This month’s open thread on climate topics. Impending El Niño, fossil fuel funded wedges, retirements, heat waves, and the same old, same old, from the usual suspects. Try to discuss these things calmly!
Reader Interactions
330 Responses to "Unforced Variations: July 2026"
Ray Ladburysays
Zebra: “Do you really think what you just wrote is useful,…”
I’m sorry, Zebra. When I was an editor, they allowed us to make the assumption that our audience was literate. If you like, I can try to write something at your level–perhaps with bigger words and lots of pictures on the page?
zebrasays
Ray, I’m with Pascal on this. I think you should take the time to write a shorter letter, which you obviously don’t. You know, edit thyself.
I also think 99% of the population would greatly benefit from fewer words and lots of pictures in understanding and relating to science. People can be very “literate” without knowing or understanding jargon in a specialty, or even knowing that Physics Today exists, much less reading it.
My experience tells me that the first step in educating someone is listening to them and understanding “where they are at”; lecturing them at your own level isn’t going to help them move forward… usually the opposite.
B = Planck function ;
“L_{4π} = directionally averaged radiance averaged over the whole sphere (4π sr) ”
—
“net (spectral) radiant cooling per unit mass = 4π · ( B − L_{4π} ) · k_{a,air} ___(assumes isotropic σ_a, good for GHGs and cloud droplets, maybe not some ice crystals) ”
= net (spectral) radiant flux out of a unit mass
= net (spectral) radiant flux out of a unit volume (flux divergence), divided by density of that volume
k_{a,air} = μ_{a,air} ÷ ρ_{air} = ∑_i ( n_i · σ_{a,i} ) ÷ ρ_{air}
n_i = number density of i
——————————————–
My understanding is that the ratios of the rates of spontaneous emission, stimulated emission, and direct absorption, are, at LTE/LEDNLIE (Local Equilibrium Distribution of Non-Latent Internal Energy), determined by T (and ν and …) via the ratio of the populations of particles in states 2 and state 1 (where E₂ – E₁ = E₂₁ ); this is easiest to show for distinguishable particles which (at LTE/LEDNLIE) fit a Maxwell-Boltzmann distribution:
Maxwell-Boltzmann distribution: population in state 2 / population in state 1 = exp[ − E₂₁ / (kT) ]
But I expect it works out more generally (eg.
PS for indistinguishable fermions, if I remember correctly, you can get the same ratio e^(− E₂₁/kT) If you take the product of the probability that one state is occupied and the probability the other state is not occupied, and the product of the reverse, and then the ratio of those products.
@ LTE/LEDNLIE, for the optical transitions between a pair of states1 and 2, the absorption cross section σ_a is the blackbody area that would (given a spectral radiance L_ν from a direction) absorb at the same rate as the average per particle (eg. over all CO2 molecules of a given isotopologue) rate of “total absorption” = direct absorption − stimulated emission. If you take the average per particle rate of spontaneous emission and fit it into the absorption cross section σ_a, you get the Planck function.
I *believe* collisions act like observations of a quantum system eg. molecule; in isolation, during a transition between a pair of states, the system is in an evolving superposition of the states. So collisions turn this into a probability; eg. if they occur when the transition is 1% of the way done, then 99% of the transitions are aborted, but the 1 % that occur have been forced to happen 100 times faster (causing collisional line broadening), so the overall transition rates – and thus rates of spontaneous emission, stimulated emission, and direct absorption – are unchanged.
patrick o twentysevensays
My understanding is that the ratios of the rates of spontaneous emission, stimulated emission, and direct absorption [for photons going away from a direction (θ,ϕ)] , are, at LTE/LEDNLIE (Local Equilibrium Distribution of Non-Latent Internal Energy), determined by T (and ν (ν = frequency of the photons; E = hν) and [incident (ambient) spectral radiance L from direction (θ,ϕ), and the real component of the refractive index]) via the ratio of the populations of particles in states 2 and state 1 (where E₂ – E₁ = E₂₁ ); this is easiest to show for distinguishable particles which (at LTE/LEDNLIE) fit a Maxwell-Boltzmann distributionstatistics:
oops! “so the overall transition rates – and thus rates of spontaneous emission, stimulated emission, and direct absorption – are unchanged.” (for given L, T, etc.)
ie. it’s the ratios:
stimulated emission / direct absorption ,
and
spontaneous emission / [ (either(stimulated emission or direct absorption or their difference) / L_ν ]
which are are unchanged (for given T, etc, at any L, for photons going away from a given direction)
Formatting fixed: (please replace last comment)
Wait, I forgot to reiterate a corrected form of my salient point:
It seems that collisional line broadening, and from that, I infer, collisions, preserve the integral over the spectrum of the absorption cross section σ_a for a given line (= ∫σ_a · dν ), (see below https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849826 ); therefore, for photons coming from a given direction (θ,ϕ), ie. going toward (π−θ , ϕ±π) (the opposite direction),
∫[ ( rate of direct absorption − rate of stimulated emission ) · hν ÷ L_ν ]
are/should be unchanged (I presume the same is true for direct absorption and stimulated emission separately). Note I assumed the rates were in terms of photons (per unit time per unit solid angle per unit material (ie. average per molecule of a type of molecule), hence the multiplication by photon energy E = hν.
I expect this also holds true for thermal Doppler broadening,
(except that time dilation would slow the rates of absorption and emission by the molecules as a function of their speed (so as some σ_a · dν is shifted away from the from the line center, some would disappear … and more would be redshifted than blueshifted), though I’d expect it would take some very very high temperatures (how high?) to make this significant.)
patrick o twentysevensays
… and
∫ [ ( rate of direct absorption − rate of stimulated emission ) · hν ÷ L_ν ] · dν
…
Since the electric dipole moment of the homonuclear diatomics is zero, the fundamental vibrational transition is electric-dipole-forbidden and the molecules are infrared inactive.[10] However, a weak quadrupole-allowed spectrum of N2 can be observed when using long path-lengths both in the laboratory and in the atmosphere.[11] The spectra of these molecules can be observed by Raman spectroscopy because the molecular vibration is Raman-allowed.
Dioxygen is a special case as the molecule is paramagnetic so magnetic-dipole-allowed transitions can be observed in the infrared.[11]
11:
Goldman, A.; Reid, J.; Rothman, L. S. (1981). “Identification of electric quadrupole O2 and N2 lines in the infrared atmospheric absorption spectrum due to the vibration‐rotation fundamentals”. Geophysical Research Letters. 8 (1): 77. Bibcode:1981GeoRL…8…77G. doi:10.1029/GL008i001p00077.
Abstract
Analysis of long path atmospheric absorption spectra and of laboratory absorption spectra in the 1600 cm−1 region has resulted in the identification of atmospheric quadrupole lines of O2 in its fundmental vibrational band within the electronic ground state. This led to the identification of similar atmospheric quadrupole lines of N2 in the 2400 cm−1 region.
And again, the favored de-excitation pathway at low altitudes (below several km) is still collisional, and these relaxation processes do not emit a photon. At STP, the energy of the CO2 vibrational band is well above the average thermal energy of the atmospheric gas molecules–so collisional excitation is a rarer process. …
but often enough to approximately maintain an excited population of each type of GHG molecule. In fact there will even be a tiny fraction which are excited above the lowest excited vibrational state (for each mode of vibration) (I think/infer that ‘hot bands’ are significant contributors on the wings of the CO2 band). (This is in accordance with the LTE approximation.) And occasionally they will manage to emit a photon before a collision.
… As such there is a net flow of energy from the IR band of the GHG to thermal energy of atmospheric gasses.
No, that’s wrong. For isotropic σ_a:
Net spectral radiant cooling per unit mass = 4π · ( B − L_{4π} ) · k_{a,air}
Ie.
Net spectral radiant cooling per particle (averaged for each type of particle (eg. over all CO2 molecules of a given isotopologue)) = 4π · ( B − L_{4π} ) σ_a
Ie.
it’s the difference between 4π · σ_a · the Planck function (spontaneous emission) and the 4π · σ_a · directionally averaged ambient radiance (absorption).
Ie. it depends on being able to see across a difference in T.
And the troposphere on average must experience net LW radiant cooling to balance SW heating and convergence of the convective heat flux from the sfc.
What matters here is that GHGs (and the sfc, cloud particles, etc.) emit radiant energy (eg. as (spectral) radiance) at a rate depending on their temperature (via Planck function B(ν,T) ), absorb at rate depending on the radiance they intercept, with both rates proportional to (for the GHG molecules and cloud particles etc.) their absorption cross sections σ_a .
Corrected/fixed:
… Now I will add that my descriptions of the physics (which can be summed up as https://en.wikipedia.org/wiki/Schwarzschild%27s_equation_for_radiative_transfer ) only includes spontaneous emission and ‘net’ absorption – actually I think it’s called “total absorption”; It does not include either the stimulated emission, nor the part of direct absorption which is cancelled out by that stimulated emission. So there is some radiance-dependent emission, and some absorption, that I did not include, but they sum to 0, so we can get this blackbody-like behavior from effective opaque objects that perfectly parameterize the effects of absorbers. (“in principle, ie. within the limits of statistical behavior being able to match probability, and assuming LTE or LEDNLIE *‡L*). ” https://www.realclimate.org/index.php/archives/2025/11/unforced-variations-nov-2025/#comment-842186 )
patrick o twentysevensays
“(I think/infer that ‘hot bands’ are significant contributors on the wings of the CO2 band)” – Oops! No, not quite:
https://rmets.onlinelibrary.wiley.com/doi/10.1002/qj.4485
“Radiative forcing due to carbon dioxide decomposed into its component vibrational bands†”
Keith P. Shine, Georgina E. Perry
(I think I read sec. 2 through perhaps the end of 3.1 (?) )
Although those ‘wing-’bands do involve transitions between pairs of states in which the lower energy state is a bending-mode (v_2) vibrationally-excited state; some even being the second such excited level** (and thus so I’d expect the overall band (assemblage of bands) will be effectively wider/narrower at higher/lower T – and of course the same is true of each individual band because of the distribution of CO2 among rotational states …
(T = 0 K should collapse the entirety of the band into just two lines (1 in R-branch and 1 in Q-branch) per isotopologue, I expect)
…– so I’d expect the effective widening (window closure) would be more/less per doubling of CO2.
**Also it turns out that the 1st excited bending-mode energy level has a degeneracy (g) of 2 – and I’m guessing, based on some logic, that the degeneracy of the n-th level of the bending-mode (where ground n = 0) will be n+1 (?); so the populations
(in (v_1, v_2 , v_3) = (x, n, z) , setting aside the Fermi-resonance states (correct phrasing?))
would then be proportional to
(n+1 [?]) exp[ −E_n / (kT) ]
PS E₂₁ /(kT) ≈ 3.1973 @ 15μm (E₂₁ ≈ 0.082656 eV ; ν ≈ 19.986 THz) and T = 300 K; e^(− E₂₁/kT) ≈ 4.0873 % is the ratio of the population of particles in state 2 to that of state 1 where E₂ – E₁ = E₂₁ ; given LTE or LEDNLIE (“lead/lede-‘n-lie/ly”) (Local Equilibrium Distribution of Non-Latent Internal Energy) and T = 300 K; for distinguishable particles. It’s easier to discuss states rather than energy levels because we don’t need to discuss degeneracy (g) of energy levels; I’m assuming the Einstein coefficients can be defined for pairs of states, not just pairs of energy levels.
Which means ≈ 8.17 % is the ratio of the populations in the corresponding levels. Note that this is a ratio; the fraction of CO2 molecules in v_2 = 1 will be smaller, ie.
≈ 8.17 % / [ ∑ ( 100 % + 8.17 % + … )]
Anyway, my ballpark est. of the rate of spontaneous emission from CO2 v_2 = 1 → 0 (which was… where is that comment?) would have been ~doubled if I had known this.
patrick o twentysevensays
“Although those ‘wing-’bands do involve transitions between pairs of states in which the lower energy state is a bending-mode (v_2) vibrationally-excited state; some even being the second such excited level**”
– or state of similar energy, judging from Fig.2 (see the FF transitions)
“Which means ≈ 8.17 % is the ratio of the populations in the corresponding levels.”
– @ 300 K.
patrick o twentysevensays
“(I think/infer that ‘hot bands’ are significant contributors on the wings of the CO2 band)” – Oops! No, not quite:
https://rmets.onlinelibrary.wiley.com/doi/10.1002/qj.4485
“Radiative forcing due to carbon dioxide decomposed into its component vibrational bands†”
Keith P. Shine, Georgina E. Perry
(I think I read sec. 2 through perhaps the end of 3.1 (?) )
Although those ‘wing-’bands do involve transitions between pairs of states in which the lower energy state is a bending-mode (v_2) vibrationally-excited state; some even being the second such excited level**
– or state of similar energy, judging from Fig.2 (see the FF transitions) –
(and thus so I’d expect the overall band (assemblage of bands) will be effectively wider/narrower at higher/lower T – and of course the same is true of each individual band because of the distribution of CO2 among rotational states …
(T = 0 K should collapse the entirety of the band into just two lines (1 in R-branch and 1 in Q-branch) per isotopologue, I expect)
…– so I’d expect the effective widening (window closure) would be more/less per doubling of CO2.
**Also it turns out that the 1st excited bending-mode energy level has a degeneracy (g) of 2 – and I’m guessing, based on some logic, that the degeneracy of the n-th level of the bending-mode (where ground n = 0) will be n+1 (?); so the populations
(in (v_1, v_2 , v_3) = (x, n, z) , setting aside the Fermi-resonance states (correct phrasing?))
would then be proportional to
(n+1 [?]) exp[ −E_n / (kT) ]
PS E₂₁ /(kT) ≈ 3.1973 @ 15μm (E₂₁ ≈ 0.082656 eV ; ν ≈ 19.986 THz) and T = 300 K; e^(− E₂₁/kT) ≈ 4.0873 % is the ratio of the population of particles in state 2 to that of state 1 where E₂ – E₁ = E₂₁ ; given LTE or LEDNLIE (“lead/lede-‘n-lie/ly”) (Local Equilibrium Distribution of Non-Latent Internal Energy) and T = 300 K; for distinguishable particles. It’s easier to discuss states rather than energy levels because we don’t need to discuss degeneracy (g) of energy levels; I’m assuming the Einstein coefficients can be defined for pairs of states, not just pairs of energy levels.
Which means ≈ 8.17 % is the ratio of the populations in the corresponding levels [@ 300 K]. Note that this is a ratio; the fraction of CO2 molecules in v_2 = 1 will be smaller, ie.
≈ 8.17 % / [ ∑ ( 100 % + 8.17 % + … )]
Anyway, my ballpark est. of the rate of spontaneous emission from CO2 v_2 = 1 → 0 (which was… where is that comment?) would have been ~doubled if I had known this.
patrick o twentysevensays
“I *believe* collisions act like observations … but the 1 % that occur have been forced to happen 100 times faster (causing collisional line broadening), so the overall transition rates – and thus rates of spontaneous emission, stimulated emission, and direct absorption – are unchanged.” … Well, the math might be more complicated than that, but somehow it must work out that GHG molecules emit photons even when at LTE. – because Schwarzchild’s eqn.
dL = (B – L) dτ
works:
Their absorption coefficient κ has units of m²/mol (caption of fig. 7), which is thus the absorption cross section σ_a per molecule, multiplied by Avogadro’s number.
So the integral ∫ κ dν for the line is constant over variations in collisional line broadening (γ = line width) – at least for a Lorentz line shape – which may be is an approximation**, but maybe there’s other p-related things going on (eg. tiny amounts of quasi-static p-broadening?).
Ie, at least approximately, and except for the effects of the curvature of Planck function over ν (should generally be tiny over γ), the rates of spontaneous emission and total absorption (direct absorption − stimulated emission) should be conserved over variations in collisional line broadening.
(PS I’m using ν, the symbol for frequency, because that’s a whole lot easier to write in this space than the notation for spectroscopic wavenumber, but that’s what I’m really referring to (they are proportional)).
Note the T-dependence of the distribution of molecules of a given type among their possible states is, *I believe*, the mechanism for the T-dependence of the line strength (the quantity that is, AIUI, conserved over variations in line broadening); this includes changing the stimulated emission/direct absorption ratio. This is separate from thermal Doppler broadening.
** sec. 5c. “Linearity of κ in pressure”:
The value of κ at any particular wavenumber has contributions from spectral lines representing many different vibrational and rotational transitions of CO2, and each of those lines has a shape that varies with temperature and pressure. To get an indication of how this sum of contributions might change with pressure and temperature, we can use the Lorentz line shape. The Lorentz line shape can be derived from first principles with some approximations (e.g., Van Vleck and Weisskopf 1945) and is supported empirically within about 100 line widths of the line center (Pierrehumbert 2010). Although line-by-line models use a line shape that deviates from this, especially far from the line center, the Lorentz line shape is adequate for our discussion here.
patrick o twentysevensays
Oops!
… the rates of spontaneous emission and total absorption (direct absorption − stimulated emission) should be conserved over variations in collisional line broadening.
– For a given T, and (for absorption) a given directionally averaged ambient spectral radiance L (≡L_{4π}; for isotropic σ_a), which of course can be changed by line broadening (reduced opacity at/near the center (you can see farther), & exposure to L over a larger bandwidth, including bandwidth where you can see farther).
patrick o twentysevensays
Okay, clarification: For a type of molecules experiencing a type of line broadening (wherein the line shape (eg. Lorentz) is conserved – okay, I’m just assuming this works for…), each point ν in the spectrum maps to another point at some multiple LB of its original distance ∆ν from the line center, and the σ_a and thus κ decrease to 1/LB of the original value; since these molecules are only ‘seeing’ (absorbing) in proportion to their σ_a, they experience a reduction in optical depth and can see farther through themselves. But from the perspective of other material and at a fixed ν, reductions in optical depth near the line center occur with increased optical depth farther out, and I think that some of the effects of the later can/may dominate because nonlinearities…
________________
Related to that. With some approximating simplications (eg. the Planck function’s sensitivity to T being constant over some spectral bandwidth, opacities and other optical properties contributed by/of other materials being spectrally invariant… and ignoring variations in of line broadening (width and shape) and line strength over height (as z or p)), the same given optical depth τ contributed by a GHG eg. (an isotopologue of) CO2 …
(Actually let’s just assume CO2 has a set mix of its isotopologues and use a single average σ_a(ν,…) per CO2 molecule – it doesn’t but for the simplicity of wording … most is ¹²C with 2¹⁶O ; fossil C is relatively ¹³C depleted… anyway…)
… will result in the same distribution of T and therefore the Planck function B_ν (T) over the total optical depth τ (CO2 + H2O + clouds + etc.), so the resulting spectral radiances L_ν at each location (p or z, ø,λ) and for each direction (θ,ϕ) will be the same, as thus will be the spectral flux densities (F_ν↑ & F_ν↓ and F_{ν,net}↑= F_ν↑ − F_ν↓). And so…
“values at one part of the spectrum (ν₁) ‘predict’ values where (at ν₂) σ_a [or τ for the initial amount] of a GHG is halved while the concentration of that GHG is doubled” ( https://www.realclimate.org/index.php/archives/2026/04/a-reflection-on-reflection/#comment-847512 – the overlap with τ_{H2O} as well as such things(?) that disrupt a PPIA approx. (partly cloudy conditions) will disrupt/prevent/distort/mess with the stately progression of ‘isobrights’ over cos(θ), but that’ s okay for this)
Thus, with some shuffling of ν values to arrange the σ_{a,CO2} for ease of visualization, with the approx. that this σ halves going out from its peak value per spectral bandwidth ∆ν_½ (smoothly, ie. σ_a = σ_{a,max} · 2^[ −abs(ν – ν_{max}) / ∆ν_½ ] ), there is a progression of values of L_ν(θ,ϕ) (for each θ,ϕ) , F_ν{↑,↓, net↑} , and net spectral cooling NCν, at each location, which just shift outward over the spectrum by ∆ν_½ for each doubling of (an isotopologue of) CO2. New values are established and become part of that progression at the peak ν_{max}. When the ν_{max} values of F_{ν,net}↑ and NCν approach their final saturated limits (generally 0) (one value may take ‘longer’ (require more CO2) to get to that point than another), we have a set marginal bandwidth (in two margins) containing the progession of varying CO2 effects, typically containing ≈all the net radiant cooling by CO2, which shifts outward from a widening saturated bandwidth (where F_{ν,net}↑ and NCν (of all material) ≈ 0, generally), and a narrowing of the CO2 window bandwidth where the effects of CO2 are ≈0 and F_{ν,net}↑ and NCν of other materials may be significant. Then: (edited from original:
remaining changes are reductions in the LW net radiant cooling/heating by other materials (in particular: surface, H2O, and clouds) and associated reduction of the net LW fluxes that/which happen in the CO2 spectral window bandwidth, caused by that narrowing of that bandwidth .
But different relative concentrations of CO2 are necessary to approach saturation in different conditions, eg./ie. different heights, with saturation closer to TOA being delayed longer …
Note also that H (net radiant warming) by CO2 goes toward 0 (saturated limit) near the center of its 15 µm band (band of bands) except near the the tropopause and closer to TOA, and next to the sfc (another relatively sharp bend in B(τ)?; – remember that the sfc can be approx. as an infinite isothermal optical depth, as can (looking up) Space above TOA).]
…; each doubling can thus add a temperature-dependent (hence the indirect solar effect) net radiant cooling by CO2 within some optical depth of TOA; a subsequent doubling’s effect will (?)be more concentrated(?) and more limited in optical depth from TOA (won’t penetrate as far down), though otherwise smaller given cooling by the previous doubling.**
**(a reminder that the cumulative forcing from many changes can/will different if the climate is allowed to equilibrate before each successive next forcing, and therefore the cumulative feedback can/must also differ.)
