Guest Post by Willis Eschenbach

There’s a new open access paper in Nature Magazine, entitled “A tighter constraint on Earth-system sensitivity from long-term temperature and carbon-cycle observations“, by Wong et al., hereinafter Wong2021. Gavin Schmidt, GISS programmer to the stars, lauds it on Twitter. The Abstract says:

The long-term temperature response to a given change in CO2 forcing, or Earth-system sensitivity (ESS), is a key parameter quantifying our understanding about the relationship between changes in Earth’s radiative forcing and the resulting long-term Earth-system response. Current ESS estimates are subject to sizable uncertainties. Long-term carbon cycle models can provide a useful avenue to constrain ESS, but previous efforts either use rather informal statistical approaches or focus on discrete paleoevents. Here, we improve on previous ESS estimates by using a Bayesian approach to fuse deep-time CO2 and temperature data over the last 420 Myrs with a long-term carbon cycle model. Our median ESS estimate of 3.4 °C (2.6-4.7 °C; 5-95% range) shows a narrower range than previous assessments. We show that weaker chemical weathering relative to the a priori model configuration via reduced weatherable land area yields better agreement with temperature records during the Cretaceous. Research into improving the understanding about these weathering mechanisms hence provides potentially powerful avenues to further constrain this fundamental Earth-system property.

So I got to thinking about their paper. The first thing that made my urban legend detector start ringing was a statement in the Abstract above that you might have gone right past, viz:

We show that weaker chemical weathering relative to the a priori model configuration via reduced weatherable land area yields better agreement with temperature records during the Cretaceous.

Translated from Scientese into English, one possible meaning of this is:

We adjusted the climate model’s tunable parameters so the output agrees better with our theory that CO2 controls the climate.

Not an auspicious start …

All of this is based around a computer model called GEOCARBSULF, which is a long-term (millions of years) carbon and sulfur cycle model used to estimate past CO2 levels. So I got to wondering … just how many tunable parameters are there in the GEOCARBSULF model?

But before I discuss the number of GEOCARBSULF tunable parameters, why is the number of tunable parameters important? There’s a famous story about Freeman Dyson and Enrico Fermi that explains this issue well. Here it is in Dyson’s own words:

We began by calculating meson–proton scattering, using a theory of the strong forces known as pseudoscalar meson theory. By the spring of 1953, after heroic efforts, we had plotted theoretical graphs of meson–proton scattering. We joyfully observed that our calculated numbers agreed pretty well with Fermi’s measured numbers. So I made an appointment to meet with Fermi and show him our results. Proudly, I rode the Greyhound bus from Ithaca to Chicago with a package of our theoretical graphs to show to Fermi.

When I arrived in Fermi’s office, I handed the graphs to Fermi, but he hardly glanced at them. He invited me to sit down, and asked me in a friendly way about the health of my wife and our newborn baby son, now fifty years old.

Then he delivered his verdict in a quiet, even voice. “There are two ways of doing calculations in theoretical physics”, he said. “One way, and this is the way I prefer, is to have a clear physical picture of the process that you are calculating. The other way is to have a precise and self-consistent mathematical formalism. You have neither.”

I was slightly stunned, but ventured to ask him why he did not consider the pseudoscalar meson theory to be a self-consistent mathematical formalism. He replied, “Quantum electrodynamics is a good theory because the forces are weak, and when the formalism is ambiguous we have a clear physical picture to guide us. With the pseudoscalar meson theory there is no physical picture, and the forces are so strong that nothing converges. To reach your calculated results, you had to introduce arbitrary cut-off procedures that are not based either on solid physics or on solid mathematics.”

In desperation I asked Fermi whether he was not impressed by the agreement between our calculated numbers and his measured numbers. He replied, “How many arbitrary parameters did you use for your calculations?” I thought for a moment about our cut-off procedures and said, “Four.” He said, “I remember my friend Johnny von Neumann used to say, with four parameters I can fit an elephant, and with five I can make him wiggle his trunk.” With that, the conversation was over. I thanked Fermi for his time and trouble, and sadly took the next bus back to Ithaca to tell the bad news to the students.

So … how many tunable parameters does the GEOCARBSULF model have? From the Wong2021 paper …

There are 68 GEOCARB model parameters, of which 56 are constants and 12 are time series parameters. The constant parameters have well-defined prior distributions from previous work, and the time series parameters have central estimates and independent uncertainties defined for each time point15.

Hmmm, sez I … 68 parameters … not a good sign.

So to see if “the constant parameters have well-defined prior distributions from previous work” as claimed above, I went to look at reference 15 listed in the above quote. It’s called “ERROR ANALYSIS OF CO2 AND O2 ESTIMATES FROM THE LONG-TERM GEOCHEMICAL MODEL GEOCARBSULF“. There, the Abstract concludes by saying:

The model-proxy mismatch for the late Mesozoic can be eliminated with a change in GYM within its plausible range, but no change within plausible ranges can resolve the early Cenozoic mismatch. Either the true value for one or more input parameters during this interval is outside our sampled range, or the model is missing one or more key processes.

Hmmm, sez I … doesn’t sound like that backs up the Wong2021 claim that “the constant parameters have well-defined prior distributions from previous work, and the time series parameters have central estimates and independent uncertainties defined for each time point15.”

So, setting aside the fact that the model has enough tunable parameters to make an elephant put on a tutu and do the Swan Lake ballet, I looked at their results. First, here is their graph of their results.

Figure 1. This is Figure 4 in Wong2021. ORIGINAL CAPTION: “Model hindcast, using both CO2 and temperature data, for precalibration and a %outbound threshold of 30% (shaded regions). The gray-shaded regions show the data compilations for CO2 (ref. 26) and temperature12. The lightest colored shaded regions denote the 95% probability range from the precalibrated ensemble, the medium shading denotes the 90% probability range, the darkest shading denotes the 50% probability range, and the solid-colored lines show the ensemble medians. To depict the marginal value of each data set, the dashed lines depict the 95% probability range from the precalibrated ensemble, when only temperature data is used (a) and when only CO2 data is used (b).”

(A short digression. Looking at Figure 1, I considered the fact that dinosaurs lived on the planet from about 245 million years ago to 66 million years ago. Mammals first appeared 178 million years ago. During that time, according to Figure 1, temperatures were between 6°C and 12°C warmer than at present. And folks hyperventilate about a further half of a degree °C warming being an “emergency” that will ruin our lives and drive extinctions through the roof? … but I digress.)

Now, their claim is that their results gave tighter constraints on the sensitivity of the planetary temperature (lower panel) to atmospheric CO2 levels (upper panel). Squinting at that graphic, I said “Hmmm …”. Didn’t look too likely.

So I did what I usually do when the authors are not conscientious enough to archive their results. I digitized the Wong 2021 temperature and CO2 data, and I graphed it up. Figure 2 shows that result.

Figure 2. Scatterplot, paleo temperatures versus the log (base 2) of paleo CO2 levels from Wong2021

Now, if CO2 levels actually were the control knob regulating the global temperature, we’d see all of the points falling on a nice straight line … but we don’t, far from it. There’s no statistically significant relationship between the temperature and the CO2 levels reported by Wong et al.

