Essay by Eric Worrall
According to Technical University Munich, poor quality climate data and oversimplified model assumptions make it impossible to pinpoint significant future climate events.
Not the day after tomorrow: Why we can’t predict the timing of climate tipping points
by Technical University Munich
AUGUST 2, 2024A study published in Science Advances reveals that uncertainties are currently too large to accurately predict exact tipping times for critical Earth system components like the Atlantic Meridional Overturning Circulation (AMOC), polar ice sheets, or tropical rainforests.
…
First, predictions rely on assumptions regarding the underlying physical mechanisms, as well as regarding future human actions to extrapolate past data into the future. These assumptions can be overly simplistic and lead to significant errors.
Second, long-term, direct observations of the climate system are rare and the Earth system components in question may not be suitably represented by the data. Third, historical climate data is incomplete.
…
To illustrate their findings, the authors examined the AMOC, a crucial ocean current system. Previous predictions from historical data suggested a collapse could occur between 2025 and 2095. However, the new study revealed that the uncertainties are so large that these predictions are not reliable.
Read more: https://phys.org/news/2024-08-day-tomorrow-climate.html
…
The study referenced by the article;
Uncertainties too large to predict tipping times of major Earth system components from historical data
MAYA BEN-YAMI , ANDREAS MORR, SEBASTIAN BATHIANY, AND NIKLAS BOERS
SCIENCE ADVANCES
2 Aug 2024
Vol 10, Issue 31Abstract
One way to warn of forthcoming critical transitions in Earth system components is using observations to detect declining system stability. It has also been suggested to extrapolate such stability changes into the future and predict tipping times. Here, we argue that the involved uncertainties are too high to robustly predict tipping times. We raise concerns regarding (i) the modeling assumptions underlying any extrapolation of historical results into the future, (ii) the representativeness of individual Earth system component time series, and (iii) the impact of uncertainties and preprocessing of used observational datasets, with focus on nonstationary observational coverage and gap filling. We explore these uncertainties in general and specifically for the example of the Atlantic Meridional Overturning Circulation. We argue that even under the assumption that a given Earth system component has an approaching tipping point, the uncertainties are too large to reliably estimate tipping times by extrapolating historical information.
Read more: https://www.science.org/doi/10.1126/sciadv.adl4841
One thing which really caught my eye is how sensitive the models are to changes to model inputs or data processing.
… The most fundamental assumption made in all these methods is that the system in question can undergo tipping for a given forcing. However, not all systems can undergo tipping, and as the methods assume tipping, they are susceptible to false positives. We now apply the three methods to time series generated by a linear model without any bifurcation but with an added mean trend, forced with red noise that increases in correlation strength (see Materials and Methods). The first and third method above predict tipping for this system (Fig. 1). For such a linear system, the ideal method would give the tipping time as infinite (i.e., no tipping time). However, the MLE method always predicts a finite tipping time, and the AC(1) extrapolation only gives an infinite tipping time for about a quarter of the cases. The generalized least squares (GLS)–based regression method is designed to account for nonstationary correlated noise, and its results do not indicate a notable decrease in system stability. Despite that, it still gives a finite tipping time for about half of the cases. …
Read more: https://www.science.org/doi/10.1126/sciadv.adl4841
The scientists were careful to suggest the error might be that models underestimate the risk of tipping points.
But the models are clearly not fit for guiding public policy.
Trying to infer behaviour from models of a system which contains degrees of freedom nobody knows about, based on patchy and unreliable data over too short a period of time, models based on such noisy data that switching to a different statistical method leads to wildly different outcomes, that isn’t science, it is superstition. An Ouija board could match the quality of data provided by such models.
Update (EW): Added “of a system” to the last paragraph, to clarify that it is the system which contains degrees of freedom nobody knows about, not the model.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Perhaps it’s not possible to precisely predict random events, especially as it relates to complex dynamic, non-linear, chaotic systems.
