Climate science from climate scientists...
This month’s open thread on climate topics. The annual ranking horse race is in full swing, discussions of extreme weather abound, Arctic sea ice is reaching it’s annual minimum, and the future of evidence-based policy making seems dim. Oh well.
Comment Policy:Please note that if your comment repeats a point you have already made, or is abusive, or is the nth comment you have posted in a very short amount of time, please reflect on the whether you are using your time online to maximum efficiency. Thanks.
John Pollack
re: https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850044
Ok, John, here we go. First, I applaud your desire to clarify definitions. That’s the first step in having a real scientific discussion, the second being agreement on what question we are trying to answer. First step:
JP: Energy in the system = the energy accessible to the climate system over the time period being considered, contained within the atmosphere, oceans and freshwater, cryosphere, and solid earth.
Energy storage = the change in the energy in the system over the time period being considered =
the integral of EEI over that time period.
z: I have no problem with the first part, but I’m not sure about your choice of the term “storage”. Are you saying that the “new” energy isn’t accessible to the climate system? That would be the conventional usage for the term “storage”, but obviously the “new” energy is partitioned (your term) in the system… similarly (if perhaps not identically) as is the “old” energy.
(That last part… the possible difference between the partition of energy in the system when EEI is “small enough to be called zero” rather than “big enough to cause the graph to look like a hockey stick”… is the motivation for my original questioning.)
All right, Zebra.
You have a point about what I am calling “storage” since it would also be energy available to the system. For the purposes of this discussion, I could call it “delta Es” to represent the change in system energy, if that seems better.
One of the implications of delta Es being a time integral of EEI is that every wiggle and wobble in the GMST temperature curve is potentially a change in the sign of EEI. This is what I meant by EEI seldom being zero, and my objection to calling it “close enough for government work.” I would expect both EEI and partitioning to be dependent on the particular time interval being examined, since energy can flow from one part of the system to another. In addition to an annual cycle, the oceans are capable of responses on a large range variety of time scales, for example. On a time scale of centuries or longer, if the average EEI is near zero, I would expect that most of the variability to be the tradeoff between ocean temperature and ice mass that I mentioned last month.
John, this is good. But I am easily distracted, so I have to ask: How are we detecting the “tradeoff” you mention, and what is the mechanism causing it? Why does the ice melt?
Whether we call it “average EEI close to zero” or “delta Es close to zero”, my problem all along has been trying to clarify my thinking about causes and effects. If we start with ice mass A and ocean temp B, what do we have at the end of the time period in question?
Have A and B changed even though we see very little delta Es?
Are we talking about global measurements or local/regional?
Is it “natural variation” or “forcing”?
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850169
“B: Changes its mind when the science at the bottom changes”
Let us make a case study over this claim..
Inamdar, Ramanathan 1998:
In summary, lapse rate changes for the annual cycle are not a valid analog for longer timescale climate changes. However, for reasons given below, the discrepancy in lapse rate changes between observations and GCMs do not alter our conclusion on the role of water vapor feedback. The observed annual cycle warming decreases with altitude, the decadal-scale warming remains roughly constant with altitude, while the GCM warming increases with altitude. When atmospheric warming is not so large as the surface warming, the enhanced emission to space is not so large (when compared with the situation when the warming is about the same at all altitudes), which in turn enhances dGa/dTs, i.e., a positive feedback. However, at the same time, in a fixed relative humidity environment, smaller atmospheric warming implies a smaller increase in absolute humidity, which suppresses some of the increase in dGa/dTs from the lapse rate feedback. Because of these two competing effects, the lapse rate has less than 10% influence on dGa/dTs.
What happened here is that after publishing a number of papers using the seasonal variation of dOLR/dTs (or dGa/dTs) as proxy for climate change, someone told them it does not even work. One needs to see how delicate the problem is. In this “proxy” the lapse rate is doing the opposite of what they only assumed it would do. It does not raise the feedback parameter in the sense of a neg. feedback, but it lowers it in the sense of a pos. feedback.
For example the feedback parameter might be 2W/m2/K vs. some 3.3W/m2/K Planck Response. One might assume that to be a total pos. feedback of 3.3 – 2 = 1.3W/m2, with maybe 1.8W/m2 and -0.5W/m2 for WV and LR feedbacks respectively. But in this proxy the LR component is not negative, but positive, and hugely positive. About the entirety of that signal is just LR, with no or only a negligible WV part, like 1.3 = 0 + 1.3! The fact that the tropospheric temperature behaves sluggish relative to Ts has been taken as evidence of a strong WV feedback – that never was.
Another pillar for pos. WV feedback would be the regional proxy:
This result demonstrates the unsuitability of using variations in different regions in our present climate as a proxy for climate change.
(Dessler et al 2008)
Ok, that does not work either, for the same reason. Turns out there is not a single piece of evidence even suggesting WV feedback was positive. They all point in the opposite direction. And that is consistent with the insight of WV being a cooling agent with given concentrations.
Anyway, the interesting part here is about how climate scientists react to “changing bottoms”: as in the instance above, Inamdar and Ramanathan did not retract their previous papers. Notably they state it would not alter their conclusion on the role of WV feedback, implying that is just what it might. In the face of lacking evidence, they instead express their strong belief in pos. WV feedback.
And despite knowing that neither the regional nor seasonal proxy works, later papers like Chung et al 2010 would yet produce both proxies as evidence. It is a pattern I have seen so many times: “yes, we were wrong, but it does not change a thing”. In real science it does..
ES: Turns out there is not a single piece of evidence even suggesting WV feedback was positive. They all point in the opposite direction. And that is consistent with the insight of WV being a cooling agent with given concentrations.
BPL: No, it is not. We know water vapor is a greenhouse gas. We know it accounts for 50% of the greenhouse effect. That would be 161 watts per square meter at the surface. Latent heat accounts for 88. Net effect: warming.
Apparently BPL does not understand the GHE. Both in S10 and K97 the WHOLE GHE is estimated to be 155W/m2.
BPL may have been referring to (global time averaged) LW F↓ @ sfc. Do we need the GHE to be one-dimensional (LW F↑(sfc) – OLR)? Or should I ascribe some of what I’ve been including in the GHE to be LW-opacity effects more generally? Remember that when the GHE is introduced in educational situations, it is often explained that it tends to reduce the diurnal T range (and F↓ @ sfc has a role in that; although one way to look at it would be to use Martin Smith’s approach and trace a batch of energy through the system from a point in time where solar heating occurs; an increased GHE should tend to prolong the residence time (hence a net increase in stored energy PS re zebra – I use storage the way John Pollack did and I’ve seen it use the same way in a textbook).
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850170
1. “how does the “established physics” explain the arrival of a MALR?”
Here for instance:
https://hogback.atmos.colostate.edu/group/dave/pdf/Moist_adiabatic_lapse_rate.pdf
It is very simple: neither the dry- nor the moist adiabat are physically dependent on GHGs. Sure you could say WV is a GHG, but it is only considered as a condensating gas, not a GH-agent. And yes, it is established physics.