So I gotta say, the data reported in the Wong2021 paper is a long, long way from establishing the claims made in their Abstract. In fact, even after they’ve carefully adjusted the tunable parameters of the GEOCARBSULF model in their favor, their results support the null hypothesis, which is that CO2 is not the global temperature control knob.

My best to everyone, dinosaurs and mammals alike,


PLEASE: When you comment, quote the exact words you are discussing. I can and am happy to defend what I wrote. But I can’t defend your interpretation of what I wrote.

5 32 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
Randle Dewees
May 30, 2021 10:12 am

Nice report, not too hard for Sunday morning.

Reply to  Randle Dewees
May 30, 2021 12:37 pm

Does anybody know how the Nature Magazine decides what papers will be “open access”?

Thomas Gasloli
May 30, 2021 10:18 am

Wong 2021–what science looks like when it is published in The Onion.

Sweet Old Bob
Reply to  Thomas Gasloli
May 30, 2021 11:04 am

Just plain Wong ?

Lance Wallace
Reply to  Sweet Old Bob
May 30, 2021 1:06 pm

Not even Wong

Reply to  Lance Wallace
May 30, 2021 3:29 pm

So is Wong white or wong ?
I’m confused by Confucius.

Reply to  Sweet Old Bob
June 3, 2021 12:04 pm

He must’ve been on the wong side of the woad.

Last edited 1 year ago by beng135
Peter W
Reply to  Thomas Gasloli
May 30, 2021 12:03 pm

Reminds me of what I just read in “Simplified Climate Modelling – Part 1.”

Reply to  Peter W
June 3, 2021 12:08 pm

Same here — prb’ly recent orders from the marxist science-bosses to now try the “carbon-cycle observations” angle.

Reply to  Thomas Gasloli
May 30, 2021 4:23 pm

One more addition is needed for the Onion:

The CO2 control knob only goes to “10”… as we have all learned from Spinal Tap, the maximum should be “11”.

Reply to  Anon
May 30, 2021 5:34 pm

+1 for the Spinal Tap reference

oeman 50
Reply to  Anon
June 1, 2021 2:46 pm

Looks like Willis carefully adapted the Men vs. Women control panels without accreditation. Good job, Willis, you avoided going to woke jail!

Joe B
May 30, 2021 10:19 am

I thank you for your informed, ongoing efforts in these matters, Mr. Eschenbach.

Pat Frank
Reply to  Joe B
May 30, 2021 3:31 pm


Reply to  Joe B
May 31, 2021 5:59 am

Loved the Fermi story. Showing what REAL science … EXPERIENCED science looks and sounds like.

Tom Abbott
Reply to  Kenji
May 31, 2021 6:42 am

Me, too. I have such great respect for Freeman Dyson, and it’s striking learning about how such a brilliant man was dressed down at one time by an even more brilliant man. Or maybe not more brilliant, but more experienced, is the way it should be put.

Yes, love the story.

Two Giants of Science.

Peter Morgenroth
Reply to  Joe B
May 31, 2021 11:15 am


Rud Istvan
May 30, 2021 10:20 am

Great job, WE! One additional observation about dinosaurs and mammals, drawn from your figure 1(a) gray ‘Foster data’. It is shown experimentally that plant photosynthesis starts to fail at 150ppm, and fails utterly below 100ppm. IF Foster’s data are to be believed, than all life on earth failed from starvation (plants eating CO2 and animals eating plants) before there ever were dinosaurs and mammals. Oopsy.

Tom Halla
May 30, 2021 10:24 am

Figure 2 is fairly conclusive. A very weak correlation between CO2 and temperature, even after stroking.

Reply to  Tom Halla
May 30, 2021 12:35 pm

Not with CO2 on a log scale and T on a linear one. In fact it’s a negative correlation

Reply to  Tom Halla
May 31, 2021 6:03 am

Hahaha ha … you said “stroking”. I suspect a LOT of stoking goes on with these climate scientists

May 30, 2021 10:26 am

You said: “There’s no statistically significant relationship between the temperature and the CO2 levels reported by Wong et al.”

It looks to me that there is a good relationship [straight line] up to log CO2 = 10, and no correlation thereafter. So, at that point something must be regulating temperature apart from CO2. Perhaps tropical thunderstorms…

Last edited 1 year ago by Leif Svalgaard
Reply to  Willis Eschenbach
May 30, 2021 7:09 pm

Hi Willis, a friend has written a significant piece on the issue of the greenhouse effect energy balance and sent me a copy for comments before he submits.

But reading through it i feel that the subject matter here is really what you have been writing here on wuwt.

Would you be willing to review the paper? It’s long.

If so may I have an email address where I can send it? My email is

You two share a very similar research interest.


Kevin kilty
Reply to  Leif Svalgaard
May 30, 2021 2:09 pm

Apply the confidence intervals from Figure 1 to both variables in Figure 2, and things get a bit hazy. Some sort of relationship might be salvaged over the fact that confidence intervals are smaller at lower temperature, but the lower temperatures are all recent observations, so adding the rest of Earth’s history didn’t accomplish anything.

Reply to  Kevin kilty
May 30, 2021 3:17 pm

The uncertainty monster is right there in their Supp Figure 3.

Screen Shot 2021-05-30 at 3.16.55 PM.png
Robert Capetola
Reply to  Leif Svalgaard
May 30, 2021 3:11 pm

Cherry-picking I see

Reply to  Leif Svalgaard
May 30, 2021 4:29 pm

Or perhaps below about 900 ppmv the CO2 level in the atmosphere is driven by the temperature of the oceans.

Jim Speight
Reply to  Leif Svalgaard
May 31, 2021 7:23 am

It may be another chicken vs. egg
Does Co2 cause temperature changes or does temperature changes cause Co2 production?

M Courtney
May 30, 2021 10:33 am

On Gavin’s linked Twitter thread the takeaway seems to be that the CO2 levels are dropping over these timescales.
Not exactly the intended message, methinks.

Reply to  M Courtney
May 30, 2021 2:44 pm

The Quaternary’s Ice Age would seem like a big warning flag. CO2 is not the problem, but it may be the answer.

Kevin kilty
Reply to  Joel O'Bryan
May 31, 2021 4:40 pm

Some years ago I sat in a geology department seminar given by a fellow whose research was trying to close the Earth surface volatiles budgets. All sorts of volatiles, CO2, Sulfur, Chlorine, etc. His take was that the subduction zones take a great deal of volatiles back into the mantle, and that the big return mechanisms, like volcanoes, are not sufficient to keep a steady surface inventory — unless, as he speculated, there are many distributed return flows that we simple don’t know about.

Suppose for a minute that the volatiles budget for CO2 is not in balance and we are continually burying CO2 in the mantle. It would explain some of the paleo CO2 concentration and temperature trends. Scary thought, long term. Longer term than I have anyway.