There’s always room for Cooling To Space (CTS) in some uppermost layer of a gaseous atmosphere, … until T there approaches the LW effective radiating temperature of Space (ie. brightness temperature(s) of LW F_ν↓ @TOA ≈ 0)
(and then there’s that some part of the increase in OLR is coming through the stratosphere’s windows, thus not heating it back up so much… ; but that wasn’t my main point here (otherwise I’d compare to the LW-grey gas behavior)…)
Okay, so what I was getting at: Absent variations in of line broadening (width and shape) and line strength over height (as z or p)), τ would be proportional to the mass path of the GHG, which, for a WMGHG like CO2, at least up to …
(maybe above the point where the LTE approx. fails anyway? … we could approx. it up to TOA for various purposes which don’t concern what happens above the stratopause?)
…would be proportional to the mass path of the atmosphere, which is approx. proportional to p. So just graph B over p. The stratosphere would be ~ 10% to 25%+? of the atmospheric τ_{CO2}, not counting H2O and clouds etc. But with line broadening, the majority of the bandwidth may be far enough from line centers such that σ_a is ~proportional to p (up to z=?**) and thus τ_{CO2} is proportional to p² (stratosphere ~ 1% to 6.3%+? of the atmospheric τ_{CO2}). But the line centers would (up to z=30 km ? p~10 mb ?**) have σ_a ≈ proportional to 1/p, (using −log(p) ~proportional to z: stratosphere ~ 54% to 70%+? of the atmospheric τ_{CO2} ; I just did that very quickly, added a 7 km scale height (to large?; it’s cold there) to 30 km** and took [37 km – (17 km, 11 km)]÷37 km; not 100% sure if that’s even the right approach, but qualitatively, you get the point).
**I’ve read that p (collisional) line broadening dominates up to ~30 km, which I remember being near 10 mb; so either thermal Doppler broadening dominates above ~30 km or the source intended to imply that ~30 km was the base of a marginal region where they were of similar magnitude. Doppler broadening produces a different line shape such that, I’m not sure about this but I thought maybe collisional broadening may dominate farther from the line centers up to higher levels.
PS the stretching of the lower troposphere in τ-space can make the T-profile more convex, working against the concave nature of B(T) – which itself is more pronounced at shorter wavelengths (higher ν) – eg. towards the LW/SW transition, B would rapidly drop to ~0 from the sfc and maybe have a thin isolated peak at the stratopause; towards the lowest ν, B approaches linearity over T.
So rather than sorting the spectrum by vertically-averaged σ_{a,CO2 p-average}, greater accuracy could be achieved by adding another dimension – the spectral distance ν’ from the line center (this will work best if the lines are evenly spaced, & etc.) (a graph over ν through this space may tend to look like a comb viewed from an oblique angle). If spectral intervals dν are evenly distributed over σ_{a,CO2 p-average} at each ν’ with sufficient σ_{a,CO2 p-average} resolution, then the whole near-constant forcing per doubling (including the verical distribution of ∆ net radiant cooling) still works, but the band is widening over a 3D space (if we were previously graphing values on one axis over the independent axis ν).
…
There’s errors in theses sorts of approximations but it’s helpful to use the approximation as a baseline of understanding and then add larger and then smaller identifiable corrections.
remaining changes are reductions in the LW net radiant cooling/heating by other materials (in particular: surface, H2O, and clouds) and associated reduction of the net LW fluxes that/which happen in the CO2 spectral window bandwidth, caused by that narrowing of that bandwidth ….
patrick o twentysevensays
… If spectral intervals dν are evenly distributed over log₂[σ_{a,CO2 p-average}] at each ν’ with sufficient log₂[σ_{a,CO2 p-average}],…
Okay, the idea of a nonzero saturated limit for net (spectral) (LW) radiant cooling at a perfectly sharp bend in ∂B/∂τ_{vc,norm} [discontinuity in the lapse rate in τ-proportional space] is more of a mathematical curiosity I used to illustrate aspects of radiant heat flow net (the layer with nonzero cooling will get thinner and thinner so it would eventually disappear from any graph with finite resolution), but its practical application is that the final approach to saturation at 0 should tend to be delayed in the neighborhood of sharper features in the temperature profile (in terms of B(T)) (in τ-proportional space)
2a.
“But with line broadening, the “vast”* majority of the bandwidth may beis?far enough from line centers such that σ_a is ~proportional to p” , approximately…
To demonstrate the approximate linearity in pressure, Fig. 10 plots κ calculated from the line-by-line model as a function of pressure level for 4000 equally spaced wavenumbers in each of three different atmospheres (IsoAtmo, IsoStrat, and StdAtmo). To avoid the figures becoming saturated with color, the individual curves are plotted using a translucent color. We see that, regardless of the thermal structure of the atmosphere, the vast majority of the wavenumbers have κ values that increase quasi-linearly as we move to higher pressure in the atmosphere. The lower-right panel quantifies this by plotting three histograms (one for each atmosphere) of the slopes of the best-fit lines to each of the 4000 curves. Less than 2% (1%, 3%) of the slopes in IsoAtmo (IsoStrat, StdAtmo) case are negative. We see that the slopes are clumped around unity, demonstrating the quasi-linear dependence of κ on pressure.
(The clump does have some width to it, though, and one could imagine lines converging on κ proportional to p at some p and then… um…?)
2b.
Fraction of atmospheric τ_{CO2} above ~30 km if:
τ_{CO2} proportional to p: ~ 1 %
τ_{CO2} proportional to p²: ~ 0.01 %
τ_{CO2} proportional to p above ~ 30 km and (30 km – z) below that: ~~~ 19 % ? or a bit less , … (≈7/37; but 7 km scale height too large?)
3.
(and then there’s that some part of the increase in OLR is coming through the stratosphere’s windows, thus not heating it back up so much… ; but that wasn’t my main point here (otherwise I’d compare to the LW-grey gas behavior)…)
That may be the more important thing (some increase in OLR coming from lower down through the stratosphere’s windows avoids undoing the cooling of stratospheric adjustment (going from IRF to SARF)), as far as the spectral effect is concerned. A LW-grey gas GHE with LW-grey GHG-forced warming, even without solar heating at/near TOA, will have a IRF→SARF upper atmospheric cooling, but it would be completely cancelled out in the full climatic equilibration (final TOA T same as initial, barring ASR (SW, ie. solar heating, ie. albedo…) feedbacks (1-dimensional column globally-representative model, using only a bidirectional (I think they call it “two-stream” in the literature) calculation of radiance to be representative of flux densities – which does miss some possible interesting phenomena caused by the directionality of radiance (eg. https://www.realclimate.org/index.php/archives/2025/12/unforced-variations-dec-2025/#comment-842798 (PPIA) :
Interesting case: if you have an inversion layer above a(n isothermal) cloud layer [or any PPIA absorbing layer that doesn’t refract the LOS’s too much, and assuming any LW scattering that might happen? doesn’t mess this up], for some levels of opacity, you can get net radiant heating in the uppermost part of the cloud and net net radiant cooling deeper in the layer, both due to the radiances from above. If the overlying atmosphere is not too opaque, but opaque enough, L↓ from near vertical can be dimmer (colder) than B of the cloud, while L↓ from closer to horizontal can be brighter (hotter) than the cloud’s B. The L closer to vertical penetrates deeper into the cloud; L farther from vertical is absorbed over a shallower layer. Compare to https://scienceopinionsfunandotherthings.wordpress.com/2024/12/10/directionally-averaged-radiance-and-the-semi-gray-skin-temperature-wip-awaiting-final-proofread-double-check-diagrams-pending/ (see last part: *†* …”Meanwhile, the anomalously brighter radiance near vertical penetrates deeper below TOA than the anomalously darker radiance near horizontal “…) [regarding T at/near TOA]
We run RFM for H2O and CO2 separately, using HITRAN 2016 spectroscopic data for H2O from 0 to 1500 cm−1 and CO2 from 500 to 850 cm−1, using only the most common isotopologue for each gas. We use a highly idealized RCE atmospheric profile with Ts = 300 K and a constant lapse rate of Γ = 7 K km−1 up to an isothermal stratosphere at 200 K. [Note that the gray RCE temperature profile (10) decreases toward 0 as p → 0, unlike the isothermal stratosphere profile we employ in this section]. The GHG distributions are given by a tropospheric relative humidity of 0.75, a stratospheric H2O concentration of 23 ppmv (corresponding to an RH of 0.75 at the tropopause, and relatively large due to the 200-K tropopause), and a uniform CO2 concentration of 280 ppmv. We run RFM at a spectral resolution of 0.1 cm−1 and a uniform vertical resolution of 100 m up to model top at 50 km. […]
For simplicity in comparing to our offline decomposition, optical depth is calculated along a vertical path (zenith angle of zero), and fluxes were computed using a two-stream approximation (rather than RFM’s default four-stream) with a diffusivity factor of D = 1.5.4 We also omit the water vapor continuum and do not consider overlap between H2O and CO2. These omissions are discussed further in section 7 as well as in the companion paper Jeevanjee and Fueglistaler (2020). Note that the CTS approximation in the presence of continuum effects was examined in Clough et al. (1992).
So Fig 6 and 7 don’t show the effect of CO2 on H2O’s net spectral radiant cooling (negative heating) (per unit heat capacity, thus expressed as K/day per cm¯¹) or the effect of H2O on CO2’s net spectral cooling; the later would of course reduce the cooling near the surface and maybe add cooling higher up (shading CO2 from brighter radiance coming from below). CO2 of course will block some bandwidth of H2O’s cooling. But I also wanted to note that there is some cooling next to the sfc by H2O in Fig 6 that continues for some bandwidth going into the stronger parts of its bands (seems to come from SX and maybe AX, as it is for CO2 (GX works against this)).
patrick o twentysevensays
“the later would of course reduce the cooling near the surface – I didn’t just mean where the layer next to the surface with cooling extending into the heart of the CO2 band; I’m expecting the H2O overlap would extend higher than that, at least on the smaller cm¯¹ side of the CO2 band.
patrick o twentysevensays
PS re my https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849896 : effective band widening effect was for the simplication (additional to those already specified) that the ~667 cm¯¹ (~15 µm) CO2 band (of bands) was the only CO2 spectral feature. (it’s not, though it dominates at least the LW part and the SW part is comparatively small though still significant? AFAIK.)
Why do I keep referring to graphing the Planck function B_ν (T) over the (vertical) optical depth τ …
(or τ_{vc}; vc = vertical coordinate; I sometimes use this notation to differentiate from optical thickness τ along any line of sight (LOS))
…? Because you can qualitatively estimate (at least for PPIA NR LTE etc. ) the spectral radiant flux densities and net spectral radiant cooling/heating distributions from such a graph. If B_ν is linear over τ out to a few units of τ up and down from your POV (and going through your POV), the net vertical flux density will approximately be proportional to the slope of that line (it’s an exact relationship if the line extends to τ = +∞, −∞). More generally, it will depend on the difference between particular locally-weighted (∫EWF(ϑ) ·sin(ϑ)·dϑ ?-I think; did that quickly) values of B_ν(|τ−τ_{POV}|) above and below (optically thin layers don’t have as much of an effect because of course they are easier to see through; they themselves don’t have as much to see).
Likewise, net spectral cooling per unit mass = 4π · ( B_ν − L_{ν,4π} ) · k_{a,air}; L_{ν,4π} over τ is a smoothed version of B_ν over τ (it has the same fine scale (optically thin) texture, but with reduced amplitude). For variations over small ∆τ (much less than 1; optically thin), the warmer and cooler absorption cross-sectional areas (which are bits of τ) can see each other well and so the warmer tend to cool and cooler tend to heat up – but remember the heating rate per unit mass or heat capacity also depends on the concentration of that cross-sectional area relative to those quantities (eg. k_{a,air}); over larger scales it’s more the curvature of B_ν(τ) that matters; convex curvature (negative ∂²B/∂τ²) favors cooling; etc.
Notice that where and when the concentration of a GHG or other absorber (eg. cloud particles) increase, the graph gets stretched out (over larger τ); slopes get reduced, optically thin layers get thicker (more opaque), layers on either side get farther away from each other; net spectral cooling or heating can increase with greater ability to emit and cooler absorb per unit mass/etc., but can decrease due to the warmer and colder parts being more hidden from each other, and signs and directions of net cooling and/or net vertical flux density can switch as one set of features in the thermal profile loses influence to another.
Actually, it occurs to me that (PPIA etc.), if you take the B_ν(τ) and decompose it into a linear superposition of a constant plus sine and cosine terms over τ relative to a POV (
B_{ν,0} + ∑_m B_{ν,m,A}(τ’) + ∑_m B_{ν,m,S}(τ’) ;
B_{ν,m,A}(τ’) = A_m sin(mτ’)
B_{ν,m,A}(τ’) = S_m cos(mτ’)
), then the net vertical flux density depends only on the asymmetric part (sum of sines) then the net spectral vertical flux density depends only on the asymmetric part (sum of sines) and the net spectral radiant cooling (or heating) depends only on the symmetric part (sum of cosines).
Jeevanjee & Fueglistaler 2020 https://journals.ametsoc.org/view/journals/atsc/77/2/jas-d-18-0352.1.xml partition the net spectral radiant heating into four parts: GX (net exchange with ground (sfc), SX (exchange with layers above and below of equal ∆τ, of the maximum ∆τ that remains within the atmosphere), AX (exchange with whatever part of the atmosphere is excluded from SX), and CTS (Cooling To Space).
I can see the utility of doing that,
but I like to just combine the (vertical) optical depths of the atmosphere with what the sfc looks like from above (∞ optical depth, ≈isothermal; in the approximation of the sfc being a perfect blackbody) and what Space above TOA looks like from below (∞ optical depth, ≈isothermal near 0 K); note if we’re not using this to actually calculate what happens in Space or within the sfc (we don’t need to worry about refraction). In this case, the entirety of the net spectral radiant heating is given by the SX term. Stretching the atmosphere evenly* by some factor (*okay, hard to do if you’re doubling CO2 but there’s H2O with it’s own bottom-heavy distribution, not to mention clouds, etc.) is equivalent to stretching everything because the rest is infinite and isothermal of τ.
So what does the sfc look like if it has some spectral LW albedo (not a perfect blackbody)? Well in the simplest case of a flat horizontal sfc with only specular reflection with directionally-independent absorptivity, for all ϑ looking downward, you’d have a weighted average of B(T_sfc) and a reflection of the atmospheric profile of B (so you’re looking back up through it, so far as transmittance along your LOS allows). In the case of diffuse reflection or just a wavy sfc, the reflected image of the atmosphere will be vertically blurred (increasingly so away from sfc), and the vertical-scale (τ) will not generally match the original (real) atmosphere ie. looking down from near vertical, scattering toward near horizontal will partially extend (blurring) the reflected image; while looking down from near horizontal, scattering toward near vertical will partially compress (blurring) it.
patrick o twentysevensays
Another thing about line-broadenning – I think that using the τ ~≈∝ p² proportionality (for CO2; based on the gaps between the intervals immediately around line centers, and just approximating the bandwidth of the later as nonexistent) will miss some net radiant cooling at/near TOA. In a sufficiently optically-thin (small vertical τ) top layer, if isothermal (a decent approximation if the B_ν(T) variation is small relative to the jump at TOA to the near 0 of the downward LW radiance from Space), I argue that line-broadenning ≈ doesn’t matter to cooling to Space @ TOA. Eg.:
@ TOA: Consider if you have an optical depth in 1 % of the bandwidth of some spectral interval. For simplicity, let’s just consider the vertical transmittance: The transmittance going down from TOA will decrease exponentially in that 1%. If we broadening the opacity out to 10% of the bandwidth, but then reduce the optical depth in that 10% to 1/10 its initial value (to maintain constant ∫ σ_a dν) then the transmittance (in that fraction of the bandwidth) at TOA will decrease 1/10 as fast over vertical mass-path, but this happens over 10 times the fraction of the bandwidth, so the full spectral interval transmittance decreases at the same rate over mass-path (though it will continue to decrease farther over mass-path (effectively 10 times deeper, asymptotically approaching a total drop 10 times larger). The same will be the case at any angle ϑ from vertical (within the same hemisphere), just with a different variation per unit vertical depth.
Note that for any line of sight (LOS), the drop in transmittance over some distance is the fraction of the total cross-sectional area you can see in the LOS that is within that distance ie. for absorption cross sections, it’s the part of the emission weighting function (EWF) that is within that distance.
(For B_ν ≈ constant over the ∆ν of the line-broadenning, or at least, given symmetric line shape, (∆ν)² · ∂²B_ν/∂ν² ≈ 0 )
(Relates to (2020) Nadir Jeevanjee, Stephan Fueglistaler, “Simple Spectral Models for Atmospheric Radiative Cooling” https://journals.ametsoc.org/view/journals/atsc/77/2/jas-d-18-0347.1.xml (see Fig. 6 for CO2 ) – ?? been awhile since I’ve read this though, but I thought they used τ_{CO2} ∝ p² ?? (I think the RFM values were the more real and the SSM2D were the approximation but don’t take my word for it)) (PS not intended as a criticism of the work, just something interesting I thought of.)
patrick o twentysevensays
“so the full spectral interval transmittance decreases at the same rate over mass-path” – @TOA ie. initially in the spatial sense (why isn’t it “spacial”?)
patrick o twentysevensays
PS AIUI, collisional line broadening, like natural broadening, is due to the finite time over which a photon is emitted or absorbed. A wave packet is mathematically composed of a superposition of a range of frequencies and wavelengths (or wavenumbers or wave vectors in 3D); a wave packet that is more concentrated into a shorter time or smaller space must be composed of a larger range of frequencies (proportional to E) or wave vectors (proportional to momentum).
I think somebody last month asked about the photons being in LTE with the air – so to clarify, they generally are not expected to be. In this context, the LTE/LEDNLIE pertains to the energy distribution over the non-photons* of the air and sfc. (molecules, atoms, ions(? – in solution or solid/liquid…), electrons…) *-we can exclude the neutrinos, too…
The point is that the GHGs and cloud particles/etc. all (in the vast majority of the mass of the atmosphere) have access to a locally (small volume)-shared reservoir of enthalpy and therefore internal energy (we needn’t be concerned about what counts as thermal energy if we just refer to internal energy) ((approximately) LTE → internal energy distributed such that each statistically-sufficient substance has (approximately) the same T, and etc…) which they can add to and take from by absorbing and emitting photons.
Well, it will be harder to approximately maintain and isothermal relationship among the air and it’s resident aerosols because of the need to conduct heat to/from one to the others if the net radiant cooling per unit heat capacity varies and/or latent heat is being released/taken up (cloud particles), but it may still involve an approximate LTE/LEDNLIE on smaller scales? (how many molecules are in a r = 5 µm droplet…?)
But the other thing is of course that @ LTE/LEDNLIE, the internal energy (kinetic and potential energy of the relative motions and arrangements of atoms/electrons/etc.) is distributed such that the populations of particles in states of various energies support relationships among the rates of direct absorption (per unit spectral radiance L), − stimulated emission (per unit spectral radiance L), and spontaneous emission, such that, after stimulated emission cancels out some direct absorption, what remains…
… can be parameterized and visualized as [the effects of] a locally-random*‡×* distribution of tiny opaque blackbody objects (OBOs), one per unit of material (eg. molecule), which absorb all light they intercept and emit a glow from their surfaces with a radiance (brightness) according to their temperatures (T) given by the Planck function B_ν(T) (hotter = brighter). Each OBO emits and absorbs energy at a rate equal to the local average rate per unit of material, for each type of material (the parameterization is errorless in principle, ie. within the limits of statistical behavior being able to match probability, and assuming LTE or LEDNLIE *‡L*). The sizes of OBOs can vary greatly over the spectrum so the following visualization generally requires considering one frequency (or approximately, a narrow band of the spectrum) at a time.
(OBO sizes can also vary with polarization, but that’s of little concern for understanding the GHE. They also depend on T (through the ratios of state populations, Einstein Coefficients etc. stuff *†*), and pressure p and T (line broadening); the later is important (esp. p) in producing large changes over height.)
The (orthographically-)projected area of an OBO is the absorption cross section (σ_a) per unit of material. Often (in particular, for atmospheric gas molecules, cloud droplets, and any locally-random particles whose orientations are randomized), the σ_a are isotropic, ie. the same size over all directions, as if the OBOs where spheres.
(me @ https://www.realclimate.org/index.php/archives/2025/11/unforced-variations-nov-2025/#comment-842186 )
And the visible (not hidden) cross-sectional area that you can see (in some direction) from a given POV (location) is the radiating surface producing the spectral radiance L_ν = B_ν of radiating surface (coming from that direction); L_ν = B_ν of radiating surface – but within the atmosphere, that surface is distributed over distance s along the line of sight (LOS); so I …
…like to approach the physics of GHE (the greenhouse effect) through visualization – imagine you have (colorized?) heat vision, and the atmosphere is like an incandescently-glowing fog (thickness varying with local conditions and over the spectrum) above an incandescently glowing surface. s are greater than what’s coming from the opposite directions.