2. That contradiction is just the entry point to a much bigger issue. Then of course the question is if one wants to explore that, or seal the door. Your notion was like: that would make a contradiction, so it can not be, case closed, nothing to see here..
My take is of course a bit different. I know already what is behind the door, and before I knew, it was a fascinating and promising subject of research. So I went after it. Knowing already that WV is a strong cooling agent, positively correlated to temperature, it just had to be a strong neg. feedback. That is a conclusion in total contradiction to all the evidence climate science so far produced, as you would agree.
So I tasked myself with the challenge of finding out how all these pieces of evidence could be wrong, if they were. And that turned out to be surprisingly simple, also because well known scientists themselves in many instances already realized they were riding a dead horse. Like these two instances..
In summary, lapse rate changes for the annual cycle are not a valid analog for longer timescale climate changes. However, for reasons given below, the discrepancy in lapse rate changes between observations and GCMs do not alter our conclusion on the role of water vapor feedback (Inamdar, Ramanathan 1998)
This result demonstrates the unsuitability of using variations in different regions in our present climate as a proxy for climate change. (Dessler et al 2008)
There are many more of these instances. But again, they are indeed all wrong. I do not want to use harsh words, so let me put it like this: never in the history of mankind have there been more ressources spent for a less optimal result.
E. Schaffer,
You respond to my questioning of you in the July UV thread
(1) The question posed was “if the air is “not totally saturated with WV,” how does the “established physics” explain the arrival of a MALR? Surely the DALR would be the mechanism when condensation isn’t present and if the temperature has yet to cool to the point where the air is “not totally saturated with WV,” how does the WV condense?”
You respond by linking to a mathematical derivation of the MALR and some nonsense that it is “very simple” because the GHE mechanisms are not a consideration.
Your response doesn’t even begin to address the question I posed!!!
The linked MALR derivation is not of great length with a quarter of it presents a derivation of the DALR. Yet it still manages to use the word “saturated/saturation” nine times. This is because the MALR is the lapse rate for “saturated” air which, as it cools with altitude, is condensing out precipitation. This, of course, assumes it is rising air but what goes up must come down. And it does so dry as a bone. Surely this is “very simple.”
And within that linked MALR derivation, we can also examine the graphic ‘Figure 1: A plot of the moist adiabatic lapse rate, Γm , in K km-1, as a function of temperature and pressure. The troposphere pressure sits 100-200mb to 1000mb with a tropopause temperature of -55°C (220K) and a chilly -80°C (195K) over the tropics and surface temperatures +27°C (300K) tropics & -30°C (245K) polar.
So let’s take a rough mid point average of all that p = 600mb, t = 250K. What would the MALR be at that pressure and temperature? Looks like it would be 8.7K/km. So why would a value of 6.5K/km be an accepted average global Lapse Rate?
Answers please on a no-bullshit postcard.
(2) I would suggest that the “contradiction” you’re at last attempting to address needs a bit of ‘diction’ here as mind-reading is not a thing (a bit like x-ray vision through doors) while those who may have read the comments in the July UV thread are likely not blessed with eidetic memory.
The issue which you termed a “contradiction” is that (A) the Lapse Rate feedback is derived as -0.5Wm^-2/°C which, if extrapolated for a 33°C GHE, would suggest a -16.5Wm^-2 contribution to said GHE while (B) you insist the contribution is about 5x larger with the LR feedback = surface WV latent heat flux = ~80Wm^-2.
Strangely, on the July UV thread, you accused me of ignoring this “contradiction” even though I has raised the issue. And I have continued to point to your failure to provide any cogent reason why your -80Wm^-2 is correct and the entirely of climatology is wrong.
So where are we with your latest explanation presented above?
You insist this -80Wm^-2 leads to “a much bigger issue” because when the negative LR feedback is combined with the positive WV feedback (a combining which climatology is want to do), if the LR feedback were a five-times-bigger than that found by the entirety of climatology, the net WV/LR feedback would be negative not positive.
I would not disagree that it would be “a much bigger issue” if ‘the entirely of climatology is wrong’ on this matter. But where is the reason to think ‘the entirely of climatology is wrong’?
You present two exemplar quotes from the literature:-
Inamdar & Ramanathan (1998) ‘Tropical and global scale interactions among water vapor, atmospheric greenhouse effect, and surface temperature’ you accuse of some scurrilous work saying of them “After publishing a number of papers using the seasonal variation of dOLR/dTs (or dGa/dTs) as proxy for climate change, someone told them it does not even work.”
Your assessment here is just jibberish.
The ‘proxy’ adopted for the atmospheric greenhouse effect is a quantity Ga, “the radiometric definition of the atmospheric greenhouse effect … which is the difference between the surface longwave emission and the outgoing longwave radiation.” Why would anyone insist adopting such a ‘proxy’ is wrong, or specifically wrong wrt “the seasonal variation of dGa/dTs”. Who is saying it “does not even work”“? The paper’s conclusion runs:-
If you are trying to present some serious criticism of Inamdar & Ramanathan (1998), you have managed a very good job of hiding it!!!!
Dessler et al (2008) ‘An analysis of the dependence of clear-sky top-of-atmosphere outgoing-longwave radiation on atmospheric temperature and water vapor’
The paper discusses lapse rate in Sec 4.2.3 [para 39-41]. It is the final sentence of these paragraphs that you quote. In this final sentence, the paper is emphasising the point that regional variations in lapse rates prevent the use of a regional lapse rate as a proxy for climate change – a global analysis is necessary.
The actual finding presented in para 39-40 is that the changes in lapse rate induced by a change in surface temperature will increase climate sensitivity relative to an unchanged lapse rate. That doesn’t appear to assist your side of this “contradiction.”
(You also mention another exemplar paper“Chung et al 2010” of which there are two potential references that I can see ( & ). Your accusation against a “Chung et al 2010” runs “despite knowing that neither the regional nor seasonal proxy works, later papers like Chung et al 2010 would yet produce both proxies as evidence.”
I fail to see anything in the two potential Chung et al (2010) papers that shows they “produce both proxies as evidence.” Neither make mention of either Inamdar & Ramanathan (1998) or Dessler et al (2008). And both are analysing climate using models and data so their “evidence” was not reliant on some prior error.)
The more you run with this silly nonsense of a massively powerful LR feedback being ignored by all of climatology, the more silly and nonsensical you make it.
“Your response doesn’t even begin to address the question I posed!!!”
If you mean the question below, I have to admit I do not understand it..
“Perhaps I should ask that if the air is “not totally saturated with WV,” how does the “established physics” explain the arrival of a MALR? Surely the DALR would be the mechanism when condensation isn’t present and if the temperature has yet to cool to the point where the air is “not totally saturated with WV,” how dos the WV condense?”
However, you brought up the notion the adiabatic lapse rate was somehow due to GHGs, which it is not, and that is the important take-away here. Moreover because WV is a condensating gas it reduces the lapse rate and the GHE with it. None of that is a secret or would be seriously disputed. Also it is not “my position” so that I would even feel the need to defend it.