Ron Long
May 30, 2021 10:38 am

Good work, Willis, and I especially appreciate your digress to talk about dinosaurs, etc. As a geologist who has walked many miles through the preserved strata of the Cretaceous, and marveled at the abundance and scale of the life preserved as fossilized bones and tree trunks, I am always on the verge of a hissy fit when someone messes with the Cretaceous. Remember the Lone Ranger? “Return with me now to those thrilling days of yesteryear…”.

Reply to  Ron Long
May 30, 2021 4:56 pm

It is clear that all the carbon in coal came from CO2 in the atmosphere. Did the carbon in limestone and dolomite also come from CO2 in the atmosphere or did it come directly from carbonaceous ions in the ocean?

Reply to  Dan Pangburn
May 31, 2021 8:02 am

I don’t know the exact numbers … but large amounts of CO2 is carried down the rivers to the oceans via organic matter (leaves etc.). Calcium weathering out of the rocks combined with CO2 creates CaCO3 i.e. Calcium Carbonate i.e. Limestone.

Reply to  Dan Pangburn
May 31, 2021 1:02 pm

keywords ‘limestone rock fossils’ revealed this “Limestone is a sedimentary rock made almost entirely of fossils”

Bill Toland
May 30, 2021 10:48 am

So this model only has 68 parameters. That’s nothing. My model has 3793 parameters and the latest model run has the oceans boiling by next Thursday. We’re all doomed, I tell you.

Reply to  Bill Toland
May 30, 2021 11:43 am

And Bill have you named your model “The Lemon”?

(If you haven’t yet registered this name, I suggest you get an application in quickly, because as Wills demonstrates here, there are many, many other candidates for this name)

Bill Toland
Reply to  Mr.
May 30, 2021 11:51 am

My model is called the Climate Reproducing Autonomous Program.

Reply to  Bill Toland
May 30, 2021 1:06 pm

You mean CRAP?!!👍

Reply to  Bill Toland
May 31, 2021 6:09 am

There appear to be quite a few climate scientists spending lots of *ahem* Autonomous time … with themselves. Is that redundant?

Reply to  Bill Toland
May 31, 2021 6:56 am

Already superseded by my Biggest Underlying Latitudinal Longitudinal Climate Reproducing Program which really puts the squeeze on them all. Particularly the grants program.

May 30, 2021 10:50 am

Ha ha, good take down Willis, this is why I stopped reading 100% modeling based papers long ago, they hurt my brain with their tuned baloney,

I hope are you are feeling well despite reading it.

Clyde Spencer
May 30, 2021 10:59 am

Speaking of models, the ‘usual suspects’ at LLNL have just published a paper
[ ] where they are basically claiming that they can judge the quality of data sets by how well a data set agrees with their models. On that basis, they further claim that the temperature of the lower troposphere has been underestimated for the last 40 years because the warmer data set agrees better with the model.

A couple of observations: They are claiming that models are more trustworthy than actual observations, and they may have provided a clue as to why models have a history of running warm.

Last edited 1 year ago by Clyde Spencer
Rud Istvan
Reply to  Clyde Spencer
May 30, 2021 11:53 am

Took a look. For sure, the usual suspects. TY.

One thing this new paper did NOT mention, but UAH actually did in developing their lower troposphere estimates, was compare their satellite estimates to many actual radiosonde temperatures at various altitudes up continental North America from Southern Mexico all the way to Alaska. The large ‘latitudinal swath’ fit is excellent when the radiosondes are averaged to reproduce the satellite lower troposphere. The Christy and Spencer paper can be found in some footnote to the penultimate climate chapter of The Arts of Truth. As far as I can tell, Mears and RAH did no such validation, and a couple of years back adjusted their results upward based on a spurius orbital drift argument that Spencer corrected in about UAH v3. Now Mears says maybe not enough?!?

Richard M
Reply to  Rud Istvan
May 31, 2021 8:09 pm

I believe Mears and Co. did do a comparison with radiosonde data about a decade ago and found they were running just a little hot. They then ignored this finding when they adjusted their results to be even hotter.

Reply to  Clyde Spencer
May 30, 2021 3:24 pm

That is along the same crap modeling science claims of Micky Mann that if a Climate model doesn’t reproduce some internal variability, like the AMO, then it doesn’t exist.

Gordon A. Dressler
May 30, 2021 11:04 am

Willis and/or WUWT editor,

My kudos to you on the most-excellent choice of photo that appears at the end of the above article. That single photo is worth a thousand million words about all “scientific” attempts to model climate toward the purpose of claiming that CO2 is the dominant driver of climate change. Of course, it isn’t, as objective scientific data clearly shows.

Of course, Willis presents a good-read, science-based, great article per his established reputation for laser focused, cut-to-the-case, often-witty criticisms of “junk science” publications. So thanks for that too.

Last edited 1 year ago by Gordon A. Dressler
Reply to  Gordon A. Dressler
May 31, 2021 4:48 am

Great photo, but I am worried about the on/off switch !

Gordon A. Dressler
Reply to  bonbon
May 31, 2021 7:38 am

THAT switch is only for God’s use.

Abolition Man
May 30, 2021 11:13 am

Thanks, Willis!
Just the thing for a Sunday morning after some huevos rancheros! Walt Disney is smiling in his grave at the reference to elephants in tutus dancing to Swan Lake!
It is kind of you to complete the work that Wong et al. failed to complete in their study! Proposing an hypothesis is all good and well, but a REAL scientist would have tested it by trying to disprove it! Instead these jamokes did everything in their power to prove their thesis and avoided any attempts at the null hypothesis! That makes them true climastrologists, not scientists!
They’re just trying to produce evidence that past temps and CO2 levels were more in line modern ones. That’s not gonna happen with the wizards of WUWT blocking their path forward!

Tom Abbott
Reply to  Abolition Man
May 31, 2021 7:00 am

“Instead these jamokes did everything in their power to prove their thesis and avoided any attempts at the null hypothesis! That makes them true climastrologists, not scientists!”

Our guy (Willis) looks at it and immediately sees the flaws in the claims, but Nature Magazine apparently can’t do the same.

Who is running these “science” organizations? It’s got to be politicians masquerading as scientists. They know the ways of politics, not the ways of science.

Happily for us, Willis *does* know the ways of science.

Last edited 1 year ago by Tom Abbott
May 30, 2021 11:28 am

Thanks, Willis.


May 30, 2021 11:29 am

“tuneable parameters”

Not only are tuneable parameters problematic, they become downright scary when you use a tuning feedback method like gradient descent optimization looking to reduce a global error.

Gradient descent optimization across a large number of parameters is essentially how AI works and allows a very large complex network to exhibit extraordinary ability to match or near match very different input series.

I’d be curious to know what process they use to tune these parameters and if its based on any kind of feedback changes using error derivatives and multiple small step change runs I’d suggest they are doing a form AI back propagation.

May 30, 2021 11:32 am

Note their graph is about right over the last 200 Myr. 3000 ppm CO2 reduction results in 3 degrees cooling. Quite a long ways from their current prediction for increasing CO2.

Reply to  DMacKenzie,
June 3, 2021 12:50 pm

Shhhh. They think the credulous only notice increases.