Yes, their peak authority the IPCC stated clearly that future climate states were impossible to predict in our coupled, non-linear, chaotic climate system.
Where have these “scientists” been for the past 2 decades?
Clearly not following the IPCC’s regular missives.
Mathematically we cannot solve the n-body problem where n>2. Sun, clouds, land, oceans. Oops, n>2. Therefore we cannot solve the future climate state. Predicting climate is impossible using math in climate models.
Mathematically we can—and have—solved the n-body problem (of gravitational attraction) by application of perturbation theory and use of associated algorithms and high-speed computers. Using such, we have achieved precise, directed navigation of spacecraft across the width of the solar system.
I believe what you meant to say is that mathematically there is no known, closed-form solution to the n-body problem of gravitational attraction.
As for extending that classic n-body problem of astrodynamics to other fields, such as climate, I sayeth not.
Of course you can solve the problem by making course corrections in the future. Climate science adjusts the data all the time to make it align with their models, then claim the models are predicting the result. They aren’t. The models are predicting the adjustments.
Well, it’s certainly true that spacecraft can navigate more precisely by performing course corrections after they are launched on a specific trajectory, but I was referring more generally to humans solving the n-body problem BEFORE launch, which is actually what is done today (knowing that such will always involve some relatively minor amount of uncertainty). I daresay that spacecraft missions that involve multiple “swingby” passes of the Moon or other planets could not happen without having solved the n-body gravitational problem of the major masses involved (Sun, Earth, Moon, Jupiter, and Saturn at least) to a high degree of accuracy prior to launch.
Let’s look at another example of very accurate solutions of the solar system n-body gravitational “problem” that don’t involve “future course corrections: astronomers, using computer programs, very accurately predict the trajectories of asteroids and meteoroid clouds as they pass through the solar system. Same thing applies to comets (especially periodic comets like Halley’s) over many years into the future, although the predictions are less accurate due to the somewhat random reaction forces acting on those bodies as they sublimate/outgas along their trajectories.
Climate change™ “course corrections” are much more nondeterministic, unscientific and arbitrary than are spacecraft course corrections, as I’m sure you agree.
I agree with ferdberple that, “Predicting climate is impossible using math in climate models.” I think you are for the most part agreeing with him. I would go even further and say that trying to simulate nature, in the sense of predicting the future, is a wrong-headed notion. One of the few exceptions, as you pointed out, is the solar system dynamics problem, because it’s so simple.
The general N-body problem, as opposed to the solar system, is a fool’s errand even on a supercomputer. Without perturbation theory techniques numerical error blows up quickly due to the inverse-square law scaling problem in near collisions. Sverre Aarseth and company have, with fifty years of experience behind them, built the “N-Body” fortran program for simulating galactic stellar motions. It seems to be generally accepted that any one simulation run quickly diverges, requiring an ensemble of runs to be averaged. I think it may as well be a Monte Carlo model–but calling it “simulation” is a matter of opinion.
The idea of numerically integrating a set of first-order differential equations to simulate nature seems to be highly overrated. The first-order initial value problem is generally subject to numerical divergence given enough time steps, and it just gets worse the more degrees of freedom there are. No numerical technique–no Runga-Kutta or predictor-corrector method of any order completely eliminates the error. How many degrees of freedom are there, and how many time steps are there in a climate model run? It just doesn’t seem possible to get anything but noise out of such an effort.
For people who would invest time trying to simulate nature, I would ask them to try and simulate a real marble bouncing on a real table, and predict its outcome. I don’t have to tell you what it would take to turn it into a serious effort. Nature at large is much more complicated than a marble bouncing on a table. I can’t take seriously any claim to simulating nature with a computer, now or ever in the future. Reasonable people when confronting a palm reader understand that you can’t predict the future. Some of those same people seem to forget that when working on a computer model.
Come again?
So it isn’t “As worse than we thought!”
No, I think the real intent is being missed here. The climate catastrophists realize that rational people will eventually recognize that a slightly milder climate is never going to cause anything but a minor benefit.
