10.1088/1748-9326/ad91ca
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.
31 Comments
Dare we trust the dire predictions of those who have made so many failed dire predictions in the past? Especially when their only solutions for the “problem” is for the world’s citizens to surrender all freedom and money to support the cult leaders in the style they’ve become accustomed.
When it comes to computers, the maxim is GIGO (garbage in, garbage out), and AI is nothing but faster computers, especially when the global warming cult controls all input.
Has AI made a failed dire prediction? I thing the ippc has pretty much been accurate in its predictions about the range of heating probabilities under various scenarios for whatever emissions end up being. Sea levels continue to rise in keeping with predictions.
Over a century ago when it was not such a hot button issue scientists were demonstrating the effects of adding CO2 to the atmosphere. Now we’re doing the experiment. Add CO2 to the atmosphere and lo and behold temps go up. Just like those guys said long ago.
“I thing the ippc has pretty much been accurate in its predictions about the range of heating probabilities under various scenarios for whatever emissions end up being.”
Perhaps in a qualitative sense, the IPCC has gotten the trend (positive) correct. However, the models are acknowledged to run warm, with CMIP6 ensembles even worse than past predictions for a given scenario. It is like a bank telling you that the interest on your savings will grow over time, but they can’t or won’t tell you how much.
When (or IF) climate models can perform as well as a tide table then we can trust the claims. As it is, climate predictions are pretty much arm waving with qualifiers like “could” or “might,” with no assigned probability and uncertainty range. The quantitative predictions have all been failures. James Hansen’s 1988 prediction, which many call “accurate,” used tricks like assuming hypothetical volcanic eruptions to lower warming for the CO2-reduced scenarios, but didn’t apply the same cooling to the “Business as Usual” scenario. A simple linear extrapolation of Hansen’s own historical data has been more accurate than his model.
https://wattsupwiththat.com/2018/06/30/analysis-of-james-hansens-1988-prediction-of-global-temperatures-for-the-last-30-years/
“Add CO2 to the atmosphere and lo and behold temps go up.”
And drownings go up when ice cream sales go up, thereby proving that ice cream sales cause drownings (Or maybe that people go swimming too soon after eating ice cream, or whatever unsupported, fanciful speculation one can think of.)
“And drownings go up when ice cream sales go up, “. Really, or are you raising a furphy? May I humbly suggest that drownings increase when more people than usual go swimming and boating as the weather gets hotter and drier under the local heatwaves caused by Anthropogenic Global Heating; which would also increase the number of ice-creams sold. And thus skin cancer would increasingly afflict the unwise, and thus ice cream causes skin cancer. Are there micro-plastics in ice-cream that might trigger bowel cancer? Food for thought.
Congratulations. You have just discovered an important principle in statistics, all by yourself! Maybe you will be more inclined to believe it since YOU ‘discovered’ it. It is called “Spurious Correlation.” That is to say, just because a correlation can be demonstrated, particularly in time-series, it doesn’t mean that there is any cause-and-effect relationship. One of the variables must be a reasonable causative factor, and the relationship must work for both increases AND decreases in the independent variable. So far, recent increases in temperature have ONLY been observed associated with INCREASES in CO2, but NOT decreases. However, the Law Dome ice core strongly suggests that, for the last 800,000 years, changes in CO2 have been preceded by temperature changes, supporting the idea that biology is driven by temperature and sunlight and CO2 is a dependent variable, not the independent variable.
Thank you for your contribution to science.
[P.S. The real furphy is the belief that swimming and boating was invented as a response to the weather getting “hotter and drier under the local heatwaves caused by Anthropogenic Global Heating.”
Weather changes have been the primary driving force for human expansion across the planet Earth throughout history.
If this prediction of a 3³C worldwide temperature increase proves to be prescient, it will drive human migrations from places like Africa, India and Southeast Asia to cooler climes such as Siberia, Alaska and Canada.
Has anyone thought about what happens when this inevitably occurs?
It is central to the concerns about the consequences of global warming. Many people have published on the topic of ‘environmental refugees.’
Then why do we hear so little about this and all the dislocations it will cause?
It depends on what you are reading.
Alas that your sarcasm belittles your scientific credibility; or you simply lack a sense of humour.
Yes; rapine, pillage and murder, commonly known as war and self-defence of national integrity.
“One recent approach is to train a neural network on historical and future climate model simulations, and then use historical observations as input to the trained network to make out-of-sample predictions based on the current state of the climate system.”