Dave Fair
May 30, 2021 11:37 am

Thanks for your work Willis, again. Crowd review beats peer review once more.

Last edited 1 year ago by Dave Fair
May 30, 2021 11:37 am

Willis, if you don’t stop revealing the man behind the curtain after every act, the climate calamity vaudeville audiences will soon go elsewhere for their fright night thrills.

Have some consideration for the ongoing financial wellbeing of the climate showbiz personalities, please.

May 30, 2021 11:53 am

The “god knob” presumes an invariant nature outside of a limited frame of reference, thus statistics, thus inference, thus plausible, thus modern science, infilled with brown matter to smooth the missing links.

Dave Fair
May 30, 2021 11:55 am

Lead Author: Ben Santer, a known fraud. The study just goes downhill from there. “Who you gonna believe, my model or your lie’n eyes.” [Clyde’s referenced study.]

Last edited 1 year ago by Dave Fair
Tom Abbott
Reply to  Dave Fair
May 31, 2021 7:17 am

If anybody ought to be sued over Climate Change Fraud, it should be Ben Santer. He is the one who lied and said there was evidence that humans are causing the Earth’s climate to change, and the Western World is currently planning on wasting Trillions of Dollars on the Climate Change Fraud based soley on Ben Santer’s say so.

The UN IPCC scientists come out with their report and the report says they cannot find a CO2 human signal in the Earth’s climate.

Then, Ben Santer, who is writing the summary of the IPCC report, which is what most people and politicians read, inserted a statement into the summary that said the human signal was in evidence, based on nothing.

Santer said the exact opposite of what the report said, and Santer’s version is the one the whole world is working on now.

Santer is responsible for this lie, and the waste it has caused, and will cause, and he should have to pay damages.

Last edited 1 year ago by Tom Abbott
Fred Hubler
May 30, 2021 11:57 am

How do they know whether they’re measuring cause or effect?

Dave Fair
Reply to  Fred Hubler
May 30, 2021 12:03 pm

Or other confounding factors not considered.

May 30, 2021 12:08 pm

Gavin Schmidt, GISS programmer to the stars

I love it!!! I have to make a stand-alone comment just to acknowledge that one!

May 30, 2021 12:13 pm

I note the likes of Griff et al are notably AWOL when Willis discusses real science.

Tom Abbott
Reply to  HotScot
May 31, 2021 7:21 am

Griff is only interested in arctic sea ice and windmills.

May 30, 2021 12:48 pm

yields better agreement with temperature records during the Cretaceous.

Those Cretaceous weather observers, studiously making temperature records, must have been better than our grandparents making temperature records from the 1910-1940’s that have to be frequently re-adjusted.

Paul Milenkovic
Reply to  Joel O'Bryan
May 30, 2021 1:23 pm

If they are considering the entirety of time since the Cambrian, hasn’t there been a goodly amount of rearrangement of continents and oceans by way of Plate Tectonics?

Wouldn’t that have a big effect on weather trends and in turn climate?

This has come up in observations of change in the Moon’s orbit and the length of the day on Earth resulting from Earth-Moon tidal interaction. There is fossil evidence of a much shorter day and month (orbit of the Moon) based coral growth rings or some such evidence, but if you extrapolated the trend, you would get an age of the Earth-Moon system in the hundred of million years, not the 4+ billion years that radioactive isotopes indicate.

The explanation is that tidal drag depends on the arrangement of oceans, and for much of geologic history, the land is pile up into one big supercontinent before it splits off, rearranges and the reforms a supercontinent? And we are in one of the split-off and rearranging epochs?

Rud Istvan
Reply to  Paul Milenkovic
May 30, 2021 2:24 pm

Dunno about the Moon, but do know about plate tectonics. Wrote up Wegener’s continental drift theory as a major example in the Recognition chapter of The Arts of Truth. No less a prominent geologist than Princeton’s Dyeffies presented compelling arguments in his last book that the present ice ages started about 2mya with the closure of what is now the Panama Isthmus, drastically changing ocean circulation patterns. After the closure, Pacific and Atlantic could only ‘communicate’ via the very cold Arctic ocean or the very cold Drake Passage.

Reply to  Rud Istvan
May 30, 2021 3:02 pm

They use a time series parameter that estimates sea floor spreading rates (tectonics), as Supp Figure 1 k. And also subducting carbonate bearing plates into the mantle as CO2 volcanic outgassing rates, as Supp Fig 1 L. Apparently CO2 outgassing from carbonate bearing plates did NOT begin subducting/outgassing until just 150 Myr ago and has been a ramp function ever since. Who knew??

These are all part of the a priori tuning that was done. total junk.

Last edited 1 year ago by joelobryan
Reply to  Paul Milenkovic
May 30, 2021 2:56 pm

The model accounts for land mass fractions as a time series parameter. You can see them in their Supplement. They also write about the problems in the Discussion,:

“However, several of the time series parameters’ distributions change substantially. Specifically, we find that changes were required in the time series for the land area relative to present (fA), global river runoff relative to present (fD), the response of temperature change on river runoff (RT), and the fraction of land area that undergoes chemical weathering relative to present (fAW/fA).”

May 30, 2021 1:14 pm

Talking of temperature control knobs.
Check out the the band of thunderstorms within the ITCZ over the Atlantic.They appear to have become so powerful that their having a impact on the jet stream up at 250 pHa.This suggests to me they must be pumping one hell of alot of heat out of the ocean. As l have suggested before in a other comment it is looking like when the jet stream goes into LIA mode it makes the mid Atlantic ITCZ thunderstorms work harder. Which increases the amount of heat been taken out of the ocean.

Allan Moluf
Reply to  taxed
June 3, 2021 6:33 pm

What is pHa? Did you mean hPa (hectoPascals – the standard atmosphere (atm) is a unit of pressure defined as 101,325 Pa (1,013.25 hPa))? (250 hPa corresponds to an altitude a bit over 10 km.)

William Astley
May 30, 2021 1:19 pm

All of the geological observations, including paradoxes need a physical explanation. The period 541 to 490 million years ago is the start of massive worldwide tectonic plate movement, reconstruction/alteration of the surface of the earth (terraforming).

There must be a physical reason, a cause to explain the start of modern tectonic plate movement 541 million years ago and why there is the appearance of negative C13/C12 ratio carbonates.

The paradigm that CO2 is the driver of the climate is an urban legend. Geology has become a cottage industry to push that fake theory and to hide observations that killed it.
Why did modern tectonic plate movement start 541 million years ago? What happened at that time to earth to start tectonic plate movement and to create the deep oceans on the planet at time? Why did advance life sudden appear on the earth during this period?

What happened in the geological record during this period? Observations not theories. Strip away the dead theories.

Geology hides/is hiding research findings which show there was a series of massive injection of primordial CH4 into the biosphere during that period. These injection started roughly 541 million years ago.
This period coincides with the appearance of advanced life on the planet and the appearance of the first deep oceans on the earth. See figure 1 in that attached paper that is a graph that shows the C13/C12 changes in this period of time.

