A Request For Peer Preview

Guest Post by Willis Eschenbach

Well, for my sins I’ve been working on a paper with the hope of getting it published in a journal. Now that it’s nearly done, I realized that I have the worlds’ best peer-reviewers available on WUWT. So before seeing if I can get this published, I thought I’d take advantage of you good folks for some “peer preview”, to point out to me any problems you might see with the title, format, style, data, conclusions, or any other part of the following paper. All of the graphics are in grayscale because that’s what the journals want.

Many thanks for any and all contributions.


The Emergent Thermostat


The current paradigm of climate science is that the long-term change in global temperature is given by a constant called “climate sensitivity” times the change in downwelling radiation, called “radiative forcing”. However, despite over forty years of investigation, the uncertainty of the value of climate sensitivity has only increased.1 This lack of any progress in determining the most central value in the current paradigm strongly suggests that the paradigm itself is incorrect, that it is not an accurate description of reality. Here I propose a different climate paradigm, which is that a variety of emergent climate phenomena act in concert to keep the surface temperature within tight limits. This explains the unusual thermal stability of the climate system.


Several authors have analyzed the climate system as a heat engine. Here is Reis and Bejan’s description

The earth with its solar heat input, heat rejection, and wheels of atmo- spheric and oceanic circulation, is a heat engine without shaft: its maximized (but not ideal) mechanical power output cannot be delivered to an extraterrestrial system. Instead, the earth engine is destined to dissipate through air and water friction and other irreversibilities (e.g., heat leaks across finite ∆T) all the mechanical power that it produces. It does so by ‘‘spinning in its brake’’ the fastest that it can (hence the winds and the ocean currents, which proceed along easiest routes).2

When viewed as a heat engine, one of the most unusual and generally unremarked aspects is its astounding stability. Over the 20th Century, the global average surface temperature varied by less than one kelvin. This is a variation of ± 0.2%. Given that the system rejects a variable amount of incoming energy, with the variations mostly controlled by nothing more solid than clouds, this is a most surprising degree of stability.

This in turn strongly argues for some global thermoregulatory mechanism. The stability cannot be from simple thermal inertia, because the hemispheric land temperatures vary by ~ 20K over the year, and hemispheric sea temperatures vary by ~ 5K.


There is no generally accepted definition of emergence. In 1874 Lewes proposed the following definition: “Emergence: Theory according to which the combination of entities of a given level gives rise to a higher level entity whose properties are entirely new”.3

For the purposes of this article, I will define emergent climate phenomena functionally and by example.

Emergent climate phenomena arise spontaneously, often upon passing some thermal or other threshold. Consider the daily development of the tropical cumulus cloud field. Upon passing a temperature threshold, out of a clear sky hundreds of individual cumulus clouds can appear in a short time.

They have a time of emergence and a limited lifespan. Dust devils form spontaneously at a certain moment, persist for a while, and then dissipate and disappear.

They form a separate whole, distinct from the surroundings. Tropical thunderstorms are surrounded by clear air.

They are often mobile and move in unpredictable ways. As a result, tropical cyclones have “prediction cones” for where they might possibly go next, rather than being accurately predictable.

They are often associated with phase changes in the relevant fluids. Convective cloud emergence involves a phase change of water.

Once in existence, they can persist below the threshold necessary for their emergence. Rayleigh-Benard circulation requires a certain temperature difference to emerge, but once in existence, the circulation can persist at a smaller temperature difference.

They are flow systems far from equilibrium. As such, in accordance with the Constructal Law4, they must evolve and mutate to survive.

They are not naively predictable, as they have entirely different properties than the substrate from which they emerge. If you lived somewhere that there were never clouds, you likely would not predict that a giant white object might suddenly appear hundreds of meters above your head.

Examples of natural emergent phenomena with which we are familiar include the behavior of flocks of birds, vortices of all kinds, termite mounds, consciousness, and indeed, life itself. Familiar emergent climate phenomena include thunderstorms, tornadoes, Rayleigh-Bénard circulation of the atmosphere and ocean, clouds, cyclones, El Ninos, and dust devils.

A Simple Example

To explain how emergent phenomena thermoregulate the earth’s surface temperature, consider the lowly “dust devil”. As the sun heats a field in the summer, the change in temperature is some fairly linear function of the “forcing”, the downwelling solar radiation. This is in accord with the current paradigm. But when the hottest part of the field reaches a certain temperature with respect to the overlying atmospheric temperature, out of the clear sky a dust devil emerges. This cools the surface in several ways. First, it moves warm surface air upwards into the lower troposphere. Second, it increases sensible heat transfer, which is a roughly linear function of the air velocity over the surface. Third, it increases evaporation, which again is a roughly linear function of the surface air velocity.

At this point, the current paradigm that the change in temperature is a linear function of the change in forcing has broken down entirely. As the sunshine further irradiates the surface, instead of getting more temperature we get more dust devils. This puts a cap on the surface temperature. Note that this cap is not a function of forcing. The threshold is temperature-based, not forcing-based. As a result, it will not be affected by things like changing amounts of sunshine or variations in greenhouse gases.

A Complete Example

The heavy lifting of the thermoregulatory system, however, is not done by dust devils. It is achieved through variations in the timing and strength of the daily emergence of tropical cumulus fields and the ensuing tropical thunderstorms, particularly over the ocean. This involves the interaction of several different emergent phenomena

Here is the evolution of the day and night in the tropical ocean. The tropical ocean is where the majority of the sun’s energy enters the huge heat engine we call the climate. As a result, it is also where the major thermostatic mechanisms are located.

Figure 1. Daily emergent phenomena of the tropical ocean.

As seen in Panel “Early Morning”, at dawn, the atmosphere is stratified, with the coolest air nearest the surface. The nocturnal emergent Rayleigh-Bénard overturning of the ocean is coming to an end. The sun is free to heat the ocean. The air near the surface eddies randomly.

As the sun continues to heat the ocean, around ten or eleven o’clock in the morning a new circulation pattern emerges to replace the random atmospheric eddying. As soon as a critical temperature threshold is passed, local Rayleigh-Bénard-type circulation cells emerge everywhere. This is the first emergent transition, from random circulation to Rayleigh-Bénard circulation. These cells transport both heat and water vapor upwards.

By late morning, the Rayleigh-Bénard circulation is typically strong enough to raise the water vapor to the local lifting condensation level (LCL). At that altitude, the water vapor condenses into clouds as shown in Panel “Late Morning”.

This area-wide shift to an organized circulation pattern is not a change in feedback, nor is it related to forcing. It is a self-organized emergent phenomenon. It is threshold-based, meaning that it emerges spontaneously when a certain threshold is passed. In the wet tropics there’s plenty of water vapor, so the major variable in the threshold is the temperature. In addition, note that there are actually two distinct emergent phenomena in Panel 2—the Rayleigh-Bénard circulation which emerges prior to the cumulus formation, and which is enhanced and strengthened by the totally separate emergence of the clouds. We now have two changes of state involved as well, with evaporation from the surface and condensation and re-evaporation at altitude.

Under this new late-morning cumulus circulation regime, much less surface warming goes on. Part of the sunlight is reflected back to space, so less energy makes it into the system to begin with. Then the increasing surface wind due to the cumulus-based circulation pattern increases the evaporation, reducing the surface warming even more by moving latent heat up to the lifting condensation level.

The crucial issues here are the timing and strength of the emergence. If the ocean is a bit warmer, the new circulation regime starts earlier in the morning and it cuts down the total daily warming. On the other hand, if the ocean is cooler than usual, clear morning skies last later into the day, allowing increased warming. The system temperature is thus regulated both from overheating and excessive cooling by the time of onset of the regime change.

Consider the idea of “climate sensitivity” in this system, which is the sensitivity of surface temperature to forcing. The solar forcing is constantly increasing as the sun rises higher in the sky. In the morning before the onset of cumulus circulation, the sun comes through the clear atmosphere and rapidly warms the surface. So the thermal response is large, and the climate sensitivity is high.

After the onset of the cumulus regime, however, much of the sunlight is reflected back to space. Less sunlight remains to warm the ocean. In addition to reduced sunlight, there is increased evaporative cooling. Compared to the morning, the climate sensitivity is much lower.

So here we have two situations with very different climate sensitivities. In the early morning, climate sensitivity is high, and the temperature rises quickly with the increasing solar insolation. In the late morning, a regime change occurs to a situation with much lower climate sensitivity. Adding extra solar energy doesn’t raise the temperature anywhere near as fast as it did earlier.

At some point in the afternoon, there is a good chance that the cumulus circulation pattern is not enough to stop the continued surface temperature increase. If the temperature exceeds a certain higher threshold, as shown in Panel “Late Afternoon”, another complete regime shift takes place. The regime shift involves the spontaneous emergence of independently mobile heat engines called thunderstorms.

Thunderstorms are dual-fuel heat engines. They run on low-density air. That air rises and condenses out the moisture. The condensation releases heat that re-warms the air, which rises deep into the troposphere.

There are two ways the thunderstorms get low-density air. One is to heat the air. This is how a thunderstorm gets started, as a solar-driven phenomenon emerging from strong cumulus clouds. The sun plus GHG radiation combine to heat the surface, which then warms the air. The low-density air rises. When that circulation gets strong enough, thunderstorms start to form. Once the thunderstorm is started, the second fuel is added — water vapor. The more water vapor there is in the air, the lighter it becomes. The thunderstorm generates strong winds around its base. Evaporation is proportional to wind speed, so this greatly increases the local evaporation. This makes the air lighter and makes the air rise faster, which makes the thunderstorm stronger, which in turn increases the wind speed around the thunderstorm base. A thunderstorm is a regenerative system, much like a fire where part of the energy is used to power a bellows to make the fire burn even hotter. Once it is started, it is much harder to stop. This gives thunderstorms a unique ability that is not represented in any of the climate models. A thunderstorm is capable of driving the surface temperature well below the initiation temperature that was needed to get the thunderstorm started. It can run on into the evening, and often well into the night, on its combination of thermal and evaporation energy sources.

Thunderstorms function as heat pipes that transport warm air rapidly from the surface to the lifting condensation level where the moisture turns into clouds and rain, and from there to the upper atmosphere without interacting with the intervening greenhouse gases. The air and the energy it contains are moved to the upper troposphere hidden inside the cloud-shrouded thunderstorm tower, without being absorbed or hindered by GHGs on the way. Thunderstorms also cool the surface in a host of other ways, utilizing a combination of a standard refrigeration cycle with water as the working fluid, plus cold water returned from above, clear surrounding air allowing greater upwelling surface radiation, wind-driven evaporation, spray increasing evaporation area, albedo changes, and cold downwelling entrained air.

As with the onset of the cumulus circulation, the onset of thunderstorms occurs earlier on days when it is warmer, and it occurs later (and sometimes not at all) on days that are cooler than usual. Again, there is no way to assign an average climate sensitivity. The warmer it gets, the less each additional watt per meter warms the surface.

Once the sun sets, first the cumulus and then the thunderstorms decay and dissipate. In Panel 4, a final and again different regime emerges. The main feature of this regime is that during this time, the ocean radiates the general amount of energy that was absorbed during all of the other parts of the day.

During the nighttime, the surface is still receiving energy from the greenhouse gases. This has the effect of delaying the onset of oceanic overturning, and of reducing the rate of cooling. Note that the oceanic overturning is once again the emergent Rayleigh-Bénard circulation. Because there are fewer clouds, the ocean can radiate to space more freely. In addition, the overturning of the ocean constantly brings new water to the surface to radiate and cool. This increases the heat transfer across the interface. As with the previous thresholds, the timing of this final transition is temperature-dependent. Once a critical threshold is passed, oceanic overturning emerges. Stratification is replaced by circulation, bringing new water to radiate, cool, and sink. In this way, heat is removed, not just from the surface as during the day, but from the entire body of the upper layer of the ocean.


A theory is only as good as its predictions. From the above theoretical considerations we can predict the following:

Prediction 1. In warm areas of the ocean, clouds will act to cool the surface, and in cold areas they will act to warm the surface. This will be most pronounced above a temperature threshold at the warmest temperatures.

Evidence validating the first prediction.

Figure 2. Scatterplot, sea surface temperature (SST) versus surface cloud radiative effect. The more negative the data the greater the cooling.

As predicted, the clouds warm the surface when it is cold and cool it when it is warm, with the effect very pronounced above about 26°C – 27°C.

Prediction 2. In the tropical ocean, again above a certain temperature threshold, thunderstorms will increase very rapidly with increasing temperature.

Evidence validating the second prediction.

Since there is always plenty of water over the tropical ocean, and plenty of sunshine to drive them, thermally driven tropical thunderstorms will be a function of little more than temperature.

Figure 3. Cloud top altitude as a proxy for deep convective thunderstorms versus sea surface temperature.

As with clouds in general, there is a clear temperature threshold at about 26°C – 27°C, with a nearly vertical increase in thunderstorms above that threshold. This puts a very strong cap on increasing temperatures.

Prediction 3. Transient decreases in solar forcing such as those from eruptions will be counteracted by increased sunshine from tropical cumulus forming later in the day and less frequently. This means that after an initial decrease, incoming solar will go above the pre-eruption baseline until the status quo ante is re-established.

Evidence validating the third prediction.

Regarding the third prediction, my theory solves the following Pinatubo puzzle from Soden et al.5

“Beginning in 1994, additional anomalies in the satellite-observations of top-of-atmosphere absorbed solar radiation become evident, which are unrelated to the Mount Pinatubo eruption and therefore not reproduced in the model simulations. These anomalies are believed to stem from decadal-scale changes in the tropical circulation over the mid to late 1990’s [see J. Chenet al., Science 295, 838 (2002); and B.A. Wielicki et al., Science 295, 841 (2002], but their veracity remains the subject of debate. If real, their absence in the model simulations implies that discrepancies between the observed and model-simulated temperature anomalies, delayed 1 to 2 years by the climate system’s thermal inertia, may occur by the mid-1990s.”

Figure 4. Soden Figure 1, with original caption

However, this is a predictable result of the emergent thermostat theory. Here is the change in lower atmospheric temperature along with the ERBS data from Soden:

Figure 5. ERBE absorbed solar energy (top panel in Figure 4) and UAH lower tropospheric temperature (TLT). Both datasets include a lowess smoothing.

