Guest Post by Willis Eschenbach

I see that a new study has been hyped with the headline:

Earth barreling toward ‘Hothouse’ state not seen in 50 million years,epic new climate record shows

I’ll spare you a link to the miscreants in question, it’s the usual alarmism. Here’s a sample.

“Now, in a new study published today (Sept. 10) in the journal Science, researchers have analyzed the chemical elements in thousands of foram samples to build the most detailed climate record of Earth ever — and it reveals just how dire our current climate situation is.”

“Dire” … here’s their alarmist graphic, showing the temperature since dinosaurs 67 million years ago (or “67 mya” as they say, which actually means 67 million years before 1950 … go figure).

Figure 1. Lead graphic from the article.

Y’all know me, I’m a data hound. Plus I don’t like science by press release. So I got the paper, “An astronomically dated record of Earth’s climate and its predictability over the last 66 million years”. It’s paywalled in Science magazine here. To get it I used the DOI and went through SciHub, but that’s just me. Here’s the money graph from their paper. 

Figure 2. ORIGINAL CAPTION: Fig. 1. Cenozoic Global Reference benthic foraminifer carbon and oxygen Isotope Dataset (CENOGRID) from ocean drilling core sites spanning the past 66 million years. Data are mostly generated by using benthic foraminifera tests of the taxa Cibicidoides and Nuttallides extracted from carbonate-rich deepsea sediments drilled during Ocean Drilling Program (ODP) and Integrated Ocean Drilling Program (IODP) expeditions. Genus-specific corrections were applied and oxygen isotope data adjusted by +0.64‰ and +0.4‰, respectively (12), with the green dot indicating the average oxygen isotope composition of the last 10 kyr. 

Average resolution for the interval from 0 to 34 Ma is one sample every 2 ky; for the interval from 34 to 67 Ma, it is one sample every 4.4 kyr. After binning, data were resampled and smoothed by a locally weighted function over 20 kyr (blue curve) and 1 Myr (red curve) to accentuate the different rhythms and trends in Earth’s carbon cycle and temperature operating on various time scales. 

Oxygen isotope data have been converted to average temperature differences with respect to today (13). Future projections for global temperature (44) in the year 2300 are shown by plotting three representative concentration pathways (RCP) scenarios (light blue, dark blue, and red dots). Gray horizontal bars mark rough estimates of ice volume in each hemisphere. Absolute ages for epochs and stages of the Cenozoic (GTS2012) and geomagnetic field reversals (this study) are provided for reference. 

The oxygen isotope data axis is reversed to reflect warmer temperatures at times of lower d18O values. Aqu, Aquitanian; Bur, Burdigalian; Cal, Calabrian; Cha, Chattian; Cret., Cretaceous; Dan, Danian; Gel, Gelasian; Ion, Ionian; K/Pg, Cretaceous/Paleogene boundary; Lan, Langhian; Lut, Lutetian; M2, first major glacial event in the NH; Maa, Maastrichtian; Mes, Messinian; NH, Northern Hemisphere; Oi-1, the first major glacial period in the Oligocene; Pia, Piacenzian; Pleist., Pleistocene; Plio., Pliocene; Pri, Priabonian; Rup, Rupelian; Sel, Selandian; Ser, Serravallian; SH, Southern Hemisphere; Tha, Thanetian; Tor, Tortonian; Ypr, Ypresian; Zan, Zanclean.

The first thing I noticed was that their main Figure was quite different from the figure in the press release. No fan-shaped predictions of the future. Hmmm, sez I … however, the figure used in the press release is indeed in the Supplemental Information.

Next, I never believe anything without running the numbers myself. So I went and got the datasets, they’re online here. The graph above is based on dataset S34.

There’s an oddity in Figures 1 & 2 just above. The 18O dataset is shown both in its original units (left-hand y-scale) and alternate units (right-hand y-scale). These alternate units are temperature anomaly in °C. Not only that, but according to Figure 1, the relationship between original and alternate units is linear. Who knew? Is it true? I have no idea, but I’ll take their data as read and see where it leads.