The C13/C12 ratio in the carbonate record is close to modern levels except during the period 541 to 490 million years ago.  In the last 15 years it has been determined that that C13/C12 ratio in the carbonate rocks deposited in the period 541 to 490 million years ago dropped to negative around negative more than a dozen times and then when positive.
The 541 to 490 million year ago geological record shows there were a series of massive injections of low C13/C12 ratio (this primordial CH4 that has been sitting in the liquid core of the planet until the core of the planet starts to crystallize which drives/starts tectonic plate actions and movements at that time).
This CH4 explains the massive hydrocarbon deposits which all contain heavy metals and explains the fact that there are deep oceans that cover 70% of the surface of the planet and that advanced life on the planet did not appear until 490 million there were deep oceans on the earth.
The CH4 is extrude from the liquid core of the planet when it crystallizes. Metals at high pressure and temperature bond with CH4 and carried CH4 down into the core of the planet. The liquid core is saturated with CH4 so when it crystallizes, the CH4 is extrude at high pressure for the liquid core.
The metals in the mantel form a sheath around the extrude CH4 which creates a tube. The tube carries the CH4 to surface of the planet. This is the force that moves the tectonic plates.
Because the earth was struck by a Mars size object, about 100 million years after it was formed, most of the CH4 that was in the mantel was lost to space as well as the early earth’s Venus like atmosphere.
The tubes extrude core CH4 is what created the earth’s deep oceans and is what is pushing around the tectonic plates now.
A methane fuse for the Cambrian explosion: carbon cycles and true polar wander
Early Cambrian time, punctuated by a unique and dramatic series of geological and biological events, has fascinated and puzzled geologists and paleontologists for more than a century [109,110]. Wide spread preservation of a detailed animal fossil record was facilitated by the almost simultaneous evolution of the ability to precipitate biominerals such as calcium phosphates and carbonates in nearly forty phyletic-level groups of animals [6,58,71].
1.3. Cambrian carbon cycles
 As compiled here in Fig. 1, one of the most puzzling and as yet unexplained features of the Cambrian Explosion is the sequence of over a dozen accompanying oscillations in inorganic δ 13C values preserved in carbonate, with typical negative-shift magnitudes of – 4%/% Delta C13/C12 and sometimes larger [9–11,20,61,65,75,76, 91,107].
The composite record shows a large negative drop starting at the base of the first Cambrian biozone, coincident with the appearance of T. pedum [19,42], followed by a general rise to positive values of about +6%/% Delta C13/C12 near the base of the Tommotian (carbon cycle I′ [65]).
Why deep oceans gave life to the first big, complex organisms
Why did the first big, complex organisms spring to life in deep, dark oceans where food was scarce? A new study finds great depths provided a stable, life-sustaining refuge from wild temperature swings in the shallows.

Schrodinger's Cat
May 30, 2021 1:50 pm

Willis, I always enjoy reading your posts and accept that you are good at interrogating datasets to extract what they are telling us. I have a challenge for you, but to be honest, it is more of a request.

The HITRAN dataset contains IR transmittance data for various greenhouse gases under a range of conditions, concentrations and even mixtures. It can be used to determine the atmospheric absorbance of longwave radiation. Happer and Wijngaarden, Smirnov, Schildtknecht and others have demonstrated that the IR absorption bands are saturated and that further CO2, or indeed, methane or NO2 emissions will cause very little additional warming. This is an incredibly important result because global warming is literally finished, but amazingly, the idea creates very little traction on this site.

It is an important finding in other ways too. If the greenhouse effect is limited by the normal spectroscopic parameters such as concentration and pathlength, then it means that the warming is beneficial and limited with no danger of overheating. This explains how our planet created the ideal conditions for the evolution of life but very high CO2 concentrations did not pose any threat to life. It is believed that CO2 reached 8,000 ppm while life was developing.

If we step back and think of the often bitter divide between alarmist climate scientists and those who disagree, absorption band saturation actually presents an honorable solution for all. It accepts that greenhouse gases are powerful enough to absorb outgoing IR to the extent that atmospheric opacity is achieved. When the concentration reaches a level when the atmosphere is opaque to IR then it does not matter how much more opacifying gas is added, transmission of IR has already ceased. This is band saturation and additional greenhouse gas addition has no more effect. This conclusion provides the unique opportunity to declare that alarmists and sceptics are both correct.

So I am asking you, Willis, to have a look at the HITRAN database and at least think about doing a study to check out the points I am making. It could be the most important study of your life.

Rud Istvan
Reply to  Willis Eschenbach
May 30, 2021 5:16 pm

There is a ‘saturation’ flaw in both (SC) Hitran and Modtran arguments. It was visually and qualitatively described in essay Sensitive Uncertainty in ebook Blowing Smoke for those unablr to do the math. CO2 can never saturate, since it is a non condensing gas whose ‘saturation window’ widens as overlapping wvf is removed by its lapse rate effect (colder means less WV, duhoh)
So, with increasing CO2 concentration its effective radiative level (ERL) just rises higher in altitude, so colder with less overlapping WV. Colder is less efficient at energetic radiative heat removal, which fully explains Callendar’s 1938 logarithmic CO2 concentration curve.

Reply to  Schrodinger's Cat
May 30, 2021 3:21 pm

YES PLEASE WILLIS, what Schrodinger’s cat said above about the HITRAN data.

Reply to  Schrodinger's Cat
May 31, 2021 6:02 am

I have spent some time with HITRAN and both MODTRAN and MODTRAN6. They are used to help explain why/how CO2 has no significant effect on climate and what actually does at . Included in that analysis are links to those codes. The applications there provide some examples of what parameters to select to get useful output.

May 30, 2021 2:09 pm

Climate change is driven by model mania.

May 30, 2021 2:32 pm

Yet more GIGO.

The Dark Lord
May 30, 2021 2:37 pm

So what ? That model matches observations …

May 30, 2021 2:41 pm

I love this line in the Wong-2021 Methods section on “Parameter Precalibration”:

Then, we rule out any combinations of parameters that yield simulations that do not agree well with the CO2 proxy or temperature data, given their uncertainties.”

Also it is worth reading the provided Reviewer comments:

number 1’s comments (start at page 6).
Rev #1:
“I have some fundamental concerns with this study listed below, from which I have to conclude that I can not recommend to publish the paper.
Then he writes immediately as the first comment: “ESS in their approach is not an output of the model, but one of the parameters(input).”

This reviewer immediately recognizes the circular logic of ESS is an input that they get as output as built into the Wong-2021 methodology. All of reviewer #1 take-down of the Wong-2021 paper is worth reading.

Reviewer #2 scolds them for not including the actual parameter values and uncertainty ranges for reproducibility.
It is mentioned that values and uncertainty ranges were used, but the reader is nowhere shown what these values and ranges are in the experiments. That makes the work essentially unreproducible, and that’s not good enough in my view. Parameters are mentioned in the text using their acronyms, but these are not always introduced/spelled out. This ought to be added, and/or reference needs to be made in appropriate places to the tables of parameter explanations.