They are apparently predicting the time to various average regional temperature thresholds, based on the unstated assumption that the models are correct when it is generally acknowledged that all but the Russian model run warm. There is no estimate for the probability that any model run is correct. The other unstated and unverified assumption is that the multi-model ensembles regress around the mean of the various model runs. There can only be one best model, and that appears to be the Russian model. If all the models running warmer are averaged together, the ensemble mean is going to be too warm. However, no one dares to acknowledge that other models are degrading the prediction by also using the warm models.
“Third, even for a robust regional forced response, internal climate variability creates uncertainty in regional warming that is generally greater than the uncertainty in global-scale warming … As a result, even IF the future climate forcing pathway were known, there is substantial uncertainty in the time until a given regional warming threshold will be reached [Or even IF it will be reached!] , with the magnitude of that uncertainty varying by region…”
Additional unstated assumptions are that the only significant influence on warming is anthropogenic greenhouses gases, which some retired researchers (Whose career paths will not be negatively impacted by not toeing the line.) have called into question, and that there aren’t any negative feedback loops that might kick in (such as a major volcanic eruption), and thus extrapolation of of present conditions WILL result in the regional (and global) thresholds eventually being exceeded. Probably the biggest risk to ANY prediction is a ‘Black Swan’ event that wasn’t even considered. If predicting the future was easy, there would be more wealthy people in the world.
I suspect these researchers are either anticipating retiring before 2040, or hope that as they produce new models, people will forget about what they said in 2024.
Of course the Russian model underestimates warming. All they’ve got to sell is oil and destruction. They’d like to melt the Arctic. They operate troll farms.
Underestimate? If you call doing the best job of matching reality to be an underestimate, then it is understandable why the models do such a poor job.
You make some good points. Models are models and climate is complicated. However, when I ride my motorcycle I appreciate that some idiot motorist on four or more wheels (truck and bus drivers are generally more aware of the road) could and indeed might take off one of my legs, so I model my riding to try to take into account the one idiot whom I might encounter among the several thousand who drive 4+\-10 wheeled vehicles.The Precautionary Principle; why should it not apply to the issue of Anthropogenic Global Heating? Whether the modellings predict a 2 degree C rise tomorrow or in 200 years, the odds are that we need to take adequate precautions and the sooner the better for civilisation. Bad luck for our wasteful standard of living, but……….?
A “nuclear winter” would cool the place off better the most volcanoes. I suppose that could qualify as negative feed-back loop caused by AGH.
There is a difference between the oft-referenced Precautionary Principle (PP) and the way people behave in the real world. The PP is predicated on an event with low-probability and high-potential consequences where the costs to prevent it are minimal. That means if the judgement was wrong, there was little committed to the unneeded effort.
Considering your motorcycle example, the odds of being in an accident are low but finite, the consequences potentially deadly, while the personal cost to avoid it is just heightened caution. PP makes sense.
However, being hit by a meteorite large enough to penetrate a house roof is very low but finite. I think the statistics are that one human and a couple cows have been prematurely sent to Cow Heaven (there is an issue about being moderated if I use a 6-letter word starting with ‘k.’) Yet, people don’t armor the roofs of their homes because that would be very expensive. They don’t wear helmets while walking around despite the possibility of saving their life and it being a minimal cost. That is, people in the real world don’t spend their waking hours worrying about low probability events, even if there are extreme consequences.
Chauncy Starr demonstrated that people tolerate risks proportional to the perceived benefit. That is, many people are injured, or worse, engaging in high-risk sports, but they willingly choose to do so. Similarly, people accept high accident rates and their consequences from driving vehicles (Even motorcycles!) because of their perceived utility.
In the case of anthropogenic warming, the votes haven’t been counted yet. There is a risk that burning fossil fuels ‘could’ result in high-consequence results, but there is no agreement on just what the probability is. Those who read the NY Times exclusively are probably convinced that we have lit the fuse to a very large cache of dynamite atop the San Andreas Fault. Those who depend on FOX News for their information are almost certainly less sure of the consequences. Most ‘skeptics’ consider the probability very low. Now, here is the difference in the application of the PP: Having a planned economy that takes us off dependence on fossil fuels, without assurance it is needed, and doing so with immature technologies (No wine before its time!) risks damaging the world economy that many consider to be more probable than the “could do irrecoverable environmental damage” protestors. So, the ‘climate’ problem is different that the classic application of PP because the costs are very high to address the concerns with an approach that is untried and of uncertain efficacy.
If the increase in CO2 is NOT driven by fossil fuels, and/or CO2 is not responsible for the temperature increases, we run an indeterminate but real risk of destroying the world economy. We aren’t even hedging our bets because we commonly destroy power plants that are prematurely decommissioned. It is a classic case of burning one’s bridges so that retreat is impossible. That is not a rational plan!