What I did with their data was to digitize the 18O data in Figure 2 above, using both the original and the alternate units. That let me back-calculate their conversion formula. Then I used their formula to convert their entire 18O dataset from the original units to temperature anomaly in °C.

Now that we have 67 million years of temperature data, allow me to get side-tractored for a moment to explain my interest and objective in analyzing the CENOGRID dataset. As I’ve mentioned before, I’m a heretic, which is different from a skeptic. A skeptic mostly challenges conclusions and methods. A heretic, on the other hand, questions the underlying assumptions. Or to put it another way, a skeptic doubts parts of things. A heretic disagrees with the fundamental assumptions that the whole edifice is based on.

The assumption at the base of this whole climate edifice is that all other variables somehow magically cancel out and that at the end of the story, the temperature of the earth is determined by variations in downwelling radiation (radiation headed towards the earth’s surface). In climate science, this downwelling radiation is known by a term-of-art as “radiative forcing”. This downwelling radiation is the sum of radiation from the sun plus the thermal radiation emitted by the clouds and the atmosphere.

In simple terms, the claim is that the change in global temperature is a function of the change in global forcing. Since changes in downwelling radiation are generally most related to changes in CO2 levels, I call this the “CO2 Roolz Temperature” theory, although of course there are other greenhouse gases (GHGs). That’s the claim driving the current insane war on CO2—the hypothesis that CO2 rules temperature, and that if we can simply drive CO2 levels lower, the world will be cooler.

And that is the claim I heretically and emphatically dispute—the idea that the long-term changes in temperature are a linear function of changes in CO2 forcing. 

Further, not only is the change in temperature said to be a function of the change in forcing. It is said to be a linear change, meaning that the change in temperature is equal to the change in forcing times some constant value. 

This constant is called the “climate sensitivity”, meaning how sensitive the temperature is to a change in forcing. Climate sensitivity is generally represented by the Greek letter lambda (λ). It is a positive number with a hotly-disputed value estimated to be somewhere between one-half and eight … here’s the history of the estimates.

Figure 3. Estimates of ECS (equilibrium climate sensitivity). Colors indicate what type of underlying data they are based on. Horizontal dashed lines show the canonical range of “climate sensitivity”, which is 1.5 – 4.5°C / 2xCO2. The units (°C/2xCO2) are degrees of warming which are said to result from a doubling of atmospheric CO2.

Now, consider that we’ve been working on this question of the value of the “climate sensitivity” for over forty years. In that time we’ve expended hundreds of thousands of person-hours, millions of dollars, and huge amounts of computer time on the question … and during that forty years, rather than either finding or even being able to narrow down the value of one dumb number … the uncertainty about the value of “climate sensitivity” has steadily increased.  

Increased! Our estimates of the “climate sensitivity” are becoming less certain rather than more certain!

During the period shown in Figure 3, entire new branches of science have sprung up, like DNA analysis and proteomics and femtosecond lasers and receptor biology, and have advanced to an unimaginable degree … and meanwhile, in climate science we’re even further from measuring the “climate sensitivity” than when we started 40 years ago. 

Climate models have gone from relatively simple and running on ordinary computers, to unimaginably complex and running on supercomputers. The computers themselves have thousands of times more speed and memory than they had forty years ago. Every year we’ve put more money and more work and more computer cycles into the question … and despite that huge investment, regarding the “climate sensitivity” we’re still increasing in uncertainty even today. 

And it’s even worse than the graph shows. The estimates from the latest “CMIP6” models are not shown in that graph, and their estimates of “climate sensitivity” are even wider in range.

To me, there is one and only one conclusion to draw from this obvious and hugely expensive failure—the underlying description of reality is wrong. 

That’s why I put “climate sensitivity” in quotes … I don’t think it actually exists. I think it is part of an incorrect understanding of climate.

And that is my heresy—I do not think that changes in temperature are a constant “climate sensitivity” times the changes in forcing. I think the climate operates in a very different manner.

To be sure, our inability to determine the so-called “climate sensitivity” doesn’t tell us what might control the temperature … but it clearly tells us it’s not the amount of downwelling radiation. 