The final point I think worth noting is in Supp Figure 7. They show a liklihood time slices at 240 Myr and 50 Myr. The 50 Myr time slice is comic gold. CO2 from their model could be 450 ppm, or just slightly less less likely ~750 ppm or ~1,000 ppm.

This study is junk science in action. Amazing that this is what now passes as science in Nature.

Screen Shot 2021-05-30 at 2.23.15 PM.png
Gordon A. Dressler
Reply to  Joel O'Bryan
May 30, 2021 2:55 pm

Joel O’Bryan,

A quite-excellent comment posting all around! Thanks 1E6.

Reply to  Joel O'Bryan
May 30, 2021 3:13 pm

The Big Final point to make is their GEOG(t) input parameter. The GEOG parameter is a time series and they graphically show what it looks like in Supp Figure 1 J. They describe the GEOG parameter as, “GEOG, the change in global surface temperature relative to present-day, assuming present-day CO2and solar luminosity, ”

In other words, they assume on input, what they want the output Global Temperature to be. This is what reviewer #1 meant when he/she wrote, “ESS in their approach is not an output of the model, but one of the parameters(input).”

They show how the GEOG(t) input parameter is inserted into the model in Equation 1 on the manuscript (see attached image):

Screen Shot 2021-05-30 at 3.09.49 PM.png
Gerald Hanner
May 30, 2021 2:42 pm

I have to laugh at the worry-warts who obsess over a little global warming. If you want to obsess, consider global cooling. Tell me how many crops you can grow when the growing season temperature is below 68 F.

Gordon A. Dressler
May 30, 2021 2:45 pm

Neil Lock posted: “The Standard Model of particle physics has either 18, or 19 parameters . . . The model is quite accurate.”

The heck, you say . . . “accurate” by whose standard of measurement?

Per , Despite its great predictive power, however, the Standard Model fails to answer five crucial questions, which is why particle physicists know their work is far from done.”

The biggest problems facing the Standard Model are (my underlining emphasis added in the following quotes taken from the above-linked website):

1) “Three of the Standard Model’s particles are different types of neutrinos. The Standard Model predicts that, like photons, neutrinos should have no mass. However, scientists have found that the three neutrinos oscillate, or transform into one another, as they move. This feat is only possible because neutrinos are not massless after all.”

2) “Scientists realized they were missing something when they noticed that galaxies were spinning much faster than they should be based on the gravitational pull of their visible matter . . . Something we can’t see, which scientists have dubbed “dark matter,” must be giving additional mass—and hence gravitational pull—to these galaxies. Dark matter is thought to make up 27 percent of the contents of the universe. But it is not included in the Standard Model.

3) “Scientists suppose that when the universe was formed in the Big Bang, matter and antimatter should have been produced in equal parts. However, some mechanism kept the matter and antimatter from their usual pattern of total destruction, and the universe around us is dominated by matter.  The Standard Model cannot explain the imbalance.

4) “The latest measurements by the Hubble Space Telescope and the European Space Agency observatory Gaia indicate that galaxies are moving away from us at 45 miles per second. That speed multiplies for each additional megaparsec, a distance of 3.2 million light years, relative to our position. This rate is believed to come from an unexplained property of space-time called dark energy, which is pushing the universe apart. It is thought to make up around 68 percent of the energy in the universe. ‘That is something very fundamental that nobody could have anticipated just by looking at the Standard Model‘ ”

5) The Standard Model was not designed to explain gravity. This fourth and weakest force of nature does not seem to have any impact on the subatomic interactions the Standard Model explains.”

But perhaps adding another 3 or 4 tunable parameters can fix all of the above. 🙂

D. J. Hawkins
Reply to  Willis Eschenbach
May 30, 2021 5:57 pm

The fact that a model explains, say, 50% of observations and claiming it’s accurate because nothing else does better, does not, in fact, make it “accurate”. It is only useful over a restricted range of circumstances. When it explains the 5 elements mentioned, then we can talk about accurate.

Gordon A. Dressler
Reply to  Willis Eschenbach
May 30, 2021 6:01 pm

Neil, thanks for the perfect segue to close this loop and return to the main thrust of Willis’ article above . . . it invites the question, does the “best” climate model win if it comes closest to (a) matching AGW/CAGW predictions (estimates?) of ECS, or (b) if it comes closest to matching observations (aka, field data) concerning “global-average” atmospheric temperature?

Gordon A. Dressler
Reply to  Gordon A. Dressler
May 30, 2021 4:26 pm

Neil, I am not on a journey to develop a Theory of Everything, so don’t wait for a related announcement from me in this regard.

As to your other comments directed to me, I hope you consider what Willis posted before me in this matter (Thanks, Willis!).

As a historical (even current) example: Issac Newton developed a terrific set of physical laws that “accurately” described the kinetics of forces influencing the positions and motions of objects. (Newton’s laws have such predictive accuracy that they are commonly used today across a range of scientific and engineering disciplines.) The along comes Albert Einstein with physical laws that “more accurately” describe forces influencing body positions and motions (and even their masses and the “curvature” of space itself) via his theories of special and general relativity.

Did Einstein improve predictive accuracy beyond the limits of Newton’s laws? Most definitely. Did doing such invalidate the useful accuracy of Newton’s laws for situations in which they are applicable? No.

Reply to  Gordon A. Dressler
May 31, 2021 5:06 am

I did, and there is no trace of Axions, nor dark matter, and even worse when they found the Higgs, a plethora of other particles predicted by a myriad of papers simply refused to show up! Lots of noise about Higgs covered that up!
There is a huge problem, very well addressed by Prof. Lee Smolin :

In a related lecture, he is confronted on accuracy, and mentioned how good Ptolemy model results were for 1000 years. Physics has gone back to Ptolemy, not just Climate.

Gordon A. Dressler
Reply to  Willis Eschenbach
May 30, 2021 6:09 pm

Depends. A thorough risk assessment (which indeed frequently assesses “dangerous” and “lethal” possibilities), as commonly performed, involves quite a bit of scientific rigor when done properly. It almost always avoids ethical judgements.

A risk assessment is a form of modeling possible outcomes from certain actions.

Reply to  Willis Eschenbach
May 30, 2021 8:09 pm

Aren’t we talking apples and oranges here. As Mr. Lock points out the standard model can be tested experimentally. The climate models not so much.

Gordon A. Dressler
Reply to  Ebor
May 31, 2021 8:50 am


Nature and the progress of time are testing the climate models for us.

So far, as concerns CO2 driving global lower atmosphere temperature (GLAT), there is a null result.

Simply put, Earth has experienced two significant periods of nearly-constant GLAT (approximately, from 1940-1975 and from 1997-present) based on accurate scientific measurements; this DESPITE global atmospheric CO2 levels having a smooth, continuously-increasing exponential trend, also accurately measured, over these same time periods . . . and there being a very significant 35% total ppm increase from 1940 to present.

Note that both NASA and NOAA define “climate” as being weather averaged over a specified geographic area for a period of 30 years or more. The span of 1940-1975 meets this criteria; the timespan from 1997 to present is just 6 years short of this.

