Institute for Basic Science
A new study, published this week in the journal Nature Climate Change, shows that the recent intensification of the equatorial Pacific wind system, known as Walker Circulation, is unrelated to human influences and can be explained by natural processes. This result ends a long-standing debate on the drivers of an unprecedented atmospheric trend, which contributed to a three-fold acceleration of sea-level rise in the western tropical Pacific, as well as to the global warming hiatus.
Driven by the east-west sea surface temperature difference across the equatorial Pacific, the Walker circulation is one of the key features of the global atmospheric circulation. It is characterized by ascending motion over the Western Pacific and descending motion in the eastern equatorial Pacific. At the surface trade winds blow from east to west, causing upwelling of cold water along the equator. From the early 1990s to about 2013, this circulation has intensified dramatically, cooling the eastern equatorial Pacific and triggering shifts in global winds and rainfall (see Figure 1). These conditions further contributed to drying in California, exacerbating mega-drought conditions and impacting agriculture, water resources and wild fires. Given these widespread impacts on ecosystems and society, the recent Walker circulation trends have become subject of intense research.
In contrast to the observed strengthening, the majority of climate computer models simulates a gradual weakening of the Walker Circulation when forced by increasing greenhouse gas concentrations (see Figure 1). “The discrepancy between climate model projections and observed trends has led to speculations about the fidelity of the current generation of climate models and their representation of tropical climate processes”, said Eui-Seok Chung, researcher from the Center for Climate Physics, Institute for Basic Science, South Korea, and lead-author of the study.
To determine whether the observed changes in the tropical atmospheric circulation are due to natural climate processes or caused by human-induced climate change, scientists from South Korea, the United States and Germany came together to conduct one of the most comprehensive big-data analyses of recent atmospheric trends to date. “Using satellite data, improved surface observations and a large ensemble of climate model simulations, our results demonstrate that natural variability, rather than anthropogenic effects, were responsible for the recent strengthening of the Walker circulation”, said Prof. Axel Timmermann, Director of the IBS Center for Climate Physics at Pusan National University and co-author of this study.
In their integrated analysis, the researchers found that the satellite-inferred strengthening of the Walker circulation is substantially weaker than implied by other surface observations used in previous studies. “Putting surface observations in context with latest satellite products was a key element of our study”, said co-author Dr. Lei Shi from NOAA’s National Centers for Environmental Information in the United States.
Analyzing 61 different computer model simulations forced with increasing greenhouse gas concentrations, the authors showed that, although the average response is a Walker circulation weakening, there are substantial discrepancies amongst the individual model experiments, in particular when considering shorter-term trends. “We found that some models are even consistent with the observed changes in the tropical Pacific, in stark contrast to other computer experiments that exhibit more persistent weakening of the Walker circulation during the observational period”, said co-author Dr. Viju John from EUMETSAT in Germany. The authors were then able to tease apart what caused the spread in the computer model simulations.
Co-author Prof. Kyung-Ja Ha from the IBS Center for Climate Physics and Pusan National University explains “Natural climate variability, associated for instance with the El Niño-Southern Oscillation or the Interdecadal Pacific Oscillation can account for a large part of diversity in simulated tropical climate trends”.
“The observed trends are not that unusual. In climate model simulations we can always find shorter-term periods of several decades that show similar trends to those inferred from the satellite data. However, in most cases, and when considering the century-scale response to global warming, these trends reverse their sign eventually”, said co-author Prof. Brian Soden from the Rosenstiel School of Marine and Atmospheric Science, at the University of Miami, United States.
The study concludes that the observed strengthening of the Walker circulation from about 1990-2013 and its impact on western Pacific sea level, eastern Pacific cooling, drought in the Southwestern United States, was a naturally occurring phenomenon, which does not stand in contrast to the notion of projected anthropogenic climate change. Given the high levels of natural decadal variability in the tropical Pacific, it would take at least two more decades to detect unequivocally the human imprint on the Pacific Walker Circulation (see Figure 1, right panel).
So models claim that natural climate change can obscure Mann-made climate change. Well, it’s a start in the right direction, but doesn’t tell us if models have any value. Except that this one is similar to actual measurement.
No one has ever shown that delta(T) = fn(CO2) in a single, credible, equation. That’s why it is BS.
We can say the same for “global warming”. Oddly they prefer to keep the bad models in the mix and obsess about the “ensemble mean” rather than focus effort on the models which work best.
Maybe someone should look at which models they found to be “consistent” with observations and see whether they are high or low sensitivity models.
Since “cloud amount” is one of the things that can not be modelled and has to be frigged with guestimated parameters, maybe that is why some models work better than others.