And that means we need a new underlying description of how the climate works. I say the climate is not like an oven, where you turn up or down the GHG gases and the oven slavishly gets warmer and colder. Instead, I say it works as follows:

When something acts to drive the temperature of the planet either up or down, the climate system ACTIVELY RESPONDS to shift it back towards the status quo ante.

A good introduction to my idea of how the climate works is my post called Emergent Climate Phenomena. There is also an index of my posts on the subject here.

For the math folks in the crowd, the mathematical equation that the Earth is supposed to obey according to the generally accepted theory is

∆T = λ ∆F

where ∆ is “change in”, T is temperature, λ is climate sensitivity, and F is forcing.

I am unaware of any rigorous examination of evidence for this claim. I discuss its derivation here. It obviously is trivially true for say a block of steel.

But for complex systems like the climate or the human body, it may not be true at all. For example, if I walk out in the sun, the total forcing “F” changes by hundreds of W/m2, and despite that my core temperature “T” barely moves … what is my “climate sensitivity”? Near zero.

And if you don’t think that happens in the climate, here’s an example. CERES satellite data lets us calculate the relationship between downwelling radiation (∆F) and the surface temperature (∆T). And for most of the world, we find that indeed, they are strongly positively correlated—when forcing goes up, temperature goes up in a roughly linear fashion.

But in a large expanse of the equatorial oceans, we find that the exact opposite is happening. Temperature and forcing not only decouple, but they move in opposite directions—forcing is increasing as the temperature is decreasing, and vice versa.

Figure 4. Correlation of total surface absorption with total surface emission.

Perhaps someone can tell me … in the blue areas in the graph, where ∆T and ∆F are negatively correlated … just what is the climate sensitivity? Minus 1.5?

The widely held but incorrect idea that ∆T = λ ∆F may arise from the fact that as Figure 4 above shows, on land in the extratropics, ∆T is indeed highly positively correlated with ∆F (red areas in Figure 1), and that’s where most people live.

But the oceans and the tropics tell a different story. The average ocean correlation of forcing and temperature is only 0.44, half the average land correlation, and as mentioned, large areas are negatively correlated …

My friend Nic Lewis suggested that decoupling of forcing and temperature might be from a delay in the response. However, a cross-correlation analysis (not shown) of the area in blue above says that there’s no lag between downwelling surface radiation and the resulting heating.

This means that while forcing controls the temperature many places, in the blue areas the only conclusion possible is that the temperature is decoupled from the forcing … which kinda knocks a hole in the underlying equation that claims that forcing roolz temperature, no? …

So with that as prologue, let me return to the CENOGRID study. We’ve seen that it contains temperature information. The change in CO2 forcing varies as the logarithm of the CO2 concentration. So if we make a scatterplot of temperature versus the log of the CO2 forcing, we should get something resembling a straight line.

Fortunately, the CENOGRID study also contains a graph of the CO2 levels for the past 67 million years. It’s Figure S34 in the Supplemental Information.

Unfortunately, they didn’t put in a table or a link to the actual values. Grrr …

Fortunately, I’m a patient man, so I digitized their graph. That gave me the CO2 data I needed to compare with the geological temperatures shown in Figure 2. Remember, we’re expecting temperature versus the logarithm of CO2 to plot as a sloped straight line …

However, Figure 5 shows the actual situation with the CENOGRID temperature and CO2 data.

Figure 5. Scatterplot of temperature versus the log of carbon dioxide concentration. Blue dots are individual data points. Colored/Black lines are the million-year Loess average of both variables. Because of the long-term averaging, and because the data only go to 1950, the point marked “present” has lower temperatures and CO2 values than the actual 2020 situation.

Temperature vs. log CO2 should plot as a sloped straight line where both either go up or go down together? … I don’t think so. This kinda defines the concept of “non-linear”. And the claimed relationship between CO2 and temperature doesn’t seem to exist.

Consider, for example, the period of the first Warmhouse (orange) to the end of the Hothouse (red). During that time, the individual CO2 values (blue dots) went from ~ 260 to ~ 1200 ppmv, more than two doublings … but the temperature barely rose.

And the situation is worse in Coolhouse 1. Over the twenty million year period from 34 to 14 million years ago, the CO2 varied from about 290 to 800 ppmv, going up and down repeatedly within that range.