But also note much more CO2 (on a volume basis) has been put into the atmosphere in this latter time span than was put into the atmosphere during the span of 1940-1975.

As Nobel prize-winning physicist Richard Feynman is famously quoted as saying, ““It doesn’t matter how beautiful your theory is, it doesn’t matter how smart you are. If it doesn’t agree with experiment, if it doesn’t agree with observation, it’s wrong. That’s all there is to it.”

Sebastian Magee
Reply to  Willis Eschenbach
May 31, 2021 2:18 am

As an engineer in power generation, with a master degree in nuclear science and technology, I can tell you for sure that the Standard Model (SM) is not used to model a nuclear reactor.

I doubt the SM can compute reactions between neutrons and Uranium-235 nuclei since the number of particles involved is 236 per reaction. CERN uses a state of the art cluster just to compute the collision of two hellium-4 nuclei (8 particles in total)

In any case if it can in fact be achieved it would require a computing cluster that would use most the whole available space in the ship.

Nuclear reactors are model with neutron transport equation using experimental cross-sections for the reaction rates, coupled with thermo-hydraulics and nuclear species radiation decay equations.

These equations have many parameters, from half-life of que species to viscosity of the fluid, and the so mentioned cross-sections which depend on the energy of the incoming neutron, but they are not free parameters, they have physical meaning and can and are measured independently. I am not aware of any free parameter in these models.

PD: this same thing happens at least with most of the parameters of the standard model, they are the masses of the particles which are measured, so they are not free parameters, their values are very tightly constrained by the experiments. Not like the climate model described in WE article.

Reply to  Sebastian Magee
May 31, 2021 9:24 am

@Sebastian – this is exactly the point I was making: parameters such as collision cross-sections etc can be experimentally determined and then, ideally, the predictions of any model that uses those parameters can be experimentally verified. This to me is a fundamental concern with climate modeling. That and the “breadth” of the models – i.e., what natural mechanisms do they not account for and how significant are they? For example, some have theorized that the apparent correlation between sun spots (as proxy for sun’s magnetic field strength) and climate (see e.g. Maunder Minimum) derives from cosmic rays playing a significant role in cloud formation but verifying that in a laboratory setting would require lots of funding that’s not made available (I’m sure that simulating an atmospheric environment and probing high energy particles interacting with water vapor would be pretty darn expensive) so, this possible mechanism (could be important but who knows?) is not incorporated in the models – sensibly I might add because how do you model an unknown? – but where does that leave the models???

Rick C
May 30, 2021 3:23 pm

Pat Frank
May 30, 2021 3:31 pm

“So I gotta say, the data reported in the Wong2021 paper is a long, long way from establishing the claims made in their Abstract.”

But Wong, et al., passed editorial and peer review at Nature Communications, the world’s ever so eminent journal of science.

May 30, 2021 3:38 pm

Well, the paper has only four authors, so that’s a plus. 🙂

May 30, 2021 3:54 pm

Mr. Eschenbach, you do another simplified take-down of an equally brief report that is just a far off as you are far right(not politically. Extracting the temperatures used show pretty dramatically that the published paper couldn’t be either precise or correct.

ps. must be self taught. I NEVER heard any analysis as sharp. It reminds me of the upper level Prof going through a derivation of the Schrodinger equation.

Rud Istvan
May 30, 2021 5:18 pm

Disney’s Fantasia.

Pat from kerbob
May 30, 2021 5:24 pm

I guess I’m just simple but don’t their graphs show that as co2 concentrations have steadily decreased the temps increased and then slightly decreased?

If CO2 controls temp doesn’t this graph show that more CO2 decreases temperature?

Randy Bork
May 30, 2021 5:46 pm

This is what is passing as ‘science’ in publications once trusted.. There is a crucial difference between ‘expert opinion’ and ‘science.’ I had thought that the journals had a passing familiarity with that distinction. It would seem, I trusted too much. Whoever is in charge of scientific rigor has been purged. It is consensus, all the way.

May 30, 2021 6:23 pm

Mr. Eschenbach, if your “urban legend detector start ringing”, I’m curious if your m.b.e. (male bovine excrement) pie detector is olfacting?

May 30, 2021 6:25 pm

Another bloviated abstract beating up on poor little CO2.

May 30, 2021 6:30 pm

I produced the attached comparison table for the Cretaceous period in a question Bob Wentworth raised.

It is quite feasible for ocean warm pools to regulate to 3 degrees warmer than present if the atmospheric pressure was 1100mb given that the oxygen partial pressure was 30kPa as observed by proxy and Nitrogen still 80kPa.

Reply to  Willis Eschenbach
May 30, 2021 9:45 pm

I do not know what relevance your link has. The Cretaceous period was more than 60M years ago.

This gives proxies for oxygen level:

Analyses of the gases in these bubbles show that the Earth’s atmosphere, 67 million years ago, contained nearly 35 percent oxygen compared to present levels of 21 percent. 

As the table shows, the tropical warm pool temperature would limit to annual average of 33C and could be as high as 35C on a monthly basis; both 3C higher than present.

Julian Flood
Reply to  Willis Eschenbach
May 31, 2021 5:22 am

Willis, of you somehow simultaneously reduced evaporation and lowered the albedo of a large area of ocean then I suspect you’d exceed the usual limiting temperature.

If only Benjamin Franklin were around, he Wight have a suggestion


Julian Flood
Reply to  Julian Flood
May 31, 2021 5:23 am

if, might… sigh…


Robert of Texas
May 30, 2021 6:54 pm

Um…So I tried connecting all those dots in figure 2 in various ways and indeed I finally got an elephant in a tutu, but for the life of me I cannot make him perform Swan Lake… I think I need more control knobs.

Climate believer
Reply to  Robert of Texas
May 30, 2021 10:25 pm

Well done Robert, I’m still trying to get the elephant…

Walter Sobchak
May 30, 2021 10:31 pm

Maybe the paper should be referenced as Wrong2021?

Izaak Walton
May 30, 2021 10:41 pm

You state that: “if CO2 levels actually were the control knob regulating the global temperature, we’d see all of the points falling on a nice straight line … but we don’t, far from it. There’s no statistically significant relationship between the temperature and the CO2 levels reported by Wong et al.”

However Wong et al. have an explicit equation (Eq. 1) that gives the relationship in their
model between CO2 concentrations and temperature. Importantly it also explicitly includes the effects of solar irradiation change and geophysical changes. So they don’t claim that CO2 is the control knob but only one of a number of effects. Hence nobody expects CO2 levels when plottted against temperature to fall on a “nice straight line”.

Reply to  Izaak Walton
May 31, 2021 12:49 am

Correct – as prior to industrialisation, the “control knob” was (largely) orbital eccentricity, with the carbon cycle shifting concentrations between land sea and air. No new carbon. Since circa 1850 man has added ~50% more to the atmosphere and an equal amount to the oceans. Now it’s the “control knob”.