I only brought up the logical follow-up question on whether WV is warming or cooling, thereby pointing out this question tends to be illicitly avoided by just stating “WV is a strong GHG”. Clouds are a strong GH-agent too, yet they are believed to be cooling. Btw. we could probably agree that this question has not been even considered, which is bad enough.
Should you reject the physics behind the moist adiabat so to avoid said follow-up question, I would not try to convince you of established physics, but rather accept that as an answer.
“So let’s take a rough mid point average of all that p = 600mb, t = 250K. What would the MALR be at that pressure and temperature? Looks like it would be 8.7K/km. So why would a value of 6.5K/km be an accepted average global Lapse Rate?”
600mb is about 4000m up. Where do you take the 250K from? Where does anything look like 8.7K/km? I have not the slightest clue how you come up with these figures. What I can tell you is the theoretic MALR is smaller then the actual lapse rate, because yes, the former is perfectly saturated, while in reality that is not so.
Let us go back a little..
me:
“You brought this perspective up before and I ignored it. Sorry for that. It is actually a great perspective, but it escalates the issue, without answering the original question. Of course it tells us there is a contradiction between the role of WV within the GHE, and WV feedback. One of the two must be wrong.
I did not say you ignored it, I said I ignored it. What you argued, and still do, is to say that said contradiction would deny that latent heat cooling by WV..
“..but your conjecture that the LR feedback = surface WV latent heat flux = ~80Wm^-2 is certainly not sound”
I say that such a contradiction tells us at least one side should be wrong, but it does not tell us which one. Concluding it must the one that is less convenient is not good reasoning either. The fact that WV reduces the GHE by said magnitude of ~80W/m2 a simple logic consquence and you could not produce a single good argument why that was not so.
So, as far as the contradiction goes, we know the one that is NOT wrong. And that is what brings the LR-feedback as assumed in the models, some -0.5W/m2/K (down from -0.84 in AR4), into the cross hair. And that is where things become truely interesting.
“The actual finding presented in para 39-40 is that the changes in lapse rate induced by a change in surface temperature will increase climate sensitivity relative to an unchanged lapse rate. That doesn’t appear to assist your side of this “contradiction.””
Yes, that is what I have explained already. Only in the “proxy” the lapse rate emulates a positive feedback, as opposed to the expected neg. feedback. It is best explained in numbers. Planck Feedback minus Feedback Parameter = Feedback might go like 3.3 – 2 = 1.3. Now the assumption is those 1.3 might consist of 1.8 and -0.5 as WV and LR feedbacks.
If you check the literature it is widely agreed that there is “certainty” over the total FB of some 1.3, but uncertainty over the exact attribution. If the WV term is larger pos., the LR term is larger neg., or vice verse. And the idea is that while there is uncertainty over this question, it would not really matter. The sum is important.
The problem is, and that is what both sources above point out, that assumption can only hold true as long as the “proxies” represent climate change. In other words, as long as the lapse rate in the proxies does the same as it would do in the long run. But that is not true, rather it is the opposite.
The LR component is emulating a pos. feedback, like > +1W/m2/K in the seasonal- and ~+0.5W/m2/K in the regional “proxies”. The scheme of total feedback equating to WV+LR feedbacks then looks extremely different, like..
Assumption:
1.3 = 1.8 – 0.5
Reality:
seasonal:
1.3 = 0 + 1.3
regional:
1.3 = 0.8 + 0.5
In other words the evidence for a strong WV feedback of 1.8W/m2/K (that is the altitude part) is largely the difference between the assumed neg. lapse rate component and the actual pos. lapse rate component, but it was never real. The erring over the lapse rate behaviour in the proxy is the “evidence” for the large WV feedback – and eventually the reason why the total WV feedback is believed to be pos., while actually being neg.
The dry adiabatic lapse rate (DALR) for a parcel of air is actually not strictly a constant value (for gas of a given composition in a given gravitational field ie. @ g ≈ 9.81 m/s²); it actually depends on the environmental lapse rate (ELR). But when a layer is well-mixed without any H2O phase changes (unsaturated air), potential temperature θ is constant in that layer, so both ELR and DALR are the same and DALR ≈ 10 K/km or more accurately, g/c_p (from memory – did I get this right) ≈ 9.81 m/s² ÷ 1004 J/(kg·K) = 9.81/1004 m/s² ÷ [ kg m/s² m / (kg K) ] ≈ 9.77 E−3 K/m … https://en.wikipedia.org/wiki/Lapse_rate#Dry_adiabatic_lapse_rate ; https://en.wikipedia.org/wiki/Table_of_specific_heat_capacities
More generally, DALR gives T as a function of θ ( https://en.wikipedia.org/wiki/Potential_temperature ) and p : https://en.wikipedia.org/wiki/Exner_function , which can be converted to a function of z given ELR, T_0 and z_0 (describes a dry adiabat)
The moist adiabatic lapse rate (MALR – a part of me wants to put a D at the end and call it a duck) does not have a single value; it is a function of T and p. It gets smaller at higher T at a given p (this is of course related to the reason for a negative LR feedback(**) and I expect it would do so at a given T at larger p (notice I didn’t say higher p so as to avoid confusion with p getting smaller at higher z) because at smaller p at a given T, there’s more H2O vapor relative to the mass of air… but following a moist adiabat from the level where condensation begins, going upward, the cooling reduces the amount of latent heat released per ∆p of ascent (and therefore per ∆z, given a value of ∂p/∂z), and eventually the moist adiabat asymptotically approaches a dry adiabat, whose value of θ can be used to characterize the ascending saturated moist air. Of course, real moist ascent is not exactly adiabatic (adiabatic = reversable = isentropic) because of the effects of curvature for the smallest droplets, but also because of the difficulty of initiating ice nucleation (I also recall reading that phase nucleation delays also play a role in the mechanical/rheological(?) properties subducting slabs of lithosphere(?)) – and also a couple other things. (un-)Mixing of different air masses (the un- pertains to precipitation) and down-gradient diffusion (ie. conduction of heat, diffusion of H2O) are not adiabatic.
** ie more latent heat is released per m of moist ascent at higher T because more H2O condenses per m of ascent. This does not necessarily or automatically mean more H2O condenses per unit time; velocity could slow down (relating to the difference between convective heat flux and convective lapse rate).