But over that twenty million years, the temperature basically didn’t vary much at all!.

In fact, that seems to be the general case. Each of the periods that they identify in their paper, from Hothouse down to Icehouse, contains times lasting millions of years when CO2 goes up and down, doubling and halving, but the temperature only varies minimally. Look at Figure 1.

Now, I thought that the hype was all in the press release. However, the authors themselves say:

If CO2 emissions continue unmitigated until 2100, as assumed for the RCP8.5 scenario, Earth’s climate system will be moved abruptly from the Icehouse into the Warmhouse or even Hothouse climate state. 

The climate “will be moved abruptly” … be afraid. Be very afraid …

First off, the RCP8.5 scenario is the hyper-alarmist scenario. It shouldn’t be used. But more to the point, even if these are separate stable climate states, where is the evidence that CO2 levels are driving the shift between the states? 

Because according to their own data, the shifts between the states don’t generally occur at the extreme CO2 values during that climate state. And in fact, the shift from Coolhouse 2 to Icehouse occurred at a higher CO2 level than the shift from Coolhouse 1 to the colder Coolhouse 2.

Next, suppose our CO2 were to go up to 700 ppmv … unlikely, but possible. In Figure 5, 700 ppmv of CO2 corresponds to three past temperatures— 5°, 10°, and 14° warmer than their “present” (average of last 10,000 years).

So which of their hothouses are they saying that we would end up in?

In short, I see absolutely nothing in their analysis to support their alarmist statement which is driving the media hype.

I must say, when I started out to look at this CENOGRID dataset, I did not expect to find persuasive evidence that CO2 is not the secret temperature knob controlling the temperature. But that is the clear conclusion from the CENOGRID study … nor is this a surprise in the larger sense.

I ask you … in what other complex physical flow system far from equilibrium is some typical steady-state parameter a simple function of only one of dozens of input and other variables? It makes no sense that all other variables would average out and only CO2 would make a difference to the global temperature.

TL;DR version?

The CENOGRID data says that changing atmospheric CO2 levels do not cause global temperature changes.

… or for the mathletes …

∆T ≠ λ ∆F

---

It’s night here on our patch of hill, six miles (10 km) from the Pacific Ocean and an hour and a half north of San Francisco. I just went outside, and it took my breath away. For the first time in a couple of weeks, I could see the stars.

The smoke here from the West Coast fires has been of various thicknesses and altitudes, but constant. One day last week it was so thick we had room lights on all day. When darkness came that night, not one of the solar path lamps had enough power to come on.

This recent week, the smoke has been up at a higher altitude. It’s been blown out over the ocean up by Oregon. From there it went out to sea and then circled back in. It has been riding on top of the “marine layer” as it’s been coming back into the coast here.

The California coast is funny. There’s a bone-cold deep-sea current that strikes the coast and upwells in a strip along the whole north coast that’s maybe 30 to 60 miles wide (50-100 km) from the shore outwards.

And there’s generally a warm moist wind that blows over the land from the warm blue north Pacific Ocean. But when the wind hits that strip of cold green water along the coast, the lowest layer of air cools way down. Being cold, it’s heavy. So when it hits the land it tends to hug the surface. And if the temperature is cool enough, the moist wind turns to fog.

This ground-level layer of cold air and fog is called the “marine layer”. It’s generally on the order of half-mile to a mile (a couple km) thick. And the Oregon smoke coming back in off the ocean is up above the marine layer here.

This has kept the worst of the smoke off of us, but the combination of smoke and fog has blocked the sun, moon, and most of all the stars for far too long.

So when I looked up tonight expecting gray, and I saw instead the Milky Way coruscating across the black velvet sky and Mars sitting on the celestial equator, I was lost in wonder at the stupendous glory of our magnificent eternity.

And I thought, “Indeed, the earth abides,” and my heart was eased.

My best regards to all, if you’re ordering the bat soup tell them to go easy on the bats, stay well,

w.

PS—As usual, I ask that when you comment, you quote the exact words that you are discussing. This avoids misunderstanding as to what and even who you are talking about.