Izaak Walton
Reply to  Willis Eschenbach
May 31, 2021 1:36 am

It makes sense in regards Wong et al.’s eq. 1. That equation says that temperature
depends on three difference functions of time. i.e.
Temp(t) = f1(t)+f2(t)+f3(t)
where only the first function describes CO2 as a function of time. What you did was plot f1(t) against Temp(t) and then you are surprised by the lack of correlation. When in fact nobody should expect to see any simple correlation between Temp(t) and f1(t).

Izaak Walton
Reply to  Willis Eschenbach
May 31, 2021 11:54 am

Nobody has ever claimed that climate sensitivity is as simple as you state. For starters the climate is clearly bistable with oscillations between an ice age state and a non ice age state. So “lambda” depends on temperature.

Secondly even if it was as simple as you suggest then you still would not expect a linear relationship between CO2 and temperature in Wong et al.’s paper since it is only one of three forcing terms in their Eq. 1.

Steve Fitzpatrick
Reply to  Willis Eschenbach
June 2, 2021 4:58 am

Yes, the implied “sensitivity to solar forcing” is very low. I think the assumed high climate sensitivity to CO2 as one parameter of the model, and a completely different (lower) parameter for sensitivity to solar forcing should have been understood by the authors and reviewers to suggest problems with the whole paper; energy should be pretty much fungible. Seems to me there is no reason for such a complicated approach as in Wong. Simple energy balance models the put the sensitivity near 1.8C to 2C per doubling of CO2. That is probably good enough that it is reasonable to discount most of the GCM’s as wildly wrong on sensitivity.

Reply to  Willis Eschenbach
May 31, 2021 7:45 am

It makes sense considering that about 50 times as much carbon in the form of carbonaceous ions is in the ocean as is in the atmosphere. . When the CO2 level finally declined enough, i.e. to about 900 ppmv, the level in the atmosphere followed the temperature of the oceans.

May 30, 2021 11:06 pm


Regards your proposed thunderstorm thermostat paper, this is the problem you will get when you submit it to any journal (see the video). So you have to cover all bases.

I had to argue for weeks that some of my paper had no references because it was A NEW THEORY. They did not understand the concept of new theories.

Short video by Tony Heller (aka Steven Goddard).


May 30, 2021 11:22 pm

Thanks Willis for this important article rebuttal.
The timescale back to 420 Mya nicely sidesteps the most problematic period for the notion of CO2 driven temperature: the end Ordovician (Saharan-Andean) glaciation during 460-440 Mya. CO2 levels increased during the inception of this glaciation and remained high throughout it.

Reply to  Hatter Eggburn
May 31, 2021 7:20 am

The common sense observation of the Andean/Saharan that the temperature not only plunged into an ice age but recovered from it. This rules out the usual excuse that the sun was weaker back then.

May 30, 2021 11:24 pm

The reason for dinosaurs during the Jurassic, is high CO2 concentrations. Because CO2 is plant-food.

More CO2 = larger plants = larger herbivores = larger carnivores.

And they are thin at one end, much thicker in the middle, and thin at the other end.


Steve Case
May 30, 2021 11:38 pm

A picture is worth a thousand words
Wikipedia gives credit to Henrik Ibsen

For some time now, I’ve been admiring the graphics that appear with each post here at WattsUpWithThat. Have the artists/editors ever been given credit? The control panel accompanying this post, just shy of 1500 words, said it all in one visualization . Did Willis slap that beauty together? Dunno, I might as well give him credit.

Steve Case
Reply to  Willis Eschenbach
May 31, 2021 1:50 am

Good Snatch (-:

May 31, 2021 1:03 am

Thanks for your diligence in showing the typical climate hyena

business as usual BS

now can I have my peanut for payment please

Steve Fitzpatrick
May 31, 2021 5:16 am

Hi Willis,
400 million years ago, solar output was about 4% lower than today, and the rate of increase since then supposed to be approximately linear. So maybe 13 watts per square meter higher (on average) at the top of the atmosphere over 400 million years. That is equal to over three times the canonical 3.7 watts/M^2 for a doubling of CO2. Do you know if this solar effect was considered in the paper?

Steve Fitzpatrick
Reply to  Steve Fitzpatrick
May 31, 2021 3:12 pm

I imagine that as a first approximation, the changing solar intensity could be converted to a equivalent (canonical) radiative forcing of 3.7 W/M^2 per doubling of CO2, diminishing the effect of ancient higher CO2. Eg, dropping from 7.4 watts per M^2 is approximately equal a 4 X increase in atmospheric CO2. IOW, 1,600 PPM CO2 200 million years ago would be expected to be not much different from today’s 400 PPM.
For what it is worth, I find these ‘arm wave’ papers are not very informative; quite the opposite.

May 31, 2021 5:49 am

I knew I saw one of these control boxes somewhere – it is a version of the City of London financial model, a variant of which is used by the FED. This very instrument was adjusted in 2008 when one of the dials, MBS, showed instability. The financial geniuses twiddling the controls did not even hear the roar of a crashing physical economy all around. The beta version was tested with the LTCM bubble in 1998. BlackRock is running the FED variant.

This time around they have an Everything Bubble control, and near it is a switch ¨Blame China¨. Someone is hitting that switch all the time.
Meanwhile :
China’s Banking Regulator Warns of Global Asset Bubble Risks – BNN Bloomberg
No wonder the climateers adopted financial control boxes. After all, the Great Reset (switch on the new model) Green New Deal requires a huge green bubble – trillion$ are at stake.
Trouble is, our very existence is at stake, if not tackled.

It is weird that the entire western world really has got to the point of believing the world economy, or climate, even the Big Bang, can be controlled by knobs and switches.
This can all be traced to mathematician Von Neumann´s Cybernetics in the 1950´s.

Cybernetics, from Greek kubernetes, to steer, i.e. the Helmsman. Looks like Mark Carney believes he is the Helmsman, steering the economy into a green swamp…

Last edited 1 year ago by bonbon
Tom Abbott
May 31, 2021 6:37 am

From the article: “So I gotta say, the data reported in the Wong2021 paper is a long, long way from establishing the claims made in their Abstract. In fact, even after they’ve carefully adjusted the tunable parameters of the GEOCARBSULF model in their favor, their results support the null hypothesis, which is that CO2 is not the global temperature control knob.”

This is what passes for science in Nature Magazine.

May 31, 2021 8:08 pm

So Wong is wrong I guess!

Michael S. Kelly
May 31, 2021 9:22 pm

Great post as usual, Mr. Eschenbach. It’s always refreshing to see a mind at work.

June 2, 2021 12:30 am

Willis, sorry to be so late to the party.
May I, gently, hoist you by your own petard? You complain that
 the authors are not conscientious enough to archive their results”
but neither do you publish the data for Fig. 2. I am sure someone could fit a nice curve to them with only 4 or 5 parameters…

June 3, 2021 7:29 am

Willis, I count 43 points in your scatter. Is there a reason you chose that degree of detail in digitizing the two graphs? Could have gone to, say, 100 points? Would that change your result at all?

June 3, 2021 12:02 pm

Gavin Shmshit? LOL. He should go back into his hole.

%d bloggers like this: