Guest Post by Willis Eschenbach
Ever wondered how a tiny rise in sea level—let’s say, the aquatic equivalent of topping up your birdbath—might be responsible for frigid winters rampaging through East Asia? No? Well, Nature Communications thinks you should, in an article yclept “Intensification of extreme cold events in East Asia in response to global mean sea-level rise“.
Here’s their premise, as bold as a flashing warning sign on your grandma’s thermostat: A global mean sea-level rise of 15–30 cm can intensify extreme cold events in East Asia, thanks to a retooled atmospheric circulatory engine, weakening westerlies, and powered-up blocking episodes.
Mind you, this is not about that old chestnut, Arctic amplification or vanishing sea ice; it’s the earth-modeling equivalent of arguing that rearranging the living room somehow makes next winter’s snow deeper.
So, how did we get here?
I did what any skeptical denizen of the scientific peanut gallery would do: pondered their use of the NorESM1-F Earth System Model, the chosen oracle of this saga. Picture it—a whopping 2200-year sensitivity experiment, climate dials set to “controlled conditions” (CO₂ steady at 400 ppm, as if the atmosphere were on cruise control). The researchers periodically poured another cupful into the global ocean, then waited to see if the weather started shivering in East Asia.
Their definition of “cold extreme”? Days below the 10th percentile surface air temperature. Their measure of blocking? A reversal index for 500 hPa geopotential height (if that sounds like NASA-jargon, it is—think traffic jams in the jet stream).
Transparency, you say? Sure, you can sift through their supplementary files and simulated archive in the EU’s Zenodo repository—model outputs in glorious bulk if you’re inclined to lose weeks in NetCDF purgatory.
But empirical data? Not a sausage. Not one single lonely observation. No time series to ground these flights of model fancy.
Let’s hit pause for a moment and return to ugly reality. Does historical sea-level rise correlate with nasty winter chills in East Asia?
Not per the record: since 1900, global sea level crept up around 20 cm, but there’s no independent demonstration that this correlated with actual cold extremes in the region. In fact, most climate analyses emphasize local, not global, fingerprints.
So here’s my summary of my critique of the study, in three acts:
1. Single Model Dependency
The findings are entirely shackled to NorESM1-F—a ponderous climate engine whose output is only as good as its assumptions. There’s no check against real-world circulation, no multi-model brawl for robustness. It’s as if we rated steak quality by asking only one vegetarian chef.
2. Plausibility—Or The Lack Thereof
A the end of the last glaciation, sea levels rose by on the order of 120 meters (12,000 cm). Is it physically plausible that nudging the average sea level upward by a mere twenty centimeters will scramble global-scale atmospheric machinery? As someone who enjoys a good weather metaphor, I say: that’s like expecting the Second Law of Thermodynamics to take requests from the audience.
And then there’s blocking dynamics. In the real world? Much larger effects emerge from Arctic warming and sea ice loss than from sea-level twitchiness. Moreover, the study does not exclude confounders like SST changes, sea ice variability, or those slow-moving decadal ocean cycles that usually run this show.
3. Where Are the Comparisons To Reality?
A single climate model struts across the stage, unopposed. No ERA-Interim, no NCEP/NCAR, no HadCRUT temperature data, no observational counterpoint.
Let’s talk logic. The paper’s “nonlinear” threshold—cold events only spring loose above a 0.625m sea-level scenario—seems arbitrary, and it’s totally unsupported by observational reality. Real-world cold spikes in East Asia have been diagnosed as the products of Arctic heat, sea ice droughts, and the peculiar ballet of oceanic cycles, not bathtub-level adjustments.
No primary source—none—shows that sub-meter global mean sea-level rise is a puppetmaster for blocking frequency or icy East Asian blasts. All such claims hitch their wagon to untested numerical projections, idealized boundary conditions, and a climate model’s fever dreams.
The logic that a shallow sea-level rise reorganizes planetary wind and weather systems?
Dramatic pause …
No. Just no.
So what does this mean?
If you’re poised to rewrite winter preparedness handbooks based on sea-level forecasts, put that pen down. The dominant factors in real-world cold extremes—Arctic temperature, sea ice, ocean temperatures—remain unaffected by the aquatic inchworm pace of mean sea-level rise.
A final thought. Sea level has gone up about 20 cm in the last 125 years. If the cold weather only starts with an additional 62.5 cm of sea level rise, that means it will begin in … divide by pi, carry the 3 …
…
… the year 2415.
Be very afraid …
As always in science, I’m open to correction—if someone drags an empirical time series from the basement that links 20 cm of GMSL with bitter Februarys in Beijing, I’ll eat my galoshes.