Maybe someone should look at which models they found to be “consistent” with observations and see whether they are better or worse at estimating the barely detectable tropical tropo hot spot.
“Consistent with” may be “coincident with”. The models have shown no skill to forecast or hindcast outside of a limited frame of reference without adjustments (“tuning”).
As ENSO models demonstrate, there are no good climate models. Today winners are tomorrow losers, because they only appear to perform better by chance.
I checked ENSO models for several years and there was no way to select a group of models that would perform consistently better on future data. Very revealing of their inability to represent what they are modeling.
I suspect you are confusing the issue that “some models are even consistent with the observed changes ” actually means these models correctly simulate the climate fluctuations. That is not proven.
If I had a wall of broken clocks, each displaying a different time yet still dysfunctional,
one would suspect that at any given time one of the clocks would agree closely with the actual time. That doesn’t mean that we should keep the “good” clock and toss out all the others because it seems to be working.
Twice a day is not accurate enough.
A man with one watch knows what time it is. A man with two watches is never sure.
“Computer experiments”, is that some kind newspeak to replace “models”?
Whatever, it sounds much more reliable than “models”, not.
Greg, When for now
“Oddly they prefer to keep the bad models in the mix and obsess about the “ensemble mean” rather than focus effort on the models which work best”
there’s nothing wrong with next time “the bad models” ranging in the better / best list.
When the facts and observations clash with the Climate Models, the Warmistas prefer to go with the Models as they give the ‘correct’ result. They know that the World is politically defective.
So model output is an input to their analysis.
If something happens the opposite to what is shown by climate models, it must be “natural variability” and the models are still correct. If it is the same as shown in models , it proves models are correct.
According to that logic climate models can never be shown to be wrong and are thus non falsifiable … and therefore worthless non-scientific computer games.
QED. Thanks Prof Timmermann.
“A new study, published this week in the journal Nature Climate Change, shows that the recent intensification of the equatorial Pacific wind system, known as Walker Circulation, is unrelated to human influences and can be explained by natural processes. ”
Also the claim that the current warming is human caused because the warming is unprecedented
Two links below
Thanks for the links.
I’ve added them to my bookmarks, now in the hundreds, on dozens of different issues. I don’t want to be presumptuous and jump to any hasty conclusions, but when there are so many studies that raise so many questions, the least a fair minded person can do is to be a skeptic.
Good observation sir
“I don’t want to be presumptuous and jump to any hasty conclusions, but when there are so many studies that raise so many questions, the least a fair minded person can do is to be a skeptic.”
And the least a rational government can do is to delay destoying its economy fighting CO2 until these questions are answered, including whether it is really necessary to fight CO2 in the first place.
The first question that needs answering is: How much net heat does CO2 add to the Earth’s atmosphere?
Even after all the decades of studying this question, we still don’t know what this number is. This is the number upon which everything else is based, and we don’t have this number. We have guesses from zero to 4.5C and even higher, but that’s all they are: guesses.
We shouldn’t base government policy on guesses.
From now on unless a climate change study factors in the impact of white Americans eating burgers I will ignore the conclusions. Clearly the natural variation in climate drivers, is only millions of times more impacting than any human endeavour. So, using Green logic, as opposed to scientific logic, humans and particularly white Americans, are the prime drivers of ocean circulations.
Can anyone advise me, where I can apply for a grant to study the human effect on solar radiation, I have found a gap in the Green activists manifesto that needs to be addressed.
I want a grant to study the effect of solar radiation on humans. While sitting by my pool. Drinking a CO2 laden beer. And, of course, eating a burger.
If even EurekAlert! acknowledges that IPCC models are a total failure which brought only blatantly wrong “projections” …what bone will remain for alarmists to gnaw ?
Perhaps one of the next bones to gnaw is a local model based on actually plausible climate change causes ?
– I’m not aware of such a study but it would be interesting to analyze how the Antarctic temperatures evolved since the satellite era with respect to CO2 concentrations :
– Indeed, the following assumption needs confirmation but it seems that most of Antarctic is well isolated from atmospheric circulations (polar vortexes seem quite steady there) and thus would constitute a suitable open sky laboratory in which it would be possible to confirm or falsify the hypothesis of CO2 induced temperatures increase (or decrease).
– If this assumption is correct, apart from long term Milankovitch cycles, the major factors of climate change in Antarctic would be solar activity and greenhouse gases.
Antarctic temperatures of the last 40 years :
The easterly trade winds are caused by the Coriolis effect. As warm moist air around the equator rises, air is drawn in from north and south to replace it. These N-S and S-N winds are deviated to the west in both cases by the Coriolis effect. They combine to add a westward component to surface winds right across the equatorial Pacific.
When the rising air hits the tropopause it spreads back out N and S giving rise to an opposite Coriolis effect at altitude: the upper arm of the Walker circulation.
The main movement of air is the rising and falling N-S movement which is known as Hadley cells. The combined residual E-W, W-E components due to Coriolis forces make up the Walker circulation.
This pushes warm surface waters westwards leading to the build up of warmer waters in the WPWP leaving slightly cooler water in the east.
So the above claim is inaccurate, it is the trade winds ( part of the Walker circulation ) which are causing the warm western Pacific , not the other way around. Probably mutually reinforcing effects in one of those pesky coupled, non-linear effects that modellers can not deal with. The things that TAR chapter 14 section 18.104.22.168 warned us were not possible to predict.
A very good, simple explanation of air circulation, Greg. Thanks very much.
I saw a science special on Jupiter the other day and have a related question about circulation on Jupiter. Jupiter has various bands of circulation and one band will be moving east to west while the band above and below will be moving in the opposite direction. I’m wondering how this circulation is created. Or is it even valid to try to compare the atmospheric circulation of Earth and Jupiter? Maybe I’m asking this question too early, as we are still in the early phase of discovery at Jupiter.
Peta, below, has decided that what you are saying is that the rotation of the earth is the cause of all this. Of course, the cause is, as always, heat transport. The coriolis force cannot “cause” the primary circulation here, as it operates at right angles to the motion of the wind, and the dot product of coriolis force and velocity (power), is then zero. The coriolis force causes the secondary circulation–the trade winds, as you state. But the trade winds do not cause the warm western pacific, they cause an enhanced thickness of warm water, which is unstable in the absence of the trades.
ROFL, their backwardness caught my attention as well. Their logic states that temperature difference between western and eastern Pacific causes the Walker circulation, which causes the ocean circulation, so following that they must claim the ocean circulation in turn causes the Earth’s rotation.
Coriolis don’t describe a force. But an effect.
The earth rotating under the sun eastward gives
sunshine heating oceans westwards.
Coriolis “force” really is a local effect – the waters deviating from equator due to temperature decline just keep the “direction” they headed before.
As a famous scientist stated for every mass in movement – it’s called relativity theory because it’s change of direction relativ to neighboring mass systems.
+ this relative movement / falsely interpreted as force / can’t compete with the real force from temperature gradient.
Actually, the Coriolis effect is but an inertial effect, not a driving force, which totally vanishes at the equator. The persistent easterly winds of the Walker circulation are more a product of the equatorial convergence of hemispheric gyres, than of any local effects.
Maybe a link to the related paper would good ( provided on EurekaAlert ).
“Reconciling opposing Walker circulation trends in observations and model projections”
Observation-model reconciliation papers have no scientific value. I don’t keep any. They also expire the moment models are changed.
They have value in showing which models are way off the mark and need eliminating for the “ensemble means” that are the basis of attempts the restructure the entirety of human existence.
I doubt any models change to the point where a high sensitivity model becomes a low sensitivity model.
There are no good models. They only appear to perform better. Eliminating the past bad ones does not improve performance on future data. They know it. That’s why no model is eliminated.
There are only “no good models”, if you expect them to demonstrate predictive skill.
Models are just tools.
When used to simulate complex natural systems, models can be powerful heuristic tools… You can evaluate how the system might respond under different conditions. The more constrained the models are, the more likely that the simulations will be realistic. Climate models are poorly constrained.
The biggest problem with climate models is that climate “scientists” seem to think they are equivalent to observations and/or hard data.
Models are just tools??
I thought tools had functional value. Let’s not try to re-define “tools” as “that which are total pieces of crap”. (^_^)
Comparing, say, a hammer, to a climate model, is something along the lines of comparing a meticulously crafted Greek sculpture to one of those square-looking chunks of metal, fashionably placed into today’s cultural landscape as a modern “sculpture”.
some more info/graphs in the S.I.
Seems that two models are way off the mark in terms of outgoing LW IR: JRA-55 and MIRRA-2 . At some stage they need to start eliminating models that are badly wrong on key factors and start looking at the ones which are not too far off the mark.
Marotzke & Forster 2015 showed that models formed a bimodal distribution in terms of model sensitivity ( despite reporting the opposite in their paper ! )
“Forcing, feedback and internal variability in global temperature trends”
Their poorly labelled fig 3b showed two clear groupings on the RHS : then longest period of their analysis. Authors refused to provide data when I pointed out this contradiction to the conclusions of their paper. I guess that must be the old “why should I provide you our data, all you want to do is find something wrong with it” .