The occurrence of moist convection can, by approximately establishing an ELR ≈ MALR, stablilizes the air to dry convection (while large-scale horizontal motion – eg. baroclinic waves / frontal cyclones – driven by geographical variations in solar heating can stabilize the air to localized vertical moist convection). It is helpful to map out the θ contours/surfaces ( https://en.wikipedia.org/wiki/Potential_temperature ). Moist adiabatic ascent crosses θ going up, so a ELR = MALR means θ increases with increasing height z. Dry adiabatic motion drags θ values with it. In order to sustain a steady-state overturning, where precipitation removes much of the condensed H2O so the cloud base is high in the descending region, the dry descent must have diabatic (ie. a net flow of heat in or out of a volume/mass of air) cooling in order to cross the θ contours/surfaces. In clear skies, a net LW radiant cooling of ~ 2 ± 0.5 K/day is typical in the troposphere:
Jeevanjee & Fueglistaler, 2020 “Simple Spectral Models for Atmospheric Radiative Cooling” https://journals.ametsoc.org/view/journals/atsc/77/2/jas-d-18-0347.1.xml
(some formatting lost in C&P:)
Hence/I assumed MA Rodger’s 8.7K/km is the MALR @ 600mb, 250K, though I didn’t verify the number for myself (didn’t have a skew-T chart). – wait, https://www.noaa.gov/jetstream/upperair/skew-t-log-p-diagrams ,
https://www.noaa.gov/sites/default/files/2022-06/skew_t_fullsize.pdf …
Of course, the climate system is not generally in a steady-state even at a climatological equilibrium, but this still broadly applies; the dry air has to cross the θ surfaces.
net flow of heat in or out of a material volume; material; a material ___ is a ___ that moves with the material (I’ve never seen the term applied to volume but it’s a logical extension of familiar usage for material surfaces). PS I could use “warming” and “cooling” to refer to the material derivative of T (DT/Dt is dT/dt following the motion of the material) whether that is adiabatic or diabatic, and might also in general use for for the Eulerian derivative ∂T/∂t (occurs at a specific coordinate location as material moves through/by), but I use heating for a diabatic process; unfortunately there isn’t a clearly distinguished option for negative heating (“colding”?) which applies at all T (otherwise freezing could work for limited cases, but…).
So why doesn’t the tropospheric ELR always follow a MALR? Well, there’s the fact that convective cloud bases are often/generally a bit above the sfc. But also, there’s the “large-scale horizontal motion – eg. baroclinic waves / frontal cyclones – driven by geographical variations in solar heating can stabilize the air to localized vertical moist convection” and the temporal cycling that stabilizes the air next to land surfaces at night (when conditions allow), etc. The significantly superadiabatic layer sometimes found next to the surface due to the ineffectiveness of convection on such small scales (is it like a cm?) and the simple fact that convection mostly does not penetrate into the land or liquid water (there is soil porosity and sea spray, but… well, vegetation does extend the surface into a volume in a sense but when you get to individual leaves… etc.) … also there’s the matter of entrainment of dry air (but won’t that limit the extent of moist convection?). But also, even where, as I understand it(?), moist convection would tend to establish a near MALR profile (deep tropics), there is some horizontal variation in conditions; moist ascent tends to concentrate in some regions, and maybe (?) some different regions are producing different moist adiabatic profiles, which may reach upward to different heights… so when you take an average… (?)
Of course, there will (tend to) be (depending on local conditions including the occurrence of other clouds) LW diabatic cooling on the tops and also somewhat along the sides of clouds, but not so much inside thick towering cumulus.
——– —
E. Schaffer, your own quotation of Inamdar, Ramanathan 1998 ( https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/1998JD900007 , p 32,192 – wow, that’s quite a tome!), my emph. and [numbered points] list format applied.:
From their abstract (my emph. ):
I didn’t read the entirety (so don’t take my word for it), but it seems like they’re considering the changes in relative humidity (and specifically arguing against a hypothesis from Lindzen?). Real world data support for the LR feedback for longer term anthropogenic climate may come from other studies (including perhaps some from the last 28 years?), not this one. The time frame is different and maybe the spatial scale (in the regional study(s?)) also has an effect on ∆LR. I would be interested in seeing an explanation for why. Maybe something to do with SST’s having a longer response time? (relatively muted seasonal changes?). There’s also of course that the Northern & Southern Hemispheres are out of phase in seasons, and matters of atmospheric circulation…
While I do appreciate your dedication on learning about the MALR, I think it is way more effective to just take away what is needed. The moist adiabat is smaller than a dry adiabat, the GHE is a function of said lapse rate, thus WV reduces the GHE and that effect is also known as “latent heat”..
Tropical storms may serve as a showpiece for how strong this effect can be. The intense evaporation / condensation there drops the lapse rate to a minimum, creating a column of hot (relative to altitude) buoyant air, which sucks in more moist air, and so on..
To understand those statements, both IR98 and D08, one needs to consider what the changing lapse rate means as a feedback, and how it specifically behaves in the proxies. In the regional and seasonal proxy the lapse rate emulates a pos. feedback. Assuming lapse rate there was acting as a neg. feedback, the resulting difference was wrongly taken as evidence of WV feedback. Only in the interannual proxy the lapse rate does as it should, making that one even more interesting and kind of the “epicenter” of the issue.
In the real world of course the lapse rate remains a neg. feedback, and that is what you have in the models. Yet there is another problem, as discussed in Santer et al 2005, or pointed out by John Christy. The tropical lapse rate is not doing what it should, ie. the “hot spot” is missing. The lapse rate should shrink, but it refuses. There are related phenomena like stagnant pan evaporation, declining relative humidity, declining cloudiness, and so on..
These instances have provoked a lot of different reactions by “critical” people jumping to conclusions, like Christy, Lindzen oder Clauser lately. And while there is definitely something going on, I an assure you all these conclusons were wrong. I would rather go with P. Minnis there and not just consider artificial cirrus clouds, but also that they might be limiting the natural water cycle underneath, while at the same time heating the climate.
I didn’t read the entirety (so don’t take my word for it), but it seems like they’re considering the changes in relative humidity (and specifically arguing against a hypothesis from Lindzen?)
And yes, the IR98 statement is a rejection of a notion brought up by Lindzen, where he suggested more evaporation could trigger more precipitation which then somehow lowers humidity levels..? Does not make much sense..
E. Schaffer,
D note that you begin to seriously try my patience.
Let us keep to the two lines of discussion initiated within the July UV comment thread. These two were (1) Why the Lapse Rate of the troposphere is roughly 6.5ºC /km, and (2) Why the science says the Lapse Rate feedback is -0.5Wm^-2/ºC when your “LR feedback = surface WV latent heat flux = ~80Wm^-2 would suggest a value 5-times-larger with a 33k GHE, thus our “contradiction”.
(1) The contention between us here is that you appear to see the 6.5ºC/km as resulting from MALR conditions while I say it is far far more complicated than that.
I do not see anything from you to demonstrate that the MALR would lead to a tropospheric 6.5ºC/km lapse rate. I have attempted to show in different why why there is a lot more going on than the appearance of a MALR.
Perhaps another attempt at explaining this complexity would be more successful.
If you recall, I suggested that the energy flux into the atmosphere from condensing WV was not the only energy flux that would have a bearing on the Lapse Rate. Let’s have another go at that.
You insist the WV energy flux is different because all the other fluxes are “put into atmosphere at the bottom and it all gets lost at the “top” (actually different altitudes within the troposphere).” And indeed, the WV carries insensible heat up into the troposphere causing heating through the height of the troposphere.
You thus insist this WV warming uniquely impacts the Lapse Rate, reducing it and thus also a strong negative feedback on the GHE.
But consider this. As well as WV, the GHE itself warms the troposphere.
Consider that Fig 7 Kiehl & Trenberth (1997) shows 324Wm^-2 of “Back Radiation”. Of course, the stupid air molecules have no sense of direction with this radiation. As well as their 324Wm^-2 coming down, there is also (almost) 324Wm^-2 going up.
But when we get to the top of the atmosphere, the IR radiation from the atmosphere out into space is only 195Wm^-2. We are missing (324 – 195 =) 129Wm^-2. Where did it go? Could it perhaps be “causing heating through the height of the troposphere” like the 80Wm^-2 WV flux? And with 129>80, more heating?
(2) (Just to explain the ‘ignoring-the-contradiction’ thing. You wrote back in July that I “brought this perspective up before and (you) ignored it.” You thus admitted to not picking up on my raising the “contradiction”“One of the two must be wrong. From there on we have different philosophies. You would want to solve said contradiction by ignoring and denying it – a well known pattern. As if contradicting perspectives would provide us with a choice what to believe. For me it was reason to analyze the contridication in find out why it exists.” [My bold] So has your analysis shown why the “contradiction”“contradiction” has been resolved saying “as far as the contradiction goes, we know the one that is NOT wrong.” The reasons given for this view is no more than that I “could not produce a single good argument why that was not so.” If you mean by “a single good argument” one which you judge to be “a single good argument” then you are probably correct. But we do not live in your solipsistic world. We exist in thr real world where science does a pretty good job of solving contradictions, the same science which you here insist is flat-wrong.
I’m happy to hear folk explain why some scientific thing is wrong. But in this matter, I see no serious or cogent explanation from you.
Our previous to-&-fro on (2) concerned two references, Inamdar & Ramanathan (1998) and Dessler et al (2008). You do not explicitly make response to my rebutal of your use of Inamdar & Ramanathan (1998).
Up-thread you quoted from Dessler et al (2008) Sec 4.2.3 [para 41] which ends saying “This result demonstrates the unsuitability of using variations in different regions in our present climate as a proxy for climate change.” This (& the Inamdar & Ramanathan (1998) quote) was apparently an exemplar of “scientists themselves in many instances already realiz(ing) they were riding a dead horse.”
I remain unable to see why this shows they would be “scientists … riding a dead horse.” Indeed, up-thread I pointed out the meaning of this Dessler et al quote and why it demonstrates no such thing. I also pointed to para 39-40 which contradicts your contention that increased evapo-transpiration (your 80Wm^-2 surface flux which will increase with surface temperature Ts) diminishes the GHE. It says “as Ts increases, so does the lapse rate, and the general effect of this is to reduce dOLR/dTs, and therefore (reduce) OLR, below what they would be if the atmosphere maintained a constant lapse rate.” Conversely, in your version of things, such adjustment of Lapse Rate would increase OLR and diminish the GHE.
Your response now to my comments on Dessler et al (2008) is surprising. You insist you have “explained already” and then regurgitate your understanding of not Dessler et al (2008) but Inamdar & Ramanathan (1998).
Regarding your understanding of Inamdar & Ramanathan (1998), the “proxy”** you talk-of is your description of the adoption of Ga as a measure of the GHE (Ga = OLR[surface] – OLR[toa]). What you appear to be saying is that you agree with their result but it is the adoption of this “proxy” that makes it all wrong.
(** Note that Inamdar & Ramanathan don’t themselves use the term “proxy”. Dessler et al (2008) does use the term but in a different context.)
Given this ‘It’s the proxy’s wots wrong!!’ assertion, the bulk of your ‘it-is-best-explained-in-numbers’ is setting out what the use of this “proxy” achieved*** with nothing describing why use of this “proxy” is allegedly all wrong.
(*** Inamdar & Ramanathan (1998) cannot provide a value for the Lapse Rate Feedback but does set out a value for the WV+LR feedback in a rather laborius way in eqs 5-to-8 which is where the ‘it-is-best-explained-in-numbers’ 3.3 – 2 = 1.3 comes from, the +1.3 for WV+LR being a problem for your insistence that WV+LR must be negative.)
Up-thread you set out your quotes from Inamdar & Ramanathan (1998) and Dessler et al (2008) adding “There are many more of these instances.” Perhaps those “many more … instances” contain papers that better illustrate your grand theorising.
I suggested up-thread that “If you are trying to present some serious criticism of Inamdar & Ramanathan (1998), you have managed a very good job of hiding it!!!!” Your attempts to now correct that situation are verging on incoherent bullshit. Translating (or more precisely ‘divining the meaning’ of) incoherent bullshit is never a precise conversion and ‘translating’ what it is you are trying to say in this interchange is not producing any outcome whatever.
I’d guess that the underlying gripe you are presenting continues to be that you still see the WV+LR feedback being calculated without consideration of the impact of the LR feedback on GHE processes other than ΔWV. But after all these words, that remains just a guess.
The “Chung et al 2010s” here & here
ES: neither the dry- nor the moist adiabat are physically dependent on GHGs.
The value of the lapse rate doesn’t, but whether the atmosphere is actually tending toward the lapse rate does depend on GHGs. Without them you get an isothermal atmosphere.
ES: Sure you could say WV is a GHG, but it is only considered as a condensating gas, not a GH-agent.
BPL: Absolutely wrong. Every model of Earth’s greenhouse effect has to include water vapor as a greenhouse agent. Every single one.
E. Schaffer: “WV is only considered as a condensating gas, not a GH-agent.”
BPL: “Absolutely wrong. Every model of Earth’s greenhouse effect has to include water vapor as a greenhouse agent. Every single one.”
Including those to which E. Schaffer ….. has referred here: Schmidt et al. 2010, Table 1: H2O (Vapor) 39.0% to 61.9 % of GH effect – “Each Absorber Being Removed (Minimum Effect) vs. That Absorber Acting Alone (Maximum Effect).
What’s the point of explaining to ES complicated details, when the guy can’t understand even the most obvious things in what he reads?
1. Yes, theoretically likely true. However, that is an extreme assumption where the troposphere was devoid of anything absorbing / emitting radiation. No GHGs, no clouds, no aerosols. Practically their concentration is irrelevant..
2. No, not wrong, but right. I was talking about the calculation of the MALR, and not climate models.
E. Schaffer: “ No, not wrong, but right. I was talking about the calculation of the MALR, and not climate models.”
Then – who cares? We make climatic projection and provide information to the society based on highly advanced, complex climate models looking combing input, and spatial and temporal variability, of MANY DIFFERENT processes affecting climate.
You can’t invalidate them by making cherry-picking a few numbers from the climate models
and putting them in one narrow concept (MALR), using massively simplifying assumptions and rejecting any criticism of them by calling the critic “nihilistic”…
On the Arctic sea ice minimum — the 2026 melt season has been unusual with the persistent warmth in the Kara and Beaufort seas. Will be interesting to see how the final September extent compares to recent years once the numbers stabilize. The evidence-based policy concern is timely given how many decisions are being made without reference to the underlying physics.
(Cont. from https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849838 … )
Trying out different wordings:
Hot and cold abs. cross sections (σ_a) will, due to the first glowing brighter than the second, have a net radiant flux from hot to cold (“from point of emission to point of absorption”**), but only to the extent of their size and number, and to the extent that they can see each other.
(** https://www.realclimate.org/index.php/archives/2024/08/unforced-variations-aug-2024/#comment-823976 )
When opacity is increased, you can’t see as far, and so, at any given location (POV), the (spectral) radiant flux densities (F_ν) and (spectral) radiances (L_ν) in various directions depend less on more distant conditions and more on conditions nearby.
I found a new source for climate science information based on this post on Bluesky:
https://bsky.app/profile/arepublic86.bsky.social/post/3ms6bzvba4c27
This resource reminds me a bit of Wood for Trees
https://www.woodfortrees.org/
with the options available for prioritizing the information to make more meaningful graphics.
https://showyourstripes.info/
This particular graphic shows the increase in hot summers over the past 130 years, and my favorite, temporary, hottest year on record, 1998 – in medium dark red.
https://showyourstripes.info/c/northamerica/unitedstatesofamerica/all
My area of SoCal is in the middle of a ten-day streak of triple digit temperatures, the new normal. Stay hydrated.
Tomas: 1 Aug. “I do not think, however, that comparing JCM with “E. Schaffer” is appropriate.
Why not? BOTH are anything-by-GHGs deniers. BOTH build up their ego on their belief that they, despite being lay people, are so brilliant that they discovered something that thousands of professional climatologists failed, or refused, to notice. NEITHER is willing to submit their dearly-held belief to the test of submitting their work to a peer-review paper.
And most importantly – by their fruits you shall know them – by trying to discredit the climate science and diverting the attention of the society from the urgency of the mitigation our GHG emissions, BOTH of them help to delay and diminish this mitigation, and make the AGW worse. Consequently, BOTH bear moral responsibility for their share of the additional deaths and suffering, and species extinctions caused by the delay/weakening our actions to reduce GHG emissions. Plus BOTH are “useful idiots” of Russia, Saudi Arabia, Iran and other petro-dictatorships, whose economies, wealth of their oligarchs, ability to wage wars, support terrorism, and repress their own population is DEPENDENT on the world continuing to buying their oil and gas.
So any differences between E. Schaffer and JCM are cosmetic, different tools, but the same goals and the same fruits of their actions, as listed above.
If anything – in his attempts to discredit science and central role of GHG mitigation – JCM went much further than E. Schaffer – dismissing climate modelling as “imaginary process mechanisms [with] rules about how things ought to be [according to their authors]“.
and blaming climate scientists for “up to 40% of planet’s land degraded”:
“It’s hard to imagine denying or actively minimizing the consequences to realclimates due to an artificial fixation and overemphasis on the outputs of trace gas and aerosol forced model estimates.” [ (c) JCM])
So, on the second thought, perhaps you are right that JCM is not as bad as E. Schaffer – he may be worse.
E. Schaffer 1 Aug: “Sure I could explain the solution of said dissent, which I happen to know”
I will not hold my breath. After all, you were incapable to answer a simple yet fundamental question to your claims: namely: “So what?”. See for instance my earlier post:
=====
Piotr to E Schaffer: “So what?
– being a passive feedback, NOT a forcing – it does not drive AGW, nor can we do anything about it. Meaning that it has NO societal value of informing the society on mitigation of AGW.
– nor does it improve our understanding of the climate, since changes in lapse rate are already calculated in climate models which “dynamically resolve or parameterize vertical temperature profiles, convection, and radiative-convective balances”.
Because of that – what difference would it make to, say out of the projected by models 2C of warming – +0.5C or -0.2C was the result of lapse rate feedback?
So far no answer other than clicking your famous ruby red slippers three times while wistfully whispering “ES: Well, it is exactly about improving our knowledge, and decisevely so! ”
===
And until you can answer the “So what?” question, what’s the point in engaging you in technical details of something the societal or intellectual relevance you yourself can’t identify?
I do understand you use a deductive approach, while I am doing induction. You would say some detail is not so important, but the bigger picture is. I would say the bigger picture is the result of all the details.
You are not alone in this, rather it seems everyone is doing so. Everyone is looking at the details as far as they fit the bigger picture. And if one detail contradicts the it, it must be wrong. It will be considered as far it fits, and if not it gets ignored, downplayed, or tampered with. Then of course you will end up with a large quantity of evidence, with poor or zero quality, and a bigger picture that is based on nothing eventually.
The problem however starts way earlier. I should add I also have some background in quality assurance. If there is even one single detail contradicting the bigger picture, that is code red already. There is no such thing as a tolerable or negligible exception. If you ignore it, it is only going to get worse, and inflationary so. Think of the 737MAX story – the rules are there for a reason.
ES, you aren’t even following your own logic. For example, last month, I brought up several “details” about your incorrect assumptions. These included not allowing for changes in the tropopause height or temperature, so that a lower lapse rate would necessarily result in colder surface temperatures. You accused me of being “nihilistic.” Ignoring the “details” that I and others raise ought to be code red for you, by your own professed standards. Instead, I expect that you will not only tolerate own your false assumptions, but work out a way to obfuscate, minimize, or ignore the objections to them.
My “own” logic is the one that scores a 100% on IQ tests. I am fine with that.
You questioned what lapse rate I mean. I clearified it was the tropospheric one. And you argued “so what”, which I referred to as nihilism. You did not bring up a single “detail” that would even suggest anything incorrect.
Yet there is one “detail” I did not respond to, because it misses the point: Yes, as the troposphere warms more to a shrinking lapse rate, it should also hold more WV. This will NOT change the neg. lapse rate feedback, but it is a vital factor in WV feedback as it enhances it.
That is an important detail if you use radiative transfer models to assess WV feedback. If you check modtran it will underestimate WV feedback just because it holds the lapse rate fixed. Modtran would give you 1.2W/m2 (US std) and 1.73W/m2 (trop) of WV feedback with clear skies. Allowing for clouds and stratosphere, these figures drop significantly, like 0.7W/m2 and 1W/m2. That is a long shot from some 1.8W/m2 of global WV feedback, and ~3W/m2 in the tropics, which are central estimates within models. Said enhancement is badly needed to increase WV feedback and get it closer to central estimates. Yet it is not nearly enough. You might get like 0.9 to 1W/m2 globally and ~1.4W/m2 in the tropics.
You might think the effect of the lapse rate reduciton is something that comes on top, a luxury problem, but actually it is already included – and overestimated. Also with it included, WV feedback is far too low.
in Re to “E. Schaffer”, 7 Aug 2026 at 6:27 AM,
https://www.realclimate.org/index.php/archives/2026/08/unforced-variations-aug-2026/#comment-850303
Sir,
I do not understand what you try to demonstrate by your comparison of some modtran simulation with a “central estimates within models”. I suppose that modtran is also a kind of model, a simple one, isn’t it? Could you clarify?
Furthermore, if in your comment of 28 Jul 2026 at 8:16 PM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850162 ,
the paragraph “Yet in my understanding, the “radiative exchange between surface and atmosphere” should not include the atmospheric window. There the atmosphere can not receive nor emit radiation, so it is not taking part in such an exchange..”
should have perhaps represented your reply to my objection of 23 Jul 2026 at 6:23 PM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849972
that you are not allowed to subtract the energy flow through the “atmospheric window” from the surface radiation, because you would thus incorrectly minimize the net radiation,
it appears that instead replying to clear arguments presented by JCM, you decided to introduce a new concept “radiative exchange between surface and atmosphere”.
I would like to ask why do you think that replacing clearly defined net radiation with your new concept is useful? What new understanding brings your magic “radiative exchange between surface and atmosphere” to climate science?
Best regards
Tomáš
ES: My “own” logic is the one that scores a 100% on IQ tests.
BPL: IQ tests aren’t graded in percent. They are arranged so that a large population gets a mean score of 100 with a standard deviation of 15. Or are you saying your score is 100?
E. Schaffer 7 Aug: My “own” logic is the one that scores a 100% on IQ tests.
Then perhaps it’s time to stop taking the IQ tests for the 5-th graders? ;-)
Or the tests Trump bragged about acing (telling an elephant from a tiger and a giraffe!) ?
John has pointed to the contradiction between what you declare about yourself in one post (an inductive mind building a big picture from careful consideration of minute details) and what you do in another post – i.e. dismissing the details of your major simplification, that if accounted for – countering your “big picture ” findings, and calling such considering of the real world “details” as ….”nihilism”, I quote:
===============
– ES: Holding other things constant, this smaller lapse rate reduces the GHE and provides a cooler surface
– JP So what? Other things aren’t constant. You get more water vapor into the lower atmosphere by warming the surface, not cooling it.
-ES “sure, but nihilism is not going to provide insights”
=================
So not only you are incapable to see the contradiction between what you say and what you do, your are still unable to see it even after John has pointed it to you.
This tells us more about your intelligence than your assurances how very intelligent you are.
By their fruits, not by their declarations about themselves, you shall know them.
@Tomas
Modtran is a radiative transfer model, like hitran, and basically gives the same output, though less precise. Although they are climate related models, they are no way “climate models” in the common sense. They try to give the best possible emulation of radiative transfer. And I am just pointing out, these models give a WV feedback only half the magnitude assumed as central estimate.
Try for yourself..
https://climatemodels.uchicago.edu/modtran/
JCMs definition of “Rlw” was a bit ambiguous. But if it meant to be the radiative exchange between surface and atmosphere, it should not include radiation that goes from the surface through the atmospheric window right into space, because there is no exchange with the atmosphere.
However, it smells like the old “back radiation = GHE” fallacy anyhow.
in Re to John Pollack, 4 Aug 2026 at 10:15 PM,
https://www.realclimate.org/index.php/archives/2026/08/unforced-variations-aug-2026/#comment-850286
Hello John,
I would like to ask if I understood correctly that you spoke about environmental lapse rate?
And yet another question came to my mind. If changes in latent heat flux and/or water vapour content in the air column change both the surface temperature as well as entire troposphere temperature profile and tropopause height, is it correct to assume that the tropopause height exhibits some temporal fluctuations? I would expect that e.g. above hurricanes with their intensive latent heat flux, tropopause height might exhibit a “local extreme”?
Greetings
Tomáš
Tomáš,
Yes, I was speaking of environmental lapse rate. The tropopause height exhibits large temporal fluctuations outside the tropics, and (usually) smaller changes within the tropics. The overall height in the tropics is set by thunderstorms, which release heat mostly in the lower levels of the troposphere, but create strong updrafts that push up the height of the tropopause. As you note, the most extreme case is above a tropical cyclone.
Just looking at the balloon soundings over the U.S. and southern Canada this evening, I see tropopause heights of 13 to 13.6 km in parts of the southern U.S. At Tampa in southern Florida, the sounding contains full tropical moisture, and is nearly saturated from the surface to the 13.3 km tropopause, where the temperature is -68C. In the desert Southwest, the soundings are close to dry adiabatic up to 600 mb, and not as steep higher up. Tropopause temperatures are around -65C. Cloud top temperatures in hurricanes can be colder than -80C. On the other hand, there is a cool cyclonic mid tropospheric circulation over central Canada. One sounding that samples this cool cyclone is dry adiabatic up to around 800 mb, weaker than moist adiabatic above that, and the tropopause is 8.8 km, where the temperature is around -40C.
ES, moist convection allows a decreased lapse rate near the surface, but results in a deepening of the troposphere, so that temperatures at the tropopause are actually colder over the tropics than at higher latitudes. That is why not considering the height of the tropopause is a “detail” that you shouldn’t ignore.
ES: if not it gets ignored, downplayed, or tampered with.
BPL: It’s a CONSPIRACY!
Piotr: “And until you can answer the “So what?” question, what’s the point in engaging you in technical details of something the societal or intellectual relevance you yourself can’t identify?”
E Schaffer: “ I would say the bigger picture is the result of all the details. ”
You still understand nothing. If you can’t answer the most fundamental question to your “big picture” namely: “So what?” – if you can’t answer this most fundamental question to your supposed “big picture” what’s the point of discussing the details of that supposed “big picture”?
And here again are these fundamental questions that you can’t dismiss with your blabbering how your evading an honest answer to these question is your presumably admirable “inductive reasoning”.
=====
Piotr to E Schaffer: “So what?
– being a passive feedback, NOT a forcing – it does not drive AGW, nor can we do anything about it.
Meaning that it has NO societal value of informing the society on mitigation of AGW.
– nor does it improve our understanding of the climate, since changes in lapse rate are already calculated in climate models which “dynamically resolve or parameterize vertical temperature profiles, convection, and radiative-convective balances”. Because of that – what difference would it make to, say out of the projected by models 2C of warming – +0.5C or -0.2C was the result of lapse rate feedback?
So far no answer other than clicking your famous ruby red slippers three times while wistfully whispering “ESchaffer: Well, it is exactly about improving our knowledge, and decisevely so! ”
===
Particularly that you based your attack on climate science on your staggering ignorance on how their models work (E. Schaffer: “W Vis only considered as a condensating gas, not a GH-agent”)
See your post above.
UAH have posted July’s TLT numbers with a global anomaly of +0.48ºC, a tad up on the June anomaly (+0.46ºC). May2026 remains the highest anomaly of the year-to-date, Jan-to-May running +0.35ºC, +0.39ºC, +0.38ºC, +0.39ºC,ºC,+0.53ºC.
The warming June→July was all NH (+0.54ºC → +0.68ºC) with cooling in the SH (+0.38ºC → +0.27ºC).