But for now? File this under “Climate Mythmaking: The Model Show.”
w.
You Know The Drill: When you comment, quote the exact words you’re discussing. I can’t defend your interpretation of what I write.
We – the human race – are so fortunate to have had scientific giants in the past who used their brains, conducted experiments, analyzed results, formed hypotheses, and then scrapped everything when there was no predictive capability or when observations falsified theories.
Nowadays not so much…what a load of carp we are subjected to in every field of scientific endeavor! Oh there may be a few pearls hidden deep in the manure piles. I give thanks for the few willing to look, who seek the truth.
NCL,
In the history of the advance of proper Science, some of the important progress was made with simple apparatus and simple thought processes but producing major significance.
I liken it loosely to music. The skill, the technical achievement, the resulting mastery of music by giants like Beethoven, Mozart, Brahms and others has been almost lost, replaced by neo-noise with amplifiers, synthesizers and repetitive drum beats drowning out melody.
Something, I know not what, has caused dramatic changes in the ways that society judges excellence. Geoff S
Einstein was very partial to Thought Experiments. Had supercomputers been extant during his time, I suspect he still would have preferred his Thought Experiments over computer models that no one person will live long enough to validate the predictions.
You’re just old fashioned in your musical tastes! Well, so am I. Especially when it comes to rap, but theirs are also other genres that I despise. Same goes for many other art forms.
As observed when Charles differently tore this new dreck apart, the saddest thing is that this new ‘study’ somehow made it into the Nature stable of ‘scientific’ journals. A complete violation of what they very publicly proclaim is their mission.
I am wondering why the ‘reviewers’ did not make the same observations and ask the same questions as Willis has?
‘Peer’ or ‘pal’ review?
Great news, they’ve cracked the formula for the butterfly effect.
Well, that’s nothing. I change my HVAC filter every April and November and it seems to provoke a huge if gradual change in the climate right across North America.
So you’re the culprit responsible for climate change, are you? Swine!
I don’t understand why a journal would accept a “study” based entirely on computer gaming. This study shows no verifications that the climate model used has any predictive power whatsoever!
The reviewers were likely climate modelers as well.
Seems they all played “World of Climate-Con” once to often.
(I wonder if they all went to this year’s convention in Brazil?)
“The findings are entirely shackled to NorESM1-F—a ponderous climate engine whose output is only as good as its assumptions.”
I am just stunned at the (mis)use of such a model for investigating a weather pattern response to sea level rise. The modeled atmosphere sits on a modeled ocean surface that is fixed in vertical position in the coordinate reference system, ignoring the ocean tides altogether for computational simplicity.
“Not even wrong” might be a good way to describe this approach, in my opinion.
I asked Grok “Fast” via X:
“I am interested in the sea surface datum used in the NorESM1-F Earth System Model for the atmosphere component of the model. Does the modeled atmosphere sit on a sea surface fixed in vertical position in the coordinate system, or does the sea surface position change vertically at each gridded location with a simulated ocean tide cycle, so as to reposition the atmosphere vertically at each time step at each grid location?”
First part of the answer from Grok:
“In the NorESM1-F Earth System Model, the atmosphere component (CAM4-Oslo) uses a hybrid sigma-pressure vertical coordinate system, where the lower levels are terrain-following (sigma coordinates) and transition to pure pressure coordinates aloft. The sea surface datum—the reference lower boundary for the atmosphere over open ocean—is effectively fixed in the model’s global vertical coordinate system. This datum corresponds to a prescribed mean sea level, which does not vary temporally or spatially in a way that dynamically repositions the atmospheric lower boundary.”
Once again – thanks Willis for an educational and entertaining post. I learned a new word, yclept – had to look it up. And, just like you, I immediately “..did what any skeptical denizen of the scientific peanut gallery would do: pondered their use of the NorESM1-F Earth System Model, the chosen oracle of this saga”. Well, that’s a lie – I don’t begin to have the chops to figure this stuff out, but I am certainly happy that I, along with others here at WUWT, have you and other contributors who do have the chops to figure it out and explain it in a way that even those like me can understand.
To misquote American philosopher Lina Lamont, WIllis is a shimmering, glowing star in the climate firmament.
Thank you for diving into the deep, deep manure pile of obfuscating blather and “supporting” supplements to arrive at the conclusion that’s become obvious about every single pseudo-science “study” making hyperbolic claims about global warming, er, “climate change.” All of them use made-up models. None of them support their claims with actual measured evidence. And that there, folks, is the emperor without clothes, the man behind the curtain, the dog that won’t hunt, the biggest hoax in history as Donald Trump succinctly put it. Models. Virtual reality, not real reality.
Just wanted to correct your calculation using the level of accuracy typically used by the climate change “scientists,” we’ll hit 62.5 cm on July 5th, 2416 (from today).
/s
bo:
Just for planning purposes, July 5, 2416 will be a Tuesday [according to Copilot, Gemini and
Grok ( though to be fair, Grok first said “Thursday” after I already had the first 2 answers. I then asked it to “think harder” —> Tues ) ] Lol
Willis is (was?) a big fan of AI. Even he probably knows the difference between Thursday and Tuesday. I hope the AI he paid for does, but of course it probably informs him that he needs to check its answer, in case it’s wrong.
Small print at bottom of Google’s AI response –
Just for fun, ask “how do i know if google ai response contains mistakes”. Bear in mind, “AI responses may include mistakes”.
Damned if you believe it, damned if you don’t. Assists the ignorant and gullible in their desire to avoid thinking for themselves.
Willis, you wrote –
I do hope you are not implying that melting sea ice affects sea levels.
If you are, I’ll just point out that the ancient Greek gentleman, Archimedes knows more about Archimedes’ Principle than either you or the National Science Foundation, as Anthony Watts pointed out a few years ago.
Adding CO2 to air does not make thermometers hotter, or create “Arctic warming” or any similar fantastic claims.
Sorry about that, but you are simply ignorant and gullible if you believe otherwise.
MIchael, I neither said that nor implied that. I was talking about effects on the temperature, which everyone but you seems to have understood.
And let me suggest that you dial back on the aggro. It makes you look like a ten-year-old having a tantrum.
w.
Willis, let me suggest that you learn Archimedes’ Principle. You say “MIchael, I neither said that nor implied that”, but as usual you refuse to quote the “that” to which you refer! You certainly did say “Much larger effects emerge from Arctic warming and sea ice loss than from sea-level twitchiness.”, which is complete nonsense .
Are you the same Willis who preaches –
You also petulantly whine –
Feel free to suggest anything you like. I assume you have some reason for thinking I value your suggestions, and I don’t blame you for keeping it to yourself. I understand you feel embarrassed about your lack of physical knowledge being pointed out – but I don’t understand why you’re blaming me for your ignorance and gullibility.
Finally, you say –
Except when you’re wrong?
It does seem you are the only reader that interpreted “sea ice loss” to mean sea-level increase. That was never typed or implied. The article and response are regarding the effects of sea ice loss on the temperature of East Asia.
As the Archidemes’ Principle has nothing to do with his post, there is no ignorance or gullibility on Willis’ part. He is not obligated to defend your interpretation of what he wrote.
I asked Grok to look up the data for me (it does in seconds what takes me days). It said “These proxies indicate sea levels were 1–2 m higher than today at the HCO (Holocene Optimum) peak [..] In China, a multi-proxy (pollen, chironomid, and ostracod) reconstruction indicates cold-month (winter) mean temperatures were ∼3°C warmer than modern (1961–1990 baseline) during 8–6 ka. Annual mean temperatures were 1–2°C warmer than pre-industrial levels“.
So, as sea level fell, temperatures also fell, especially in winter. It seems unlikely that a rising sea level will have a dramatic downward effect on winter temperatures in Asia. I asked Grok to comment and it said: “September 2025 Nature Communications paper titled “Intensification of extreme cold events in East Asia in response to global mean sea-level rise.“ [] relies on climate model simulations rather than direct empirical observations or proxies []
Turning to instrumental observations and historical proxies, the available data do not support the idea that rising sea levels are likely to drive cold extremes in Asia. Instead, they reveal a consistent empirical pattern: periods of higher sea levels (and associated global ice melt) correlate with warmer winter conditions in East Asia, while falling sea levels align with cooler winters. This holds across both the Holocene record you referenced and recent decades of observed sea-level rise. []
In summary, empirical data—both proxy and instrumental—indicate the opposite of the argued mechanism: rising sea levels, via ice melt and thermal expansion, empirically associate with warmer (not colder) Asian winters over millennia and decades. Occasional cold extremes reflect atmospheric variability, not a sea-level-driven trend. The claim thus seems unlikely on observational grounds.”. [my bold]
Video games. These dudes are not scientists. They are gamers reporting on the results of last weekends tournament.
cut off their funding.
Junk science. Those are the words we were looking for.
On a specific point. Not one climate ‘model’, NorESM included, produces the ENSO sloshing of the Pacific as an emergent phenomenon. Since that oscillation dominates the weather worldwide, its absence should make any researcher, however mediocre, ask the question why we would take anything coming out of those models serious.