As predicted by the theory, the absorbed solar energy goes above the baseline until the lower troposphere temperature returns to its pre-eruption value. At that point, the increased intake of solar energy ceases and the system is back in its steady-state condition.

Prediction 4. The “climate sensitivity”, far from being a constant, will be found to be a function of temperature.

Evidence validating the fourth prediction.

Figure 6 below shows the 1° latitude by 1° longitude gridcell by gridcell relationship between net downwelling radiation at the surface and the surface temperature.

Figure 6. Scatterplot, CERES net downwelling surface radiation (net shortwave plus longwave) versus Berkeley Earth global surface temperature. The slope of the lowess smooth at any point is the “climate sensitivity” at that temperature, in °C per watt per square metre (W/M2)

The tight correlation between the surface temperature and the downwelling radiation confirms that this is a valid long-term relationship. This is especially true given that the two variables considered are from entirely different and unrelated datasets.

Note that the “climate sensitivity” is indeed a function of temperature, and that the climate sensitivity goes negative at the highest temperatures. It is also worth noting that almost nowhere on the planet does the long-term average temperature go above 30°C. This is further evidence of the existence of strong thermoregulatory mechanisms putting an effective cap on how hot the surface gets on average.

Prediction 5. In some areas, rather than the temperature being controlled by the downwelling surface radiation, the surface radiation will be found to be controlled by the temperature.

Evidence validating the fifth prediction.

Figure 7 below shows the correlation between net downwelling surface radiation (net shortwave plus longwave) and surface temperature. As expected, over most of the land masses the correlation is positive—as the downwelling radiation increases, so does the surface temperature.

Figure 7. Correlation between monthly surface temperatures and monthly surface downwelling radiation. Seasonal variations have been removed from both datasets.

However, over large areas of the tropical ocean, the temperature and downwelling surface radiation are negatively correlated. Since decreasing downwelling radiation cannot increase the surface temperature, the only possible conclusion is that in these areas, the increasing temperature modifies the number and nature of the overlying clouds in such a way to decrease the downwelling radiation.


1) The current climate paradigm, which is that in the long run, changes in global surface temperature are a simple linear function of changes in forcing (downwelling radiation), is incorrect. This is indicated by the inability of researchers to narrow the uncertainty of the central value of the paradigm, “climate sensitivity”, despite forty years of investigations, millions of dollars, billions of computer cycles, and millions of work-hours being thrown at the problem. It is also demonstrated by the graphs above which show that far from being a constant, the “climate sensitivity” is a function of temperature.

2) A most curious aspect of the climate system is its astounding stability. Despite being supported at tens of degrees warmer than the moon by nothing more stable than evanescent clouds, despite volcanic eruptions, despite changes in CO2 and other GHG forcings, despite great variations in aerosols and black carbon, over the 20th Century the temperature varied by only ±0.2%.

3) This amazing stability implies and indeed requires the existence of a very strong thermoregulation system.

4) My theory is that the thermoregulation is provided by a host of interacting emergent phenomena. These include Rayleigh-Benard circulation of the ocean and the atmosphere; dust devils; tropical thermally-driven cumulus cloud fields; thunderstorms; squall lines; cyclones; tornadoes; the La Nina pump moving tropical warm water to the poles and exposing cool underlying water; and the great changes in ocean circulation involved with the Pacific Decadal Oscillation, the North Atlantic Oscillation, and other oceanic cycles.

5) This implies that temperatures are unlikely to vary greatly from their current state because of variations in CO2, volcanoes, or other changing forcings. The thresholds for the various phenomena are temperature-based, not forcing-based. So variations in forcing will not affect them much. However, it also opens up a new question—what causes slow thermal drift in thermoregulated systems?


1 Knutti, R., Rugenstein, M. & Hegerl, G. Beyond equilibrium climate sensitivity. Nature Geosci 10, 727–736 (2017). https://doi.org/10.1038/ngeo3017

2 Lewes, G. H. (1874) in Emergence, Dictionnaire de la langue philosophique, Foulquié.

3 Reis, A. H., Bejan, A, Constructal theory of global circulation and climate, International Journal of Heat and Mass Transfer, Volume 49, Issues 11–12, 2006, Pages 1857-1875, https://doi.org/10.1016

4 Bejan, A, Reis, A. Heitor, Thermodynamic optimization of global circulation and climate, International Journal of Energy Research, Vol. 29, Is. 4, https://doi.org/10.1002/er.1058

5 Brian J. Soden et al., Global Cooling After the Eruption of Mount Pinatubo: A Test of Climate Feedback by Water Vapor,Science 26 Apr 2002, Vol. 296, Issue 5568, pp. 727-730, DOI: 10.1126/science.296.5568.727

Anyhow, that’s what I have to date. There are few references, because AFAIK nobody else is considering the idea that emergent phenomena act as a global thermostat. Anyone who knows of other references that might be relevant, please mention them.

Finally, any suggestions as to which journal might be willing to publish such a heretical view of climate science would be much appreciated.

My best to all, the beat goes on,


As Always: I can defend my own words, but I can’t defend your interpretation of them. So if you comment, please quote the exact words you are discussing so we can all understand what you are referring to.

4.8 28 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
April 28, 2021 10:01 am

I would publish on a pre-print service first, just to be sure. Also, I don’t understand journals wanting grayscale, I would submit everything in colour, I haven’t seen a journal wanting grayscale since the 90’s.

Kevin kilty
Reply to  Willis Eschenbach
April 28, 2021 11:26 am

Nature is quite a high bar, Willis. They slapped me and my five co-authors around hard on a submission some years ago and wouldn’t bother with us at all — told us don’t come back basically. Some of their objection was undoubtedly a stance that our lead author took that I advised against and that they disliked, but Nature, the editors and reviewers, will not consider certain topics or viewpoints. So, be prepared for a rejection is all I am saying. I see you are getting some suggestions for prior work below, which will aid you a lot.

I’ll have a read of this later, and pass on what I can. As I have said, I like this hypothesis of yours, but it seems somewhat similar to the IR iris thesis of Lindzen, which I am sure you know of.

Rud Istvan
Reply to  Kevin kilty
April 28, 2021 1:49 pm

Similar but not same. See my comment below. Lindzen argued via Tstorm indirect impact on amount of warming (hence IR) high cirrus formed from thunderhead moisture detrainment. WE is arguing direct via humidity washout and insolation/clouds

Gee Aye
Reply to  Kevin kilty
April 28, 2021 10:24 pm

Nature has door guards. Only a small percentage of submissions go to review with the subject specialist editors culling most papers.

Kevin kilty
Reply to  Gee Aye
April 29, 2021 8:00 am

I understand, but my warning is that hinting at certain things, especially things the editors dislike, is a sure way to get cancelled early. The best policy is often to let the argument and data speak for themselves. i.e. Willis’s discussion about future warming I would ixna…

Eric Vieira
Reply to  Kevin kilty
April 29, 2021 1:44 am

One important thing here is the claim that climate sensitivity is not a global constant, but is variable and locally dependent on temperature. That is new and quite important, since this “debunks” a lot of climate models which depend on a “fixed” ECS and gives an explanation (among a lot of other things) why they fail.

Kevin kilty
Reply to  Eric Vieira
April 29, 2021 8:06 am

This is a good point, but it needs a lot more development. A person could have looked at the pattern of warming two decade ago and found that it was largely confined to specific places like western North America. The paleo record probably shows the same. We came out of the last ice age in fits and starts, here and there, and the timing of the climb from the Little Ice Age seems dependent on region as well.

Reply to  Kevin kilty
April 30, 2021 7:53 am

Yes, different regions have their own internal storage of warm energy in water and some regions have their own internal storage of ice that cools by reflecting (albedo) and cools by ice thawing (ice was formed from IR out at an earlier time)
Willis has a good description of tropical cooling, but he does not mention the Polar cooling. Warm water flows to the polar regions and cold water flows back.

Reply to  Eric Vieira
April 30, 2021 7:48 am

One important thing here is the claim that climate sensitivity is not a global constant, but is variable and locally dependent on temperature ADD and dependent on water and ice. 

Reply to  Willis Eschenbach
April 29, 2021 6:42 am

Willis, have you thought about crowd-funding regarding the colour figures? Compared to Peter Ridd’s requirements this should hardly be a problem.

Rainer Facius

Reply to  Willis Eschenbach
April 29, 2021 9:36 am

Willis – Nature will NEVER take your paper.

a. They are snobby, and you lack their credentials and establishmentarianism.

b. You are proposing a theory outside the consensus, which is completely taboo.

They will just take your paper, and sit on it for a year – thus stopping you from applying to anyone else.


David L. Hagen(@hagendl)
Reply to  ralfellis
April 29, 2021 7:24 pm

Willis. Don’t presume they won’t. Your emergent phonomena is a major breakthrough.

Philip Rose
Reply to  David L. Hagen
May 1, 2021 1:52 am

Agree. Try to condense to letter form first, with just max 2 figures, balance supporting to come later. Should get through. Once wedge in crack will open and you will get full peer review. Important for WE to get his discovery (yes) out there first. Brilliant!

Reply to  Willis Eschenbach
April 29, 2021 11:01 am

I think everything is better in colour Willis (except “It’s a Wonderful life”) but I cant judge how much you are losing by the greyscale. Ae you able to publish here at WUWT figure2 which I guess would give us a good example as to whether greyscale in this case is inferior


Kevin kilty
Reply to  Pauleta
April 28, 2021 11:17 am

I can’t recall what the policy is for Nature, but I suspect they will not consider for publication a paper that has appeared elsewhere, and I imagine that includes pre-print.

Reply to  Kevin kilty
April 28, 2021 4:29 pm

Is there any progress since October last of the H&W 2020 pre-print paper finding saturation of CO2and water vapour in the atmosphere discussed here at the time?

Kevin kilty
Reply to  Herbert
April 29, 2021 8:09 am

I have not kept up with progress of the Happer and Wijngaarten paper. For me the preprint is generally all I have interest in — I just want to know the data and thinking behind something.

Reply to  Kevin kilty
April 29, 2021 1:22 pm

What people believe is not of any practical value. What happens is.

Pre print exists to allow the paper to get pre pub comment and protect the IP.

Pre print is not considered as a publication that then makes it the copyright property of the publisher in some contractual way. You CAN offer a pre pub for publication, I have, several times. It’s non consensual science so hard to place.

FYI I am advised by academics that it is common in academe to submit to several journals at once and simply lie about this, so you can withdraw if more than one accept. I also think this is reasonable, for all the obvious reasons of serialised delay, ……….. as long as the acceptance results in the other submissions being terminated.

Elsevier SSRN is where my two papers sit, being improved until I can find a publisher that is not scared of papers testing theories about the natural World that prefer the application of physical laws to test a definite theory by actual observations. Deterministic science 101. Surely not!

I have not had one suggestion/question from the pisting, but 200 reads for each. People seem more interested in publishing what they believe than reading what others have actually proven in the new “consensual science”

Lovelock’s GAIA included bio control. i suggest in my “test the limits macro approach” that you can burn the planet off and nothing much happens, the oceans have control.

What’s on the dry, low heat capacity land is largely irrelevant in the scale of things, as the Dinosaurs showed, the carbon just got recycled into new “bio diversity”, that nature just creates by accident but doesn’t care about and that constantly go extinct to be replaced by the new better adapted model. Evolving surface mold. Whatever works best at the time evolves to exploit the carbon and water, and by eating plants and each other.

Anyone of our rather too smart for its own good evolutionary pathway, looking in a quantified way at the planet as a joined up / coupled system of transfer functions, subected to positive and negative perturbations from the range of possible causes, can see it is not an open system sensitive to perturbation such as small changes in the lapse rate. That only happens with partial approaches that are simply wrong – for this very obvious reason that is somple to ddemonstarte on tye umbersscience knows..

It is self evident the oceanic response to SST change and its secondary effects on albedo dominate, and the lapse rate changes of whatever GHE there is are a tiny effect within one transfer function, controlled within the strongly fed back system.

Hope that helps somehow?

“1.1       Natural Feedback and Cyclic Control:In addition to the above causes and effects, extreme “surface” events, such as external asteroids or internal super volcanoes on land, primarily affect the low heat content atmosphere in the short term, and have not been of sufficient nor sustained effect to cause significant deviation from the long term climate cycles in the geological record. That is because the dominant control of planetary climate is the varying heat content of the oceans, 1,000 times that of the atmosphere, that cover 70% of the Earth’s surface and control global surface temperatures at sea and on land, and whose powerful negative feedback to SST change also limits the upper and lower bounds of the ice age cycles. The land modifies that dominant control to produce continental climate effects. The atmosphere’s primary function is as a medium to support the level of water vapour required to cool the ocean surface by evaporation and to transport the latent heat in the vapour to the Tropopause by convection, that then contributes to the formation of clouds as the vapour condenses. The clouds reduce solar insolation through the change in total cloud albedo. 
This oceanic response is the dominant negative feedback control of the Earth’s equilibrium heat balance. Currently this is at 100W m-2 of convective transport of evaporated latent heat [14], plus the albedo effect at 50W m-2[15] , of which the variability is estimated to be as high as 10% per deg K at the equator where it is highest. Another estimate from Roy Spencer published by John Christy suggests 2.6W m-2 deg-1 for the average global effect. This dominant response to changing SST is largest in the tropics, where the majority of the energy is both deposited by the Sun and given up by the oceans, the evaporative effect changing at around 10% per degree at 28 deg c.

1.1.1     Tropical Feedback As a Limiting ControlThe interglacial rise itself is observed to continue at a steady pace until arrested relatively suddenly. This is suggested to be due to the rapidly increasing negative feedback from the rising humidity in the tropical atmosphere, as Tropical oceans reach daytime SSTs of around 28 C, from the 23 C or less of the glacial phase. As a result the exponentially increasing power of tropical region’s powerful evaporative control rises to a level capable of minimising further increases from the cause of submarine volcanic heat. Achieving this saturated tropical climate at the equator, characterised by precipitation rather than temperature, appears to define the stable upper level of ice age cycles. The 4 degree higher polar temperatures observed in the Eemian are thus more likely to be the result of a tropical climate extending towards the poles to lose the excess heat at a greater rate, rather than from higher SST’s of above 28 deg K at the Equator. This is supported by the evidence of more exotic fauna over a wider latitudinal range during the Eemian, such the well known remains of Hippos, Lions and Elephants across Northern Europe dated to the last interglacial event. Also note the extra 4 degrees was well controlled in the record.”

Pre pub here, but this is the early version. http://dx.doi.org/10.2139/ssrn.3259379

John Tillman
April 28, 2021 10:13 am

This explains the unusual thermal stability of the climate system.

“Unusual” compared to what? Past climates? Climate on other planets? Possibly “marked” or some other term than “unusual”.

When viewed as a heat engine, one of the most unusual and generally unremarked aspects of the climate system is its astounding stability. 

Standard practice would include a relevant literature review section. Your references provide a starting point for such a survey.

Last edited 8 months ago by John Tillman
John Tillman
Reply to  Willis Eschenbach
April 28, 2021 10:50 am

Literature on climatic homeostasis is legion:


A source for some literature, to include textbooks as well as papers:


Last edited 8 months ago by John Tillman
John Tillman
Reply to  John Tillman
April 28, 2021 11:12 am

A goal of literature search is to argue for the originality of your hypothesis.

As far as you know it is, but that’s why a literature search is SOP, in order to ascertain that assertion.

Perhaps compare and contrast with Lindzen and other similar hypotheses:


Last edited 8 months ago by John Tillman
John Tillman
Reply to  Willis Eschenbach
April 28, 2021 12:56 pm

De nada. Good luck!

John Tillman
Reply to  Willis Eschenbach
April 28, 2021 11:09 am

Stable, warm temperature on centennial scale isn’t unexpected in an interglacial.

Again, unexpected by whom? Consensus “climate scientists”?

John Tillman
Reply to  Willis Eschenbach
April 28, 2021 12:55 pm

There are homeostatic mechanisms, for sure. But they’re not powerful enough to keep glaciations from happening. However within each climatic state, self-regulating factors operate.

Stability for centuries and millennia is but a blink of the climatic eye. Earth’s climate has been everything from global ocean of magma under a metallic atmosphere to iceball.

Andy May(@andymay2014)
Reply to  Willis Eschenbach
April 28, 2021 1:53 pm

HI Willis, excellent and thought-provoking post. I would definitely include the granddaddy of all journal articles on thermo-stability, Newell and Dopplick:
Newell, R., & Dopplick, T. (1979). Questions Concerning the Possible Influence of Anthropogenic CO2 on Atmospheric Temperature. J. Applied Meterology, 18, 822-825. Retrieved from http://journals.ametsoc.org/doi/pdf/10.1175/1520-0450(1979)018%3C0822%3AQCTPIO%3E2.0.CO%3B2

Robert W Turner
Reply to  Willis Eschenbach
April 28, 2021 2:12 pm

Leave out as much qualitative language as possible and replace it with simple quantitative statements.

This explains the unusual thermal stability of the climate system.
This explains the thermal stability of the climate system – ca. 1.5 degrees K for the past 10,000 years.

Reply to  Robert W Turner
April 29, 2021 2:16 pm

I was looking for this reply 🙂

Things like “lowly” dust devil are all unnecessary and the current version of the paper is full of this.

Also …references? You’re probably a couple of hundred short.

I don’t think the paper reads like a scientific paper as it is. It’s more an opinion piece and every argument you make must be backed by a prior scientific result to have scientific merit.

Reply to  Willis Eschenbach
April 29, 2021 11:59 am

We have already explained in an AMS presentation last January, what determines the long-term stability of global surface temperature of a planet using NASA data and Earth’s geological records. There are 2 factors defining the baseline (long-term) temperature of a planet: TOA solar irradiance (i.e. distance from the Sun) and total surface atmospheric pressure. As long as these factors are constant (or nearly constant), the baseline global surface temperature will remain stable. Watch this video for details:

Implications of a Semi-empirical Planetary Temperature Model for a New Understanding of Earth’s Paleoclimate History and Polar Amplification

Reply to  Willis Eschenbach
April 30, 2021 8:07 am

The stability of the climate system is real, it is not unusual, it is not unexpected, it is simply not understood or accepted by mainstream peer reviewed consensus.
I believe Willis has a best understanding of the Tropical Climate Stability.
IR out in polar regions produce sequestered ice. Tropical energy is transported to Polar Regions by warm tropical ocean currents. That energy powers the conversion of water to ice. Ice is sequestered on land in Polar Regions and high mountains. Ice spreads and reflects and thaws and provides additional cooling that is not considered by Willis.

Philip Rose
Reply to  Willis Eschenbach
May 1, 2021 2:07 am

Try “remarkable”, it’s why you,re writing. Mention the poles, as those are where the regulating currents are going.

Pat from Kerbob
April 28, 2021 10:27 am

Wish i could help, as I’m in violent agreement with your posts.
But sadly lacking such skills or background.

Reply to  Pat from Kerbob
April 28, 2021 11:40 am

I would only embarrass myself.

Pat from kerbob
Reply to  Hotscot
April 28, 2021 4:17 pm

My people!!

Reply to  Hotscot
April 29, 2021 8:07 am

I went ahead and embarrassed myself anyway. Clarity is always good, even if the muddiness is on my side.

April 28, 2021 10:31 am

Prediction 3. Transient decreases in solar forcing such as those from eruptions
On longer scales this would also apply to changes in CO2.

This would explain the finding some years ago by econometric analysis that CO2 effects were transient.

Reply to  Ferdberple
April 28, 2021 1:23 pm

“On longer scales this would also apply to changes in CO2.”

The one variable that Willis cannot measure is the slight change in emergent timing each day. Can satellite data determine if an exta 5 minutes of daily thunderstorms have happened over the past 50 years? In theory, imo, the global avrage temperature would rise just slightly, but no where what has been expected from our added CO2.

April 28, 2021 10:39 am

Prediction 4. The “climate sensitivity”, far from being a constant, will be found to be a function of temperature.
Prediction 3 is a special case of Prediction 4.

Reply to  Willis Eschenbach
April 28, 2021 3:48 pm

Willis ignore the “special case”. I misread prediction 4.

Joseph Zorzin
Reply to  Ferdberple
April 28, 2021 11:23 am

Possibly a function of many variables? I should think- not being a scientist- that more variables would result in more stability?? So, if one changes substantially, like “carbon pollution”- it’s effect is not on the same scale as its change?

April 28, 2021 10:45 am

I would say that ‘emergence’ is another way of saying ‘the devil is in the details’, meaning you can explain it after the fact, but not before the fact because higher level generalizations&approximations of theory&data smooth over those details.

April 28, 2021 10:50 am

Isn’t it merely that the properties of water with the fixed points at which it changes state and the latent heat involved what provides the underlying stability?

John Tillman
Reply to  Kalsel3294
April 28, 2021 11:23 am

Self-regulation is to be expected on a water world, yet Earth’s present climate system switches among different states, ie glacial maxima, stadials, interstadials and interglacials.

Robert W Turner
Reply to  John Tillman
April 28, 2021 2:20 pm

The climate system clearly exhibits stochastic resonance. Long periods of stability in an attractor state between short periods of rapid change to a new attractor state brought on by a resonance of noise and first order variables.

Reply to  John Tillman
April 30, 2021 9:10 am

The climate system does not change states. The climate system has internal responses and the perceived different states are just different phases in the loner term cycles. When much tropical warm water flows into the Arctic, Increased IR out producing sequestered ice on the continents initates ice ages and the the cold phase of the cycle follows while the ice spreads and thaws and depletes. More IR out from the Arctic cools the climate systems, later when the ice reflects and thaws.

Bill Treuren
Reply to  Kalsel3294
April 28, 2021 10:10 pm

That’s key, imagine a planet where all is identical other than atmospheric pressure is say double our world. From what I see the upper temperature achievable of the ocean would be higher than the notional 31C we tend to observe, or alternatively, a lower pressure atmosphere would yield a lower “maximum ocean temperature”.

The point being that it the equatorial ceiling temperature that sets the stage for the width of the Hadley cells etc.
This is a sensational paper but Willis it’s a tree falling in a forest with no ears. May my pessimism not be rewarded.

Reply to  Bill Treuren
April 29, 2021 2:22 am

If the planet imagined also has identical reserves of water, water vapour should vary the atmospheric pressure with the same effect it does here.

Reply to  Kalsel3294
April 30, 2021 9:24 am

The cooling depends on the circulation of the water. If the primary circulation is around the equator the planet would be warmer as earth was fifty million years ago. If tropical water is circulated in polar regions to thaw sea ice and get additional IR out and if there was land in the polar regions to sequester ice, it could be as we are now.

Chris Nisbet
April 28, 2021 10:51 am

Willis – I think we have a typo here…
“and millions of work-hours being throw at the problem”.
Should be ‘thrown’, not ‘throw’.

Last edited 8 months ago by Chris Nisbet
Gee Aye
Reply to  Chris Nisbet
April 28, 2021 10:34 pm

Actually better to remove that phrase altogether even if you have evidence to show that it is true (which you don’t). The amount of effort spent doing something, no matter how right or wrong, is irrelevant to the argument being made.

Gee Aye
Reply to  Willis Eschenbach
April 28, 2021 11:15 pm

Number of publications would be a proxy for hours

April 28, 2021 10:52 am

Remove personal pronouns. The royal “we” and “our” are preferred over “my”.

Joseph Zorzin
Reply to  Ferdberple
April 28, 2021 11:25 am

the royal consensus? :-}

Clyde Spencer
Reply to  Ferdberple
April 28, 2021 12:32 pm

That has been the style for publications in the past. However, that has been loosening up in recent years. It also varies with the publisher.

Reply to  Ferdberple
April 29, 2021 9:52 am

Most papers are collaborative efforts, so end up as being ‘we’.
I think Willis is working in solitary mode.

April 28, 2021 10:56 am

Thanks Willis,

Quick editorial comment. The showpiece scatterplot at the top of net downwelling vs surface temperature has the caption “The slope of the lowess smooth at any point is the climate sensitivity at that temperature”. Some people might take issue with reference to a “slope” at a “point”. Perhaps this wording could be revised.

To hammer home the concept of emergent phenomena reinforcing thermoregulation and system stability you could clarify what you propose is the primary cause of the system (temperature) stabilising at the current level. Is it an hydrostatic equilibrium conceptual framework? I realise this may be outside the scope of the paper.

Congrats on presenting the draft document.

Last edited 8 months ago by JCM
Reply to  JCM
April 28, 2021 11:27 am

The “slope at a point” of a curve is well defined if the curve satisfies certain requirements which, in non-mathematical lingo, ensure the curve is “smooth.” Construction of a lowess curve meets those conditions, and more.

Reply to  Willis Eschenbach
April 28, 2021 12:15 pm

While that may be technically true, I think the graph and caption is unclear about how I can use it to find climate sensitivity. Most readers will skim over most articles and if the showcase plot isn’t immediately clear it will be dismissed. my two cents. regards.

Last edited 8 months ago by JCM
Reply to  JCM
April 28, 2021 12:36 pm

Maybe i’m over analysing, but the more I look at it the more confused I am. All I see is that “net downwelling surface radiation” is some function of “surface temperature”, and vice versa. Where is the lamda “climate sensitivity” listed. It’s in the slope at a 1D point location? What is “net downwelling surface radiation”, exactly?. Maybe it’s just me over complicating this, I seem to be the only one having this problem.

Last edited 8 months ago by JCM
Reply to  JCM
April 28, 2021 1:08 pm

I see it’s in the text “net downwelling radiation at the surface”. So what I’m seeing is that for a change in “net downwelling radiation at the surface” there is an associated change in temperature. This is not measured at a point, it is a delta between two values. The figures should be labelled carefully and precisely. The graph shows that temperature no longer increases with increases to net downwelling radiation above 625 W/m2 or so.

Reply to  Willis Eschenbach
April 28, 2021 6:22 pm

ok I see what you’re getting at. I might suggest that in your paper you discuss a bit about CERES EBAF data and what it really is. It is satellite brightness temperatures transformed to flux by calibrating to GEOS-5 simulation. GEOS-5 relies on temperature and humidity outputs from GCMs, or at least it used to be.

Reply to  JCM
April 28, 2021 6:34 pm

In a nutshell, any satellite data does not measure flux directly. They merely measure the spatial variation of intensity of radiation or reflection from different wavelengths and polarities, usually expressed as a brightness temperature. This raw data must then be run through a model to output physically meaningful units and it’s not just a direct linear transform. This process should be considered in detail prior to the analysis in order to fully convey the limits of the data, uncertainties, and dependencies.

Last edited 8 months ago by JCM
April 28, 2021 10:58 am

You said: “As the sunshine further irradiates the surface, instead of getting more temperature we get more dust devils.”

I am not aware of any studies that actually show this.

John Tillman
Reply to  Leif Svalgaard
April 28, 2021 11:26 am
Reply to  Leif Svalgaard
April 28, 2021 11:31 am

perhaps ‘more heat’ but ‘higher temperature’ ?

Rud Istvan
Reply to  Leif Svalgaard
April 28, 2021 11:49 am

Willis, a suggestion. Drop dust devils and stick with thunderstorms. Two reasons. 1. They are familiar to all. Nobody in Florida knows what a dust devil is. 2. They cannot be modeled by climate models.

To do a good job on Tstorms, grid cells need to be 4km or less (doable with regional weather models out a few days). Due the computational constraints imposed by CFL, this is not possible to do in global climate models for a century—several orders of magnitude computational intractability. As a rule of thumb, NCAR says halving a gridcell x-y results in 10x the computation intensity thanks to CFL constraints on numeric solutions to,partial,differential equations. The typical CMIP5 resolution was 280x280km at the equator (my previous posts on models here). Just checked; the typical CMIP6 is 250x250km at the equator. Still off by ~six orders of magnitude on the biggest baddest superduperest super computers that exist today.
So Tstorms are parameterized. But as you nicely show, there is NO one Tstorm parameter since it varies during the day and by season.

Ben Vorlich
Reply to  Rud Istvan
April 28, 2021 12:45 pm

In parts of the UK these are known as Hay Devils. Presumably because they are more common in summer when there was dry hay to play with.


Charles Rotter(@jeeztheadmin)
Reply to  Rud Istvan
April 28, 2021 12:58 pm

Or call them Dry Waterspouts.

Robert Shirkie
Reply to  Charles Rotter
April 28, 2021 10:03 pm

I have called them “Saskatchewan Waterspouts” since I saw one stirring up alkali dust from the partly dessicated lake bed of Old Wive’s Lake during one of the droughts that periodically descends on the prairies.

Regarding the tropical thunderstorm system, when I fly over the Carribean, I watch thew with awe… so high and so turbulent. Any image of the Earth from space shows a line of them following the sun across the ocean. There might be a way to measure the difference in longitude from the sun’s position to a given width of cloud formation and compare any variation with sea surface temperature changes.

Reply to  Rud Istvan
April 28, 2021 1:27 pm

I agree: drop the dust devils.

Reply to  Willis Eschenbach
April 28, 2021 6:23 pm

I would think that anyone reading your paper would know what a dust devil is, even Floridians. I have a dust devil antidote. I once hiked to the top of Mt. Sacagawea near Bozeman Montana. Just as I got to the peak I saw a dust devil (it might not have been a big enough swirl to qualify as a dust devil in some official sense). If someone wanted to study dust devils, perhaps a mountain peak might be a good place to wait for them to form.

bruce ryan
Reply to  Willis Eschenbach
April 28, 2021 8:30 pm

Perhaps the term could be “thermals”? Certainly understood by pilots and bird watchers.

Reply to  Leif Svalgaard
April 28, 2021 4:33 pm

“Electric Dust Devils
The electrical character of dust devils and tornadoes is rarely mentioned. In fact, researchers only recently began to examine the electrical nature of dust devils in an effort to understand what is happening on Mars. Mysteries still surround electrical activity in our atmosphere. For example, the Earth has a vertical electric field, in the order of 100 volts per meter in dry air, whose origin is unknown. And scientists do not know what causes the most obvious electrical phenomenon in the atmosphere –’ lightning.”
Electric Dust Devils – holoscience.com | The ELECTRIC UNIVERSE®

Rud Istvan
Reply to  jmorpuss
April 28, 2021 5:42 pm

I know this is a mere comment detail. But with respect to lightening we have known the basic physics since Helmholtz elucidated it in 1888. It results from the Helmholtz double layer static dielectric effect (in Tstorms from static friction between any two phases of water — think carpet shuffling and doorknobs), and is the basis of all commercial EDLC capacitors, a $multibillion industry in which I have several very basic US (and Korea, and Japan, and Russia) issued storage materials patents.

Reply to  Rud Istvan
April 28, 2021 11:50 pm

And you Rud can explain how that electrostatic charge happens to be there in atmosphere.

I my self see it as very simple to explain, and maybe it could give WE some idea about his work.

Oh well you seem to have explain it… sorry, missed it!

It is simply due to the thermal flux and thermal currents.
The thermal gradient, “produces” an electrical gradient, due to the high energy exchange.

Sure, you know what electricity is!


Last edited 8 months ago by whiten
Reply to  whiten
April 29, 2021 1:42 pm

Oh, Rud still keeps believing of him being a good man… Hilarious… !!!

cheers, not!

Last edited 8 months ago by whiten
Reply to  whiten
April 29, 2021 2:00 pm

Willis, I hope for the best that you at least try to understand the concept of “vanity”…
The favored sin, as per Devils advocate.


Let it be without you playing the “prostitution game”… please… do consider!

Frack them Whores for everlasting!
Do not commit to their wrong, by validating their deceptive platform structure as with some merit in truth or reality… by some pleading or appeasement.

Well, still find kindness in your self, to
forgive such a blatant, direct request!

Don’t Fracking do it!

Respect to whatever you decide!


Reply to  Rud Istvan
April 28, 2021 4:17 pm

The Mars Dust Devils may add to the climate there since thunderstorms don’t exist.

Reply to  Leif Svalgaard
April 28, 2021 3:59 pm

I think what Willis mean is that : past a certain point, there will be no more temperature increase. ie. it will get to 45 degrees C, and then instead of rising to 46 or 47, there are dust devils created

April 28, 2021 10:59 am

The tight correlation between the surface temperature and the downwelling radiation
Isn’t this an expected result independent of emergent effects? Not sure this advance your theory

Reply to  Willis Eschenbach
April 28, 2021 4:02 pm

Yes, my eyes missed the negative correlation.

I would recommend blowing up the scale on the top right of the graph because it tends to get lost.

April 28, 2021 11:01 am

This is great Willis. I forwarded it to Adrian Bejan and others.

BTW, off topic, fountain of youth (TRIIM-X) trials ongoing.

April 28, 2021 11:11 am

One item I didn’t see was the percentage of the planetary energy budget subject to emergent phenomenon.

If it is 1% then it can be dismissed. If it is 50% then it seems easy to accept the effect cannot be ignored.

Rud Istvan
Reply to  Ferdberple
April 28, 2021 3:40 pm

Ferd, great suggestion, but maybe not reliably calculable because of uncertainties.
BUT, I went back to my analysis of Trenberth’s famous energy balance 2009 paper in essay Missing Heat in ebook Blowing Smoke. Added up all the stuff he claimed due to emergent phenomena (clouds, Tstorms, other emergent convection stuff…). Answer appears to be 206w/m^2 on 341. SO, up to 60%. QED using only their stuff. Willis, use as you may. Even Tstorm alone per Trenberth 2009, clearly labeled separately as such, is 80/341 or 24% by itself. NOT Trivial.
My essay critique of this paper/diagram was that the uncertainties summed to several times the net energy imbalance estimate, so utterly useless.

April 28, 2021 11:17 am

“the long-term change in global temperature is given by a constant called “climate sensitivity””
This is a long-running issue, but you need to define what you mean by climate sensitivity. They won’t (or shouldn’t) let you avoid that by just putting it in quotes. The ordinary, well-defined meaning, is equilibrium climate sensitivity, and that is not to be found by the methods used here. There is a defined notion of transient climate sensitivity, but in each such case you need to define an associated scenario – eg steady warming over 70 years.

It’s no use claiming that “climate sensitivity” is dependent on temperature unless you show how to quantify it.

Reply to  Nick Stokes
May 2, 2021 10:13 am

I think one challenge is that the usual “climate sensitivity” is global, whereas Willis is presenting a local version. This makes comparisons difficult.

I would also cite the Sherwood et al. 2020 paper on climate sensitivity somewhere (doi.org/10.1029/2019RG000678) as the most recent overview paper of the mainstream climate scientists. Note that the “standard” climate community approach does include the likelihood that some kinds of tropical clouds provide negative feedbacks, so I’d be curious about a more in-depth comparison of that.

Finally, it would be useful for your claim to discuss how the climate shifted between glacial and interglacial temperatures, and how global temperatures were substantially warmer before 3 million years ago. One issue is how tropical warming is balanced with extra-tropical warming – e.g., you can get a lot of “global” warming or cooling with very little tropical temperature change.

As an alternative to Nature, I might suggest Earth Systems Dynamics or some other similar journal. One benefit of ESD is the public, interactive review.

(as an aside: I think the analysis is really neat, I’m a little more dubious about the broader claims, so if I were writing this paper, I would make it much more targeted, but i recognize that’s not what you are interested in)

[fixed typo in email-mod]

Ron Long
April 28, 2021 11:21 am

Willis, I wonder what the effect is of lightening discharge in these tropical thunderstorms? The lightening bolts are 20,000 to 25,000 deg C, four or five times hotter than the surface of the sun. The internal circulation of a cumulo-nimbus cloud generates the accumulation of this energy, and it discharges either cloud-to-cloud or cloud to surface (some videos suggest surface to cloud?). Large tropical thunderstorms present an impressive lightening display for hours. The additional complication of more intermediate latitudes is the accompanying discharge of hail, which is truly and oxy-moron (or moron-moron?) as the 25,000 de C is accompanied by frozen water. Good luck with your publication.

Lance Wallace
April 28, 2021 11:21 am

I reviewed your document and put the word file on Dropbox.


If the link doesn’t work, I can send the Word document to your email if you provide it.

Lance Wallace

David L. Hagen(@hagendl)
Reply to  Willis Eschenbach
April 29, 2021 6:47 pm

Willis Please email draft doc to me at David L Hagen @ gmail dot com.

Rud Istvan
April 28, 2021 11:25 am

Willis, Lindzens ‘adaptive infrared iris’ published if I recall correctly in 2010 is an early emergent thermoregulatory hypothesis involving thunderstorm generated high cirrus. You probably want to include it. Judith Curry and I did a paired posting on it over at Climate Etc a few years back. Contains reference to the paper. (Just useher search tool on adaptive iris.) She interviewed Lindzen about the paper’s history, I commented on a then new climate model modification in Germany that incorporated the adaptive iris and resulted in a significantly lower ECS.

Reply to  Rud Istvan
April 28, 2021 12:24 pm

Rud, I agree. Lindzen’s Adaptive Iris theory is similar, but not identical.

Willis, I would concentrate on tropical thunderstorms since that is what can change the radiative budget of the planet and would be a negative feedback to an enhanced greenhouse effect. Or is it more than a negative feedback? Does it set an effective limit on how hot the plant can get? If CO2 causes warming, thunder storms just start earlier and reflect more sunlight, to counter act all of the greenhouse effect.

The same phenomena occurs on summer days in areas with a continental climate, so you should mention that it’s not just limited to the tropics. I remember eating an early dinner in a restaurant on the top floor of a skyscraper in Chicago in the summer in the mid 1980s. From that vantage point, I could see a large thunder storm pop up and move across the countryside.

You should consider asking Richard Lindzen to coauthor the paper with you. You still get the credit for the idea but it’s more likely to be published and he has lot’s of experience with writing papers for the climate community. I have Richard email address. I would be happy to share it with you. Charles can give you my email address.

By the way, relative humidity at ground level often goes down durning a thunderstorm. I think it is due to water vapor condensing on cold rain droplets, and to the fact that the falling rain brings air with it. That air was saturated at cloud level but it’s absolute humidity was low. When it mixes with warm air near the ground the result is drier air.

Last edited 8 months ago by Thomas
Reply to  Thomas
May 5, 2021 12:03 pm

“…I would concentrate on tropical thunderstorms since that is what can change the radiative budget of the planet and would be a negative feedback to an enhanced greenhouse effect. Or is it more than a negative feedback? Does it set an effective limit on how hot the plant can get? If CO2 causes warming, thunder storms just start earlier and reflect more sunlight, to counter act all of the greenhouse effect.”

What I read in the paper, the emergent phenomena make CO2 TOTALLY IRRELEVANT!!! How? He already said that, “Thunderstorms function as heat pipes that transport warm air rapidly from the surface to the lifting condensation level where the moisture turns into clouds and rain, and from there to the upper atmosphere without interacting with the intervening greenhouse gases. The air and the energy it contains are moved to the upper troposphere hidden inside the cloud-shrouded thunderstorm tower, without being absorbed or hindered by GHGs on the way.” In other words, a thunderstorm punches a hole right through the atmosphere, transporting the required amount of heat to the required places, regardless of the composition of that atmosphere!!! And Willis, I think you should emphasize that harder, louder, in bold, with italics, and a couple of foot stomps just for good measure.

I recall reading something in the last few months, right here on WUWT IIRC, explaining just how remarkable is the stability of this planet’s climate, and used as an illustration an internal combustion engine running to that steady state even under varying load! I think it mentioned the cruise control on my automobile as well. Was that your work, Willis? Seems to me the writing style I recall was more like Monkton. Gently plagiarize some quotes from that, use a few paragraphs worth but beef them up to standards for an academic paper, just to emphasize what a big deal this is!

Willis, I don’t mean to be discouraging, but based on the comments I have read so far it looks like this paper is about 20% complete with respect to being ready for submission. But I agree completely with this hypothesis, and I want to see it pursued to completion! Thanks for all your hard work!

Oh, one more thing, in Figure 6, Downwelling Radiation vs. Temperature, it is such a tight correlation just by Mark I eyeball, but wouldn’t it be even more impressive to show an r-squared?

John Tillman
Reply to  Rud Istvan
April 28, 2021 1:54 pm

Above I suggested discussing that hypothesis among other similar instances in a literature review section.

Rud Istvan
Reply to  Willis Eschenbach
April 28, 2021 5:48 pm

Willis, your point is part of what gets you ahead of Lindzen. See a more detailed mechanism comment below for details.

Bubba Cow
April 28, 2021 11:27 am

Hi, Willis.

Very nice and written well enough that I could follow along and envision the actions of your descriptions and predictions. No small feat, that – you are a talented writer; please keep it up.

I’m not a climate scientist (whatever that is), but I am a retired scientist in biomedical and I read a bunch in that arena.

a nitpick and a suggestion

nitpick – you have written everything in the 3rd person (excellent) except for 2 places in 1st person. Keep it all in 3rd person – shows objectivity rather than subjectivity.

Evidence validating the third prediction –
Regarding the third prediction, my theory solves – change to this theory …

4) My theory is that – change to the proposed theory is …

suggestion – have a few illustrations particularly of clouds and your hypotheses. Don’t leave it readers’ imagination – show and tell.

Best of luck.

Reply to  Bubba Cow
April 28, 2021 4:18 pm

Keep it all in 3rd person
Yes. Reads much better.

Larry Nelsn
April 28, 2021 11:29 am

As always fine work Willis. Here are my humble quibbles on a first quick read.

Reis and Bejan quote. Take the dash out of “atmo- spheric”.

In your Prediction 3, add the word “volcanic” so it reads ” those from volcanic eruptions”. My brain stuttered a moment while I figured out that you were referring to geologic rather than dermatological eruptions.

I stumbled on the phrase: “If you lived somewhere that there were never clouds”. It gains some grace if reads “If you lived in a cloudless place”.

Thanks for letting me nibble around the edges of a paper I am grateful you have chosen to prepare.

Paul Johnson
April 28, 2021 11:32 am

Could you include a plot of “climate sensitivity” versus temperature as described in Figure 6? You might also rename it “temperature response” to avoid confusion with CO2 effects.

April 28, 2021 11:37 am


I’m no scientists but I suggested some time ago this was the way science should be heading, for an open, public peer review system.

I guess it’s difficult for confidential matters or those with IP issues, but if most other science was peer reviewed publicly, perhaps it would free up the time of regular reviewers to provide more than a cursory glance at papers.

Don’t the Donnelly’s run an open type of peer review?

Gary Ashe
Reply to  Hotscot
April 29, 2021 5:39 pm

Yes they do, their papers are there.

H. D. Hoese
April 28, 2021 11:48 am

“Stratification is replaced by circulation, bringing new water to radiate, cool, and sink. In this way, heat is removed, not just from the surface as during the day, but from the entire body of the upper layer of the ocean.” I like that, many marine papers I read don’t know seem to know about nocturnal operations.

Also like this one. I think Reid taught me physical oceanography. DiMarco, S. F., M. K. Howard and R. O. Reid. 2000. Seasonal variation of wind-driven diurnal current cycling on the Texas-Louisiana continental shelf. Geophysical Research Letters. 7(7):1017-1020. 
Shelf currents have a strong onshore component most of the night, but offshore most of the day. It was not apparent from the wind.

H. D. Hoese
Reply to  Willis Eschenbach
April 28, 2021 6:28 pm

I know a now retired fisheries biologist who worked his way up in a federal fisheries lab from a helper, only had an junior high education. This was a relatively recent surprise to me when I asked him where he went to school. His name on important work. It’s not the impressiveness of your resume, but the quality of your work!

Gary Ashe
Reply to  Willis Eschenbach
April 29, 2021 5:46 pm

You figure them out quite well Willis, the actual difference is you notice the real world effects you experience, academic’s do not, the difference is the curiosity factor, you ponder your experiences, academics dont even notice the phenomena.

The snod-ification of academia will get you tho, as you say your lack of anything formal as per qualifications,

April 28, 2021 11:57 am

Wills, this is a keeper for me to discuss with youngsters the amazing observable weather phenomena that naturally occur, and the wider purposes and effects they produce.

Just 2 non-technical suggestions from me –
When you write

If you lived somewhere that there were never clouds, you likely would not predict that a giant white object might suddenly appear hundreds of meters above your head.

Could it be better phrased – “If you one lived somewhere that there were never clouds, you one likely would not predict that a giant white object might suddenly appear hundreds of meters above your one’s head.”

I say this only because you’re intending submitting to academic journals, where passive literary styles may be more in keeping than everyday narrative styles. ?

Also, is there any chance you could submit this work under a pseudonym, given the certain automatic denial of the probity of your work by “establishment” peer reviewers?

Reply to  Willis Eschenbach
April 28, 2021 5:28 pm

 . . . peer review restructuring would include double-blinding (neither the author nor the peer-reviewers identified during the review)

This should be an absolute, non-negotiable condition of any work being submitted for publication.

Biases just don’t get a co-pilot seat at the screen that way.

Reply to  Willis Eschenbach
April 28, 2021 9:27 pm

passive voice
Willis, academia loves verbs that end in “ing” because they make BS sound like fact.

The active voice is preferred when you have facts to convey and are prepared to back them.

Reply to  Ferdberple
April 29, 2021 8:37 am

“when you have facts to convey and are prepared to back them.”

THAT would be an anomaly, the journals wouldn’t know what to do! 🙂

Bill Parsons
Reply to  Ferdberple
April 30, 2021 2:38 pm

It isn’t just academics who love passive voice. Banks and other institutions like it because it defers responsibility: “It was found that an error occurred in your account.” (or) “An error has occurred in your account…” Instead of the more direct “Stan Smith made an error in calculating your account. He’s been fired and the error corrected.”

Passive voice is universally considered a weaker sentence structure than active. A commenter (above?) suggested a revision as follows:

“4) My theory is that the thermoregulation is provided by a host of interacting emergent phenomena.”

“my theory” yuck – instant rejection.

“A host” and “phenomena” unnecessary tautology


It is proposed that thermoregulation is provided by interacting phenomena.

Agree that Willis should get rid of the first person throughout his essay, but substituting a passive construction taking two steps forward and one step back. Simpler is better. Try:

“Thermoregulation is provided by…”

I’m not a science / math person, but I do read technical papers sometimes, and the ones that lard on the passive voice structures really do no credit to objectivity, style or direct speech.

Willis and several other writers on WUWT are able to build strong technical arguments with many difficult underlying concepts… and draw conclusions that logically follow. Keep it simple for everyone who reads, including the non-scientist. If that’s off-putting for the style police who don’t like clarity and directness, or who dislike conclusions that don’t support AGW alarmism, who cares? The object isn’t to butter them up, it’s to write well and create a good analysis.

Eric J
April 28, 2021 12:08 pm

Hi Willis –

I have a couple of comments about your excellent draft.
It’s been a while since I dealt with academic papers, but I was always impressed with my professors who used a deliberately understated tone or voice in their papers.

You say:

“This lack of any progress in determining the most central value in the current paradigm strongly suggests that the paradigm itself is incorrect, that it is not an accurate description of reality.”

Since this is the introduction, and you don’t want to lose your audience immediately, I suggest something more neutral like “This lack of any progress in determining the most central value in the current paradigm is a significant problem, limiting its utility and may even suggest that the paradigm itself is incorrect. I.e. that it is not an accurate description of reality.” Your later conclusions still convey the full force of the argument.

Prediction 4 graphic has the caption: “Figure 6. Scatterplot, CERES net downwelling surface radiation (net shortwave plus longwave) versus Berkeley Earth global surface temperature. The slope of the lowess smooth at any point is the “climate sensitivity” at that temperature, in °C per watt per square metre (W/M2)”

That is fascinating. I looked at the shape of the curve to try to estimate the slope at various points. I wonder if you could create a second, related graphic that actually is the slope, i.e. the derivative of the above figure. It doesn’t look to me like the “climate sensitivity” is in the same units as I’m used to seeing in the other paradigm, like “ECS range of 1.5-4.5”, and possibly there could be a scale that relates your data to the other method of presentation?

Further on, you say “Since decreasing downwelling radiation cannot increase the surface temperature, the only possible conclusion is that in these areas, the increasing temperature modifies the number and nature of the overlying clouds in such a way to decrease the downwelling radiation.” May I suggest “Since decreasing downwelling radiation cannot increase the surface temperature, the more likely alternative is that the increasing temperature modifies the number and nature of the overlying clouds in such a way to decrease the downwelling radiation.” When people tell me “the only possible conclusion” I always start looking for others. Stick to your argument.

These are just small nits in what I think is a clear and coherent presentation. Thank you.
I hope it gets some circulation and acceptance.

– Eric

Geoff Sherrington
Reply to  Willis Eschenbach
April 28, 2021 5:08 pm

Woerth including? That is debatable and troublesome.
Your graph shows climate sensitivity reaching zero. Many people have been dogmatic that sensitivity cannot be zero, so you risk having some of them turning off and stopping their reading of this rather important paper. Geoff S

Rud Istvan
Reply to  Geoff Sherrington
April 28, 2021 5:52 pm

IF thermoregulation is real, then forcing sensitivity MUST reach zero.
Logically necessary to include.

Reply to  Rud Istvan
April 28, 2021 9:07 pm

Perhaps “asymptotic to zero” is more exact. Each successive increment in forcing may be countered more and more by the various emergent dynamics Willis discusses (eg the thunderstorms and the ocean layer overturning).

It would be an empirical issue as to whether the emergent mechanisms could “take on a life of their own” and once triggered by warming, in the end, produce net cooling,

My sense is that happens locally (eg the cooled ocean in the wake of the recent hurricane), but over time and space is not sustained for a global system.

Reply to  kwinterkorn
May 5, 2021 12:17 pm

Well, I’m with the people that say, “Calculating an average temperature of the globe is nonsense.”, and as such you cannot calculate an average climate sensitivity, neither ECS nor TCS. This graph just hammers home that point.

David L. Hagen(@hagendl)
Reply to  Rud Istvan
May 11, 2021 5:25 am

More importantly, Willis shows that the derivative (slope, forcing sensitivity) GOES NEGATIVE at the highest temperatures, (not just approaches zero). That appears to correspond with Richard Lindzen’s “Iris Effect”. Lindzen, R.S. and Choi, Y.S., 2021. The Iris Effect: A Review. Asia-Pacific Journal of Atmospheric Sciences, pp.1-10. https://link.springer.com/article/10.1007/s13143-021-00238-1

Eric J
Reply to  Willis Eschenbach
April 29, 2021 5:49 am

I thought it would be worth including because it helps clarify the temperature/feedback relationship that is your central point. Now I’m not sure. There were some comments that pointed out the Tropics vs Global issue and the question of negative feedback on a global scale, and others that spoke about ambiguous definitions for climate sensitivity, ECS/TCS/xCS.

I come to believe that any mention of “climate sensitivity” will detract from your effort. It will lead to endless debates about whether your definition of climate sensitivity matches theirs, which one is correct, and who doesn’t understand the other. This isn’t central to your theory. You don’t need/want an ECS figure anyway. It can’t be calculated because the feedbacks are temperature-dependent and therefore do not lend themselves to a scalar value. Maybe a second-degree polynomial function, best-fit to your data, could give an approximate effective feedback value for a given temperature (and location), with no claim that this represents a global equilibrium anything.

I would therefore remove the mention of climate sensitivity from the caption of Figure 6 (and everywhere else). Maybe just invent/define a new term to stay away from the inevitable word games that will result from using that old term in this new way.

Eric J
Reply to  Willis Eschenbach
April 29, 2021 7:26 am

Second comment/reply. The larger problem is the difference is overall conception of the climate system.

“Climate sensitivity” as a unitary global figure depends on the idea that the energy balance is unaffected by emergent phenomena (and most other things besides downwelling energy and GHG).

Your hypothesis on the other hand posits that as the earth warms these emergent systems regulate the max, and ultimately the average, temperature. They don’t appear at the same times, or at the same places. And the frequency/duration of their appearances will increase along with temperature. The heat pump mechanisms of the Earth (wind/currents) will then distribute the heat more uniformly over time. Your estimate of “climate sensitivity” would necessarily approach 1.0, by definition. The other paradigm must be higher. These are two different world views and the term climate sensitivity can’t mean the same thing in both. I recommend you not buy into the old paradigm and avoid the term and the old concepts entirely.

Reply to  Willis Eschenbach
May 5, 2021 12:14 pm

Yes, I think you should include this.

Reply to  Red94ViperRT10
May 5, 2021 1:16 pm


Let me make a complete 180 on this. Do not include this. Furthermore, as someone else said, remove all references to λ from any graph you produce for this paper. Why? Because as I already noted elsewhere, this paper shows that the AGW wishcasting as currently framed, a change in temperature resulting from a change in GHG concentrations, does not exist! So the climate sensitivity you talk about at the beginning of your paper, and as laid out in the references, is the change in temperature resulting from a change in atmospheric CO2 concentration. And as such, it’s irrelevant.

But most importantly, the Figure 6 graph from whence you calculated this “sensitivity” does not even include CO2 concentration, so that sensitivity cannot be obtained from taking a derivative. What’s in the graph? Net downwelling surface radiation is on the x-axis, but that may or may not depend on the concentration of CO2 in the atmosphere, as far as I know there is not research, data or proof as to how much the downwelling surface radiation changes with respect to a change in CO2 concentration. And the y-axis shows the resultant temperature at each given surface radiation value. So the “sensitivity” you have graphed via taking the derivative shows a different sensitivity.

Now, you can still leave your discussion of sensitivity at the beginning of the paper (though make it a little less confrontational as someone else suggested), but the rest of the paper then lays out the proof that that sensitivity is irrelevant to (maximum) surface temperatures, at least over substantial bodies of water.

Thought question… doesn’t this emergent phenomena also dictate that the temperature within the eye of ALL hurricanes, regardless of location on the globe or time of day, is always the same? Right at the limit which instigates this particular emergent phenomena (I’m going with, a hurricane is just an organized accumulation or gathering of multiple thunderstorms.) Right?

April 28, 2021 12:13 pm

4) My theory is that the thermoregulation ….

perhaps the less subjective
“Proposed hypothesis is that the thermoregulation ….”

April 28, 2021 12:32 pm



Robert of Texas
April 28, 2021 12:34 pm

Good read, a few thoughts to consider:

1) The emergent phenomenon you describe applies mostly to large bodies of water – landscapes may lack the necessary water or have significant interfering mountains that complicate the system you describe.

2) “As the sun continues to heat the ocean, around ten or eleven o’clock in the morning a new circulation pattern emerges to replace the random atmospheric eddying. As soon as a critical temperature threshold is passed, local Rayleigh-Bénard-type circulation cells emerge everywhere.”

I was under the impression that large-scale cloud formations occurred mostly in bands over the oceans, so again a complication that isn’t addressed. (or I am wrong)

3) There is no mention of angle of incidence other then the Sun “rising” from morning to noon – latitude also plays a big part. It seems to me that the angle that the Sunlight hits clouds (especially ice particles) would play a part in the amount of light reflected to Earth below.

4) The ocean serves to buffer temperature changes in other ways – in the morning it takes up heat and starting around noon to evening it releases extra heat. This buffering of heat is seldom ever mentioned but helps result in a more moderate climate. Rocks and soil do the same – this can lengthen the amount of time that water is being used to transport heat high up.

Maybe all of this is tangential to the purpose of your paper, but it would show that you considered such ideas rather than didn’t think of them.

Roger Taguchi
April 28, 2021 12:36 pm

Conclusion number five (5), specifically where you say ” or other changing forcings” is flat out wrong. You ask the question: “what causes slow thermal drift in thermoregulated systems?” which has already been answered by Milankovitch: https://earthobservatory.nasa.gov/features/Milankovitch You can see the effect in the ice core record due to the minuscule change in radiative forcing of ~0.45 W/m2

Roger Taguchi
Reply to  Roger Taguchi
April 28, 2021 12:38 pm

In other words Willis, you hypothesis fails to regulate the climatic effect of very small changes in TSI due to orbital variations.
You also need a different word than “emergent” because the subject phenomena of your hypothesis have been around for millions of years, and have failed to prevent the advance and retreat of glaciers. If they “regulate” the climate, they don’t do a very good job of it.

Last edited 8 months ago by Roger Taguchi
Fred Souder
Reply to  Roger Taguchi
April 28, 2021 1:14 pm

Small changes in TSI may cause shifts in storm tracks. Over time, all that is needed is a small place (like in northern Quebec) to accumulate a few more inches of winter snow than melt in the summer. 5 inches of ice per year? Over 1000 years, you now have an expanding region of 1/2 mile high ice, which will now accumulate ice year round – because it has a surface elevation of 2000 ft, which can then more quickly reach 10,000 feet, the pressure of which spreads the ice out and south. I would argue that it would take a major emergent phenomena to melt all this, causing our brief interglacials.

John Tillman
Reply to  Roger Taguchi
April 28, 2021 1:59 pm

Most Milankovitch cycles don’t rely on change in TSI, but in insolation due to orbital and rotational mechanics.

Roger Taguchi
Reply to  John Tillman
April 28, 2021 3:28 pm

Jut to be clear, it is the change in orbit that causes the TSI to change.

Reply to  Roger Taguchi
April 28, 2021 1:17 pm

Where did you come up with that crap, Taguchi? The total solar insolation at 65° N (where all the land is) varies from 450 W/m^2 to 550 W/m^2 over the course of the Milankovitch cycles.

Roger Taguchi
Reply to  Meab
April 28, 2021 3:29 pm

And tell us Mr. Meab, when 65N increases, doesn’t 65S decrease?

Reply to  Roger Taguchi
April 28, 2021 4:00 pm

Hai Taguchi-san, but it is the 450 to 550 north of 65° that causes ice formation north of 65°.

Fred Souder
Reply to  Thomas
April 28, 2021 7:27 pm

Maybe. I would argue that it is the increased local precipitation more than the subtle change in insolation which causes the ice to build up. It would not have to be much, because we are talking thousands of years to accumulate.

Reply to  Fred Souder
April 29, 2021 7:07 am

Fred, Very good point. There was a theory published in the 1970s that posited an ice free Arctic ocean would cause more precipitation on land, and the gradual accumulation of snow, which would eventually become glaciers.

Reply to  Roger Taguchi
April 28, 2021 11:27 pm

No, Taguchi. When the orbital eccentricity elongates, the Earth spends more time near aphelion and less time at perihelion (Kepler figured this out 400 years ago). Winters in one hemisphere will be much longer and much colder and summers will be much hotter but also much *shorter*. The percent change in insolation is about 4 * eccentricity. Presently, eccentricity is about .017 so the Earth receives about 6.8% more insolation during the Southern Hemisphere summer (because the Earth is closer to the Sun during SH summer at present). The Earth’s eccentricity gets as high as .0679 so the difference between the hemisphere’s insolation will rise to about 27%. That’s huge.

The hemispheres will still be out of phase BUT insolation over land areas counts the most in warming the Earth. 2/3 of the Earth’s land area is in the Northern Hemisphere – that’s why the global average temperature actually goes up in present times during the NH summer despite the fact that the Earth is closest to the sun during the SH summer. So, when the Earth’s axial precession puts NH winter at perihelion at a time with maximum orbital eccentricity, the whole Earth cools dramatically – enough for a global glacial advance as opposed to just a NH glaciation which can happen during other Milankovitch cycle conditions.

Taguchi, the impact of the Milankovitch cycles does NOT result from the tiny change in radiative forcing that you falsely alluded to.

Roger Taguchi
Reply to  meab
April 29, 2021 8:29 am

You are wrong Meab. When the orbit elongates, the TIME spent at the DISTANT aphelion increases. Total energy drops because it is TSI times time. The TSI increases at perihelion, but since the time drops, the amount of energy decreases. It is an unbalanced difference.

Last edited 8 months ago by Roger Taguchi
Reply to  Roger Taguchi
April 29, 2021 4:48 pm

That’s what I wrote, learn to read. Quoting ” When the orbital eccentricity elongates, the Earth spends more time near aphelion and less time at perihelion (Kepler figured this out 400 years ago).”

You are pulling stuff out of your nether regions. It’s NOT the global TSI that creates glaciation, it’s the TSI over land. You know, where grounded glaciers grow. Sheesh.

Roger Taguchi
Reply to  Meab
April 29, 2021 5:39 pm

You post: “The total solar insolation at 65° N (where all the land is) varies from 450 W/m^2 to 550 W/m^2 over the course of the Milankovitch cycles.”
Ignoring what happens in the Southern Hemisphere where MORE ocean area with a lower albedo than land areas.

April 28, 2021 12:58 pm

I certainly root for this to be published in the most distinguished journals in existence. Failing that, how about aiming for popularity, in household magazines, trade journals, hobby mags etcetera? THEY will not easily give this exposure, but the battle is for the heart and mind of the “common man”, so publish where the “common man” reads.
One caveat; the graphs may need some more, er, colloquial captioning for such an audience, but the information as is should be comprehensible to most folk, given a little more background. (How does a model work, and how that ‘senstivity’ thing fits into it. One or two paragraphs?)
For this particular audience, this is near perfect, far’s my limited opinion matters.
P.S. Colour them pics, then make’em move. Apparently that is how today’s kids ‘read’.

Larry Brown
Reply to  paranoid goy
April 28, 2021 2:07 pm

Hi Willis – Great article – my suggestions are grammatical/clarity related only
You say “They are not naively predictable, as they have entirely different properties than the substrate from which they emerge. If you lived somewhere that there were never clouds,” – I suggest you say “If you lived somewhere where clouds did not exist — “

You say “The sun plus GHG radiation combine to heat the surface, which then warms the air. The low-density air rises.” – I suggest you define GHG here – then you don’t have to do it again

You say, “Thunderstorms function as heat pipes that transport warm air rapidly from the surface to the lifting condensation level where the moisture turns into clouds and rain, and from there to the upper atmosphere without interacting with the intervening greenhouse gases.”  — here you say greenhouse gasses – if you defined GHG earlier it is proper to here just say GHG.  Nitpicky stuff but ——
You say “Prediction 4. The “climate sensitivity”, far from being a constant, will be found to be a function of temperature.” I suggest you change to say “The “climate sensitivity”, far from being a constant, will be a function of temperature.”
You say “Prediction 5. In some areas, rather than the temperature being controlled by the downwelling surface radiation, the surface radiation will be found to be controlled by the temperature.” – same suggestion as above – recommend rewriting to say, “In some areas, rather than the temperature being controlled by the downwelling surface radiation, the surface radiation will be controlled by the temperature.”

April 28, 2021 1:10 pm

Refreshing to see old fashioned predictions.

April 28, 2021 1:17 pm

I heartily approve of your effort to publish, and I think that I understand what you have written. It is refreshing to see a reasonable hypothesis and theory for this complex system.
Along the lines of the Terry Pratchett character, however, to call me a peer is to call for more sawdust on the floor. In my sordid past, I have done engineering technical writing and editing, and have some suggestions along those lines not suited to this comment format.
My suggestions may not be appropriate for where and how you may intend to publish. Let me know, here, if and/or how I should send these to you, and correspond with you in a different way.

April 28, 2021 1:20 pm

Willis, as one of your biggest fans, I hate to sound brutal on this, but believe me, the reviewers will be even more brutal, and forget Nature, even if you submit as Eschen Willisbach. Those immature, phony-scientists are still trying to get out of their political short pants since leaving university. I wish I had more time to work on this as I know I can help. In fact, this month I, and my colleagues, just had two papers accepted in seriously good journals after some fairly healthy reviewer critiques. I think the content is great but it’s just not written in Journal style and, as such, makes it a bit difficult (for me) to follow.

I just picked a paper that I thought was well written (when I first read it), for PNAS, that you could use as a template, whether or not you agree with the content: Feldman et al. (2014).


I think if you rewrote the paper in that style, it would look fantastic.

If you’re not pissed off at me, I could help further, just not this week.

Reply to  Willis Eschenbach
April 28, 2021 8:23 pm

I knew you wouldn’t be but, you know, how beautiful one’s baby is can sometimes be a sensitive subject. I actually do get pissed off at some reviewer’s comments if they’ve obviously been a bit lazy.

Steve Z
April 28, 2021 1:21 pm

Overall, an excellent paper. But there is one nit to pick:

[QUOTE FROM PAPER]”Note that the “climate sensitivity” is indeed a function of temperature, and that the climate sensitivity goes negative at the highest temperatures. It is also worth noting that almost nowhere on the planet does the long-term average temperature go above 30°C. This is further evidence of the existence of strong thermoregulatory mechanisms putting an effective cap on how hot the surface gets on average.”[END QUOTE]

Since most of the paper deals with the formation of thunderstorms over tropical oceans, which tend to prevent them from becoming too hot, the statement about “long-term average temperatures” never going above 30 C should be limited to areas over or near oceans, not anywhere “on the planet”, which can include some desert areas (Sahara or parts of southwestern USA) where average daily temperatures can exceed 30 C. Even over some bodies of water nearly surrounded by land, such as the Mediterranean Sea or Gulf of Mexico, average daily temperatures can exceed 30 C in summer.

Also, what is meant by a “long-term” average temperature? A diurnal average? A monthly average? A yearly average? The time period of the average should be specified.

Rick C
April 28, 2021 1:23 pm

Maybe a nit pick – on dust devils – “First, it moves warm surface air upwards into the lower troposphere.” The surface air is already in the lower troposphere isn’t it? So “in” instead of “into”.

Figure 5. label vertical dashed line “Pinatubo”.

Izaak Walton
April 28, 2021 1:31 pm

Well for starters I think that most climate scientists would disagree with the first statement that
The current paradigm of climate science is that the long-term change in global temperature is given by a constant called “climate sensitivity” times the change in downwelling radiation, called “radiative forcing”. 

The parameter or climate sensitivity is something that emerges from different climate models as a simple number that allows people to hide all the messy details. For example in your model there is still a climate sensitivity parameter only you are claiming that it is zero.

Secondly you appear to be confusing global and local effects. The climate sensitivity parameter as used is an average over the entire globe. So your discussion after prediction 4 is missing the point entirely. At best you would need to average your climate sensitivity at each grid cell to come up with an average climate sensitivity for the globe and then look to see if that global average varied with average global temperate.

In addition to which climate scientists already know that climate sensitivity varies with temperature. For example in Roe’s “Feedbacks, Timescales and Seeing Red” the climate
sensitivity parameter ignoring feedbacks is just the derivative of the Stefan-Boltzmann law with respect to temperature giving (his Eq. 3) 1/(sigma T^3).

There are also fairly fundamental issues regarding your model. Firstly the question arises as to timescales over which your emergent effects work. You describe a series of events such as thunderstorms, dust devils etc all of which emerge over the course of a day and so can only operate on timescales of 24 hours. Thus they should work to stop seasonal changes — your model clearly predicts that seasons shouldn’t happen since your thermo-regulatory system would act to counteract the reduction in forcing. But clearly seasons happen and result in local temperature variations of more than 0.2%. Hence the question would be why can your emergent effects stop global variations of more than 0.2% but not local seasonal changes.

Your model is also incomplete. It only discusses how energy is moved around the climate system. It does not say anything about how energy enters or more importantly leaves the earth. I could just as easily say that your model explains why global warming is more pronounced at the poles since all the thermoregulatory systems you describe take energy from the equator and shift it polewards. This would explain the fact that the arctic is warming faster than the equator, keep all of the mechanisms you discuss intact and allow for significant global warming. If you want to claim that you thermoregulatory system actually cools the earth then you need to couple it to a model of radiation and discuss the greenhouse effect in detail since that is the only way for energy to leave.

Also you do not show any evidence that the mechanisms you discuss are not already included in current climate models. Clearly they are not explicitly in them due to the size of the grid cells but if a modeller were to claim that such feedbacks are implicitly included as parameterised values for convection and cloud feedback or something similar and therefore there is nothing new in your model since it is already in current simulations how would you disprove them? You need to create a mathematical model for your feedbacks so that you can estimate the size of them and then compare them to the size of various parameters in the climate models.

Your final sentence shows that the entire premise is wrong. You ask
what causes slow thermal drift in thermoregulated systems?”
to which the answer many people would give might be the rise in greenhouse gases. Nothing you describe stops an increase in greenhouse gases from producing a slow thermal drift. Dr. Spencer’s satellite measurements show a temperature rise of 0.13 degree rise per decade. How do you know this is not caused by an increase in CO2? You also make the misleading claim that temperatures have varied by less than one degree Kelvin. Most people would say instead that temperatures have risen by about one degree Kelvin. This difference is crucial to whether or not your mechanism is working. If there is a slow steady rise over a century then clearly your thermoregulatory mechanisms are not doing their job. If the temperature on the other hand has gone up and down over the past century while remaining roughly constant then your mechanisms might be working but then you have to explain how effects that only last a day work to slow temperature changes over a 50 year period but not any faster.

Curious George(@moudryj)
Reply to  Izaak Walton
April 28, 2021 4:58 pm

“Climate sensitivity is something that emerges from different climate models as a simple number that allows people to hide all the messy details.”

To hide messy details, it has to be built in – not to emerge from the same messy details.

Fred Souder
Reply to  Izaak Walton
April 28, 2021 7:47 pm

I don’t believe that the emergent phenomena as described by Willis would have anything to do with seasonal suppression. The primary region for this proposed effect is the ITCZ. Thunderstorm formation anywhere outside of this is typically not due to the temperature reaching a high point, but rather lifting mechanisms: fronts, orographic lifting, advection, etc. Sure, there is some, but not the overwhelming majority. The transport of this excess energy from the tropics toward the poles via humidity or enso or whatever will increase the rate of energy leaving the system as a whole. Now, you can argue that the rate of energy transport from these emergent phenomena does not exactly equal the energy change radiated from earth, and I’ll listen.

You say “If there is a slow steady rise over a century then clearly your thermoregulatory mechanisms are not doing their job.” Or it could be that they are. These are not mutually exclusive. The mechanisms are primarily describing what happens in the tropics, and that region has not warmed.

Finally, I didn’t catch that part where Willis claimed that the rise in CO2 didn’t cause the rise in temperature, but I was reading it fast. I’ll read it more carefully now.

Izaak Walton
Reply to  Willis Eschenbach
April 28, 2021 9:24 pm

If your feedback mechanisms are not 100% effective then they do not work as you claim they do and still leave room for a small change in temperature over the course of a century to be due to greenhouse effects. You state that the temperature has changed by about 0.2% since 1900 and similarly the forcing at the tropics has changed by less than half a pecent over the same time period. So an imperceptible change in the timing of the cumulus field would just as easily be changed by the forcing resulting in the temperature rise.

And again unless you can put some numbers into your model it will just get ignored. Since you can easily still be right and for CO2 to be driving the climate. Suppose for example the global temperature is given by:
T(t) = T0+ A cos(b*t) Exp(-c*t)
where T0 is the set point, A the magnitude of some perturbation and c is a decay term. Your paper only says that c is positive while leaving open the question of what determines the set point T0. All a climate scientist has to do is say that once you average over suitably long time scales the value of A, b and c are irrelevant and the only equations of interest are those for T0 which is determined by things like the amount of CO2 in the atmosphere, the ellipticity of the orbit, solar intensity etc.

Reply to  Izaak Walton
April 28, 2021 9:34 pm

Willis’s emergent mechanisms are not described as eliminating climate variations, including seasons and decadal (or longer) climate changes, but in moderating them—-and most importantly, putting bounds on them.

Those bounds mean No Tipping Point from CO2 forcing lies ahead. This is just the opposite of the claims of climate alarmists, whose models forecast runaway warming from the course of rising CO2 we are on.

This is an enormously important point. The Western world is getting ready to commit unnecessary economic suicide based on the models Willis’s theory contradicts.

Izaak Walton
Reply to  kwinterkorn
May 1, 2021 5:23 pm

It is very hard to see how emergent mechanisms such as Willis describes operates on a decadal timescale. His model involves things like thunderstorms or tropical clouds that naturally have a periodicity of about 24 hours (i.e. the water heats up during the day causing clouds to form in the afternoon). Thus any such mechanisms react almost instantly and on a timescale of hours to days to a change in temperature. Therefore if they exist and they work they should stop it getting colder in winter so the fact that in most places it gets colder in winter shows that such mechanisms are of very limited strength.

Nor do Willis’ mechanisms put any useful bound on global temperatures. Tropical temperatures might be limited to 30 degrees but the average temperature of the planet is only 15 degrees or so and hence there is still a maximum of 15 degrees of global warming possible until Willis’ mechanisms operate globally. So there is still room for catastrophic global warming even if Willis’ theory is correct.

Reply to  Izaak Walton
May 5, 2021 1:42 pm

How can it be catastrophic? This mechanism is already in full affect in the tropics, and all kinds of creatures inhabit the tropics, in fact the biodiversity and/or concentrations of living organisms on this planet is likely highest in the tropics. If anything, a whole lot of warming on this planet would just make bigger tropics. And if that worries you, relax, this planet has been there before as well, and still supports life just fine. If anything, global warming would make life a beach everywhere!

David Blenkinsop
April 28, 2021 1:35 pm

I like the description of thunderstorms as heat pipes that cool the ground, partly by way of cool rain hitting the ground. Of course the raindrops themselves at whatever temperature have heat content, caloric content, so the falling rain represents a downward component of heat related energy. This all seems to me very much in contrast to simple depictions of greenhouse theory, where virtually the only downward power components are held to be either solar input or else downwelling IR, with seemingly no special attention to a regulatory cooling effect overall.

April 28, 2021 1:44 pm

Willis thanks for a very interesting article! I think there is an additional related phenomen that perhaps should be checked in the future. When a thunderstorm forms large amounts of water is condensed into microscopic droplets with zero CO2 content at a high altitude and thus low temperature with an extremely large surface area. My view is that the thunder storm efficiently locally removes CO2 from the atmosphere thus increasing radiating heat loss over the sea. Over land the CO2 is released close to the ground when the water evaporates.
Perhaps something to look into in the future.
To me the mechanisms you describe feel very plausible.

Rud Istvan
April 28, 2021 2:23 pm

Willis, a suggestion concerning Tstorm mechanism discussion, which you may wish to add. You note clouds cool by increasing albedo. But some don’t—high wispy ice cirrus warms because transparent to incoming SWR but opaque to outgoing LWR—part of Lindzens adaptive iris.

IMO there is a second and arguably more important Tstorm thermoregulatory mechanism. The condensation at altitude in a thunderhead releases the heat of evaporation in large part above the greenhouse radiative altitude threshhold, radiatively cooling directly. For sure this is true for hail and grauppel. The condensation falling as rain lowers troposphere specific humidity, reducing the water vapor feedback. That cools indirectly by lowering the positive WVF.

I studied this second issue back when writing The Arts of Truth (climate chapter) and then Blowing Smoke (eassys Humidity is Still Wet, and Cloudy Clouds). In both CMIP3 (AR4) and CMIP5 (AR5) the amount of modeled ocean rainfall is about half that which can now be ‘known’ directly from the ARGO near surface ocean salinity measurements. Means two things. First, you have second and third Tstorm cooling mechanisms at least as powerful as reduced insolation. And second, observational proof that Tstorms are incorrectly parameterized. See my above comment for the more powerful statement you ‘prove’ by showing they cannot be parametrized at at all since vary over the day and season. In the conventional climate model sense, parameters are the ‘constants’ in some larger climate function. See my post here some years ago ‘The Trouble with Climate Models’ for details.

Curious George(@moudryj)
Reply to  Rud Istvan
April 28, 2021 5:03 pm

“releases the heat of evaporation in large part above the greenhouse radiative altitude threshhold, radiatively cooling directly.

N2 and O2 cannot cool radiatively, but liquid or solid water can. I wonder if models take it into account.

Rud Istvan
Reply to  Curious George
April 28, 2021 6:00 pm

They do, but incorrectly by assuming constant relative humidity, observationally proven false.

Fred Souder
Reply to  Rud Istvan
April 28, 2021 7:58 pm

Rud and Willis,
I’ve wondered about this. All the energy of evaporation or sublimation is converted into latent energy, and yes this cools the surface, but then an equal amount must then be released on condensation or precipitation. During condensation or precipitation, I have always assumed that the bulk of this energy is then thermalized, but I always wondered what fraction of it is radiated. Do you have a handle on what % of the energy released on condensation is radiated? Or, is it that all energy is transferred to the surrounding molecules, warming them, and then it is radiated?

Rud Istvan
Reply to  Fred Souder
April 28, 2021 8:38 pm

If the release is above the radiative layer, it all goes to space. The equivalent radiative layer, determined by CO2, is a function of CO2 concentration and altitude. Is why CO2 never saturates.

Bob Wentworth
Reply to  Rud Istvan
May 7, 2021 6:07 pm

If the release is above the radiative layer, it all goes to space.

I would think that needs to be an exaggeration, on at least two counts:

  1. Global convective circulation transports high altitude warm air towards higher latitudes where it eventually descends to the surface producing surface warming.
  2. High altitude warm air will radiate in all directions and will, in turn, absorb longwave radiation coming from all directions. The net effect is, of course, net radiative heat transfer upward (as is true at all altitudes). But, there’s no magic that say that at some point radiative heat transfer becomes so efficient that all added heat will be instantly radiated away to space.
Bob Wentworth
Reply to  Fred Souder
May 7, 2021 6:16 pm

During condensation or precipitation, I have always assumed that the bulk of this energy is then thermalized, but I always wondered what fraction of it is radiated. Do you have a handle on what % of the energy released on condensation is radiated? Or, is it that all energy is transferred to the surrounding molecules, warming them, and then it is radiated?

It’s all thermalized, transferred to the surrounding molecules.

The warm air radiates at a rate related to its temperature. (It also absorbs longwave radiation from below and above.)

The efficiency of radiative heat transfer from the air increases at higher altitudes, as the atmosphere above becomes more “optically thin.”

But, there is no point at which all added heat is instantly and magically whisked away by being radiated to space.

Some of the heat is eventually returned to the surface, as convective circulation transports air to higher latitudes where it returns to the surface.

See also my comment to Rud.

Bob Wentworth
Reply to  Rud Istvan
May 7, 2021 5:43 pm

assuming constant relative humidity, observationally proven false.

Could you unpack or cite what evidence you’re referring to? (I don’t have any assumptions about this either way; just curious.)

AC Osborn
Reply to  Curious George
April 29, 2021 4:35 am

“N2 and O2 cannot cool radiatively,”
Is that actually a true statement?
Just because they do not radiate in the LWIR doesn’t everything radiate away it’s energy if it is above 0k.
In the case of N2 and O2 it is in the microwave frequencies.

Granum Salis
Reply to  Rud Istvan
April 28, 2021 9:53 pm

The condensation falling as rain lowers troposphere specific humidity, reducing the water vapor feedback. That cools indirectly by lowering the positive WVF.”

I can see that specific humidity would be lower in the upper troposphere, where the condensation occurs, but there must be a lot of evaporation from raindrops as they fall and warm. This should cool the surviving droplets but increase humidity (as compared with before the rain) at that lower altitude.

David Blenkinsop
Reply to  Willis Eschenbach
April 29, 2021 12:32 pm

You wrote:

3. Cold rain and cold wind. As the moist air rises inside the thunderstorm’s heat pipe, water condenses and falls. Since the water is originating from condensing or freezing temperatures aloft, it cools the lower atmosphere it falls through, and it cools the surface when it hits. In addition, the falling rain entrains a cold wind. This cold wind blows radially outwards from the center of the falling rain, cooling the surrounding area.”

Yes, I distinctly experienced such a cold wind one time back in the mid 1990’s here in Saskatchewan. From my tractor at the time, I could see a really big lenticular shaped cloud formation above the hills to the northeast of me, maybe 10 miles away, or maybe more like 15 miles to the center of this thing. When I stopped my machine to step off for a bit, the wind coming direct from there felt very cool indeed.

Later when I got in my car and drove through that area, there were steel bins strewn all around. Also I saw one old farmhouse ripped in half, and another newer house sitting intact in an otherwise completely obliterated farmyard. “Plough wind” damage, they called it.

April 28, 2021 2:32 pm

I will give it a closer look later, but I want to say from the start the opening statement is very strong. The fact that decades of new and better data have failed to improve the uncertainty in our estimates of climate sensitivity is proof that the underlying model is wrong. This is a basic principle of science. Just as the very precise observations of Tycho Brahe failed to improve the uncertainty about the size of the orbit of Mars, until the underlying model was changed from a circle to an ellipse by Kepler, our estimate of climate sensitivity will not become more precise until the underlying model is similarly rectified. This is the central issue.

April 28, 2021 2:46 pm

Willis, I really wish more journal articles read as easily as your writing, but sadly, that is not the case. There seems to be a stylistic bias requiring more turgid and less easily comprehensible writing style. An example from this article is your statement “doesn’t raise the temperature anywhere near as fast”, from which you could remove the “anywhere near” making it more in accord with scientific journals, but less interesting. There are many such stylistic adjustments that might be made, to make this important work read more in the way to which journal editors are accustomed. I have never been the lead author on a journal article, so I am perhaps a poor judge, but others on WUWT might agree with me here that such stylistic changes may improve the likelihood of publication.
Also, the consensus elite (?) scientists will no doubt be looking for every nit to pick, so the way you have used “climate sensitivity” in the context of the processes in a single day will be an easy target. I understand why you have done so, in that climate is simply weather writ large, yet being content to call it temperature sensitivity may be a safer option. Again, the thoughts of someone having been published in a lot of journals would carry more weight than mine.

Clyde Spencer
April 28, 2021 2:51 pm

Lowess fitting provides an aesthetically-pleasing representation of the general trend of data. However, it has a disadvantage of not being particularly useful for interpolation or short-range extrapolation. That is, it all has to be done manually with the graphics. Furthermore, it does not provide a quantitative estimate of the goodness-of-fit for purposes of comparing models or supporting any claims of utility.

Therefore, I’d suggest adding, as appropriate, function fits, from a statistics package, to your figures. Either fit all the data, or your Lowess fit. That way, you can claim how well your hypothesis is explained by your independent variables. For example, you might get by with a 3rd-order polynomial fit for Fig. 2, and Fig. 6 looks like it might be fit well with a logarithmic or 2nd-order polynomial.

M Courtney
April 28, 2021 3:03 pm

Obvious but important point.
The predictions are supported by evidence. But you do not discuss the counter-evidence.

Even if the discussion is a single sentence of “As yet there are no observations that contradict this interpretation”, there ought to be some consideration of the alternative.

This is a discussion of the issue not an article intended to sway opinion.

April 28, 2021 3:06 pm

Great effort, Willis. My only contribution from experience is to eliminate any qualitative descriptors and ensure that you have a ‘complete’ literature search.

April 28, 2021 3:10 pm

I think LOWESS (LOcally WEighted Scatter-plot Smoother) should be caps like CERES.
Willis, don’t feel bad if you paper doesn’t make Nature. They cater to the university academia Ph.D track who have spent such big taxpayer $ on test gear that they essentially HAVE to publish in Nature. There are many good journals in the meteorological realm.

April 28, 2021 3:19 pm

Global warming is mostly about polar amplification.
So it seems global warming or cooling is mostly the tilt of Earth’s axis.
And since our axis tilt is reducing, long term, we heading towards, global cooling.
So we have recovered or are recovering from Little Ice Age and within few centuries or
less, we could expect a return to something like the Little Ice Age.
And a continuation of our 5000 year tread of cooling.

Reply to  gbaikie
April 28, 2021 5:04 pm

I agree. Global warming during Ice Age Terminations is extremely rapid compared with cooling during Ice Age onset. I believe this is due to the exponential growth of melt ponds at high northern latitudes when absorption of solar radiation by water surfaces exceeds a specific threshold determined by solar elevation angle. It is a good predictor, see: http://blackjay.net.au/wp-content/uploads/2020/04/IceAges-1.pdf

April 28, 2021 3:47 pm

Once the sun sets, first the cumulus and then the thunderstorms decay and dissipate. 

High rates of precipitation occur at any time over the warm pools. They are triggered by convective instability and that is not linked to daily cycle. It takes about 30 hours to recharge the CAPE.

Attached was observed at the moored buoy at 0N 156E when it was in a warm pool. The heaviest rainfall occurred on the evening of day 3. Also heavy falls early morning on day 7.

Fred Souder
Reply to  RickWill
April 28, 2021 8:02 pm

Willis has the advantage of having lived in the tropics, as did my father who suggested that in Singapore you could set your watch by the timing of the afternoon thunderstorms. Obviously, this is a general statement and there are a variety of features that can create heavy rain in the tropics.
Also of note, is that precipitation rates in tropical systems like hurricanes often increase overnight. Curious.

Reply to  Fred Souder
April 29, 2021 12:13 am

Land is a different situation to open water. The land is much more responsive to sunlight and clouds than the ocean.

You can go and look at the buoy data yourself to see the big downpours occur at any time. It is reasonably well known in the literature as well.

April 28, 2021 3:48 pm

Not bad, but can you explain why you mixed CERES with Berkeley, when CERES has its own temperature (skin_temp, adj_skin_temp)?

Reply to  Willis Eschenbach
April 29, 2021 9:08 am

Oh, this dataset has skin_temp:


This is actually EBAF4.2, I think.

Kip Hansen(@kiphansen2)
April 28, 2021 3:52 pm

w. ==> Figures 2 and 3 both scatter-plots, show classic signs of non-linearity — with fractal-like banding. There are two potential causes of this: 1) The physical phenomena themselves are non-linear and have fractal-like expressions in the real world or 2) the fractal-like numerical results are a outgrowth of the non-linear mathematical expressions.

Some explanation has to be advanced for this obvious feature in the plots.

Geoff Sherrington
Reply to  Kip Hansen
April 28, 2021 5:19 pm

Possibly rounding of values contributes some pattern (if they are actually rounded). I do not know, just theorising. Geoff S

Kip Hansen(@kiphansen2)
Reply to  Willis Eschenbach
April 29, 2021 7:23 am

w. ==> Lat/Long might produce a regular banding, but I think not the distinctive “strange attractor-like” patterned banding showing in your two figures. I’m going to try to find and read Rud’s fractal productivity paper in which he finds the “same result”.

Rud Istvan
Reply to  Kip Hansen
April 28, 2021 6:44 pm

Kip, a comment from someone who actually published major peer reviewed on fractal implications long ago. See https/doi.org/10/002/smj.4250130705/
or search Istvan, fractal productivity. Same result.

Kip Hansen(@kiphansen2)
Reply to  Rud Istvan
April 29, 2021 7:02 am

Rud ==> I’d love to read it but none of my usual ten ways of finding an paper that is “in hiding” have turned it up, including a doi search. If you have a direct link, can you please post it here?

Kip Hansen(@kiphansen2)
Reply to  Rud Istvan
April 29, 2021 7:18 am

Rud ==> I haven’t been able to locate your paper, but I suspect that “same result” in your comment implies that there are some fractal-like non-linearities in the real world function being measured?

I have seen this fracticality time and again in the oddest natural phenomena — larval instars in flour beetles, cloud production, etc.

April 28, 2021 3:54 pm

Germaine, I suppose, but small potatoes:
1) “Here is Reis and Bejan’s description” This should be followed by a colon.
2)  “I will define emergent climate phenomena functionally and by example.” Needs a colon there instead of a period; and, would be better as “…climate functionally (in bold) and by example:”
Very best wishes to you on this!

April 28, 2021 3:56 pm

Willis, you heretic! (I mean that as a compliment).
Apart from the nit-pick that Benard either has an accent or doesn’t, what stood out for me
was some strange things I saw in your scatterplots.
Now, you have plotted 1×1 degree grid cells, over 20 years. And it seems (but your
words do not confirm) that you have plotted each individual grid cell for each
individual month as a separate point. That’s over 15 million data points. I
usually only do scatterplots over a few dozen, or at most a few hundred data
points. And my experience is, that even if there’s a trend line or curve, the
deviations from that trend don’t have any particular structure. They are pretty
much random. I wouldn’t expect that to change if you up the number of data
points. Unless there are secondary effects that you aren’t accounting for.
But your scatterplots show an internal structure, even well away from the LOWESS smooth line! Your Figure 2 (which I call the “Swan’s Neck”) has some really strange patterns below the line. Like several of what look like bridges, reaching out towards what look like islands. Reminiscent of the Mandelbrot set. Could we have a new discovery here, the “Eschenbach pattern?” Well, I count ten of those bridges, and some dubious ones either side; so, it may merely be that it wasn’t the right thing to do to plot individual months. It might have been better to average over the years first. If, when you do that, the odd-looking details disappear, that would tell us something.
There’s a similar internal structure in Figure 3 – there’s a duck in there somewhere – and Figure 6, “the Eel.” So, my best guess is that you’re probably not plotting the data in quite the right way.

Kip Hansen(@kiphansen2)
Reply to  Neil Lock
April 28, 2021 4:13 pm

Neil ==> Yes, I noticed the same. The question with the fractal/chaotic/non-linear banding “Is it it the real world or in the maths?”

Reply to  Neil Lock
May 5, 2021 2:04 pm

I thought Figure 3 looked like the skeleton of the Loch Ness monster.

April 28, 2021 4:01 pm

One of the consequences of water vapour responding to surface temperature is that there is persistent high level cloud at lower latitudes in the Southern Hemisphere when the water vapour peaks.

I prefer real heat fluxes at the top of the atmosphere rather than non-physical downwelling longwave but the sum of reflected SWR and OLR at the ToA show a high positive correlation with moisture. It also shows the uptick at the high end over the warm pools. The slope for a year is up but there are three months when the slope is negative – the cold surface months. So water vapour itself has an overall regulating role.

The other thing to note is that there is an upward trend in the area of warm pools this century.

Reply to  RickWill
April 28, 2021 4:34 pm

I have named the gradient of the regression line for the ToA heat loss v TPW as the Atmospheric Water Cooling Coefficient.

The attached chart shows how the slope changes with sea surface temperature. As noted above there are three of the twelve months when it is negative. It is highly responsive to sea surface temperature. The annual average AWCC for the year considered was 1.4W/sq.m/cm.

Crispin Pemberton-Pigott
April 28, 2021 4:24 pm


Another example of a discussion of emergent phenomena:



Lindzen’s point is that there is a powerful emergent phenomenon that is temperature caused and correlated.

As I understand it, your major point is that the reason the phenomenon is temperature-related is that the “forcing factor” – which is to say base to final temperature – is correlated with the initial temperature. By that is meant “the Climate Sensitivity” which is another way of saying the GHG-based warming to be expected from an increase concentration. Is that a fair characterization?

The trouble you may get into with reviewers is that the phenomenon you describe would emerge without any GHG’s at all. If you double the CO2 concentration in your graphical representation, exactly the same emergence would be seen, and the result would – I wager – be essentially identical. The reason the CO2 concentration would make no difference is a) water vapour swamps any feeble GHG effect, b) there isn’t much CO2, c) CO2 doesn’t form clouds when it rises.

Strictly speaking, I don’t think you can argue that the effect of temperature is a “climate sensitivity” because CS is defined as a temperature variation with respect to GHG concentration (factored to CO2e) provided in units of “feedback Watts per sq M”.

You have shown (quite well) that there is a correlation between the starting temperature and the (permitted) rise in degrees before powerful shading and reflecting is invoked. The upper limit on temperature is capped.

GHG warming theory says that if you were to plot such “capping” lines as you have in hand, increasing the concentration of GHG’s will lift all points of the whole line, (unequally, it is admitted, but everything will go up, they say). GHG forcing is the same as increasing the radiative power at source.

But that is not what is happening. Whatever the starting temperature, the upper limit is fixed by a combination of water vapour properties, air pressure and the fraction of the Earth that is covered by water. It is inherent in the construct.

If you were to produce the same final chart showing the slope of the curve is related to the starting temperature, but bin the CO2 concentration data by date, you might be able to show there is a correlated <i>changing peak temperature</i>. One chart for 350 ppm, one for 375 and another for 400 might show “something”. Or not. If there is no correlation between the position of that line and the CO2 concentration, you have an publication that extends your Thunderstorm Hypothesis paper showing not only a correlation of clouding-over-time with morning sea surface temperature, but that the phenomenon works over an large range of initial values.

If you were to combine this (which is essentially a heat rejecting mechanism) with Lindzen’s Iris (which is essentially a heat venting mechanism) you can account for heat gain or loss and terminal temperature whether it is cloudy or not.

Best regards

April 28, 2021 4:37 pm

An excellent study Willis, ahead of its time – emergent phenomena is where climate research is needed.

Here’s another interesting study from 2016 that you might consider a brief mention:


They show a model system where temperature regulation again emerges spontaneously.

We demonstrate the emergence of spontaneous temperature regulation by the combined action of two sets of dissipative structures. Our model system comprised an incompressible, non-isothermal fluid in which two sets of Gray-Scott reaction diffusion systems were embedded. We show that with a temperature dependent rate constant, self-reproducing spot patterns are extremely sensitive to temperature variations. Furthermore, if only one reaction is exothermic or endothermic while the second reaction has zero enthalpy, the system shows either runaway positive feedback, or the patterns inhibit themselves. However, a symbiotic system, in which one of the two reactions is exothermic and the other is endothermic, shows striking resilience to imposed temperature variations. Not only does the system maintain its emergent patterns, but it is seen to effectively regulate its internal temperature, no matter whether the boundary temperature is warmer or cooler than optimal growth conditions. This thermal homeostasis is a completely emergent feature.

April 28, 2021 4:43 pm

Willis, thunderstorms are evidence of how clouds shutter the light from the surface and are the result of evaporation, for sure. But every day https://epic.gsfc.nasa.gov/ shows us pics that the major cloud producers are advective weather fronts, not mostly convective thunderstorms. I think that with clear blue sky the sunlight mostly sees an ocean Albedo of .06. Then the sunlight heats the ocean surface until enough water vapor forms .8 Albedo clouds, driving the planetary albedo to 0.3 again. It does this over and over in randomly shaped patches of the Earth’s surface the size of Europe moving over the surface by Coriolis forces. The net result is an average between ocean Albedo of .06 and Cloud Albedo of .8 and cloud cover of 60%, all varying accoding to Clausius Clapeyron…

Curious George(@moudryj)
Reply to  DMacKenzie
April 28, 2021 4:52 pm

DMK, did you analyze it quantitatively? Please share your results.

Reply to  Curious George
April 29, 2021 7:13 am

George, you can do your own little spreadsheet….
So start with the usual Te= [ So/4(1- albedo)/5.67e-8]^0.25 add 33 degrees average lapse rate, not quite technically correct but we’re trying to get a feel for reality here. You should get around 255 C plus 33 gets you 288 C average surface temp.
Now do the same calculation as if the Earth was a water world, albedo say 0.1, and again as cloud covered world, say albedo = 0.8. See the temperature difference.
Temperature controls how much water evaporates off the wet parts of the planet to make the clouds that changes the planetary albedo….and so albedo of about .3 and 65% cloud cover ends up being the balance point.

Last edited 8 months ago by DMacKenzie
Fred Souder
Reply to  DMacKenzie
April 28, 2021 8:11 pm

I would argue that while most of the earth is indeed a fairly random chaotic mess of advection and storm systems, these all “wash out” on average. Storm systems do follow a semi-regular pattern (see Lezac’s Recurring Cycle), and while some areas are cloudy, others are clear. The total Albedo averages out to be fairly constant. However, the ITCZ’s, on the other hand, where the most direct insolation occurs, are regular. So, you have subtropics to polar where there is no thermoregulatory effect, but in the tropics you do.

Reply to  Willis Eschenbach
April 29, 2021 6:59 am

There is also the issue of “how long clouds last”, kind of the back end of the emergent issue, that is probably important for your paper.
Just want to point out that CRE is a measure of how “in balance” or “out of balance” the radiation effect is. Since the calcs are typically done when everything is in balance the calcs only show a couple of watts feedback. But in reality, the difference in Albedo between an ocean surface planet and a full cloud cover planet is a range of 60% of incoming sunlight being reflected back into space or not. That’s huge, not a couple of watts positive as some (not yourself) indicate. You make a good point “clouds over ice make little difference”…yes look at https://epic.gsfc.nasa.gov/ today…clouds over ice, its the ice that makes little difference, quite a normal occurrence, plus being far out on the cosine of the zenith angle of incident sunlight, makes little difference compared to the albedo range anyway.

April 28, 2021 4:50 pm

Figure 3 looks like you found a Loch Ness Monster.

Reply to  tygrus
April 28, 2021 5:18 pm

At 94 i usually have difficulty in understanding these submissions.

But yours made sense. Thank you.

VK 5ellmje

Reply to  tygrus
May 5, 2021 2:13 pm

The skeleton of a Loch Ness Monster.

Geoff Sherrington
April 28, 2021 5:38 pm

Don’t let this sidetrack you, you will be busy enough.
Wondering about latitudinal effects. If the processes emerge because the surroundings have become too hot, would there not be a different distribution of those big clouds? Would they not be going flat out all the time in the warmer regions of the globe, but be rare or absent where is it not much cooler?
The basic question is one we have briefly covered before about 5 years back: Regulators typically work by using a comparison. If the measurement departs from the comparison, the mechanism emerges and drives the system back towards the comparator or reference, whatever you like to call it. So, if this is what happens in Nature, what is the reference temperature and how is it created? I have a worry that in your description the reference value perhaps has to change with latitude and with seasons.
Basically, it would be neat if there was an explanation of what is meant by expressions like “when it becomes too warm, clouds appear” becoming “when temperature exceeds the reference value for this location, as determined in the following manner, clouds appear”.
OTOH, maybe you do not need to consider this sort of philosophical point because you are reporting what Nature does, not so much why it does it. The earth has had a long time to settle down to systems that keep it stable. If it had not settled down we mght have had a very different earth and no us. It is fundamentally the inexorable operation of physics, chemistry, biology etc, the hard sciences, that have combined to create what we are able to habitate.
Geoff S

Geoff Sherrington
Reply to  Willis Eschenbach
April 28, 2021 10:50 pm

Thank you, provider of full service. Geoff S