If no climate computer model can exactly simulate climate what value is there in comparing model results to observation? None. The assigning value to the model closest to current (and hindcast) observations is just another iteration of the Texas Sharpshooter Syndrome.
Not really. Texas Sharpshooter Syndrome would be changing the data record to fit expectations based on model output. That also happens in spades.
The value is in evaluating how well the models are simulating the climate. If the observations consistently tracked near the mean of the model ensemble, they would be accurately simulating changes in average temperature.
However, the observations consistently track as cool as, or cooler than, 95% of the model ensemble… A random walk could do this.
That is exactly right. And all of the model/observation comparisons show the same thing that John Christy’s figure shows. The envelope of model projections is rising much faster than the observations. In fact, the observations are always on the verge of emerging below the envelope of model projections. And yet, I read all these elaborate arguments about how the observations are “consistent with” the projections. It’s just not so.
They are “consistent with” the projections because they haven’t quite dropped “off the curve.”
Here is Christy’s graph, showing dropping off the curve:
Observations are after-the-fact data. It is beyond hilarious that model hindcasts cannot replicate the major movements in decadal surface temperature trends of the past.
The modelers’ rely so much on ‘greenhouse’ gas forcing that their best attempts with parameter adjustments cannot replicate past facts without blowing up their model outputs. Maybe that is the reason CliSci is so focused on altering past data; if the models don’t reflect past data, just change the data?
It is my understanding that the models that produce lower ‘greenhouse’ gas sensitivities more closely track past temperature trends. Is that true?
I like the conclusion that even though observations run contrary to model projections, the observation “does not stand in contrast to the notion of projected anthropogenic climate change.”
Nothing can refute the existence of
Godanthropogenic climate change.
Exactly, so the models are non falsifiable and thus non scientific. QED.
AGW is non-scientific because it is unfalsifiable even when the hypothesis (the models) is falsified.
Again, the problem is that they are treating the models like real data, until they fail… Then they explain away the observations
Javier – do you have a 30 yr cooling trend?
From the article: “This result ends a long-standing debate on the drivers of an unprecedented atmospheric trend, which contributed to a three-fold acceleration of sea-level rise in the western tropical Pacific, as well as to the global warming hiatus.”
How can it be unprecedented if it is a natural process? Now, I suppose it could be an unprecedented event, but these scientists can’t know that because they have only been observing the Earth’s weather and climate for a very few years in the scheme of things.
One overstatement after another is what modern climate science has become.
I’m with Greg on this – they’ve got Cause & Effect the wrong way round.
The primary driver is Earth’s rotation causing the East/West trade winds
These ‘sweep’ warm surface water to the western side of the ocean.
This allows cooler water to come to the surface at the eastern side of the ocean, Upwelling.
(This creates a ‘pile’ of warm water in the West, which eventually collapses > warm water rushes eastwards = an El Nino)
1) their assertion that rising CO2 will cause the trade winds to decrease in strength is crazy – it goes against all other predictions of ‘greater storminess’
2) Assuming what they say is true, (trade wind decrease) then LESS warm water will accumulate in the west, not more.
Its all wrong but no matter. The Computer said it was so, so it MUST be true.
I do love the little aeroplane in their picture –
It’s doing well for a 737 innit = at least flying sideways instead of pointing straight down.
It would be mean & horrible to suggest that Climate Scientists should have a go at ‘Altitude Deficiency’ – at least once in their lives.
T’would be a nice compliment to the Science Deficiency they inflict upon us all though
How do today’s Met people work to give us the three day forecast. Is it the
old fashioned way, by observation and experience of the staff, or do they
feed the data into a PC and hope that in the short term it might be close to the
We use the computer runs.
Those short term computer run weather models NWS uses several different ones) can be in agreement, disagreement, spot on, way off, and close enough. I don’t know what Wunderground uses, but where I live, Wunderground is more accurate. That may be because NWS does not consider micro-climates within large counties. Longer term predictions are also modeled by some, or compared to similar year data (analog years) by others. Prediction accuracy plummets as predictions are made further out than just a few days.
We know that but we’re not going back to forecasting freehand. The workload places upon meteorologists has become prohibitive.
workload placed upon
“…an unprecedented atmospheric trend,…”
I would not bet my house on that.
A minor thought experiment: Would a warming world not result in increased evaporation, leading to more uplifted humid, warm air at the equator, everything else being equal? Would not the additional uplifted humid, warm air at the equator be displaced by increased N/S in-rushing air? Would not that increased in-rushing air at the surface be deflected West by the Earth’s rotation? Would this not enhance Walker circulation?
You tell me; inquiring minds got to know.
Sooooo…anthropogenic warming continues except when it does not. This piece of research cost how much?