Those in the US old enough to remember the elimination of AT&T’s monopoly on long distance phone service back in the mid-1980s will remember AT&T’s strategy to combat the upstart MCI. They relied on FUD — Fear, Uncertainty, and Doubt.
What if your calls don’t go through? What if their network isn’t reliable? Will the sound quality be good enough? Do you want to take that risk? Isn’t it worth it to be sure?
We’re now into the FUD arguments. Will the 150-yr record of slow linear sea level rise suddenly accelerate? Could the AMOC suddenly stop and plunge Europe into an Ice Age? Could pigs start flying? What if???
The Precautionary Principle personified! Fear of the unknown. Luddite’s operational protocol!
Not only the ATT break up but Pacific Bell being broken up into Baby Bells that ATT then bought out. Now it’s all ATT
According to The Day After Tomorrow, it is real.
*** BS alert **
The most important lesson in Climate Science:
Never make an unfounded assertion on a
timeline that expires before you do.
Bob Kutz 2018
Besides that maybe they should try for a
Seldon Crisis.
That’s because there are no real permanent climate tripping points. If there were, we’d either be in a perpetual Hot House or Ice House state…and we aren’t. We get warmer (+3°C higher than today without tipping) then we get colder (-10°C colder than today without tipping) then we get warmer again for the last 880,000 years. CO2 has been higher (3,500-7,000ppm without tipping) and it has been lower (180ppm without tipping). Milankovitch cycles dictate that the next glaciation in the 110,000 year cycle is on the horizon and approaching. We should be preparing for the new Ice Sheet build-up in the north and the migration that will ensue
Yeah, but in the long run we’re all dead. We should be preparing for the sun’s impending red giant phase?
I certainly agree that we have (well, probably our great-great-great grandchildren’s grandchildren) have to be prepared for an end to the benign and currently improving climate conditions when the earth reverts to a glaciation and we will lack adequate fossil fuels. There won’t ever be any climate harm from slight warming, but try growing crops on ice.
Or under ice
I was just outside grilling a couple of filet mignons, while drinking cerveza with lime in a chilled glass, and I came to the recognition that 94F at 20% relative humidity is not that hot.
This sounds like a tipping point to me.
Naw. You’re thinking of tipping places. They’re different.
Sounds like a tippling point to me 😉
Could easily become a tripping point your mother warned you about.
I hope you tipped one for us all.
A tipping point implies a time frame so the word “permanent” is not meaningful.
The current fear mongering primarily concerns the loss of Greenland and Antarctic ice. That would put a lot of humanity under water. But it would not be permanent because it has happened before and Earth recovers.
I predict the next tipping point will be obvious around 2400 and it will not be melting ice but permanent ice formation over high latitudes land in the NH. Basically the permafrost will be advancing south. That will gradually lead to oceans dropping at 40mm per year. A large proportion of habitable land now will be under permanent ice.
And when that happens, humans, being stupid, will keep building as the water recedes, totally forgetting that eventually it will come back. And the cycle starts over again…
Why will it be stupid to build and grow crops where we have access to the sea’s resources on much-needed, newly available land while other land is becoming uninhabitable?
What’s the useful lifetime of the typical house or building? A century? A tiny fraction may last longer but not many. That land will remain dry for a score of millennia. What sense would it make to declare it off-limits to construction?
There’s an opposite error to short-sightedness.
The green blob will be demanding that we burn coal and dump our EVs for ICE cars.
My eyes tend to glaze over with math, but from what I understand about chaotic relationships, small changes can produce wildly different results. So if one is unsure of the exact relationship between climate phenomena, and unsure of the data, making predictions by examining the entrails of an animal would seem as appropriate, and accurate.
But the western nations still have to spend multi trillion dollars preventing “it”.
And it is as effective as throwing virgins into volcanoes.
Such a waste of perfectly good virgins. Sad.
Funnily enough Tom, the IPCC agreed with your take on the matter.