The spirit of the PP is that when convenient, people CHOOSE (not be forced to) to walk or ride a bicycle instead of flying or driving somewhere. People CHOOSE a lifestyle with a small carbon footprint, not demand a planned economy for everyone, run by bureaucrats who have never heard of “Spurious Correlations.”
“the ‘climate’ problem is different that the classic application of PP because the costs are very high to address the concerns with an approach that is untried and of uncertain efficacy”.
That may be the case, but the stupidity of H sapiens, a tribal animal now with weapons capable of causing more than the minor mayhem such as occurred between 1939-1945, is the threat . Global warming would fine if we had the population and muskets of 1750, but we don’t, and historically when the going gets tough, the tough get going. Hence I regard the risk associated with Anthropogenic Global Heating to be very serious as such a heating will entail societal stress and consequent large-scale movement of populations as heating continues. The ultimate cost may well exceed the cost of any purported remediation of the problem. Quite when the frog will find its feet getting cooked is the subject of numerous models, but the odds are that the frog will find itself in warmer water, despite access to ice-cream and sun-tan lotion, at some time in the next hundred years if things continue as they are.
As we all prefer to use increasing amounts of energy for our increasing ease, comfort and amusement then our grandchildren are in for an increasingly uncomfortable world. That’s my opinion; yours may differ and the proof, as they say, will be in the pudding. I won’t be around to see it, which is a pity, as it is an interesting argument involving both science and human behaviour.
Personally, I think that the risk of collapsing civilization from modern (particularly nuclear or biological) warfare is much greater than the potential risks of environmental damage. And, we don’t need the pressure of environmental damage to create refugees. Humankind has a long history of warfare, before and even after muskets were phased out. Resources to support increasing population and technology have often been at the core of wars, (How many people are aware of the role the silver mines of Athens played in the loss of the Peloponnesian War, or the role tungsten played in the Korean War?) Wars can result for other reasons, but resources are never very far out of the picture.
While well-meaning (but misguided) people are protesting uncertain climate changes over the next century, almost no attention is being paid to the risk of a nuclear exchange involving Russia, which would be a war that would be over in a matter of hours. The Doomsday Clock is closer to midnight than it has ever been, and people are gluing themselves to pavement because someone has said the next generation of children won’t know what snow is. (BTW, there was a news article today showing snow in the Sahara.) People have got their priorities mixed up.
Richard Feynmen famously said that if predictions do not match observations then the predictions are wrong, it doesn’t matter who makes them.
Purely on that statements validity this article and all others using the current highly inaccurate climate models should be returned to to the authors with a note saying do research and stop guessing.
A look at the consensus papers of Cook etc, with anything like critical thinking shows massive flaws. Data that can be used to prove or disprove in equal measure proves nothing except that those who “peer reviewed” it know nothing about science, statistics or critical thinking. That Cook etc says 60%+ of research that attributes no cause to climate change agrees with his idea is just adding 60%+ inaccuracy to his claims.
That plant biologists, farmers, glasshouse manufactures and others know that more CO2 is better for the majority of plants but climate scientists says lots.
That those in the insulation industry and physicists know about insulation “saturation” yet climate scientists don’t how this applies to CO2 isn’t credible.
That beer and wine drinkers around the world know that as their liquids warm they lose CO2 while climate scientists claim that warming oceans, rivers and lakes absorb CO2 contradicts real world observations, and why is Henry’s Law not applied anywhere in climate “science”?
Why are we told that sea level rise is increasing when real life measurement says the opposite?
Why are we told about Pacific islands getting drowned when research that measures them says they are growing in area?
Research has shown that water vapour and clouds have 85% and 35% more effect on global temperatures than CO2, why isn’t that published?
Why is it that our whole advanced civilization has blossomed due to warmer temperatures is not mentioned?
Why is that everything we have is courtesy of fossil fuels is ignored?
That climate science is based on physics and chemistry and yet experts on these topics are mocked and ignored is very telling.
That biology, ecology, geology and many other credible sciences have data that is relevant to the climate discussion but are ignored is stupid and actively discrediting all science.
That wind and solar can’t replace fossil fuels is obvious yet you keep publishing articles that proclaim it can does no good to your reputation.
Shame on you for publishing this climate change drivel
Scientists say oceans will absorb less CO2 as the climate warms. That will mean more in the atmosphere and more warming. It’s not obvious at all that wind and solar can’t replace fossil. It’s happening right now. Current breakthroughs in storage tech are even taking down natural gas. Line by line your assertions are taken down. Doesn’t matter, you or someone else will repeat them as infinitum.
However, it is obvious that the seasonal changes in atmospheric CO2 is driven by biology and that the steady human contribution is only about 4% of the annual CO2 flux. That is probably why when the COVID shutdowns in 2020 resulted in anthropogenic declines of as much as 14-18% in April alone, there was no perceptible change in the Fall-Spring ramp-up phase, either in the slope of the graph or the peak in May. The curve of the CO2 ramp-up for 2019-2020 is indistinguishable from the previous and subsequent years.
https://wattsupwiththat.com/2021/06/11/contribution-of-anthropogenic-co2-emissions-to-changes-in-atmospheric-concentrations/
https://wattsupwiththat.com/2022/03/22/anthropogenic-co2-and-the-expected-results-from-eliminating-it/
I observe daily that the Earth is flat. Some Ancient Greek chap predicted that the Earth should be a sphere based merely on some local geometrical calculations, and assumed such local measurements applied to all the parts of the Earth that he had not visited. Some other folk observed that the sun goes round the Earth; some bloke predicted that it shouldn’t and constructed a theoretical system in which the Earth went round the sun.
Ah well; we know better now.
At least we think we know better now. The problem is the zealots who ‘know’ what the unchanging truth is and see skepticism as a weakness instead of the core advantage of science over reactionary trial and error.
We are at 1.6 C now. If you wait 20 years to confirm that and we reach 3 C humanity and most of life on Earth will be on the road to extinction. 3 C ignites a “methane termination event” that will raise Earth’s temperature by 10 C. Look up Permian Extinction.
However, it is obvious that the seasonal changes in atmospheric CO2 is driven by biology and that the steady human contribution is only about 4% of the annual CO2 flux. That is probably why when the COVID shutdowns in 2020 resulted in anthropogenic declines of as much as 14-18% in April alone, there was no perceptible change in the Fall-Spring ramp-up phase, either in the slope of the graph or the peak in May. The curve of the CO2 ramp-up for 2019-2020 is indistinguishable from the previous and subsequent years.
https://wattsupwiththat.com/2021/06/11/contribution-of-anthropogenic-co2-emissions-to-changes-in-atmospheric-concentrations/
https://wattsupwiththat.com/2022/03/22/anthropogenic-co2-and-the-expected-results-from-eliminating-it/
You have made a couple of unsupported and improbable claims.
As to the Permian Extinction, there have been numerous hypotheses on that, none of which seem to be the final answer to the question. One of the newer ones suggests that it was the cooking off of volatiles in coal beds that was responsible for the Great Dying.
https://scitechdaily.com/coal-burning-in-siberia-250-million-years-ago-led-to-climate-change-caused-the-earths-most-severe-extinction-event/
https://scitechdaily.com/mass-extinction-traced-to-ozone-depletion-fossil-pollen-sunscreen-evidence-emerges/
https://scitechdaily.com/the-great-dying-rapid-warming-contributed-to-abrupt-collapse-of-forest-mire-ecosystems/
The Paleocene-Eocene Thermal Maximum (PETM) was a rapid warming of about 10 deg C that was also a time of great biological diversity and rapid evolution of mammals, not the least of which was primates that led to humans.
@Dan wrote “If you wait 20 years to confirm that and we reach 3 C humanity and most of life on Earth will be on the road to extinction.”
—
More likely we will just move underground or emigrate to orbital space homes or other bodies in the solar system, even if we have to live under domes.
It looks like you have been reading ‘science’ fantasy.
When I was in the army, I spent a month in what was then known as Point Barrow (I can’t pronounce the new name). We had a local Inuit(?) driver driving a tracked vehicle called a weasel, to take us out onto the ice. I had plenty of time to talk with the driver. He had been in the National Guard and traveled to the ‘Lower 48.’ However, he missed the tundra and ice and came back to Barrow. I experienced something similar the two times I visited New Zealand. NZ is beautiful, but it just didn’t feel like ‘home.’ People get imprinted with the things they grew up with and the first colonizers on Mars will probably get homesick. That is, they will deal with depression.
That is assuming that we can solve the problem of getting lots of people on the planet without them receiving overdoses of radiation, particularly if there are coronal mass ejections from the sun during the trip.
Please, learn what’s going on by also observing other things like the Sun, growing anomalies in Earth’s magnetic field etc. All materials can be freely used for republishing: http://www.book198.online/docs/Catastrophic-Consequences-Geomagnetic-Polarity-Reversal-Can-Be-Prevented
This article has received a little wider exposure with a citation for “Read More” at:
https://wattsupwiththat.com/2024/12/31/claim-artificial-intelligence-can-improve-climate-models/