… … … UAH July anomalies
… … … … Global … … NH … … …SH
2023 … +0.56ºC … +0.58ºC … +0.53ºC
2024 … +0.73ºC … +0.85ºC … +0.61ºC
2025 … +0.36ºC … +0.49ºC … +0.23ºC
2026 … +0.48ºC … +0.68ºC … +0.27ºC
The ranking of warmest Global TLT Julys in UAH puts 2026 +0.48ºC strongly in 3rd place, ahead of 4th 1998 +0.38ºC, 5th 2025 +0.36ºC, 6th 2022 +0.32ºC, 7th 2020 +0.30ºC, 8th 2016 +0.26ºC,
…
And I should also admit a “wrong glasses” moment with the July ERA5 SAT numbers.
I reported a July global anomaly of +0.70ºC which was actually +0.67ºC making July 2026 (shock-horror) slightly below both July 2023 (+0.72ºC) and July 2024 (+0.68ºC) but with all three still well above 4th-place 2025 +0.45ºC, 5th-place 2019 +0.40ºC, et al.
For those interested in the on-going temperatures of 2026 & the potential for 2026 becoming hottest-year-on-record (which is looking less than likely with ERA5 SAT), I have added a graphic of year-on-year daily & 30-day average ERA5 SAT comparing 2023, 2024 & 2026 to the collected graphics at The Banana!!! Watch.
Plotting temperature against CO₂ rather than against time has a property I haven’t seen exploited: it puts the deep-time reconstructions, the observed record, the CMIP6 SSPs and the CMIP7 ScenarioMIP pathways on one set of axes. Normally you have to look at those in four different places, and you can’t see how they relate to each other while they’re apart.
Together, a few things become visible. Where the modern excursion sits relative to the relationship the two variables held for the last 66 million years. How far a projection is from that relationship, and in which direction. And how the two scenario generations line up — Medium and SSP2-4.5 are nearly indistinguishable through the end of this century, then diverge on mitigation timing rather than on ambition.
That last one may be the most immediately useful, since it’s a concrete answer to the suggestion that the new scenario set represents a change in expectations rather than a normal update.
My latest effort at combining and comparing these records and projections can be found here, https://justdean.substack.com/p/climate-sensitivity-the-most-contestable
I’d welcome feedback and being told if it is less useful or novel than it looks.
Dean Rovang, proposing: “ Plotting temperature against CO₂ rather than against time – a few things become visible. Where the modern excursion sits relative to the relationship the two variables held for the last 66 million years. How far a projection is from that relationship, and in which direction”
A few points:
1. your approach removes the very variable that may explain this “modern excursion” – given the thermal inertia of system the very recent in geological terms spike in Co2 didn’t have enough time to have the full effect on temperature.
2. sensitivity of T to changes in Co2 would be different in periods when your have ice in polar regions and when you don’t (different room for ice/snow albedo amplification of the CO2 effect)
3. in the last 66 mln the concentrations and the role of non-CO2 GHGs may have been different – which is ignored if we plot T vs CO2 only.
4. when plotting over large range of Co2 – one should plot T against logCO2 rather than CO2
And last but not least …. it has already been done:
– in short time scale – since preindustrial see for instance visualisation: <a href= "https://www.youtube.com/watch?v=IXHOc2rmSPM", https://www.youtube.com/watch?v=IXHOc2rmSPM
– over geological time-scale – I see Judd et al. 2024 as most reliable – showing on the same graph of BOTH T and Co2 against time, and T vs Co2 graph, showing a nice correlation in Paleozoic and Cenozoic, and discussing potential reasons for the lack of correlation in Mesozoic (possibly my points 2-4 and some others ) – see graph 4b from Judd et al. 2024 the end of : <a href=" https://skepticalscience.com/new-favorite-graph-opposite.html", https://skepticalscience.com/new-favorite-graph-opposite.html
To Tomas re: https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850229
thanks
To the thread more generally: I can’t resist discussion of water cycling so I will drop some more thoughts here.
As climate warms and tropospheric temperatures move to lower pressures, the cooling rate (K day−1) increases. ok.
Atmosphere is in approximate equilibrium where the radiative cooling is balanced by the convergence of the upward flux of energy by atmospheric motions.
energy.
Allow for some warming to happen:
warmer parcels holding more water vapor, per unit mass.
Each kilogram of ascending air able to condense more water vapor, per unit distance.
Masses of air rising some distance containing an increasing fraction of water vapor. A local, per-unit mass effect. Local parcel scale thermodynamics. Depending on temperatures and slope of saturation vapor pressure curve. ok
Contrast that with what moist adiabatic lapse rates do not depend on (global scale energetics):
Total atmospheric mass flux
Total condensation
How much upward flux of energy is occurring for the globe.
In steady state, the integral of condensation (precipitation) is equal to the evaporation, called the latent flux.
These are constrained by energy budgets, not local temperatures. We are interested in the convergence of upward flux by atmospheric motions to close the balance of radiative cooling of atmosphere.
We can constrain the rate of radiative cooling of atmosphere and latent flux in steady-state using surface budgets.
Surface net radiation = flux of sensible and latent heat.
Increasing temperature 1K may be associated with a change of LW down – LW up, say perhaps +1 W/m2, assuming both terms increase and the net is +1.
A representative increase of 1 W/m2 per K in available surface energy.
so we allow +1 = H + LE, and for simplicity partition 10-20% to sensible heat flux and 80-90% latent heat flux at current temperatures (by the slope of saturation vapor pressure curve and psychrometric constant, called equilibrium partitioning.)
That is up to 0.9 W/m2 increasing latent flux per K, or say approximately 1% increasing moisture cycling per K.
It may be less if atmospheric solar absorption should increase and take away from surface budgets, and it may be more if increasing surface solar absorption is associated with global warming.
So we have 1% increasing energy throughput on atmospheric motions per K against ascending parcels carrying up to 7% more moisture per unit mass…
With warming, each kg of ascending air is capable of condensing substantially more water over a given vertical distance than would be implied by the small increase in globally integrated latent heat flux
The rates are reconciled in part by a weakening of the convective mass flux (kg per m2 per s). Each kg of ascending air condensese more water and releases more latent heat, which is why moist lapse rates become shallower. Through the energy budgets we can see the constraint: fewer kilograms of moist air are ascending per unit time, because the column-integrated global latent flux does not rise at the same rate.
It can also be visualized as an increasing depth of convection, and increasing water vapor duration. Or as a decreasing frequency along with a possibility of increasing intensity.
The relaxation of moist adiabats with warming is a local thermodynamic property of the air that rises; and the heat flux is an energetic integral over how much air rises.
The latent heat flux is about how many kilograms of air are rising per unit time and how much latent heat each kilogram releases. An integral over the entire global circulation.
The moist adiabatic lapse rate is a local thermodynamic property: it depends only on how much latent heat a single saturated parcel releases as it rises through a given vertical distance.
More on the specifics of radiation in this context is available in The Vertical Profile of Radiative Cooling and Lapse Rate in a Warming Climate from Hartmann
https://journals.ametsoc.org/view/journals/clim/35/19/JCLI-D-21-0861.1.xml