As long as it isn’t beef. /s
I Predict: No tipping points will occur, ever, due to minute quantities of CO2 in the atmosphere.
I predict a tipping point if the atmospheric CO2 levels fall to about 30% of the present level. All plant life would struggle to survive and animal life would soon struggle without plants for nourishment.
We have to thank India and Chine for doing most of the heavy lifting with regard restoring the CO2 balance in the atmosphere. Plants have proven to be their own worst enemy by eating themselves into oblivion in past periods of Earths history.
“, the uncertainties are too large to reliably estimate tipping times by extrapolating historical information.”
Uh.. the holy grail of climate science.. being able to make reliable predictions.
Maybe they should focus on something simpler first and just try to fix the uncertainty for the role of anthropogenic CO2 for global warming, BEFORE trying to making dubious statements about the far more complex changes in climate.
Everybody can easily see, how successful climate scientists were with their key question over the last 45 years.. if I were them I too would not like to talk about global warming anymore:
from
https://cpo.noaa.gov/increasing-certainty-of-climate-sensitivity-in-models/
Move the y-axis up by 1K (so the zero is where the one is currently) and add the word Crap in front of “Model range of ECS” and it might even be close.
Climate “Science” is unique in that its practitioners actually know less about the climate today than they did in 1979.
“Why we can’t predict the timing of climate tipping points”
Cuz there are none. Short of the Denmark Strait plugging with ice…stopping the 2 km high underwater waterfall that kickstarts north Atlantic ocean currents…. Hard to predict…
Or the sudden opening of the Panama Isthmus, which would totally change the Gulf Stream, among other things.
I’ve wondered about that before. Would it have to be plugged totally (i.e. down to the sea bed, or close. Or would 2 or 3 meters of surface ice do it? Is there any proxy- or navigation data from say 1850-ish on this?
Once it starts to slow the current flow, ice would build to the bottom fairly quickly (a few decades) since flow of warm water back northwards would be restricted.
“An Ouija board could match the quality of data provided by such models.”
My bold. Careful with the word data. Climate equations work with numbers and output them, but they are not data in my universe.
Data from the perspective of scientific input into policy making, but point taken thanks.
Shouldn’t it be “a ouija board”, not “an”?
An wonder that you ponder that.
Well…the profits of doom predicting a tipping point are correct. According to live science.com “Even if an errant asteroid, nuclear holocaust or other disaster don’t doom Earth, in another 1.75 billion years the planet will travel out of the solar system’s habitable zone.” Ten years may be inside prediction uncertainties, but theoretically it is correct to predict that the earth will reach a tipping point. Man or other earth organisms may or may not have a 1-99% possibility of contributor to reaching that tipping point. Just wanted to go on the record with my prediction.
And the tipping point depends on the X line. A point can be a year… or a million.
Hansons tipping point: Manhattan will be underwater by yr 2000. Whoops.
‘Gore’s tipping point: The Arctic will be ice free by 2015. Whoops again.
Those who live by the tipping points should also professionally die by the tipping points.
I’ve said it before, and conditions on the ground have only proven my point more valid as the years advance, putting Manhattan under water is a goal greatly to be desired.
The very many measurements of weather and climate properties that I have seen over the last 30 years do not point to any tipping point. There is evidence of a change in 1975 in the Australian data that I study most, but it is a change, not a tipping point with evident danger.
So what are the past tipping points anywhere on the globe that these imaginative activists are using for models? Or are their fears based entirely on projected tipping points from their imaginations?
Geoff S
Im going to play devil’s advocate here: even though we cannot reliably calculate tipping points or even accurately pinpoint the timeframe of disaster we cannot rule out tipping points or that the onset of disaster might in fact come sooner.
So, as you see, this will be the defense and they will use the precautionary principle to keep pushing the climate measures. This also conveniently ignores certainty/ uncertainty. A win/win for the alarmists.
But what about the risk of a visit from hostile aliens? Shouldn’t we be spending trillions on a network of solar defence stations just in case? After all in a galaxy of a hundred billion stars, it’s a near certainty we will encounter the bad guys one day.
Or we could save our money for genuine problems whose imminence is backed by evidence.
“But what about the risk of a visit from hostile aliens?”
After my brilliant 1997 climate prediction: “The climate will get warmer, unless it gets colder” … in 2020 I predicted an invasion of aliens from the planet Uranus. Which was half right. There was an invasion of aliens in 2021, but they were from South America and Latin America, not Uranus
I soon began having a nightmare about space aliens landing in Washington DC, like in the movie The Day The Earth Stood Still. They demanded to see our leader, so we brought them to see Joe Biden. Thinking that was a trick — old dementia Joe, and cackling Kamala the word salad queen, could not possibly be our real leaders — so they destroyed Washington DC with a Uranus Death Ray. Then I woke up.
Some (25,000 or so) were from China which is fairly close to Uranus
“I soon began having a nightmare about space aliens landing in Washington DC,”
Have you been inserting illegal substances in Uranus again ???
Just probing
I don’t usually wake up until something bad happens.
There is one element that trumps all others: a cultural persistant meme that humans are damaging the planet and take it out of balance. Tied to overpopulation due to the use of hydrocarbons but ancient in myth. This underpins the whole focus on Co2 and the climate alarm. Without that it would never have become as big as everybody can see that the science doesnt really point to alarm and that uncertainty can never be equated away. The money involved in this meme supports its continuation. A part comes from those who are hungry for power and are involved in supra national organisations and NGOs. The other part comes from decisions made by politicians who are swayed by them and their feeling of guilt as portrait in the meme. And the msm media is fully on board.
Without the money and the guilt the house of cards collapses. You need both.The guilt also makes it hard to make a proper risk assessment ( Lomborg, Currie). No emotion involved. Feeling trumps logic ( i know, im married 20 years)🙂.
So, aliens, who cares? Asteroids? Nothing we can do. And no guilt involved.
The whole outcome the Climate Syndicate desires to achieve is the reduction of the population to less than 1B and a socialist economy ruled by the One World Order.
They have said so.
My climate tipping point will be when there is no longer enough snow in SE Michigan to need a snow shovel. Getting very close. Just 10 minutes of snow shoveling all of last winter. I will then retire my snow shovel, have it bronzed and display it on my living room wall.
To celebrate, we will break into our DOE subsidized Aluminum Carbon Capture Facility (a case of unopened beer cans in the garage) and celebrate until we are so drunk we tip over and fall off our bar stools.
That is the tipping point I dream of: When the climate in the Detroit suburbs is just like the climate in San Diego in the 1970s.
What does “the uncertainties” refer to in this context? The IPCC tosses the word about quite a lot in their tomes, but they clearly aren’t referring to real measurement uncertainty. But climate science in general has little if any understanding of the subject.
Anyone with basic statistics knowledge should understand that extrapolations from regression is thin ice at best, the fitting is only valid over the range of data.
Ah uncertainty! The future is certain, it’s the past that is so uncertain. How much will the 1930s cool in the next decade, for example?
There are a few things even about the future that remain uncertain. I mean, the exact timing of the catastrophic tipping points. We know that there are tipping points and we know that they will be horrifying, but will it come in a month or a decade?
Well, as the saying goes:’ preductions are hard, especially about the future’. An interesting line considering the issues with hindcasting in climate science..
You took some of what I was going to right out of my fingers!
Somehow mathameticians and statisticians have convinced climate scientists that they can remove uncertainty in physical measurements. The only conclusion one can make is that neither group has been held accountable for the results they arrive at.
If I design a bridge and find the separate uncertainties of the span and the pilings, I don’t reduce the combined uncertainty by two. Heck I don’t even combine the uncertainty using RSS, I simply add them! Why would I do that? If one can answer that, you might have an appreciation for uncertainty.
Extrapolating the future from a linear regression is a fools errand when components are known to be non-linear. One can only do that extrapolation if there are multiple, linear causes whose relationship remains linear, AND WELL CHARACTERIZED.
The earth has experienced many changes such as major extinction events, asteroid collisions, and orbital events and is still viable for life. With feedback relationships not well known, I sincerely doubt that tipping points can even be known, let alone when.
Extreme value analysis (EVA) versus root sum square (RSS) analysis.
Glad someone else understands.
There are only 2 certainties: Death and Taxes.
Everything else is intransience.
CO2 will heat the surface all things remaining equal. I’ve heard this repeated by many of the luke warm authors that post to this site.
Yet, all things never remain equal. The climate system has near infinite degrees of freedom. This means all things never stay equal. They are free to take any of the available pathways, and will tend to the lowest energy state for reasons we do not understand.
Climate forecasting is hubris. It is chicken bones and tea leaves. Mathematically impossible today and likely impossible forever.
But you can use broad probability heuristics. Like, at some ‘point’ there will come another ice age.
Using that we can also predict the climate will cool in the future. Milancovic and other cycles, but not exactly. Unless you insist Co2 is the climate forcer. But chances of that are close to zero considering all the factors in Earth’s atmosphere..
But you can use broad probability heuristics. Like, at some ‘point’ there will come another ice age.
=======
prediction requires both an event and a time. Predicting there will be some event without specifying a time is not a prediction because in time all things are possible.
Entropy is a b**ch!
Wouldn’t that be
Entropy is a ihctb
I forecast/predict that at some point today I will inhale and at some other point I will exhale and it is a certainty that I will not do both simultaneously to the femto second.
Any computer program that can predict the future climate could just as easily predict the stock market. There would be no climate scientists because they would have cashed in and been living on a tropical island. Missing winter not one bit.
Or the lottery.
“uncertainties are too high to robustly predict tipping times.”
Or even if there is such a thing outside of getting hit with an asteroid. Here is a tree stump on the NW Canadian Arctic coast that grew there 5,000 yrs ago. It is a White spruce, the same species that grows in the taiga forest which today lies over a hundred km to the south. A tree of this girth today would be over another 100km further to the south. This puts the global anomaly 5,000 yrs ago at 3-5 °C greater than today and guess what? … there was no tipping point in evidence! The forest simply retreated a few hundred km south.
So let me safely guarantee and spare all from worry (stop terrorizing the children) we are not going to have a climate tipping point until we slip into the next glacial maximum in a few millennia. But do look up! I can’t guarantee a bolide won’t crash into us!
Excellent! Ever wonder why this evidence has been memory holed by climate science?
Must be an Inconvenient Truth
“Why we can’t predict the timing of climate tipping points”. Easy, there aren’t any. Duh.
Story Tip.
Tipping point for electricity supply coming soon to Germany..
‘Handelsblatt’ Reports: Photovoltaics “Causing Problems Electricity Distribution Grids (notrickszone.com)
I can certainly believe that the collapse of Germany is Habeck’s favorite project, but did he admit that? 😀
“Why we can’t predict the timing of climate tipping points”
But the climb it ex-spurts ( Hanson, Gore, prince Charlie Chump) have been doing it for years & it’s normally just 10yrs away.
It’s sort of like why we can’t predict when the space aliens will arrive.
Or when commercial fusion power plants will go online.
Or when humans will return to the Moon.
Tipping time or tipping point; why should either be assumed if it cannot be observed in nature? Why would the two terms be synonyms?
We can’t predict weather more than about 3 days in advance, so why would anyone have the hubris to say they can predict climate 30 years to, say, 100 years into the future???
As for predicting climate tipping points, no one can predict what they cannot objectively define.
Bingo, they won’t/can’t/don’t define what constitutes a tipping point, yet spread FUD about them happening down the road..
Mark Twain wrote that the same people that confidently predict events 500 years in the future cannot successfully predict events next week.