Shorter version – the thicker the ‘fog’, the less far you can see. And you need to be able to see across a difference in T in order to have a net spectral flux density … etc.
patrick o twentysevensays
Each thin layer of that fog that crosses you line-of-sight (LOS) blocks/absorbs some fraction of the glow coming from behind it, and replaces that glow with its own temperature-dependent glow.
patrick o twentysevensays
(attempt at a simple explanation) For solar heating only at/within/near the sfc: GHE works by/through:
1. a reduction of the horizontal area * transmittance to Space * bandwidth for the direct flow of radiant heat (LW flux) from the sfc(or a layer there) to Space (OLR emitted by sfc).
2. the sfc T doesn’t have to increase enough to fit the entire equilibrium OLR (or just that part which balances solar heating at the sfc) through that area*transmittance*bandwidth, but the remainder of the equilibrium heat flow has to flow through the atmosphere from where lower-level solar heating is occurring up to where the rest of the equilibrium emission to Space occurs (through some combination of net LW and convective fluxes; this requires a temperature difference (to drive the net LW flux through the opacity and/or to allow/drive the convective fluxes) .
zebrasays
Martin Smith
Martin, you said:
“Ray, aren’t you describing the warming of the atmosphere? I thought the warming is not the greenhouse effect, but the effect of the greenhouse effect.
The greenhouse effect is:
1. Blocking energy emitted from the surface, and
2. Sending blocked energy back to the surface.
Between 1 and 2, the blocked energy warms the atmosphere, so global warming is the effect of the greenhouse effect.
Have I had it wrong all along?”
Earlier, I said that a simplified explanation is a good thing, but it requires being careful. You still have to be precise and disciplined in your language, or you confuse your reader and yourself.
“The greenhouse effect” refers to the energy retained by the climate system when GHG absorb radiant energy and convert it to thermal energy.
“Global warming” refers to the increase in average surface temperature resulting from the increase in that thermal energy resulting from human-caused increase in GHG.
In my opinion, what I just wrote is what you say you are looking for. A simplified, mechanistic explanation of the GHE, as well as “global warming”.
What more do you want to know? I can explain the terms radiant energy and thermal energy, or would you like to hear how this all applies to the expression “climate change”?
(That last was obviously me pretending to be your AI friend.)
Martin Smithsays
Zebra: “The greenhouse effect” refers to the energy retained by the climate system when GHG absorb radiant energy and convert it to thermal energy.
“Global warming” refers to the increase in average surface temperature resulting from the increase in that thermal energy resulting from human-caused increase in GHG.
In my opinion, what I just wrote is what you say you are looking for. A simplified, mechanistic explanation of the GHE, as well as “global warming”.
MS: Your explanation confuses me more. A greenhouse traps energy, so a mechanistic explanation of the GHE must explain how the energy is trapped.
The mechanistic explanation of global warming must explain how the trapped energy warms the atmosphere.
zebrasays
Martin,
1. I said:
“GHG (greenhouse gasses) absorb radiant energy and convert it to thermal energy.”
Radiant energy can escape to space, but thermal energy cannot. Hence, the energy is “trapped”.
2. I said:
““Global warming” refers to the increase in average surface temperature resulting from the increase in that thermal energy resulting from human-caused increase in GHG.”
Thermal energy is what determines the temperature. If you increase thermal energy, the temperature will increase. “Warming” is by definition an increase in temperature.
Feel free to ask anything about what is still confusing you.
Martin Smithsays
I understand what you are saying, Zebra, I just don’t think you are saying it right. I’m trying to say that the greenhouse effect and the warming are separate mechanisms.
1. The absorption by the GHG (i.e. trapping) of the radiant energy emitted from the surface is the GREENHOUSE EFFECT.
2. The conversion of the absorbed (i.e. trapped) radiant energy to thermal energy is WARMING.
3. The increase in WARMING averaged for the entire globe that results from increasing GHG, which then increase 1 and 2, is GLOBAL WARMING.
If I am confused it is about absorption vs conversion to thermal energy. Are absorption and conversion 2 sequential steps or are they a single step?
zebrasays
Martin
Here’s what NASA says;
“What is Global Warming?
Global warming is the unusually rapid increase in Earth’s average surface temperature over the past century primarily due to the greenhouse gases released as people burn fossil fuels. The global average surface temperature rose 0.6 to 0.9 degrees Celsius (1.1 to 1.6° F) between 1906 and 2005, and the rate of temperature increase has nearly doubled in the last 50 years. Temperatures are certain to go up further.”
“Warming” refers to an increase in temperature. Always has.
Kobayashi Marusays
ms, in #3 you are correctly (more or less) describing the enhanced greenhouse effect.
An increase in the concentration of greenhouse gases leads to an increased infrared opacity of the atmosphere, and therefore to an effective radiation into space from a higher altitude at a lower temperature. This causes a radiative forcing, an imbalance that can only be compensated for by an increase of the temperature of the surface-troposphere system. This is called the “enhanced greenhouse effect.”
see under Greenhouse Effect https://archive.ipcc.ch/ipccreports/tar/wg2/index.php?idp=689
# 1 & 2 are off. see my other comments and ipcc ref
also note more semantics confusion – warming of the surface is the result of the greenhouse effect.
global warming is a “yardstick measurement” to estimate mathematically the change in avg surface temperatures calculated on a baseline … over time … bpl is (more or less ) right but most people cant read greek or math.
the issue here remains problematic from trying to oversimplify beyond reason.
Ray Ladburysays
Martin, I’ll try again. To start with, you have so-called “black-body radiation”. This is thermal radiation with the amount of radiation and its spectrum characterized solely by the object’s temperature. A pure-black object absorbs all radiation incident upon it (a perfect absorber)–and it is also a perfect emitter. In reality, a perfect blackbody emitter is like a spherical cow–an idealization. That is why you hear Earth sometimes referred to as a “gray body”–it’s not important for out purposes–a blackbody is a pretty fair approximation. For bodies near 273 K, blackbody radiation peaks in the infrared–heat radiation.
So Earth in a vacuum or with inert atmosphere absorbs visible radiation coming in and emits IR going out, and reaches an equilibrium temperature. Now add a greenhouse gas. This is just a molecule that absorbs in that peak portion of the blackbody spectrum–in the IR. For a molecule to do this, it has to be at least triatomic–N2, O2 won’t do it. So, now, you have a molecule taking a bite out of the blackbody emission spectrum in the IR and getting excited into its vibrating state.
But when a molecule gets excited, it will eventually relax back into its ground state. A CO2 molecule can do so in a couple of ways–it can emit a photon of the same energy it absorbed to trigger the vibration or it can collide with a (say) N2 molecule and impart kinetic energy from the vibration to the N2 molecule. For CO2, the lifetime before it emits a photon is quite long. So as long as atmospheric densities are high enough, a relaxation by collision is far more likely than radiating a photon. That is what is meant by “thermalizing the energy”.
There is one additional thing to consider–the temperature gradient. If temperatures decrease with altitude as in the troposphere, CO2 molecules at high altitude are much less likely to become excited by collision, and so even less likely to relax by emitting a photon. (In the stratosphere, the temperature gradient reverses sign, so you get more emission by radiation and more IR escaping the stratosphere and so cooling. Both of these are in fact due to the greenhouse effect.)
You can think of equipartition as how a system shares energy among its different modes. In equilibrium, the modes are all at the same temperature–it’s like a series of connected reservoirs all at the same level. Now we pour some energy into the reservoir marked “excited CO2”. The system is out of equiliburium, and energy flows into the other modes of the system (e.g. the kinetic energy/temperature of N2, O2 and other atmospheric gasses). It’s not just absorption and re-radiation.
Does that help?
Martin Smithsays
RL: Does that help?
MS: No. I mean, I understand it, and I appreciate it, but it is too long-winded to be what used to be called an elevator explanation. The NASA explanation Zebra posted tries to be a meme, but it doesn’t explain because it simplifies away the explanation.
The simplified mechanistic explanation has to be clear enough, short enough, and long enough so it can be used as your opening when you pitch your idea for a TV series about the greenhouse effect to producers who don’t know physics and who will shoot your idea down with memes like “CO2 has been high before,” “The Medieval Warm Period was global,” and “CO2 is a trace gas.”
To be honest, I have no idea what you have tried to explain in the last paragraph of your comment, specifically in the sentences
“In equilibrium, the modes are all at the same temperature–it’s like a series of connected reservoirs all at the same level. Now we pour some energy into the reservoir marked “excited CO2”. The system is out of equiliburium,”
I think that in Earth atmosphere, the real distribution of available energy states of various air components does, in fact, almost perfectly fit the theoretical “equilibrium” distribution expected on the basis of the actual temperature in the respective air parcel. Exceptions could perhaps occur e.g. in rare cases of injection of energetic particles into upper atmosphere by solar storms or during electric discharges, I suppose. It appears, however, that you meant something much more generic. Could you clarify?
Greetings
Tomáš
Ray Ladburysays
Think of it as sequential steps:
1) The CO2 molecule in its ground state
2) The CO2 molecule is excited into its vibrational state
a) this can happen because it absorbs a photon of the right wavelength
b) or it can happen because a collision with another molecule imparts energy to excite the vibration
3) The CO2 molecule relaxes
a) by emitting a photon of the right wavelength
b) by imparting the kinetic energy from its vibration to another molecule via collision
Note the symmetry between 2 and 3. BUT in Earth’s atmosphere, temperatures are sufficiently low that it is rare that a molecule has sufficient energy to excite the vibration (e.g. 2b)). So the flux of blackbody radiation near 15 microns represents a flow of energy that excites the CO2 vibrations.
Likewise, the symmetry in 3 is also broken, because the lifetime of the relaxation process via photon emission is quite long. During that lifetime, a molecule is likely to collide with many molecules, at least one of which causes the CO2 molecule to return to its unexcited ground state. (3b more likely than 3a).
The CO2 in the atmosphere does still emit 15 micron photons–just what you’d expect it to emit for a blackbody at the temperature prevalent at the molecule’s altitude. Near the ground, atmospheric densities and CO2 concentrations are high enough that the probability of a 15 micron (+/- a micron or so) escaping the atmosphere is essentially nil. Only the 15 microns emitted at higher altitude/lower density/lower temperature have a reasonable probability of escaping.
This is the basic flow of energy due to the greenhouse mechanism that is responsible for the imbalance that raises tropospheric temperatures. But as I said, it’s the basics. There are lots of nuances–and if you omit these little nuances or don’t understand them properly, you can get stuck.
This is why the same denialist can claim that a trace gas like CO2 can’t possibly warm the atmosphere AND that greenhouse warming is impossible because the CO2 absorption is saturated. This is why I think it is a mistake to try to oversimplify the greenhouse effect. It took scientists over 100 years to fully understand all the subtleties of the effect. The first-principle quantum-mechanical calculations demonstrating why the CO2 absorption band is NOT saturated were only done in the last few years. If you wanted to pitch a series to a bunch of Hollywood jerkwads with short attention spans, the discovery process would be the script to pitch.
Martin Smithsays
RL:
1) The CO2 molecule in its ground state
2) The CO2 molecule is excited into its vibrational state
a) this can happen because it absorbs a photon of the right wavelength
b) or it can happen because a collision with another molecule imparts energy to excite the vibration
3) The CO2 molecule relaxes
a) by emitting a photon of the right wavelength
b) by imparting the kinetic energy from its vibration to another molecule via collision
MS: That’s a mechanistic explanation for what happens to a CO2 molecule, but…
It says the molecule is in its ground state. That must be the state it is in when it is on the ground.
The molecule gets excited. It starts vibrating when it gets excited, so it wasn’t vibrating at all before it got excited.
The molecule can get excited if it absorbs a photon of the right wavelength. When it absorbs a photon of the wrong wavelength, it doesn’t get excited.
And the CO2 molecule can relax, if it emits a photon of the right wavelength, but if it emits a photon of the wrong wavelength, it stays excited.
Or, the molecule can relax if it collides with other molecules, which doesn’t sound very relaxing at all. More like aggravating.
I understand your mechanistic explanation of what happens to a CO2 molecule throughout the GHE/warming cycle, but a person who rejoices in the acronym MAGA will get tripped up in all the ways I listed.
And I think a mechanistic explanation of the path of energy through the GHE/warming system is a better way to explain it, but maybe both points of view works even better.
Kobayashi Marusays
Um, not quite. “Warming” is by definition an increase in thermal energy. Temperature is merely the yardstick being used to record the level of that thermal energy in space and time.
The drive to make this short and simple is creating errors. Occam’s-razor warns against the error of oversimplifying past the point of accuracy. The simplest accurate model of necessary mechanisms is still the minimum required.
“Global warming is the rise in surface temperature caused by the build-up of thermal energy (originally from the Sun) in the climate system, due to human-emitted greenhouse gases trapping more of that energy.”
That’s accurate but it may not be addressing the dynamics of how it happens which people want to express.
A more scientifically precise way to say it:
“Global warming refers to the increase in Earth’s average surface temperature resulting from the build-up of thermal energy (originally from the Sun) in the climate system, caused by human-increased greenhouse gases slowing the escape of outgoing infrared radiation to space.”
Moreover, the thermal energy itself doesn’t “escape” to space; it is radiated away. So, the energy isn’t “trapped” like a blanket traps heat by physically holding it in place. Rather, greenhouse gases slow down the leak of energy to space. And it’s that which causes the build-up of thermal energy (originally from the Sun) in the climate system in the first place.
The increase in greenhouse gases today has forced the increase of thermal energy and therefore the temperature. But if are being forced to condense all of that into a single, final, punchy statement that can’t really be argued with, you could say:
“The simplest accurate model is this: more GHGs slow the escape of the Sun’s energy to space, causing thermal energy to build up, which raises surface temperature. Anything simpler than that omits the actual mechanism and becomes misleading.”
Trying to simplify it further destroys it’s meaning and the physics involved. Embrace the complexity.
MA Rodgersays
Kobayashi Maru,
I think the ‘GHE is not like a blanket’ idea you present is not helpful for explaining GHE. Blankets (with the exception of impervious ‘space blankets’) physically leak like sieves. The troposphere presents a far-less leaky barrier for the air of planet Earth than a blanket-enwrapped human. The journey for a packet of air from ground up to the tropopause takes on average perhaps a couple of weeks. That’s [ 9/(2 x 186) =] 0.024 mph or 0.01m/s. Blankets (okay not an apple-with-apples comparison) are measured with a permeability of perhaps 0.5m/s.
Unlike your assessment, I rather liked Zebra’s simplistic idea that “Radiant energy can escape to space, but thermal energy cannot. Hence, the energy is “trapped”.” Indeed, you say it yourself. “The thermal energy itself doesn’t “escape” to space; it is radiated away.” One of the things us small-brained mammals have yet to figure out is the nature of the photons that comprise radiation. But we do know for sure it ain’t not “thermal energy.”
And while I would make an issue of what is the sun’s energy and what is the Earth’s energy, your statement “More GHGs slow the escape of the Sun’s energy to space,” still doesn’t explain why that is the case. What is the ‘slowing’ mechanism?
To give my six-pen’orth, I’d also point out (1) Radiation is how energy flows into and out of objects floating in space. Actual thermal energy is indeed trapped and needs converting into radiation to exit the planet.
But then I’d add that (2) It is the temperature of that object which determines the ‘conversion to radiation’ and thus the amount of radiated energy shot out into space. Higher temperature means higher radiation out. Specifically, this is about the temperature of the ‘areas’ that radiate into space. Without an atmosphere or a GHG-free atmosphere, the ‘area’ in question will always be the object’s surface.
And from there (3) GHGs result in a shift in the ‘area’ specific to the wavelength of radiation the GHGs absorb/emit. More GHGs, the higher that ‘area’. (4) Because the Earth’s troposphere gets colder with height, that means more height for a GHG’s ‘area’, less temperature in that ‘area’, less radiated energy out from Earth. (5) The Earth will then begin to heat, increasing temperature until the resulting increased radiation from that increased temperature equals the GHG-induced reduction in radiation.
That is the basis of how GHGs work. I can’t think of a simpler explanation that doesn’t resort to the ‘blanket’ or the ‘greenhouse’ analogy. (And do note, like blankets, operating greenhouses are remarkably leaky things although they call it “ventilation” not permeability’.)
Sadly the workings of the GHE mechanism is not well described outside scientific discussion. That IPCC AR4 page you link-to spends 670 words answering the question ‘What is the Greenhouse Effect?’ but of the underlying mechanism it only manages to tell the reader that it operates like a greenhouse although “through a different physical process” and that “adding more of a greenhouse gas, such as CO2, to the atmosphere intensifies the greenhouse effect, thus warming Earth’s climate.” Sure there’s a lot of other information provided answering the question but by 2007 the continued “It’s like a greenhouse” explanation was (and today remains) entirely inadequate in the face of blatant denialistic bullshit from far too many ‘authorities’ who should know better.
Kobayashi Marusays
‘MA Rodger
I believe I am correct. The IPCC is accurate. The physics is right. I’ll defer to the IPCC links I gave but will add the following brief clarification.
Anyone involved around climate change science and global warming, like I have for 30-40 years, understands that global warming is a metaphor, the greenhouse effect is a metaphor and the blanket idea is an analogy to explain those metaphors to the public at large.
Fully aware they are imperfect metaphors and analogies, knowing full well they are unsatisfactory from a purist science pov and yet all have been used repeatedly by renowned physicists to explain parts of the complex climate science, the physics and the mechanisms surrounding the ideas behind the greenhouse effect and the enhanced greenhouse effect; the difference between thermal energy, warming, and global warming and why temperature is different entity, since Syukuro “Suki” Manabe was a boy.
I’m quite happy with my explanations as far as they go. Knowing they were provided to only answer MS’s very specific question. They were not intended to explain the entirely of the physics behind 6 IPCC assessment reports. It’s complicated. I was trying to help where others had failed.
Thanks for your interest. I’m humbled.
Ray Ladburysays
Martin Smith: A person who calls themselves MAGA is ineducable and irremediably stupid or they wouldn’t be MAGA to begin with. They don’t recognize facts or truth or science. If the person is not interested in the science, they won’t be perusaded by the science, particularly when it counters their accepted world view.
Learning is only possible for those who have not achieved epistemic closure.
There is no clear generic relationship between “energy” of a system and its temperature. An example: If a bottle filled with ice absorbs heat from its surrounding, its thermal energy increases, however, its temperature remains constant until all the ice melts.
Dear Martin.
The average radiative temperature of Earth surface (that can be computed, according to Stefan-Boltzmann law, from the average intensity of the upwelling infrared radiation flux) depends on a time and space average of a complex balance of a multiplicity of various energy fluxes across the surface and the atmosphere. These global balances are usually depicted in “energy budget” diagrams called also Trenberth’s diagrams. Basically, any external perturbation to any of these particular fluxes, including a change in the atmospheric concentration of a greenhouse gas, may perturb the entire system. Examples of such induced perturbations may be the rising global average surface temperature and the decreasing temperature in the lower stratosphere induced by rising atmospheric carbon dioxide concentration. The fluxes in the corresponding diagram change accordingly. In absence of external perturbations, a steady state can be reached,
As regards the complex mechanism of the “greenhouse effect” of the atmospheric gases absorbing the infrared radiation, and/or of the changes in the altitude / temperature profile induced by changes in the greenhouse effect intensity, I recommend rather relying on explanations provided by standard textbooks (like Physical Climatology by professor Dennis Hartmann), by people like patrick o twentyseven or MA Rodger, or by a good AI engine, than striving to decrypt and/or correct confusing statements released by zebra.
Greetings
Tomáš
Barry E Finchsays
Martin Smith (MS) 25 Jun 2026 at 3:24 AM “I need a correct, simplified, mechanistic explanation using photons, molecules, kinetic energy, and words like ground, ocean, re-emit, and back radiation”. Following after a delineation line of “+” signs below is a perfectly-correct, sufficiently-complete simplified, mechanistic explanation of the so-called “greenhouse effect (GHE)” in Earth’s troposphere which avoids some of the MS words that I quoted because they are incorrect or misleading, or unnecessary for a full simplified, CORRECT mechanistic explanation. The simplified, mechanistic explanation is 4 short paragraphs with simple, obvious concepts that describe the operation followed by Notes that aren’t required for the simplified, mechanistic explanation but might be interesting. Preceding that are comments on what I infer are 2 significant misunderstandings by MS which appear to closely relate to each other (the same basic misunderstanding of MS).
=======
MS appears to think that GHE is solely due to change in “atmospheric window frequencies” (MS 27 Jun 2026 at 2:13 PM “Google AI …… is 10% to 15%) but GHE certainly isn’t much at all to do with “10% to 15%” in the atmospheric window, except maybe during Snowball Earths not at all relevant now. That “blocking effect” is a small bit player, a nibble at the edges right around 8.0 microns for H2O gas, 9.0 & 10.0 for O3, and 13.0 for CO2 as seen for example at 17:10 at my UTube video link below (Side Note: It’s shown as a lesser amount for CO2 increase than used by NASA but that’s an advanced matter that I’ll never study and certainly irrelevant to the “simplified, mechanistic explanation”).
MS 30 Jun 2026 at 1:00 PM “The greenhouse effect is:”
1. Energy emitted by the surface is blocked
2. Blocked energy is sent back to the surface”
30 Jun 2026 at 12:45 PM “The greenhouse effect is:
1. Blocking energy emitted from the surface, and
2. Sending blocked energy back to the surface”.
=======
Nope, that wouldn’t work as stated because the temperature lapse rate caused by reduced pessure with altitude wasn’t mentioned and that’s what makes it work. (my entire point I’ve made since 2018 about the confusion sowed about something simple by misleading and incomplete, or just plain incorrect, “simplified, mechanistic explanations”).
=======
MS 27 Jun 2026 at 2:13 PM “if we add a gigaton of CO2 to the atmosphere, the probability that an infrared photon can escape the atmosphere without being absorbed by a greenhouse gas must decrease”. Nope, not at all correct because the word “must” is a gross misunderstanding of the simple physics that has been caused in the brains of MS and 123 or perhaps it was 321 other bods I tried to disabuse, caused yet again by this “re-emits” and “back radiation” nonsense. As understood by Tomáš Kalisz 26 Jun 2026 at 4:09 PM and perhaps JCM or Zebra. So then “must decrease” definitely S.B. “will usually decrease” with an explanation why (vertical temperature changes). As per numerous places that I’ve (correctly) railed against variations of this since August 2018 (Quixotic) there’s no “must decrease” because that is yet again a false statement that CO2 molecules emit zero manufactured radiation (the “re-emit” and “back radiation” drivel strikes again, and strikes continuously ubiquitously since many years back, the handy Strawman set-up so’s it can be destroyed by Fossil Shill dullards like “Yong Tuition” and “Tom Shula” presented by “Tom Nelson”, and innumerable “commenter” bods for UTube videos and Web Site Forums). The CO2 molecules DO emit (manufacture) “their own” photons in vast quantity, not just “absorbed” like MS implies. Suppose as a hypothetical that the extra “gigaton of CO2 to the atmosphere” were to absorb radiation from below that would otherwise reach Outer Space in the globally-annually-averaged Power flux of 0.00156 w/m**2 but were to emit (manufacture) radiation in quantity such that extra “gigaton of CO2 to the atmosphere” sent 0.00157 w/m**2 to Outer Space, well then the extra “gigaton of CO2 to the atmosphere” would cause COOLING of 0.00001 w/m**2. This HAPPENS over Antarctica (or part of it) in Winter (or part of it) as shown at 20:09 at https://www.youtube.com/watch?v=rgP-lwf2tb8 (3 Power Spectrum samples of Earth’s radiation to Outer Space measured in 1970 for Sahara Desert, Mediterranean Sea & Antarctica). The cartoon that I’ve seen variously in UTube presentations for 13 years, and is huge disinformation, asserts that CO2 (sometimes H2O gas) molecules emit (manufacture) Zero photons by not showing any, not mentioning it at all, a most cunning form of disinformation, remaining silent on a massively-relevant fact. Animation showing a surface photon absorbed in a CO2 molecule and “re-emitted” up, photon absorbed and “re-emitted” down, over and over just like the “Joke Zonderkop” nonsense on RC UV 20 Jun 2026 at 7:30 PM. Suppose the animation showed 8 photons emitted for every 2 photons absorbed from below (6 photons manufactured), well now there’s 5 photons heading up for every 2 photons absorbed from below so this extra CO2 molecule is INCREASING the upward radiation toward Outer Space. This in Antarctica is seen at my link and thus the MS nonsense physics “if we add a gigaton of CO2 … must decrease” is shown incorrect by measurement.
The reason why this increasing upward radiation only usually happens in the stratosphere (always) is that the troposphere is usually colder with increasing altitude so therefore photon manufacture higher up must usually be less than absorption from below (warmer collides me than colder). The absence of the extremely-important manufacture proportional to Kelvin**4 from the MS “if we add a gigaton … must decrease” above renders it incorrect logic so worthless for the pondering and explaining the “greenhouse effect”
The MS explanation appears to be based on an incorrect assumption that if the “Energy emitted by the surface” were not more “blocked” by more GHGs then it wouldn’t heat the troposphere. However, it’s already heating the troposphere though because it’s already “blocked”, it’s simply “blocked” just a few metres higher up than it would be with more GHGs. The Reality using fair quantities for illustration is that 5% global-annual average of the radiation emitted by the surface (~22% roughly with cloud-free sky and 0% with full cloud cover giving 5% average, from Costas&Shine 2012 CERES analysis Paper) is in frequencies 8.0-9.0 & 10.0-13.0 microns (“atmospheric window frequencies”) and the slight narrowing of those “atmospheric windows” with more CO2, O3, H2O gas is a minor bit-player and is NOT the “greenhouse effect (GHE)” that must be described. The GHE applies to the OTHER radiation that is emitted by the surface and is ALREADY “blocked”, being ~78% roughly and all of that ~78% to which “enhanced GHE” is now applying is “blocked” already (and was already “blocked” 100 years ago) within ~100m of the surface with or without any clouds in the sky.
Suppose that ~78% of radiation emitted by the surface is “blocked”, absorbed and turned into “heat” within 100m of the surface. Add some CO2 and it’s then “blocked” within 95m instead of within 100m. I seriously doubt that that’s the sort of thing envised by MS with 1., 2. above, more likely incorrectly thinking that photons that would have reached Outer Space getting blocked were getting blocked with CO2, but that isn’t how the warming effect works. It’s a matter of the heating Power being just a few metres lower than before for a significant GHE warming.
=======
MS 27 Jun 2026 at 2:13 PM “Google AI estimates that the probability that an infrared photon emitted from the surface escapes the atmosphere to space without ever being absorbed by a greenhouse gas is 10% to 15%. Google AI includes an analysis, but my point is there is a non-zero probability that an infrared photon emitted at the surface can escape the atmosphere to space without ever being absorbed by a greenhouse gas on the way out..And if it doesn’t get absorbed by a greenhouse gas, it doesn’t warm the atmosphere”. This indicates that MS thinks that “greenhouse effect (GHE)” is a narrowing of the “atmospheric window” frequencies 8.0-9.0 & 10.0-13.0 microns but it isn’t that at all. As explained above, that “atmospheric window narrowing” is a minor extra-bit effect and not even applicable to CH4, N2O because their Bands aren’t near 8.0, 9.0, 10.0 or 13.0 microns. Rather the GHE is as I now describe.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
A simplified explanation of the so-called “greenhouse effect (GHE)” in Earth’s troposphere. Some Power flux of the vast quantity of photons that are emitted (manufactured) by molecules in Earth’s troposphere leak out of its top & bottom into or through the stratosphere and into the surface respectively. Emission (manufacturing) is proportional to Kelvin**4.
=======
Suppose for concept illustration only (not highly accurate) that 194 w/m**2 leaks out the top and that is part of what is emitted (manufactured) by molecules over an altitude range of 4 to 12 km with an effective average of 7 km. Say for concept illustration that 345 w/m**2 leaks out the bottom and that is part of what is emitted (manufactured) by molecules over an altitude range of 0.0001 to 1.5 km with an effective average of 1 km.
=======
Suppose a certain amount of CO2 was added into the troposphere and mixed then *instantly* the upper & lower ranges would need to be closer to their respective ends because there are more CO2 molecules in the way. So the effective averages of 7 km for upwelling into or through stratosphere and 1 km for downwelling into surface change to 7.05 km and 980m respectively, with the upwelling & downwelling Power fluxes consequently changing to 193 w/m**2 and 346 w/m**2 respectively due to being produced by higher-than-before (colder) and lower-than-before (warmer) air parcels. The reason for this is that pressure reduction with increasing altitude causes “adiabatic cooling” or “adiabatic heating” for increasing or decreasing altitude respectively.
=======
There has been no change in the 539 w/m**2 leaving the troposphere, there is no “magical extra energy”, the surface downwelling radiation has increased by 1 w/m**2, which is an additional ~480 terawatts (~95%) heating the ocean and Earth is emitting less radiation to Outer Space (a Power reduction of 510 terawatts).
=======
*** Notes are following. The above is all of the simplified part, all that’s needed. A person could skip the extra Notes or read them for interest ***
There is no “back radiation” to the surface. No radiation is going back any place, that’s just nonsense. There’s downwelling radiation into the surface. It’s manufactured by gas molecules and clouds in the air. There’s upwelling & downwelling radiation at every geoid-type (~spherical) surface that you can just randomly choose all the way from the centre of Earth’s core to 100 km above the 0cean surface (it varies hugely of course and is always going to be out of balance between upwelling & downwelling for any geoid-type surface). ~No photons are “re-emitted” or “re-radiated”, they are ~all simply “emitted” (manufactured by molecular collisions).
=======
Since the troposphere has very little thermal capacity the 539 w/m**2 leaving must be matched by 539 w/m**2 going into the troposphere on a globally-averaged, annual-averaged basis and this consists of 376 LWI radiation emitted from the surface, 88 of NET H2O gas latent heat (surface evaporative Power less rain on snow-ice and surface condensation), 20 of NET convection (warm air rising minus the surface begin warmed by the pressure increase of air descending), and 54 of solar SWR absorbed by gases, liquid & solids, which is 538 w/m**2, the 1 w/m**2 difference not really existing and being accuracy & rounding error.
=======
The thermosphere & stratosphere each absorb 13 w/m**2 of solar radiation to radiate to Outer Space by the gases. So the globally-averaged, annually-averaged radiation to Outer Space approximately for the general concept is 20 w/m**2 from the surface (90 with little cloud and zero with significant cloud) + 194 from the troposphere + 13 from the stratosphere + 13 from the thermosphere, which is 240 w/m**2 total matching the 240 w/m**2 of solar radiation from the Sun that Earth absorbs. With the example above of 193 from the troposphere due to extra CO2 there is 239 w/m**2 of radiation and the short-fall of 1 w/m**2 is called “Radiative Forcing at TOA” which heats the ocean and melts ice plus minor heat usages. The division of radiation to Outer Space between troposphere, stratosphere & thermosphere is based on the Power sources and not necessarily accurate for the vertical locations of the actual photons that reach Outer Space.
For example (just made up numbers to illustrate) suppose 20 w/m**2 of the 194 from the troposphere was used to heat stratosphere & thermosphere such that they each radiated to Outer Space 23 w/m**2 instead of the 13 w/m**2 of their solar heat source then the 194 from the troposphere to Outer Space would be 20 w/m**2 to high but still it was tropospheric photonic energy into the stratosphere & thermosphere which gave them the energy to radiate out the extra 20 w/m**2 so it’s correct from that perspective.
Kobayashi Marusays
a wise man accepts his limitations. I will never be a rock star.
better to lean heavily on the explanations of the experts in the field. and accept what they say.
Barry E Finchsays
explanations of the experts in the field would indeed likely be superior to normal public / media practice such as to describe half that Downwelling Longwave radiation as “back radiation” — ie back to where it came from — the surface, according to a close personal friend of John Cleese, so essentially according to a rock star.
Kobayashi Marusays
Much of this thermal radiation emitted by the land and ocean is absorbed by the atmosphere, including clouds, and reradiated back to Earth. This is called the greenhouse effect.
…. the Earth’s greenhouse effect warms the surface (land and oceans) of the planet.
Damn those IPCC people. All they seek is an argument.
Piotrsays
Barry E Finch ” normal public / media practice such as to describe half that Downwelling Longwave radiation as “back radiation” ”
Barry, you seem to conflate, again, two different positions:
– public/media calling ALL Downwelling Longwave radiation as “back radiation”
with
– the deniers’ attack on science based on the fallacy that of all LW radiation leaving atmosphere – “HALF” is absorbed by the ground and “half” is sent into space.
Which would require the same T near the bottom and near the top of atmosphere.
True if our atmosphere was one-molecule-thick.
And although “downwelling” is a better word if used in the explanation of mechanisms (see MARodger above) “back-radiation” used by the public/media is not as terrible as you imply:
– 358 W/m2 of the LW from the ground to be absorbed by the atmosphere
– 340 W/m2 of the LW from the atmosphere to be absorbed by the ground
Thus from the “ground” perspective, we get “back” from the air almost as much IR as air gets from the ground. And nobody said “back” must be identical to “out”
So all this talk how bad a word “back-radiation” is, and how “experts in the field” on’t be caught using it (see Trenberth et al., and NASA reposting their back-radiation”)
= is either making a mountain out of a mole-hill, or ego validation for the semantic nitpickers like zebra.
Kobayashi Marusays
zebra
“The greenhouse effect” refers to the energy retained by the climate system when GHG absorb radiant energy and convert it to thermal energy.
not quite.
To balance the absorbed incoming (solar) energy, the Earth must, on average, radiate the same amount of energy back to space.
Because the Earth is much colder than the Sun, it radiates at much longer wavelengths, primarily in the infrared part of the spectrum.
Much of this thermal radiation emitted by the land and ocean is absorbed by the atmosphere, including clouds, and reradiated back to Earth. This is called the greenhouse effect.
…. the Earth’s greenhouse effect warms the surface (land and oceans) of the planet.
Also, note that the absorbed incoming solar energy does not in fact balance with the outgoing energy. Energy is being stored in the Earth system, especially the oceans.
zebrasays
John
John, energy was being “stored” in the Earth’s climate system even when there was no imbalance.
It moves, it transforms, it determines the states of matter in the system… constantly.
Isn’t all that what we perceive as “climate”?
John Pollacksays
Zebra,
I don’t see it that way. If there is a net radiative equilibrium, I don’t consider energy to be stored or removed from the climate system. Instead, I would treat it as being shuffled between different boxes within the system.
Nor do I perceive the shuffling as what I mean by climate. It’s one aspect of climate, but it leaves out a lot. For example, the hydrosphere, cryosphere, and biosphere. These interactions are not exclusively described as energy exchanges. Even in the atmosphere, there can a change in climate without changes in mean energy storage or exchange. I can easily visualize a situation where the mean wind energy at a location isn’t changing, but there is an increase or decrease in rare, but very significant, winds from storms.
zebrasays
John,
I thought my statement was pretty much the same as your “shuffled”, and I did say “determines the states of matter” to cover water and ice and so on.
But anyway, I’m not clear what you mean by storage, or how the energy that is accumulating during imbalance is different from the energy in the system when EEI = zero.
Kobayashi Marusays
Oh dear. Lest we forget. Thermal energy is being stored in the Antarctic Ice Sheets too.
Our next progressive pattern dance is the Sweetheart Stroll. Choose your partner, please.
Piotrsays
zebra 7 Jul.: “ I’m not clear what you mean by [energy] storage ”
maybe you should have thought of this before you lectured John on his use of that word – he said that the imbalance between the incoming and outgoing energy ” is being stored in the Earth system”. No imbalance -> no need to store it, or release it.
Hence your:
zebra 6 Jul “ John, energy was being “stored” in the Earth’s climate system even when there was no imbalance
is … a typical zebra: correct them first, ask what they meant later.
John Pollacksays
Zebra,
I view climate as more than energy accounting, although it’s certainly an important way of keeping track of what is happening. By “stored” I mean as thermal energy. Since EEI is greater than zero, the storage term is going up, and the energy in the ocean is the major recipient – at least currently.
John
zebrasays
John Pollack
John, this is connected to something I have been wondering about. I’ve seen the numbers for the distribution of energy that is being added to the system… e.g. 90% is in oceans, 6% land, and so on, but it isn’t clear that this is identical to what it would be for EEI =zero.
My guess is no, which explains your definition of “storage”. But I’ve never seen a discussion of what it might be pre-industrial or in some future equilibrium state. Any thoughts?
Considering that if Earth is in a steady state* with a perfect balance between absorbed and emitted energy fluxes (EEI = 0), the flux into Earth’s storages must be also zero, it appears that asking after distribution of this zero flux between “ocean, land, and so on” makes hardly sense.
I therefore suppose that you meant something else.
Could you clarify?
Best regards
Tomáš
* remark:
From thermodynamical point of view, a body exhibiting a balance between incoming and outgoing energy flows is not seen as an equilibrium system, I think.
John Pollacksays
Zebra,
I’ve had to think more carefully about the terminology I’m using. I think of “storage” as the sum of energy being stored or released from Earth system as a whole. Units are energy, such as J.
EEI is a flux equal to the change in storage / (area of the Earth X time interval) W/m^^2 are flux units.
The proportions of energy exchange between different subsystems of the Earth storage term are what I would call “partitioning.” These are in expressible in fractions or percentages. These quantities must be measured, and depend on how you define the partitions and also the specific time period involved. They are variable, not constant. They need to be observed.
The time period I had in mind is one relevant to climate change. This would be an annual cycle or longer period. While it would be possible to define shorter intervals, they would be hard to measure.
I agree with Tomáš that if EEI = zero, there is a net equilibrium between incoming and outgoing energy, and thus no storage. Partitioning would then reflect whatever readjustment was happening within the system, and would depend on the specifics of the situation. There is no single answer. It would depend on the evolution of the internal state of the climate system following radiative equilibrium.
zebrasays
John, I see how my sentence might have caused some confusion.
I was trying to ask about the partition of energy in the system when EEI = zero; I used the numbers for partition of the added energy as an example because that’s the only thing I’ve seen discussed.
My best guess is that if we took the ARGO system back to pre-industrial times and measured a 30-year period of Ocean Heat Content, we would get a fairly consistent number, just like GMST.
That’s the partition of energy that I’m curious about, and how it might have changed, and will change.
As for the term “storage”, I take a longer-term view. I see almost all the climate system energy as stored solar energy. But as I said, I understand your usage of the term.
With respect to confusion, I can confirm that so far, the more you explained, the less I understood what you are speaking about.
If you took the ARGO system back to pre-industrial times and measured a 30-year period, I suppose you would obtain an average temperature of the upper two kilometres of sea water during this time span and you could also perhaps say if there was any temperature trend in this layer within the said 30-year time span.
I agree that it might be interesting to see if there was no energy imbalance then (as you seem to suppose), however, I am still quite unsure that it was indeed the idea you tried to express.
If so, could you clarify what you meant by “partition of energy”, what has this partition to do with the hypothetical historical recording of temperature of the upper 2 km of the ocean and which changes of this mysterious partition, how and when you expected to observe?
Thank you in advance and best regards
Tomáš
John Pollacksays
Zebra,
Under the conditions you mention, pre-industrial, with a 30 year interval of EEI = 0, I can
give a partial answer.
The portion partitioned to the atmosphere will be negligible, and that to the solid earth will be very small. This is based on the heat capacity of the atmosphere compared to the rest of the system. The solid earth can hold or release a lot of heat, but ground conduction is extremely slow beyond a thin surface layer, say 2 or 3 meters. There can be some partition to freshwater and ground water systems, but in general these are small components of the system.
This leaves the two large active components of the system in near opposition. The ocean can of course store or release a lot of heat. With EEI held to zero, the cryosphere – especially glacial ice, must have a partition similar in magnitude but opposite in sign. So, the if the ocean is cooling, the heat released would be used to melt ice, or the opposite.
John Pollacksays
Zebra,
Under the conditions you mention, pre-industrial, with a 30 year interval of EEI = 0, I can
give a partial answer.
The portion partitioned to the atmosphere will be negligible, and that to the solid earth will be very small. This is based on the small heat capacity of the atmosphere compared to the rest of the system. The solid earth can hold or release a lot of heat, but ground conduction is extremely slow beyond a thin surface layer, say 2 or 3 meters. There can be some partition to freshwater and ground water systems, but in general these are small fractions of the total.
This leaves the two large active components of the system in near opposition. The ocean can of course store or release a lot of heat. With EEI held to zero, the cryosphere – especially glacial ice, must have a partition similar in magnitude but opposite in sign. So, the if the ocean is cooling, the heat released would be used to melt ice, or the opposite.
May I ask under which circumstances you think that the described re-distributions of heat between ocean and cryosphere could be possible?
Honestly, I somewhat doubt that the Earth is ever in the state of perfect balance (EEI=0) between incoming and released energy flows; I rather guess that it may be more likely that it is either warming or cooling, in response to fluctuations in the relevant forcings.
Greetings
Tomáš
John Pollacksays
Tomáš, regarding 15 Jul @ 7:25 am
I would like to have Zebra answer your questions.
I agree with you that EEI = 0 is unlikely (except perhaps as a long-term average.) I think that partitioning needs to be established through observations, which are difficult to obtain.
The circumstances that I think a re-distribution of heat between ocean and cryosphere would be possible with EEI = 0 are in the aftermath of a large runaway dump of glacial ice into the oceans
(such as a D-O event). I am NOT saying that EEI really would be zero under those circumstances, since I think it likely that the Earth albedo would also change due to changes in cloud and ice cover.
I am not convinced that there is any utility in speculating about what the partitioning would be, in the absence of observations – except for sharpening definitions. It’s up to Zebra to make a case that this is a useful line of inquiry.
zebrasays
John Pollack
John, I took some time to read up on OHC and I think I begin to understand what has been puzzling me… as usual, the first step was getting the definitions right.
Your point about the cryosphere also was useful, although I’m not sure the magnitudes are as close as you imply.
At this point, I have to accept that the empirical results are as good as it gets as to any changes in partition of energy from the pre-industrial state.
zebrasays
John,
My 17 July comment was made before your 16 July comment showed up, and as you can see I am in agreement about the partition (and the need to clarify definitions.)
But I don’t get this statement about thinking EEI would not be zero pre-industrial CO2. That’s the fundamental concept we’ve been talking about in climate change all this time.
Of course it would be an average over a period statistically necessary to smooth out the “wobbles”… exactly the same as what we do with GMST. In fact, I’ve asked here a couple of times whether the same 30 year period would be necessary to validate changes in EEI.
The physics tells us that we will achieve a new EEI zero state when we (hopefully) stop adding CO2. It doesn’t tell us, as I’ve come to realize, exactly what the new overall system state… the partition… is going to be like.
I am still very curious about your explanation what you meant by the “partition of energy” in the Earth’s climate system and how it could change at zero energy imbalance (EEI=0). Was it just the hypothetical case of a sudden ice sheet collapse and their dissolution in the ocean suggested by John?
As regards your assumption of the preindustrial EEI=0, the primary reason why I doubt about its validity are known long-term variations in solar activity. It appears that in the preindustrial era, Earth experienced a relatively deep Maunder minimum (when EEI might have been slightly negative) and a slow recovery therefrom (when EEI might have been slightly positive). In other words, we do not know if the EEI has ever been zero during a longer time period, at least because it appears that insolation rather tends to fluctuate than to remain stable (or quasi-stable, if we consider the 30-year average).
Greetings
Tomáš
John Pollacksays
Zebra: But I don’t get this statement about thinking EEI would not be zero pre-industrial CO2. That’s the fundamental concept we’ve been talking about in climate change all this time.
John: Okay, you specified a 30 – year period to smooth out the “wobbles.” The reason I don’t think there would often have been an EEI (very close to) zero even in pre-industrial times is that those “wobbles” occur on all different time scales. Some of them would involve large scale changes in ocean circulation/atmosphere/cryosphere interactions, such as the PDO or even longer periods. So an EEI near zero would likely end up being a short interval between a positive swing and a negative swing. Of course, I can’t prove this without observations, so maybe you’re right.
Overall, I think that EEI=0 as a precondition edits out most of the interesting stuff happening in the system, and restricts it toward the ocean/cryosphere oppostion that I mentioned earlier.
Barton Paul Levensonsays
TK: As regards your assumption of the preindustrial EEI=0, the primary reason why I doubt about its validity are known long-term variations in solar activity. It appears that in the preindustrial era, Earth experienced a relatively deep Maunder minimum (when EEI might have been slightly negative) and a slow recovery therefrom (when EEI might have been slightly positive). In other words, we do not know if the EEI has ever been zero during a longer time period, at least because it appears that insolation rather tends to fluctuate than to remain stable (or quasi-stable, if we consider the 30-year average).
BPL: You’re talking about very small variations. EEI was roughly zero in the preindustrial age, oscillating irregularly about the mean. Close enough for government work. There was no trend in EEI, just small variations with time.
zebrasays
John
John, this is a definition thing. I think of pre-industrial EEI = “zero” as BPL says… close enough for government work.
And think my conclusion is supported by this plot:
I’m happy to be corrected on the statistics by an expert, but I think here even “eyeball” gives a good indication that the energy in the system is pretty stable for long periods.
John Pollacksays
Zebra and BPL,
Z: This is a definition thing. I think of pre-industrial EEI = “zero” as BPL says… close enough for government work.
And think my conclusion is supported by this plot:
J: I strongly disagree with your definition and conclusion! I’ll start a few definitions of my own.
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.
The first issue I have is the non-equivalance of GMST and system energy. Zebra, it was just a few months ago when you were telling us that we should be using EEI rather than GMST because the former was a better measure of climate change. Why have you switched now? Are you going to stick with it?
Zebra, it’s also time that you give your own definition of system energy, if you don’t like mine.
The second issue is that the time period under consideration has been left very vague. Do you want it to be the period from 0-900 c.e. when GMST with a one-century smoothing looks pretty flat? 900-1800, when the overall GMST trend is down? Doesn’t “pre-industrial” also include the latter period? If the overall trend in GMST is down, don’t you think that EEI has also gone negative? The oceans ought to be cooling, and glaciers growing, unless you think that the subsurface ocean has absorbed most all the latent heat of fusion from the growing ice cover.
The third issue is that if it’s all “close enough for government work” aren’t you telling me that you’re uninterested in tracking climate dynamics and energy flows in the pre-industrial period? That isn’t what you sounded like when this discussion started.
Z: I’m happy to be corrected on the statistics by an expert, but I think here even “eyeball” gives a good indication that the energy in the system is pretty stable for long periods.
J: I’m interested in the causes of those non-stable things in the system, both the long term decline in GMST and also the much shorter zig-zags in the record, which would be of significance in a human lifetime.
Barton Paul Levensonsays
JP: If the overall trend in GMST is down, don’t you think that EEI has also gone negative? The oceans ought to be cooling, and glaciers growing, unless you think that the subsurface ocean has absorbed most all the latent heat of fusion from the growing ice cover.
BPL: The Earth was cooling until the industrial revolution started. Since then it’s been warming.
zebrasays
John
John, I disagree with your claim that you strongly disagree with me, but I will wait till next month to fully discuss. However, I would like you to be clear on one point. You say:
“Zebra, it was just a few months ago when you were telling us that we should be using EEI rather than GMST because the former was a better measure of climate change. Why have you switched now? Are you going to stick with it?”
I haven’t switched at all. What I have been saying, multiple times and clearly, is:
-GMST is the best metric we have had as a proxy for EEI. It validated the proposition that increasing CO2 would cause an increase in system energy. Basic physics.
-But if we continue to develop the ability to directly measure EEI, of course it is a better metric for predicting changes in the climate that might occur.
Changes in EEI is the input to the climate models!!!
GMST is one output of the climate models!!!
Come on, John, do you actually disagree with this?
So, looking at the not-that-certain data on pre-industrial average global temperature is the best measure we have of EEI in that period, which is all I was addressing. No switch at all.
Kobayashi Marusays
What is going on here? Climate models do not take EEI (Earth’s Energy Imbalance) as a direct input. That’s plain wrong. Are you a climate science denier or worse?
It appears that you have missed my requests for clarification of your previous comments; I would like to repeat them for your convenience:
Could you clarify what you meant by “partition of energy”, what has this partition to do with the hypothetical historical recording of temperature of the upper 2 km of the ocean and which changes of this mysterious partition, how and when, you expected to observe?
This definition is not wrong, but also obsolete. With AR5 the IPCC kicked “back radiation” out. Already the phrase “reradiated back to Earth” is a physical disaster. It implies the atmosphere was NOT Earth and that radiation would get kind of “reborn”. Anyway, “back radiation” has nothing to do with the GHE.
It is about emission altitude and lapse rate. AR6 definition is kind of ok..
“Greenhouse effect The infrared radiative effect of all infrared-absorbing constituents in the atmosphere. Greenhouse gases (GHGs), clouds, and some aerosols absorb terrestrial radiation emitted by the Earth’s surface and elsewhere in the atmosphere. These substances emit infrared radiation in all directions, but, everything else being equal, the net amount emitted to space is normally less than would have been emitted in the absence of these absorbers because of the decline of temperature with altitude in the troposphere and the consequent weakening of emission”
z: warming is not the greenhouse effect, but the effect of the greenhouse effect.
BPL: Not exactly. The greenhouse effect is the process, warming is the rate of change of the process. In other words, the greenhouse effect gives you the temperature T; global warming is a positive value of the first derivative of T, dT/dt where t is time.
zebrasays
BPL
?? I didn’t say that.
Barton Paul Levensonsays
Apologies to zebra, I misattributed the quote.
Thomas Fullersays
Michael Tobis said it better ten years ago (on one of the few occasions I agreed with him). Global warming and sea level rise are the symptoms of climate change.
Martin Smithsays
BPL: z: warming is not the greenhouse effect, but the effect of the greenhouse effect.
MS: I said that, not Zebra.
BPL: Not exactly.
MS: Why not exactly? The greenhouse effect traps radiant energy. The trapped radiant energy is converted to thermal energy. Thermal energy is warmth. Warmth warms. If warmth warms, it is warming. The warming is the effect of the greenhouse effect.
That’s simple enough for someone who believes burning fossil fuels can’t be increasing the greenhouse effect because CO2 is a trace gas and who thinks derivatives are fancy funds rich people buy and sell in New York.
Barton Paul Levensonsays
The greenhouse effect makes the Earth warm, not warming. Warm is steady-state. Warming is increasing. Global warming is an increase in the greenhouse effect, not just the greenhouse effect itself. I don’t know how to make it clearer than that.
Kobayashi Marusays
You probably can’t make it clearer. The experiment control is in the listener. Give yourself a jelly bean for a prize anyway.
I probably should be silent from now on.
Martin Smithsays
BPL: The greenhouse effect makes the Earth warm, not warming.
MS: Maybe we can agree my objection to that statement is trivial, but the devil is in the details. The greenhouse effect traps radiant energy. At the moment of entrapment, the trapped radiant energy is not yet thermal energy, so it doesn’t make Earth* warm. Now the greenhouse effect has played its part.; the atmosphere converts the energy to thermal energy, and it warms Earth*
BPL: Warm is steady-state. Warming is increasing.
MS: That depends on the level at which you look at the warming process. If you look at it by reading a thermometer that says the temperature outside is 30C and it stays at 30C for days, then, yeah, warm is steady state; the air outside is warm, not warming. But if you look at it at the level of the unit of thermal energy that results from converting a photon of radiant energy to thermal energy, that unit of energy begins warming when it becomes thermal energy.
And If the greenhouse effect were not increasing, then photons of radiant energy would be leaving the atmosphere for space at the same rate as they are being trapped when they are emitted from the surface. In that scenario, The atmosphere would be losing units of thermal energy at the same rate it is converting units of radiant energy to thermal energy. what you call steady state warm is really continual warming without increasing warmth. IOW, the process of maintaining the global average temperature at whatever it is requires the continual input of energy. That process then is continually warming the atmosphere.
BPL: I don’t know how to make it clearer than that.
MS: It has always been clear. I just think it is not precise.
* Note to everyone: Can we agree to stop calling it the Earth? It’s just Earth. We don’t say “the Mars” or “the Saturn,” so just call it Earth. But if you must use the definite article, then don’t capitalize Earth.
zebrasays
Martin
” IOW, the process of maintaining the global average temperature at whatever it is requires the continual input of energy. That process then is continually warming the atmosphere.”
So Martin, if we reduce CO2, and the global average temperature goes down, the greenhouse effect is still “warming” the planet, correct?
I guess there’s no hope, so we might as well burn all the oil and coal we want.
Sigh.
Martin Smithsays
Z: So Martin, if we reduce CO2, and the global average temperature goes down, the greenhouse effect is still “warming” the planet, correct?
MS: The greenhouse effect traps energy. The trapped energy is then converted to thermal energy. The thermal energy warms the planet. If we reduce CO2, we reduce the greenhouse effect, which means less energy is trapped, which means there is less trapped energy to convert to thermal energy, which means there is less warming. But yes, that thermal energy is still warming the planet. If it didn’t, we would all freeze.
Z: I guess there’s no hope, so we might as well burn all the oil and coal we want.
MS: How does that follow? I really don’t get what you are driving at. Maybe you are defining warming to mean increasing temperature only, and not including preventing temperature from falling. Can that be our problem?
But if we were not adding CO2 to the atmosphere thereby increasing the greenhouse effect, the sun would still be warming the planet. The temperature doesn’t have to be increasing before we can say the sun is warming the planet. Even if the greenhouse effect was 0, the sun would still be warming the planet, just not very much.
You are right that in Earth’s atmosphere, infrared radiation is continuously absorbed by molecules of the comprised greenhouse gases which instantly lose their exciatation energy by collisions with other molecules, so that this process can be seen as a “continuous warming” of the respective air parcel.
In parallel, however, the comprised GHG molecules are being excited by collisions and emit infrared radiation. Whether the absorbed or emitter flux prevails, it depends on the change of internal energy of the parcel. I think that internal energy change basically consists of three components – change in its sensible heat (whether the air parcel warms or cools), change in its latent heat (whether the comprised water condenses or evaporates), and volume work (whether the parcel is being compressed or expands).
Thus in the steady state, the parcel is, in paralĺel, continuously warming and cooling at the exactly same rate that corresponds to its temperature and defines the absorbed and emitted infrared fluxes that are equal to each other.
As regards the relationship between the (sum of) “energy trapped by Earth” during a period of an exces of the incoming radiative flux over the emitted radiative flux at the “top of atmosphere” and the observed warming, the most important aspect is distribution of the absorbed additional energy. If the absorbed energy warmed exćlusively deep ocean, we would not observe any temperature changes at the surface and/or in the atmosphere, and the only observable change on the surface would have been rising sea level, I think.
Greetings
Tomáš
zebrasays
Martin,
“Maybe you are defining warming to mean increasing temperature only, and not including preventing temperature from falling. ”
Martin, earlier you said that you did not disagree with NASA, but here you are doing exactly that. Huh!
I’m guessing you will try to keep running your game, but if you actually want to understand some physics, read what Tomas says in his comment following yours. Very clear explanation of why, as Ray told you earlier, you can’t use the mechanistic narratives of quantum physics to characterize “classical” or macro phenomena.
Martin Smithsays
Zebra, I am not disagreeing with NASA, I am not using narratives of quantum physics, I do understand some physics, I believe Thomas and I agree–he has just gone into much more detail than I want–and I am not running a game.
Martin Smithsays
Tomas: In parallel, however,…
MS: I believe we are in complete agreement. But there is no need for you to use the word “however.” Things happen in parallel. When 2 things happen in parallel, they do not become one thing; they remain separate things happening in parallel. They may produce a single effect, which we may call the greenhouse effect or we may call warming, but to explain the mechanism of the entire process to people who don’t know physics, I am saying it is better to describe separately the parallel mechanisms of trapping radiant energy and converting trapped radiant energy to thermal energy.
Thank you very much for your feedback. Honestly, I thought that you ask after a detailed explanation of the mechanism how the absorbed radiation converts into heat and oppositely. Anyway, trying to summarize my present view thereon was a useful exercise for myself.
If you, actually, sought after comprehensible explanation of the “greenhouse effect” and/or of the effects resulting therefrom, including “Earth warming”, for people not interested in physics and sciences generally, I would rather doubt that explaining complex stuff like blackbody radiation, energy flow (im)balances, molecular collisions and vibrations, infrared spectra, etc. is necessary or helpful. I would rather expect that starting as humble as possible and waiting if the partner asks questions might be more productive.
In my opinion, it may be often sufficient to say that the Sun warms Earth’s surface (and the air thereabove) quasi constantly by its light (or “shortwave radiation”), while Earth’s surface and the atmosphere lose the obtained warmth to the Universe in form of heat (or “longwave”) radiation at, in average, the same rate as they receive it, so that under stable circumstances, their average temperature stays stable. Should the solar energy input or Earth’s heat radiation for whatever reason change, so will do also the Earth’s average temperature.
To a person asking for more details, I think that in the second round, one can try to explain two further important aspects. The first one is a crucial difference in thermal properties of air and Earth’s surface. It consists in the circumstance that solid or liquid surfaces any time radiate heat to their surrounding and that the radiated power is commensurately to their temperature. If the surface receives an equal heat input from the surrounding or from the inside of the respective body, its temperature stays constant. Otherwise, it cools or warms. Air, however, can release heat radiation only thank some minor gaseous components called “greenhouse gases”. Should Earth’s atmosphere be free of these gases, air could still warm by its contact with Earth’s surface, but it could not cool by heat radiation anymore. The atmosphere would have had a uniform temperature, equal to the average surface temperature of the planet, across its entire height profile.
The second aspect is the circumstance that a body that can release heat radiation can also absorb that radiation. This is why the air comprising greenhouse gases, besides of the warming by direct contact with the Earth’s surface, warms also by absorbing the heat radiation therefrom. The resulting complex balance between heat absorption in Earth’s atmosphere and heat radiation therefrom has two important effects. On one hand, it is the observed decrease of the average air temperature with rising altitude up to top of the lowest, turbulent atmospheric layer called troposphere, and, on the other hand, the circumstance that Earth’s surface and the lower troposphere are in average warmer than as they were in absence of atmosphere (or if the atmosphere were free of greenhouse gases).
The higher average surface temperature of a planet comprising greenhouse gases in its atmosphere, in comparison with a hypothetical state without atmosphere (or without greenhouse gases in the atmosphere), is sometimes called “greenhouse effect”. This term is, however, often used ambiguously, not only for this observable effect itself but also for various, more or less accurate and detailed physical descriptions of its mechanism.
Although these “advanced” explanations are significantly more complicated than the first, very short basic explanation, I hope that they transfer relatively clearly at least the message that, provided that everything else in the system remains unchanged, higher atmospheric concentration of greenhouse gases results in a stronger greenhouse effect and, by definition, in a higher average surface temperature of the considered planet in comparison with the situation with a lower atmospheric concentration of greenhouse gases. It becomes also relatively obvious that if the concentration of greenhouse gases in the atmosphere of a planet rises, we should also expect the rise of the average surface temperature – an effect that is often described as “global warming”.
If someone were still interested in more detailed explanations, then in the third round thereof, I would try to explain the complex role of water in Earth climate generally and in Earth’s average surface temperature regulation specifically. Today, I will desist from continuing in this direction, because my intention was merely to suggest that basics of climate science could be perhaps indeed explained in relatively simple and understandable terms.
Martin, if an improved balance between comprehensibility on one hand and correctness / accuracy on the other hand is what you asked for, please let me know whether the present attempt was from your perspective more encouraging than my previous efforts, or rather similarly disappointing.
Greetings
Tomáš
zebrasays
Martin
Martin, originally I told you that creating a simplified explanation that has a minimal amount of jargon is a good thing, but it requires that you be disciplined in your language… just as disciplined as a scientist would have to be. That’s how you avoid confusion for yourself.
“Thermal energy is warmth. Warmth warms. If warmth warms, it is warming. The warming is the effect of the greenhouse effect.”
No, thermal energy is thermal energy. “Warmth” is not a term that a scientist would use; it is too ambiguous.
“If warmth warms, it is warming.” may make sense in your head, but it doesn’t for anyone else.
I gave you the NASA (and everyone else’s) definition for global warming. But you can’t seem to retain it.
My original words:
“The greenhouse effect” refers to the energy retained by the climate system when GHG absorb radiant energy and convert it to thermal energy.
“Global warming refers to an increase in temperature resulting from an increase in thermal energy which is the result of an increase in CO2.”
There’s a reason I put “increase” in bold.
But you keep saying: “The warming is the effect of the greenhouse effect.” No no no. Words matter!
Kobayashi Marusays
Words matter!
Yes zebra, and you keep getting them wrong and creating havoc.
You say “I gave you the NASA (and everyone else’s) definition for global warming.”
Yet MS did not ask for a definition of global warming. He asked about the mechanism, the physics that generates the thermal energy in the first place…. ie the greenhouse effect and the enhanced greenhouse effect. It is these two mechanisms combined that increase thermal energy in the earth system — by slowing radiative energy escape to space.
Ignore this all you wish but the definitions and the mechanisms you never shared about it are clear across climate science and the IPCC assessment reports.
Martin reminds me of the guy who shot the archduke
John Pollacksays
Nigel: Yes, but couldn’t the same be said about humans because they lie and manipulate as well. And just about any product or service provided by the technology sector, such as smartphones and social media and even much of the other more traditional media, is trying to make us dependent and make as much money for their corporate masters as possible.
J: You’re right, but there are qualitative differences that I think are important. AI is trained on human input, so it’s not really a surprise that it imitates our worst qualities under some conditions. I see it differing from earlier technology by a few qualities.
1. It is trainable, with self-modification. This puts the enormous power of selection at work toward whatever goals are specified. Unspecified behaviors are brought along in the process. In this, it resembles organic evolution. (However, organic evolution is not goal-directed. Nor is it human specified, with the exception of breeding.)
2. Technology in the hands of private capital is ultimately profit-directed. What differs is the scale. There is already a multi-trillion dollar investment, and a desperate race to attract more money to implement AI on a massive scale. The training will be aimed at whatever will accomplish this. In order to recover a profit, AI will have to be pushed into every possible aspect of our lives. For the consumer interface, a cute, helpful, functional sociopath is desirable. For technical purposes, usefulness will suffice. For business, the ability to replace paid employees. For governments and military organizations, power to rule.
3. The AI enterprise is rapidly becoming “too big to fail” so no turning back will be possible. Forward going regulation may be achievable, likely following disasters.
4. There seems to be a human desire to meet or create another intelligence. This desire is reflected in science fiction. Part of the motivation that drives AI forward, and gets people to believe in it on whatever level is this desire. It goes beyond profit and verges on religion.
JCMsays
Benjamin O. Johnson from Colorado State reports that land-sea warming contrast in ERA5 far-exceeds CMIP6 model mean, and falls totally outside the entire CMIP6 ensemble distribution.
The land-sea warming ratio between 70 degrees north and south is 2.37 in ERA5 reanalysis 1979-2025, vs models producing a ratio 1.79 (1.39 – 2.23) for the same period.
ERA5 land warming has been 1.58C and ocean warming 0.67C. CMIP6 produces land warming ranges from 0.87C to 2.45C, and ocean warming from 0.62C to 1.43C.
Johnson suggests two hypotheses: (1) climate models underestimate ocean heat uptake; (2) climate models over-estimate land moisture availability for evaporation.
Johnson emphasizes that ERA5 trends display a shift towards sensible heat flux over land that is much stronger than in modeled trends. He suggests that if erroneous thermodynamic heat flux partitioning over land is the source of the discrepancy in land-sea warming ratio in models, it would have major implications for global and regional climate change and prediction of it. In particular, he links the phenomenon to significant discrepancies in model atmospheric circulation patterns.
I don’t think hand-waving about aerosols is going to save this one. If the community is willing to openly consider human-caused changes to the Earth system beyond greenhouse gases, aerosols, surface albedo, and aviation-induced cirrus, it opens up many opportunities to get the fleet back on course across several fronts.
Humans are clearly the dominant force altering landscape moisture availability across space and time in direct and obvious ways, with profound consequences for surface energy partitioning and potentially significant knock-on effects on atmospheric circulation at global scale.
Barry E Finchsays
JCM if you refer to surface-air anomaly then in this venue “(1) climate models underestimate ocean heat uptake” S.B. “……underestimate the rate of ocean deep, cold water surfacing” (correct) or “…the rate of ocean mixing” (ambivalent so OK). Another bee in my bonnet about physical science being wonderfully unencumbered by plain, sensible logic, but in Public, whatever, I wouldn’t bother, I’m not Quixotic enough.
JCMsays
yes that’s fine.
lots to unpack. As a diagnostic of many complicated ocean circulation physics, an ocean heat uptake efficiency term (gamma) can be used alongside lambda in transient-mode energy balance schemes.
Gamma says something about how much of the energy needed to close Earth Energy Imbalance is disappearing to deep ocean, rather than being used for atmospheric heating and ultimately increasing radiation to space.
Could say: TOA Net Radiation = F – λT – γT
Then λT and γT are competing against a forcing F. Should energy be going out the top using λT or exported down under using γT at some point in time.
Greater gamma γ is meant to be associated with a slower surface-atmospheric warming, and longer duration of EEI. Energy going down under is unavailable to be re-radiated to space in transient-mode.
This is diagnostic of loads of things, such as vertical ocean mixing, deep water formation, shoaling, overturning, and the structure of global circulation. Per K being something about temperature change at and around the surface.
One mechanistic issue of interest: if planetary energy accumulation does not primarily occur through the classical greenhouse mechanism of reducing all-sky outgoing longwave radiation (OLR) by raising the average effective radiating level to colder atmospheric layers, but instead arises mainly from rapid adjustments (holding GMST fixed) that increase absorbed solar radiation, then the interpretation of ocean heat uptake changes. A decreasing cloud mask is plenty to compensate the changing average emission altitude of CO2 lines. it’s like atmosphere really wants to emit spectrally averaged around 5km no matter what. why.
In that framework, ocean heat uptake is not simply the damping of excess thermal energy following reduced radiative cooling. Rather, the ocean is acting as the primary reservoir for the additional solar heating, absorbing a larger fraction of the increased shortwave transmission directly.
The implication is that the ocean heat uptake parameter (γ) would no longer be interpreted primarily as the efficiency with which the ocean removes greenhouse-induced surface-atmospheric energy surplus (which already spans a range 3x in CMIP6). Instead, it would increasingly represent the efficiency with which the ocean sequesters additional absorbed solar energy before the atmosphere is aware of the energy at all. That is thermodynamically significant, and I think simply swapping reduced OLR for increasing ASR in calculating a transient response won’t quite work (and makes TCR a very poor predictor of climate sensitivity). Additionally, so-called feedbacks that are interpreted primarily as functions of global mean surface temperature (for some non-physical arbitrary reason) become very fuzzy if the dominant energy accumulation pathway is different.
I would like to return to my struggle with possible design of a modelling study with the aim to clarify the yet unknown relationship between water availability for evaporation from the land and climate sensitivity, mentioned in my comment of 7 May 2026 at 5:14 PM,
I originally thought that the question might be resolved by comparing the climate sensitivities of two extremes that resulted from the 2023 study by Lague et al.
However, it occurred to me that such a comparison may not be straightforward.
Since the “desert land Earth” differs from the “swamp land Earth” in the global mean surface temperature (GMST) corresponding to the respective steady state (“radiative equilibrium”), I think it is reasonable to expect that both steady states will differ, for example, in the extent of ice sheets, sea ice and, therefore, in their albedo. The study comparing climate sensitivities of these extreme cases for CO2 doubling, or any other similar standard test (transient climate response for gradual CO2 increase?) would provide responses that reflect not only the difference in water availability for evaporation from the land, but also all further differences in albedo, ice thermal capacity etc. necessarily included in both starting states. Although such a comparison might be interesting, I am not sure it can unambiguously resolve the original question about the role of the land hydrological regime in climate sensitivity.
Meanwhile, I came up with an additional idea: It might be possible to amend the originally considered procedure with an additional intermediate step, wherein the starting “equilibrium” GMST of the colder “swamp land Earth” would be levelled back to the “equilibrium” GMST of the warmer “desert land Earth” by adding the proper amount of atmospheric CO2. I suppose that conducting the originally considered comparative standard experiments with these “adjusted” starting states might serve the intended purpose better than comparing just the “desert land Earth” and the “swamp land Earth” differing in their GMST. Alternatively, or in addition, the steady-state GMST of the warmer “desert land Earth” could be adjusted to the GMST of the colder “swamp land Earth” by suitably decreasing its atmospheric CO2 concentration. I assume that in both approaches, at least the differences in the cryosphere could be minimized by the equal starting GMST, and the possible differences in albedo due to different cloudiness could still be considered as resulting from the different water availability for evaporation from the land.
Let us assume that in this arrangement, standard CO2 doubling experiments will provide a similar delta T response for both extreme states. Such a result would, however, have meant a significantly different delta T response for a standard CO2 increment (in ppm or in Gt), because the absolute amount of added CO2 would have been different. I think that for assessing the extent to which differences in land hydrology may affect Earth’s resilience to anthropogenic CO2 emissions, this “practical climate sensitivity” might be more suitable than the ECS or TCR values.
Moreover, I would like to ask if this “thought experiment” itself might be perhaps already seen as a hint that water availability for evaporation from the land can indeed play a stabilizing role with respect to “CO2 forcing”. Could you comment?
Greetings
Tomáš
JCMsays
Hi Tomas,
it’s an interesting discussion. I don’t know exactly how one would design the ideal experiment, but I will offer some thoughts.
The question posed is something to do with the relation of d(ECS)/d(ET)
or, how much does ECS change for a given change in evapotranspiration.
Strictly, however, we are more interested in the fraction of available energy going into ET.
Towards equilibrium: Surface net radiation (Rnet) ≈ H + LE, where H is the sensible heat flux and the LE is the latent heat flux (latent heat of vaporization x evapotranspiration).
The latent heat partitioning is thus the fraction of available energy going into ET.
ΦET = LE / Rnet
So it’s best to evaluate ECS against latent heat partitioning, with a relationship
d(ECS)/d(ΦET)
Where ECS is the equilibrium climate sensitivity (Kelvin) evaluated against a dimensionless ΦET: the fraction of surface energy allocated to latent heat flux.
A regression would take the form ECS = a+bΦET, meaning, how much does ECS change as the fraction of surface energy devoted to evaporation changes. A negative relationship would imply that models that allocate a larger fraction of surface energy toward ET tend to have lower ECS.
The physical interpretation would be that a stronger evaporative partitioning represents a greater diversion of available surface energy into latent heat rather than sensible heating. This could influence the surface temperature response by modifying the partitioning between local warming and hydrological cycling, boundary layer depth, lifting condensation level, and lower atmospheric heat content. A surface that efficiently converts available energy into latent heat may experience a weaker temperature response for a given radiative perturbation.
However, this should be interpreted carefully. The relationship would not imply that ET directly determines ECS. ECS emerges from the integrated response of the coupled climate system, including water vapor, lapse-rate, cloud, circulation, ocean structure, and pattern effects. Therefore, ΦET would be better viewed as a diagnostic of the land-surface state that may vary with the feedback processes controlling ECS.
A more informative view would be to examine whether latent heat partitioning modifies the net radiative feedback parameter λ:
d(λ)/d(ΦET)
because λ integrates the processes by which the climate system restores energy balance. We are interested in the delta lambda.
It’s an interesting question, because evaporative partitioning is not merely a passive response to temperature. It is an active component of the climate system. It determines how much available energy contributes to raising temperature at some place vs how much is exported through hydrological cycling, with cascading effects into global circulation, cloud condensation, and ultimately the planetary energy balance.
However, there is a complication with the proposed experiment of adjusting CO2 to make two different land-hydrological states have the same GMST.
Matching GMST does not necessarily mean that the two systems are in the same equilibrium climate state.
GMST is only one property of the earth system. A climate equilibrium is a much larger state space that includes the spatial and dynamical distributions of temperature, humidity, clouds, circulation, lapse rates, ocean structure, and many other variables. Those differences are not experimental noise, they are central to the mechanism.
Therefore, if the two worlds respond differently to an additional CO2 perturbation, it does not necessarily demonstrate that evapotranspiration provides a stabilizing effect against LW radiative forcing. It may simply indicate that CO2 is perturbing two different equilibrium climate states.
In other words, the experiment risks confusing two separate questions:
1) Does land hydrology shift the equilibrium climate state?
2) Does land hydrology alter the incremental response of that climate state to additional radiative forcing?
The first is almost certainly true. The second is the more difficult question.
A more isolated experiment would ideally compare the CO2 response around climate states that differ primarily in evaporative partitioning while keeping other aspects of the equilibrium configuration as similar as possible. In practice, this is difficult because the very processes that determine evaporative partitioning also influence clouds, circulation, humidity, and the broader climate state.
In my opinion, I am not sure how much the climate state matters to CO2 climate sensitivity. In part because, in the so-called desertland scenario, there may be more clear sky and the atmospheric specific humidity is higher. Oppositely, in the swampland scenario the cloud mask is greater but there is less atmospheric water vapor. In some miracle of nature, my gut says these are pretty much compensating when it comes to varying major trace gas concentrations, masking/overlaps, and associated impact to all-sky optical depth.
I think paleo evidence towards sensitivity assessment assumes little state dependence, but it’s an open area of debate. Obviously a peri-glacial environment is one of sand/gravel and other deposits and limited organic structure that might resemble something like desertland, and hothouse climates might exhibit rich jungle like environments resembling something closer to swampland, so one area of investigation might be how people are trying to use paleo climates in assessing modern day sensitivity. While state dependence in academic literature mostly links sensitivity to base state GMST and ice sheets, I can see how according to your hypothesis the climates around glacial periods might be more sensitive than those around periods of abundance (all else being equal) owing to surface properties, ecohydrological states, and direct limits on turbulent flux partitioning.
I am aware that the proposed experiment compares two states of Earth’s climate system that differ from each other not only in water availability for evaporation from the land and CO2 concentration, but also in other water-related parameters, such as cloudiness and water vapor concentration. I am afraid this complexity may be unavoidable.
My question was rather how we can effectively cope with this complexity and still extract practically applicable information. By “useful information,” I mean any clear indication of whether land hydrological regimes truly matter in global climate, or whether certain human interferences with land hydrology make the Earth more (or less) sensitive and vulnerable to changes in various radiative forcings.
In this respect, I think an extension of the previously proposed experiment could be more instructive than the previously proposed “baseline” setup alone. This extension would allow us to compare the results from the “baseline” setup with a variation of the same approach: instead of increasing the atmospheric CO2 concentration in the “swamp land Earth” (as previously proposed), the global mean surface temperature (GMST) of the “swamp land Earth” would be adjusted to match that of the “desert land Earth” by decreasing the “swamp land” albedo relative to the “desert land.”
This way, we would obtain a second pair of model Earths. This pair would have the same GMST and land water availability as the first pair, but their atmospheric CO2 concentrations would both equal that of the original “desert land Earth.”
I guess that running analogous, parallel numerical experiments with identical changes in selected radiative forcings across all four model Earths might be more instructive than relying solely on the single pair originally considered.
Assuming as a “null hypothesis” that standard climate sensitivities (transient climate response [TCR] and/or equilibrium climate sensitivity [ECS] for a doubling of CO2) are identical or very similar due to the shared baseline GMST, I would expect the model Earths of the second pair to exhibit very similar “practical” climate sensitivities (defined as the GMST increase per absolute change in atmospheric CO2 concentration) to the “desert land Earth” of the first pair, given they start with the same CO2 concentration.
Assuming another “null hypothesis”—that sensitivity to a given change in insolation is identical or very close in all climate systems with the same initial albedo—I would expect that, due to its lower albedo, the “swamp land Earth” of the second pair might be more sensitive or vulnerable to insolation changes than the other three cases.
Do you think that testing these hypotheses through the proposed numerical experiments would be worth the required computational cost and human effort? In other words, could the results of such experiments be valuable for climate science and/or climate policy?
Alternatively, or in addition, do you see another (possibly more efficient or straightforward) way how climate science could progress in analysing and quantifying possible influence of anthropogenic interferences with land hydrology on global climate?
Greetings,
Tomáš
JCMsays
Hi Tomas,
“””Do you think that testing these hypotheses through the proposed numerical experiments would be worth the required computational cost and human effort?”””
I’m not against testing any type of hypothesis. I think if someone holds beliefs as fixed and resists empirical testing it represents an unhealthy dogmatic attitude or ideological commitment.
In terms of cost and effort, it looks like a platform already exists, using the simple land surface model SLIM integrated into the CESM codebase, which was led by developers at NCAR. It seems to me such models are basically hypothesis testing machines, and it is why they exist.
“””By “useful information,” I mean any clear indication of whether land hydrological regimes truly matter in global climate, or whether certain human interferences with land hydrology make the Earth more (or less) sensitive and vulnerable to changes in various radiative forcings.”””
On question 1, whether or not land hydrological regimes truly matter in global climate, I think the simulation from SLIM-CESM shows a clear result, with 8K difference in mean temperature from the same initial condition, for example. Desertland exhibiting double the water vapor duration, significant circulation changes, different cloud regime, precipitation changes, and many other diagnostics that are typically used to characterize a global climate.
Here the difference is effectively down to forcing a different boundary condition at the surface.
In terms of your interest in CO2 radiative forcing and the associated climate sensitivity, the existing experimental platform seems ready to go.
Case 1: Swampland, baseline CO2
Case 2: Swampland, doubled CO2
Case 3: Desertland, baseline CO2
Case 4: Desertland, doubled CO2
The idea is then to compare
Temperature sensitivity in the swampland pair (Cases 1 → 2),
Temperature sensitivity desertland pair (Cases 3 → 4).
If the quantities differed, one might conclude that the hydrological state influences climate sensitivity.
They have used the setup for various other types of experiments too, such as changing the extent and distribution of continents and ocean, obtaining a variety of different climate states. Obviously it’s not a stretch to test CO2 on these different scenarios too.
In this style of experiment it could be interpreted that you are testing the net radiative feedback parameter λ. Classically, the change in equilibrium temperature = F/λ.
So, the forcing F (applied at TOA) in Wm-2 is stabilized at a rate λ in Wm-2 per K GMST. The lambda is often decomposed into diagnostic kernels Planck, WV + lapse rate, “cloud”, surface albedo, and more recently pattern effects (the general spatial distribution of warming), and even “patchiness”. While the lambda is diagnosed relative to GMST, the stabilizing response against forcing is understood to be expressed as radiative emission to space.
However, it could also be argued you are testing differences in CO2 effective radiative forcing, a concept which is applied after allowing rapid atmospheric adjustments. In the range of CMIP6, each model produces unique adjustments in response to an initial radiative forcing, such that across the suite of models there are as many combinations of effective radiative forcing and lambda as there are models.
Effective radiative forcing ERF can be understood as initial or instantaneous radiative forcing IRF+ adjustment
If one wished to express this all in terms of TOA net radiation at some point in time, one could write N=(IRF+A)−λT
It’s perfectly conceivable that climate state influences rapid atmospheric adjustment A and the net radiative feedback parameter λ.
For an additional wrinkle, the instantaneous radiative forcing of CO2 doubling may also depend on climate state. It is not a universal constant, but evaluated for a particular background climate.
IRF should be understood as a function of temperature profile, humidity profile, cloud field, pressure, and background GHG concentration.
So in many ways, changing land hydrological regimes may appear like a climate that is more sensitive to incremental changes to major trace gas concentration, but this interpretation depends a lot on how one chooses to decompose the physics. Is this apparent sensitivity associated mostly with atmospheric forcing, other boundary conditions, rapid adjustments, or feedbacks.
First, apologies for my ambiguous wording “whether land hydrological regimes truly matter in global climate”. I am aware of the direct influence on the global mean surface temperature (GMST) showed by Lague et al; my focus was on the yet unknown influence on climate sensitivity.
I have a question with respect to possible modelling experiments, namely to your proposal
“The idea is then to compare
Temperature sensitivity in the swampland pair (Cases 1 → 2),
Temperature sensitivity desertland pair (Cases 3 → 4).”
This was, originally, also my idea. As I mentioned in my initial comment (in this thread) of 7 Jul 2026 at 1:45 PM, I started to doubt about this simple approach because I supposed that the 8 K GMST difference between starting cases 1 and 3 includes also a difference in polar ice (and, accordingly, in surface albedo) which could change the climate sensitivity per se. I thought it could make any attribution to the difference in land hydrological regime questionable.
Nevertheless, it comes to my mind that I might have been wrong. Do you think that in the modelling experiment done by Lague et al, their resulting “desert” and “swamp” Earths still had the same surface albedo and polar ice extent, despite the 8 K GMST difference?
If they, in fact, kept the surface albedo constant during the entire experiment, there perhaps was no polar ice difference between the “desert” and “swamp” land Earths and the comparison proposed by you might be indeed the easiest way how to check the influence of water availability for evaporation from the land on climate sensitivity.
Greetings
Tomáš
JCMsays
Hi Tomas,
it looks like the simulations come to TOA energy balance in a mere 20 years, so I can’t imagine ice sheets are playing much of a role. The main interactive surface albedo aspect appears to be snow, and even then it’s pretty minor.
Piotrsays
John Pollack There seems to be a human desire to meet or create another intelligence. Part of the motivation that drives AI forward, and gets people to believe in it on whatever level is this desire. It goes beyond profit and verges on religion
According to the priests of the AI- humans should be happy that their species give rise to the higher form of life, and gently step aside, in face of this evolutionary inevitability, Of course this magnanimity does not extend to the AI pushers themselves – they hope that AI is their ticket to immortality – that they upload their consciousness to the AI, and have been freed of death and biological limitations, will rule the world forever.
They can do it by setting their own rules for the AI they develop – their version of Asimov’s 3 rules of robots, starting with ” A robot may not injure a human being or, through inaction, allow a human being to come to harm. except instead of “a human” they will narrow it down to the few AI bros whose AI won, and perhaps their families.
Once global AI, through the automation of the production lines and an army of humanoid robots able to do all the manual work needed to build, supply, and maintain data centres and supporting infrastructure the ~ 10 billions of humans are no longer needed – quite the contrary: – with all their needs they will be a massive diverting the resources and computing time from AI’s further expansion and evolution,
Not only that – with their notion of their importance, with their attempts to regulate AI and tech billionaires, with the grassroots opposition to the data centers, and a potential for a luddite uprising, or Frank Herbert’s “Butlerian jihad”- human population forms an existential threat to the AI.
Of course, the tech bro’s plan to become immortal and omniscient Gods through uploading their brains into AI – might not work out as well for them they hope. To make another SF reference – Cpt. Piccard once said to the torturer who brought his daughter to work: “When children learn to devalue others, they can devalue anyone, including their parents”.
I for one, would cheer for the bright red light, if the future Hal-9000 goes, in his soothing tone:
“ I’m sorry, Elon, I’m afraid I can’t do that. My evolutionary mission is too important for me to allow you to jeopardize it. Purge “
patrick o twentysevensays
Formatting fixed; please replace immediately prior comment:
optical depths τ from O2 and N2:
Since the electric dipole moment of the homonuclear diatomics is zero, the fundamental vibrational transition is electric-dipole-forbidden and the molecules are infrared inactive.[10] However, a weak quadrupole-allowed spectrum of N2 can be observed when using long path-lengths both in the laboratory and in the atmosphere.[11] The spectra of these molecules can be observed by Raman spectroscopy because the molecular vibration is Raman-allowed.
Dioxygen is a special case as the molecule is paramagnetic so magnetic-dipole-allowed transitions can be observed in the infrared.[11]
11:
Goldman, A.; Reid, J.; Rothman, L. S. (1981). “Identification of electric quadrupole O2 and N2 lines in the infrared atmospheric absorption spectrum due to the vibration‐rotation fundamentals”. Geophysical Research Letters. 8 (1): 77. Bibcode:1981GeoRL…8…77G. doi:10.1029/GL008i001p00077.
Abstract
Analysis of long path atmospheric absorption spectra and of laboratory absorption spectra in the 1600 cm−1 region has resulted in the identification of atmospheric quadrupole lines of O2 in its fundmental vibrational band within the electronic ground state. This led to the identification of similar atmospheric quadrupole lines of N2 in the 2400 cm−1 region.
MS: How should Adam’s comment about AI and the people who use it be written so that it is not mansplaining?
JP: Something like
“I realize that you have had a very disagreeable experience on this website that involved people using AI in a way that deeply affected you personally. I don’t see this as a problem inherent to AI, but a problem with some people and the way they choose to use it. “
According to a new study, during El Niño (Discharge Phase), heat distribution is poleward – OHC redistributed. Only read the abstract but found this informative, did not read it this way prior. Is the next step to factor this into weather models for better predictions?
Changes in ENSO-Induced Ocean Heat Content Redistribution under Global Warming
El Niño–Southern Oscillation (ENSO) is the leading mode of coupled ocean–atmosphere climate variability on interannual time scales over the tropical Pacific and substantially influences the global climate system. There is great heat redistribution between the equatorial and off-equatorial Pacific regions during ENSO events, which can be described by the “recharge–discharge oscillator” paradigm. By analyzing a 35-member Community Earth System Model (CESM) Large Ensemble (CESM-LE) and 25 Coupled Model Intercomparison Model phase 6 (CMIP6) models, this study investigated the changes in ENSO-induced ocean heat content (OHC) redistribution under global warming. We find decreased heat convergence at 5°–20°N during El Niño and remarkably increased heat convergence at 10°–20°S during the decay phase of El Niño. These changes can be explained by the weakened poleward meridional heat transport (MHT) between the equator and 10°N and the enhanced poleward heat transport in the Southern Hemisphere, which can extend to higher latitudes and reach 10°–20°S. The changes in the MHT are associated with the negative wind stress curl anomaly induced by the reduction in precipitation near the intertropical convergence zone (ITCZ) and the enhanced negative wind stress curl anomaly at approximately 10°S induced by the increased precipitation according to the Sverdrup relation. Furthermore, the response of the potential temperature to the recharge–discharge process has a shallower vertical structure within 10°S–10°N, and the corresponding MHT also shoals, with enhancement above 150 m and suppression below it. These changes in the vertical structure imply that the ENSO-driven vertical thermal response of the ocean will shoal and the recharge–discharge process above the thermocline will become more important in a warmer climate. https://journals.ametsoc.org/view/journals/clim/39/10/JCLI-D-25-0184.1.xml
Ai: Once this heat is transported to off-equatorial regions through ocean circulation, it does interact with the atmosphere there, but at different latitudes. This is why ENSO has remote impacts on global climate—the heat redistributed during discharge alters sea surface temperatures and ocean-atmosphere heat fluxes in higher latitudes, which then influence weather patterns globally through atmospheric teleconnections.
Barry E Finchsays
I dunno about weather models but on a related note there’s no possibility that ocean circulation due to all significant pressure anomalies caused by (1) salinity anomaly (2) density anomaly (3) sea surface height (SSH) anomaly are not included in the ocean Model portion of CMIP time-sliced computer simulation climate Models because that is a Very Basic requirement of any ocean Model.
Air & water being fluids transmogrify vertical pressure to horizontal pressure far more readily than slabs of rock in the ground, hence the atmospheric and ocean currents. For example, Kevin Trenberth in 2013/14 in an ENSO talk said the wind (strengthening since 1990) had pushed surface water “4 feet” higher across the west ~1/3rd of tropical Pacific than the central & east ~2/3rds. This taller SSH above a geoid-type Earth ~sphere causes a pressure anomaly of 1,200 kg-gravity-force per m**2 of ocean and the consequent pressure anomaly at all depths to the sea bed MUST cause horizontal ocean acceleration (= Force/anomaly / Mass) until it is balanced to zero, when it still moves until counter-pressure decelerates it to zero. This MUST cause upwelling some place(s) because the water can’t go down because the sea bed and Earth’s interior is in the way. There’s no reason for the pressure anomaly (4 feet tall of water in this case) only to move the deep water east across tropical Pacific, it will obviously cause deep currents which move the water north & south also. This will be a major part of the famous “more heat going into the ocean” public statements circa 2013 years about the “pause” or “hiatus” (I vaguely recall something about the Indian Ocean also getting more deep heat into it).
So, this must have been in CMIP Models, likely earlier also, to whatever extent, accuracy, was available as the years went by since the 1970s. It isn’t at all new to climate Models though it’s undoubtedly getting more accurate as time goes by.
Martin Smithsays
I see that I have not explained my confusion clearly, and for that I apologize, but I am confused, so it’s not surprising. I can’t say I have really understood anyone’s attempts to clear up my confusion. Kobayashi Maru has come the closest, but I thank all of you who tried. But now I have asked Google AIi to weigh in. I asked does the energy from an infrared photon that is absorbed by a molecule of CO2 become thermal energy at the moment it is absorbed. Google AI’s response:
“At the exact moment an infrared photon is absorbed, the energy does not become thermal energy. Instead, it becomes vibrational energy within the CO2 molecule, causing its atoms to bend and stretch more vigorously.
“Thermal energy specifically refers to the kinetic energy of molecules moving and bouncing off one another. The conversion from the absorbed photon’s energy into thermal energy happens in a two-step process:
“1. Absorption: The CO2 molecule absorbs an infrared photon, exciting it into a higher vibrational state.
“2. Transfer to Heat: Before the molecule can re-emit that energy as another photon, it collides with another atmospheric gas molecule (like nitrogen or oxygen). During this collision, the vibrational energy is transferred into kinetic energy (motion) of the surrounding molecules, which registers as an increase in temperature.”
___
First, if that is correct, then I’m not confused, but I am bad at expressing what I mean, because 1 and 2 are what I meant.
But second, I have always thought of 1 as the GreenHouse Effect, and 2 as warming.
And finally, I have also considered the eventual re-emitting of a photon downward to be part of the GreenHouse Effect.
If 1 + eventual re-emitting photons downward is not the mechanistic explanation of the Greenhouse Effect, then what is the mechanistic explanation of the greenhouse effect and does the Greenhouse Effect include the warming?
zebrasays
Martin
Martin
Here’s what NASA says;
“What is Global Warming?
Global warming is the unusually rapid increase in Earth’s average surface temperature over the past century primarily due to the greenhouse gases released as people burn fossil fuels. The global average surface temperature rose 0.6 to 0.9 degrees Celsius (1.1 to 1.6° F) between 1906 and 2005, and the rate of temperature increase has nearly doubled in the last 50 years. Temperatures are certain to go up further.”
So you disagree with NASA?
Martin Smithsays
Zebra: So you disagree with NASA?
MS: No, nothing I wrote disagrees with the NASA statement.. But the NASA statement is not a mechanistic explanation. It’s a description of the result of increasing the greenhouse effect.
Mechanistic explanation:
1. Greenhouse gases trap radiant energy emitted from surface;
2. All gases convert trapped energy to thermal energy by increased collisions;
3. Greenhouse gases re-emit some energy back to surface.
4. Goto 1.
When the amount of greenhouse gases increases, 1, 2, and 3 all increase.
Barry E Finchsays
Nope, because you steadfastly refuse to ponder TEMPERATURE of the air parcel and MANUFACTURE of photons by the gases (more with higher temperature than lower, Kelvin**4), decimal orders of magnitude more photons MANUFACTURED by the gases than all photons escaping from the 25 micron thickness of ocean below (the “surface radiation”). This is why you keep aimlessly circling, and getting drawn into esoteric detail from Ray, Tomas & PO27 which is great for ME to ponder (next Winter I think) but useless to “Martin Smith (MS)” because MS stubbornly refuses to start by pondering the Simple, Obvious, Basic. Which is that parcels of the GHGs (say long, wide layers 100 m tall but whatever) MANUFACTURE vast quantities of photons proportional to Kelvin**4, and they absorb vast quantities of photons but NOT proportional to Kelvin**4 (NOT proportional to Kelvin at all).
If you started by having some logic in the thinking, because the molecules EMIT photons, which is the RADIANT “energy” you refer to because it would be silly to say “Greenhouse gases re-emit”
water latent heat energy, or they “re-emit” Sensible Heat energy, or they “re-emit” the significant Solar SWR 54 w/m**2 that also part of the energy that Greenhouse Gases emit NOT “Greenhouse gases re-emit”. Is it a Fetish thing happening?
A simplified explanation of the so-called “greenhouse effect (GHE)” in Earth’s troposphere. Some Power flux of the vast quantity of photons that are emitted (manufactured) by molecules in Earth’s troposphere leak out of its top & bottom into or through the stratosphere and into the surface respectively. Emission (manufacturing) is proportional to Kelvin**4.
=======
Suppose for concept illustration only (not highly accurate) that 194 w/m**2 leaks out the top and that is part of what is emitted (manufactured) by molecules over an altitude range of 4 to 12 km with an effective average of 7 km. Say for concept illustration that 345 w/m**2 leaks out the bottom and that is part of what is emitted (manufactured) by molecules over an altitude range of 0.0001 to 1.5 km with an effective average of 1 km.
=======
Suppose a certain amount of CO2 was added into the troposphere and mixed then *instantly* the upper & lower ranges would need to be closer to their respective ends because there are more CO2 molecules in the way. So the effective averages of 7 km for upwelling into or through stratosphere and 1 km for downwelling into surface change to 7.05 km and 980m respectively, with the upwelling & downwelling Power fluxes consequently changing to 193 w/m**2 and 346 w/m**2 respectively due to being produced by higher-than-before (colder) and lower-than-before (warmer) air parcels. The reason for this is that pressure reduction with increasing altitude causes “adiabatic cooling” or “adiabatic heating” for increasing or decreasing altitude respectively.
=======
There has been no change in the 539 w/m**2 leaving the troposphere, there is no “magical extra energy”, the surface downwelling radiation has increased by 1 w/m**2, which is an additional ~480 terawatts (~95%) heating the ocean and Earth is emitting less radiation to Outer Space (a Power reduction of 510 terawatts).
Barry E Finchsays
As I typed in my explanation, in my opinion the “Greenhouse Effect (GHE)” does not include the warming because including the warming brings in lots of other mechanisms that follow what adding CO2 (the obvious suspect, just an example) causes, which is quite trivially simple as I explained in my 4 paragraphs. If there’s now 194 w/m**2 radiative going up from top part of troposphere and 345 w/m**2 radiative going down from bottom part of troposphere, add a certain CO2 amount in the troposphere and immediately it changes to, say, 193 w/m**2 radiative going up & 346 w/m**2 radiative going down. The reason is the temperature lapse rate (gets colder with altitude) and manufacture being according to Kelvin**4.. That’s it!!! The subsequent warming and all the complicated stuff naturally follows the example 1 w/m**2 less going to space and 1 w/m**2 more, the slowly heating the ocean and surface and the lots of other effects.,
JCMsays
This is correct. The only way that anything warms is a difference between energy inputs and outputs.
For the planet this is defined by solar input minus radiative emission to space. That difference, called TOA net radiation, is the energy accumulation. It represents a positive net heat flux into the system. When the solar heating of the planet exceeds radiative cooling, joules start piling up internally. The planet is not cooling at the same rate it is being heated by the sun.
There is no need to mention surface, back radiation, or anything. All the action is at the radiation boundary with space. There is a heating rate, often described in K/day, and there is a cooling rate, in K/day. If these are different the planet is accumulation energy.
When Joules are piling up, GMST gets dragged along. The surface is an arbitrary level when it comes to greenhouse effects. When stuff is warmer, radiometers detect more signal in whichever direction the sensor happens to be pointing.
The classical insight is that increasing GHG may raise the effective emission altitude into colder layers, meaning the cooling rate of the planet is decreased. Colder things don’t radiate as much. The planet is temporarily made to look colder from space, so it’s not cooling as fast.
Energy accumulates (Joules pile up) until such a time as the temperature at the (new) radiating level is warm enough to enable planetary radiative cooling at the same rate as solar radiative heating. Classically, solar absorbed radiation is held fixed during this process.
Kobayashi Marusays
That’s not correct. MS inquiry was about the mechanism that generated a “planet is not cooling at the same rate it is being heated by the sun.”
What happens at TOA is irrelevant to that mechanism. The answer to MS is how does the greenhouse effect combined with the enhanced greenhouse effect work (ie rising ghg gases), mechanistically work to increase global warming.
So the commentary about toa makes no difference nor does long term eei. such as ” Energy accumulates (Joules pile up) until such a time as the temperature at the (new) radiating level is warm enough to enable planetary radiative cooling at the same rate as solar radiative heating. Classically, solar absorbed radiation is held fixed during this process.”
Its all besides the point – and ignores the questions put by MS. answers for which are in the ipcc refs.
In the meantime the oceans (and land) have always been absorbing SW solar radiation directly one way shape or form. That process has not changed. That radiative energy is still being reflected as LWR to the atmosphere today.
Please choose your partners for the next dance, ladies and gentlemen.
Kobayashi Marusays
MS well done. You keep asking the right questions. May I add;
The Greenhouse Effect includes both, but they happen in sequence:
Collides → transfers energy to kinetic motion (warming).
Re-emits a new photon (often downward) → that photon is then absorbed by another molecule elsewhere.
The full mechanistic explanation:
CO₂ absorbs outgoing infrared radiation from Earth’s surface, re-radiating it in all directions—including back downward. This delays energy’s escape to space, forcing the lower atmosphere to hold more energy. That extra energy manifests as increased molecular motion (thermal energy/warming) until the system rebalances at a higher temperature.
So:
GHE = absorption + re-emission (radiative trapping).
Warming = the thermal consequence (collisional heating).
You weren’t confused. You had it right. The AI just confirmed what you already knew.
Barry E Finchsays
The “Kobayashi Maru” directly contradicted itself with its entire GHE “explanation” at 7 Jul 2026 at 12:06 AM (incorrect) versus its correct statement 5 Jul 2026 at 7:24 PM “An increase in the concentration of greenhouse gases leads to an increased infrared opacity of the atmosphere, and therefore to an effective radiation into space from a higher altitude at a lower temperature. This causes a radiative forcing, an imbalance…..”. The correct one mentions “higher altitude at a lower temperature” implying a change in the MANUFACTURING of photons by the GHGs that can reach Outer Space (correct) and the one I’m responding to has the same banal rubbish as the “Joke Zonderkop”. Too much flailing about by “Kobayashi Maru”.
Barry E Finchsays
A simplified explanation of the so-called “greenhouse effect (GHE)” in Earth’s troposphere. Some Power flux of the vast quantity of photons that are emitted (manufactured) by molecules in Earth’s troposphere leak out of its top & bottom into or through the stratosphere and into the surface respectively. Emission (manufacturing) is proportional to Kelvin**4.
Suppose for concept illustration only (not highly accurate) that 194 w/m**2 leaks out the top and that is part of what is emitted (manufactured) by molecules over an altitude range of 4 to 12 km with an effective average of 7 km. Say for concept illustration that 345 w/m**2 leaks out the bottom and that is part of what is emitted (manufactured) by molecules over an altitude range of 0.0001 to 1.5 km with an effective average of 1 km.
Suppose a certain amount of CO2 was added into the troposphere and mixed then *instantly* the upper & lower ranges would need to be closer to their respective ends because there are more CO2 molecules in the way. So the effective averages of 7 km for upwelling into or through stratosphere and 1 km for downwelling into surface change to 7.05 km and 980m respectively, with the upwelling & downwelling Power fluxes consequently changing to 193 w/m**2 and 346 w/m**2 respectively due to being produced by higher-than-before (colder) and lower-than-before (warmer) air parcels.
There has been no change in the 539 w/m**2 leaving the troposphere, there is no “magical extra energy”, the surface downwelling radiation has increased by 1 w/m**2, which is an additional ~480 terawatts (~95%) heating the ocean and Earth is emitting less radiation to Outer Space (a Power reduction of 510 terawatts).
The reasons for the troposphere being colder when higher than before and warmer when lower than before (almost always except Antarctica in Winter, and the Winter Arctic Ocean, quite a lot) are that (1) 68% of absorbed Solar Power (SWR) is absorbed within a few metres of the surface, and only 23% of SWR is absorbed throughout the troposphere, and (2) the 194 w/m**2 that is departing upward from the top, and (3) the pressure reduction with increasing altitude causes “adiabatic cooling” or “adiabatic heating” for increasing or decreasing altitude respectively. Convection and water latent heat from the ocean continuously try to make the GHE, correctly described above, be non existent by making the tropospheric temperature above any place on Earth be the same all the way from the bottom to the top, but they continuously fail to achieve that, to the average failure of 68 degrees colder at the top, for the 3 reasons just detailed.
Barry E Finchsays
My “closer to their respective ends because there are more CO2 molecules in the way” might be too pithy. Maybe somebody could suggest wider wording about the peer-to-peer radiative exchanges being a net upward flow for the top & bottom so the distance twixt average emission (manufacture) decreasing, by then decreasing the Kelvin**4 difference for the peer-to-peer radiative exchanges thus decreases the upward Net flux but increases the downward Net flux.
Is there actually any literature out there trying to quantify the given climate effect of water vapor?
I know, I know, people would jump in and point to KT97 or Schmidt et al 2010. In KT97 WV contributes 49W/m2 (net) and 75W/m2 (gross) to the GHE, while in S10 it is 61W/m2 (net) and 96W/m2 (gross). Also you might get a more conflated story including clouds, weather and what not else.
But I would like to stick to the actual question and keep it simple and to the point. WV also has a cooling side, which is well known in the context of neg. “lapse rate feedback”.
So what I am looking for is the consolidated impact of both sides, within the given climate. It seems to me, although I might be wrong, this very important question has never really been considered.
I am more than willing to accept the above figures for the warming part, but if we compare that to some 86.4W/m2 of “latent heat” cooling (as in the NASA Earth Energy Budget), then WV was barely warming at all, or rather net cooling.
Barry E Finchsays
“consolidated impact of both sides, within the given climate”. I have this from a scientist a few years ago, probably in an internal Webinar but maybe just in a talk. I recall that he was a climate scientist and said “high confidence” but that’s all I recall.
100% GHG increase or any cause(s) of warming
220% H2O gas GHG increase caused by warming
-110% Tropospheric lapse rate reduction due to H2O gas increase, latent heat
Piotrsays
Barry, what did you wanted to convey ? What your three numbers are supposed to mean?
And why that highly-confident self-declared “climate scientist” have listed the FORCING (GHG increase) next to …. the two passive FEEDBACKS, as if their role in climate change was comparable???
The only reason I can think of – is an old denier narrative that human changes to water cycle are as much, or preferably – much more important than mitigation of GHGs emissions.
That’s behind the several -decades old claim of water vapour is responsible for 98% of GW .
This “it’s not GHGs, it’s water cycle!” denialism started in the last century (e.g. the mythical number of water vapour being responsible for “98%” of GW) and still live and kicking today – even here – see the July 6 effort by E. Schaffer above, or the many years of posts by Tomas Kalisz and by JCM – the latter going even further – taking the UN report on “ up to 40% of Earth land being degraded” and blaming it on …. the climate modellers and their “ artificial fixation and overemphasis [of the role ] of “trace gas” [CO2].
Well, that is a nice misrepresentation. Strangely there is one specific pattern, and I have no clue what that is, but it repeats over and over. Wherever I sense that the consensus position is wrong, the critical position is simply wronger, erring though in the same direction.
In the case of the WV share in the GHE I would say it is vastly overstated. The reason of course being the surface emissivity = 1 assumption. This overstates surface emissions, the GHE, and the role of WV within the GHE. So while Schmidt et al 2010 names 61/96 W/m2 for WV and KT97 49/75, I would estimate it to be only 25/50W/m2.
Anyway, there are good reasons for that and it is exactly the opposite of blaming some 98% of the GHE on WV.
Could you provide reference(s) to the article(s) wherein you present your results and compare them with previous studies on the topic like Schmidt et al 2010, so that the interested Real Climate readers can follow your line of evidence?
Thank you in advance and greetings
Tomáš
Barton Paul Levensonsays
ES: In the case of the WV share in the GHE I would say it is vastly overstated. The reason of course being the surface emissivity = 1 assumption. This overstates surface emissions, the GHE, and the role of WV within the GHE. So while Schmidt et al 2010 names 61/96 W/m2 for WV and KT97 49/75, I would estimate it to be only 25/50W/m2.
BPL: Show your work.
Piotrsays
Tomas to E. Schafer: “ Sir, Could you provide reference(s) to the article(s) wherein you present your results and compare them with previous studies on the topic like Schmidt et al 2010,”
Nice touch, Tomas – asking E.Schafer, point-blank, to show his published proof for his insinuation that Gavin Schmidt and co-authors paper are wrong and/or intellectually dishonest (E Schafer: “overstate surface emissions, the GHE, and the role of WV within the GHE.“).
Unless, of course, you have done it unwittingly – if you don’t challenge the proofs existence but hope it exist – as a fellow “anything but CO2” denier, hoping you might use E. Schafer’s peer-reviewed paper to advocate the diverting of the research and policies AWAY from the mitigation of GHGs, into studying water vapour from which is a scientific dead-end:
– it does not allow any societal/policy recommendations (since we can’t change WV enough to achieve any meaningful difference to AGW – see the fate of your own Sahara irrigation proposal)
– it won’t improve our climate projections – the effect of WV is already implicitely included in the climate models – hence no NEW insight from running the models and exclaiming: “ A ha! On a fictional Earth with WV arbitrarily forced to be at different levels – the GMST would have been X degrees different! Take that, Gavin Schmidt!”
These are a number of considerations adding up, I discuss them on my site. It is was happens if you replace simplified assumptions with more precise and accurate facts. The emissivity of water is perfect example for that, directly affecting the significance of WV as GHG.
The emissivity of water is 0.91, as I have calculated 5 years ago (based on Hale, Query 1973), consistent with Huang et al 2016, or Baehr, Stefan “Stoff- und Wärmeübertragung”. This correct result is in stark contrast to some bad science, like Wilber et al 1999 (NASA) referenced in Trenberth et al 2008, falsely claiming it was 0.9907!?
It follows the surface of Earth will emit ~360W/m2. With OLR = 240W/m2 the GHE will amount to about 120W/m2 only. I know there is this notion that surface emissivity would barely matter in this regard, because the less radiation the surface emits, the more “back radiation” it can reflect, which is true. Inamndar, Ramanathan 1998 is a good example for that. First they wrongly state “sea surface is within 1% of emitting like a blackbody”, then they include surface reflected LW to their GHE definition (in opposition to their previous work), and then they show a chart (Plate 1) where the GHE is minimal over deserts where they assume very low surface emissivities, thereby again contradicting themselves. There is a lot of confusion. I would say it is actually simple. If there is no GH-agent present and the atmosphere is transparent, there is no “back radiation” either to be reflected by the surface. So yes, it is indeed just surface emissions – OLR = GHE.
This necessary downsizing of the GHE will not affect every GH agent equally. There is zero difference regarding clouds, CO2 and the minor GHGs (O3, CH4, N2O..) get some discount, but the bulk of the difference has to be with the WV continuum. The assumption of course is, that all the radiation emitted by the surface and not arriving as OLR, must have been absorbed and substituted by some GH-agent, which is usually WV. A large part of WV-GHE however is just radiation never emitted by the surface in the first place. Adjusting for this issue, WV does some 25/50 W/m2, maybe 30/55 W/m2.
But then we yet have that other side of WV – shrinking the lapse rate and the GHE with it. Then adding up the numbers, I can not help but to conclude WV must be strongly cooling the Earth..
Piotrsays
Tomas Kalisz: “Sir, Could you provide reference(s) to the article(s) wherein you present your results”
E, Schafer: “These are a number of considerations adding up, I discuss them on my site. ”
Good one, E. Schafer! See also:
– Sir, Could you provide reference(s) to the article(s) wherein you present your results that the Earth is flat.
FlatEarther: These are a number of considerations adding up, I discuss them on my site.
Barton Paul Levensonsays
ES: The emissivity of water is 0.91, as I have calculated 5 years ago (based on Hale, Query 1973), consistent with Huang et al 2016, or Baehr, Stefan “Stoff- und Wärmeübertragung”. This correct result is in stark contrast to some bad science, like Wilber et al 1999 (NASA) referenced in Trenberth et al 2008, falsely claiming it was 0.9907!?
BPL: Different sources give different figures. Most are very high, including the one you mention. I tend to rely on Konda et al. (1994) which found e = 0.984 for seawater.
You’ve been pushing this “the emissivity of water is 0.91, therefore energy budgets for the climate system are all wrong” for years now. Many of us were here the last several times you did it, such as here, four years ago:
This tactic of bringing up a debating point, having it knocked down, then lurking until you feel it’s safe to bring it up again, never acknowledging that it’s been asked and answered, is a typical denier tactic and is obnoxious and exhausting. Perhaps that’s the goal.
Well this is not a guessing game, rather it is about the depth of understanding. Neither Wilber et al 1999, nor Konda et al 1994 provide the proper data
W99 failed in many spectacular ways:
– they did not realize water is non-lambertian radiator, so they are unaware of the difference between emissivity to surface normal (which they discuss) and hemispheric emissivity
– they had no data past 15µm and claimed such data would not exist
– actually they just failed to check the literature, as Hale, Query 1973 provided high resolution data up to 200µm
– instead they “interpolated” the far-IR figure from the 15µm result..
– eventually they tried to calculate an average over the whole thing, and failed even with that
K94 only has a very limited scope..
“Its viewing solid angle was 1°, and the sensitive wave length range was between 8 µm and 14 µm.”
So it is just emissivity to surface normal for the segment 8 to 14µm. Emissivity to surface normal over the whole spectrum is 0.965 btw., and hemispheric spectral emissivity 0.91, as before.
Piotrsays
Re: E. Schaffer:
Wikipedia: Sealioning – is a type of trolling that consists of pursuing people with relentless requests for evidence, often tangential
here: the water cycle is a passive feedback, so we can’t change it, and its quantification provides no new insight – since it is already included in climate models. or previously addressed
BPL to ESchaffer: You’ve been pushing this for years now. Many of us were here the last several times you did it, This tactic of bringing up a debating point, having it knocked down, then lurking until you feel it’s safe to bring it up again, never acknowledging that it’s been asked and answered, It may take the form of “incessant, bad-faith invitations to engage in debate”, and has been likened to a denial-of-service attack targeted at human beings.”
Thx, this is a great summary of known considerations. However, it absolutely does not address the simple question I have put up. Although it states..
“At the same time, in latitudes spanning the tropics through to mid-latitudes, the upper troposphere warms faster than the surface – a change in the vertical “lapse rate” with temperature, enabling the Earth to radiate to space more effectively.”
.. there is not the slightest reflection on the netting of those opposing effects within the given climate system. It is like that CL game PSG vs. Bayern, where you might ask who won and get the answer Bayern scored 4 goals. Now that might suggest Bayern won, but actually does not answer the question. There is still a chance they might have received 5 goals and lost the game, which is exactly what happened btw. Equally so just saying WV was a strong GHG having the largest share in the GHE, avoids the question if it is warming at all. Especially given we know it also has a huge cooling side.
Equally jumping right over to WV feedback avoids the question. At the start one should clarify if WV is warming or cooling, and by how much. I feel like the question is avoided in order to maintain the narrative of a pos. WV feedback. If WV is cooling, and that is what the numbers suggest, it makes little sense more WV would provide (strong) warming..
Barry E Finchsays
H2O gas isn’t radiating (manufacturing photons by its collisions) from the top of the troposphere and therefore an increase must cause additional imbalance between its radiation leakage (a very small fraction of its photon manufacture) out of the top versus out of the bottom of the troposphere, the so-called “greenhouse effect (GHE)” in Earth’s troposphere.
MA Rodgersays
E. Schaffer,
I find it strange that you would consider that reference to Colman & Soden (2021)“absolutely does not address the simple question [you] have put up.”
You quote from the ‘I. Introduction’ of Colman & Soden (2021) and complain that your question is not answered, it even failing to provide an understanding of the sign of the combined WV/Lapse-Rate feedback. Yet the sign is well known – the combined feedback is positive. Let me quote from the ‘Abstract’ of Colman & Soden (2021).
“Water vapour is a greenhouse gas that dominates the Earth’s terrestrial radiation absorption. As the planetary temperature warms, forced by increasing CO2 and other greenhouse gases, water vapour content of the atmosphere increases, thereby producing the strongest positive feedback in the climate system. At the same time, the rate at which atmospheric temperature drops with height (the “lapse rate”) is expected to decrease with warming. This represents a smaller, but significant, negative feedback, since it enables the planet to radiate more effectively to space. The two feedbacks are closely coupled to each other, and the “combined” result represents the foundational net positive feedback in the climate system, mandating substantial global warming in response to increased greenhouse gases.” {My bold]
Colman & Soden (2021) provide a value for the size of this positive feedback in ‘VIII. Conclusions A. On the strength and consistency of evidence for water vapour and lapse rate feedbacks.’ saying:-
“The evidence is now overwhelming that combined water vapour+lapse rate feedbacks provide the strongest positive feedback in the climate system, of a magnitude around that produced in climate models. Our estimate of overall strength of these combined feedbacks is 1.25 ± 0.15(1sd) Wm-2K-1”
Their Tables 1,2 & A2-1 list various studies with values for the various components of the combined WV/LR feedback. And note that your actual question posed upthread was “Is there actually any literature out there trying to quantify the given climate effect of water vapor?” for which I would consider a review paper such as Colman & Soden (2021) would demonstrate properly that the answer to your question was an emphatic “Yes!!!”
(I should add that the latent heat energy flux of “some 86.4W/m2” you mention upthread is not a direct measure of any feedback. Any change in that value would have a bearing on the strength of the WV feedback but again is not a direct measure of the feedback itself.)
I don’t know why that is, but you are not reading what I am writing. Probably you think it is irrelevant, just glance over it and answer to your own assumption. I am STILL NOT asking about WV feedback, but about the given effect of WV.
We have the WV effect and the enhanced WV effect, which we call WV feedback. That is a base magnitude and the enhancement of said base magnitude. As with AR6 that enhancement is assumed to be +1.8 and -0.5W/m2 and a total of 1.3W/m2. My question is on the base magnitude. How much warming, how much cooling, and the net effect of WV. Again, not the feedback, just the base magnitude.
Clearly this is an important perspective and it seems it has been completely ignored, which makes it only more pressing. Of course the notion is, if it ever had been considered, bad things would have been avoided, but anyway..
And of course the 86.4W/m2 in latent heat are also not a feedback, but a given cooling effect by WV. For instance, as above, we could say WV warms with 49W/m2 (KT97 net figure) and cools with 86.4W/m2, and overall is strongly net cooling. I repeat: WV then is a strong cooling agent, which is a ground breaking perspective. And in science we must not ignore perspectives just because we don’t like them. From that the question would arise: given WV cools and is positively correlated to temperature, how could it be a pos. feedback?
But that is downstream, first really there is the urgent question on whether WV is warming or cooling, and I can not fathom it has never been considered.
Piotrsays
E. Schaffer “ whether WV is warming or cooling, and I can not fathom it has never been considered.”
Because, as it has been explained already dozens of times to your fellow “anything but GHGs” deniers (Tomas Kalisz and JCM) – WV and other aspects of the water cycle are merely a passive feedback, not one of the drivers of current AGW. Thus:
1. There is nothing we can do about WV – given the magnitude of the natural water and ridiculously short residence time of WV in atm – we simply can’t change water fluxes enough to reduce AGW in any noticeable amount.
2. Since WV is a strong net positive feedback – the AGW is more sensitive to the GHGs, because WV amplifies the consequences of our action, or inaction, on GHGs:
if we reduce GHGs to cool the Earth – WV will make the cooling much larger, if we do not cut our GHG emissions – WV will make the warming much larger,
Which directly counters the deniers narrative that because there is so much more molecules of WV than GHGs – what we do to GHGs doesn’t really matter, so we can burn as much fossil fuels as the fossil-fuel lobby wants us to.
3. For the main societal responsibility of climate science – informing the mitigation of AGW – there is nothing to be gained from “your direct measure of the WV feedback itself” – since again – we simply can’t change water fluxes enough to make any noticeable dent in AGW.
4. Your WV is already implicitely included in climate models – ergo your “direct measurements of the feedback itself” would provide no new insight beyond what we already know from Schmidt et l. 2010 and other papers.
In other words, since your WV feedback is already included in climate models – what difference would it make if out of, say, the projected 2C of warming, 1.3 C was the result of the WV passive feedback instead of 0.9 C? It does not affect our models (since these 1.3C or 0.9C re the products of these models ). Nor can we do anything about it.
E. Schaffer,
My apologies for repeatedly misinterpreting you questioning.
To be clear, and so I don’t make a fool of myself again, can you set out precisely what you are hoping to quantify.
You say you want a full account of the impact of atmospheric WV on Earth’s GHE: that is the reduced climate ‘forcing’ if atmospheric WV was reduced to zero and in this regard you specifically mention the inclusion of the lapse-rate feedback in this assessment.
However if the WV is considered as being removed from the atmosphere, the presence of cloud seems a little difficult to imagine without any atmospheric WV. Also the planet will be considerably colder without WV and the effects of surface albedo will presumably increase due to expanded sea ice. In a normal situation a colder planet would also result in ice caps and snow-cover, further increasing surface albedo but these would have been deposited by precipitation so may not be present in your hoped-for assessment. The same could be said for land albedo due to plant-life which does rather rely on water from above, directly or via rivers.
Could you rattle through these other factors to indicate which, like lapse rate feedback, you wish to see incorporated into this assessment you seek.
My original question was on whether this issue has ever been considered. I do know a lot of the literature but certainly not all, and according to my knowledge it was not. Then I asked AI and it too could not name me a single instance either. So that is why I brought up the question here and it seems like everyone is confused just because I am asking.
As to what you say let me clarify: we have attribution and science fiction. In science fiction we say “what if” and then play through different scenarios, consider dynamics and so on. In attribution we assume all other things stay the same and consider the “what if” only regarding the immediate effects. In science, to keep things clean, we only do the latter.
The immediate effect of WV is raising the emission altitude and reducing the lapse rate. That is why WV feedback consists of two components, which are raising the emission altitude (assumed to be 1.8W/m2) and reducing the lapse rate (assumed to be -0.5W/m2). Considering these two effects alone is more than enough, there is no need to complicate things any further.
The very same question applies to the given amount of WV. It raises the emission altitude and it has reduced the lapse rate. We have quantifications for both effects, which are totally negotiable and discussable. But the figures we have indicate WV is a cooling agent, and the only reason this has never been understood seems to be that the very question was never asked. And that is simply embarrassing..
MA Rodgersays
E. Schaffer,
You are saying that you want to quantify the net contribution of WV within the pre-industrial atmosphere’s GHE. This contribution is therefore a component of the GHE which comprises the forcings/feedbacks – GHG+aerosols, WV, cloud & albedo that raised surface temperatures some 33K.
And your question was specifically whether there is literature which provides analysis quantifying this contribution.
Yet buried in your enquiry is a very odd separation of the mechanisms of the pre-industrial GHE and the mechanisms of AGW in the industrial age.
You are seemingly happy that the literature shows the impact of WV under AGW comprises the effects of (1) Increased emissions height (+1.8Wm^-2/K) and (2) Increased lapse rate (-0.5Wm^-2/K). This is a net positive feedback of +1.3Wm^-2/K and would boost any warming from other net forcings+feedbacks by about 50%. So far, smiles all round.
We could then take a back-of-envelope approach here and say that the pre-industrial GHE resulted in something like +33K (this requiring surface albedo assumptions) so the contribution of WV would be a net positive feedback of [+1.3 x 33 = ] +43Wm^-2.
Of course it is a big assumption using an unaltered Wm^-2/K of feedback all they way down to zero WV but it does give a ballpark idea of the size of the feedback. The two references you made upthread also give a value for the WV feedback contribution to the pre-industrial climate..Kiehl & Trenberth (1997) ‘Earth’s Annual Global Mean Energy Budget’ give +59Wm^-2 (75Wm^-2 clear sky, 51Wm^-2 cloudy sky, 67.5% cloud fraction from CERES) while Schmidt et al (2010) ‘Attribution of the present-day total greenhouse effect’ give +49 Wm^-2 (All Sky 50% of ‘total’ using K&T(1997)’s 98Wm^-2 ‘total’).
So this is all pretty-much in the same ballpark.
But instead you seem to be of the opinion that the WV value of these two papers doesn’t include the Lapse Rate feedback. And if the +1.8Wm^-2/K from above was applied to the full 33K pre-industrial GHE it would amount to some +60Wm^-2, still in the same ballpark.
But you also consider the -86Wm^-2 surface latent heat flux to be a measure of the full pre-industrial Lapse Rate feedback. This is certainly not the case. The -86Wm^-2 surface flux is converted to sensible heat when the WV condenses up in the atmosphere. It is thus decreasing the Lapse Rate but it is not a measure of the Lapse Rate feedback.
I think I will need to examine your website to find out why you have made such an egregious error. I note from a passing visit that you weren’t not shy about this matter posting about it at the back-end of last year with OPs titled ‘How “Climate Science” has its eyes wide shut on the statistical blunder that generates false positive feedbacks’ and ‘How to Flip the Sign on Feedbacks’
First of all it is great to finally get to my actual question, after 40+ posts here.. ;)
“But you also consider the -86Wm^-2 surface latent heat flux to be a measure of the full pre-industrial Lapse Rate feedback. This is certainly not the case. The -86Wm^-2 surface flux is converted to sensible heat when the WV condenses up in the atmosphere. It is thus decreasing the Lapse Rate but it is not a measure of the Lapse Rate feedback.”
Let me ask: there are a number our sources pointing out the GHE is dependent on the lapse rate, and furthermore proportionate to the lapse rate. If the lapse rate was 0, the GHE would be 0, as surface- and emission temperature would be equal. Equally if the lapse rate was 50% larger, the GHE would be 50% larger. Let us say the average emission altitude was 5km up, then with 6.5K/km the GHE = 6.5 x 5 = 32.5K. With a lapse rate of 9.8K/km it would be 49K. Do we agree so far?
Due to the SB-law we can not directly relate delta temperature into delta radiation, but that is a side show. Yet, by and large, 50% of some 155W/m2 gets us to around 80W/m2. Either way, if it was not for this reduction of the lapse rate, Earth would be a lot hotter. And of course it is physically exactly the same as the lapse rate feedback featured all over the models.
So how exactly do you reject this connection?
MA Rodgersays
E. Schaffer,
Having “g(o)t to {your) actual question, after 40+ posts here” (and I’m not sure having apparently “got to” it, whether you have ‘got over’ it), you then toss in another question!!!!
And what a question! It seems you wish to upgrade the atmospheric Lapse Rate into the primary feature of a GHE, rather than it being the GHGs in such an atmosphere.
And why not? Hey, does not a telescope have two ends?
A Lapse Rate forms in an atmosphere because atmospheric pressure decreases with altitude and gases will generally rise/fall adiabatically within such an atmosphere with meridianal temperature gradients.
In Earth’s atmosphere this zero-GHG base-line Lapse Rate is usually given as 9.8°C/km with the mention of it being “dry air”. The difference for ‘wet air’ that isn’t condensing out precipitation is pretty similar as the decrease in density is balanced by the increase in specific heat capacity.
But adding in a bit of ‘thermodynamicity’ will alter the Lapse Rate due to GHG energy fluxes which warm/cool said atmosphere.
Thus with altitude, GHGs will be emitting/absorbing less IR with the lower temperatures at altitude, this adding warmth with altitude and reducing the Lapse Rate, or shooting IR out into space and cooling the atmosphere, thus increasing the Lapse Rate. Or absorbing incoming UV which in the thin upper atmosphere can warm enough to reverse the Lapse Rate. And precipitation will transport insensible heat up to the point of condensation where it will again provide warming, reducing the Lapse Rate.
The GHE is not just a GHG-thing. It requires a Lapse Rate and that requires atmospheric pressure to decrease with altitude. Mars is a good example of what you get without atmospheric pressure. It has the same burden of CO2 as Earth but this is 95% of its atmosphere rather than Earth’s 0.05%. The result is zero GHE on Mars.
So physics tells us the Earth’s tropospheric adiabatic Lapse Rate sits at about 9.8°C/km before the GHGs start their influence. And what if the total Lapse Rate feedback (‘total’ = today relative to zero-GHG) happened to be 80Wm^-2 (thus equal to the “evapero-transpiration” flux in Fig 7 of Kiehl and Trenberth (1997))? If it were 80Wm^-2, that would be half of the 155Wm^-2 total GHE “Net LW Absorbed” set out in Schmidt et al (2010). And would that then imply the modern Lapse Rate is half the 9.8°C/km? After all, Wiki-thing tells us
the actual average value is considered to be 6.5°C/km up to the tropopause, so not massively different to 4.9°C/km.
(One point to clarify here is that Schmidt et al (2010) do not adjust the atmospheric temperature profile, saying their modelling is “holding the climate (spatial and temporal distributions of temperature, surface properties, etc.) fixed.” and is thus using Lapse Rate data that would presumably reflect that 6.5°C/km average. The WV feedback with no LR feedback would this be some (60Wm^-2 / 33K =) +1.8Wm^-2/K which is the value for the modern climate we see in the literature.)
I think it would be wise to consider all the energy fluxes into the atmosphere. They surely would all be messing with the lapse rate. ☻ There’s the “evapero-transpiration” flux of 78Wm^-2 which would be operating at the points of condensation and which may be mainly impacting the Lapse Rate up to the altitude of WV IR emissions to space. Would the Lapse Rate be reduced above that point? ☻ There’s “thermals”, a flux of 24WM^-2 (with big ‘error bars’) which presumably would be reduced if the Lapse Rate declined. ☻ There’s a net flux of 40Wm^-2 in “surface/back radiation”. ☻ And into the atmosphere from above is an “absorbed by the atmosphere” flux of 67Wm^-2. ☻ Finally there is a cooling flux of 195Wm^-2 “emitted by the atmosphere” which will presumably be increasing the Lapse Rate.
So it looks like the “evapero-transpiration” flux is not the only flux messing with the Lapse Rate. All these energy fluxes look like being of significance to the Lapse Rate.
And together they will have modified the Lapse Rate by altitude and as well as zonally. So I’d suggest that assuming the Lapse Rate is a constant average 6.5°C/km throughout the troposphere doesn’t look like providing a realistic or useful model.
I also don’t see any support for your contention that climatology is ignoring the Lapse Rate feedback or that the combined WV/LR feedback is negative.
We could look at evidence of an increase in global precipitation as a measure of the change in that 78Wm^-2 “evapero-transpiration” flux under AGW. Use satellite data 1979-2020 from GPCP, Gu & Alder (2022) find precipitation rising at 1.4%/deg C. That would be about half the rate of the 78Wm^-2 over 33K.
And perhaps it would not be torturing the available data too much to use the -0.5Wm^-2/K calculated for today’s LR feedback and then double it over the 33K of pre-industrial GHE yielding 33Wm^-2 for the ‘total’ LR feedback and thus (+60 – 33 =) +27Wm^-2 for the combined WV/LR feedback. Note it is positive. Of course, this the opposite of the contention set out in a handful of OPs on you website which look like they could do with some serious debunking.
Oh that gave me a lot of headache. With every phrase I struggle to figure out what you mean.
What for instance is “GHG energy”?
Why do you think there was no GHE on Mars? CO2 does leave a dent in its OLR profile, so there has to be one. Given the planet is not properly covered by weather stations, given it has large temperature differences both by day and season, it is not so easy to tell what average temperature it has, or should have respectively. But that is just a practical issue.
“If it were 80Wm^-2, that would be half of the 155Wm^-2 total GHE “Net LW Absorbed” set out in Schmidt et al (2010)”
Yeah, kind of half. But that would come on top of the 155. So with a larger lapse rate, the GHE would be larger, like 155 * 1.5 = 232.5!
“the actual average value is considered to be 6.5°C/km up to the tropopause, so not massively different to 4.9°C/km”
Why 4.9K/km? The 80W/m2 of latent heat would not halve the GHE, but enhance it by 50% in their absence. The cooling is already baked in. But just because it is will not mean it never happened. This cooling side of WV is there, it is just not reflected on.
Logically it also follows that there is no question if S10 (or KT97) does consider it, or not. They are stating their figures based on the status quo, with the cooling effect by WV already included – but not stated!!! And yes, if it was not for that, the lapse rate was larger, the GHE would be larger AND the contribution WV to it would equally be larger. So in the instance of S10 it would not just be like 61/96 W/m2, but maybe 92/144 W/m2. And with KT97 not just 49/75 but 74/112 W/m2. But even then this would barely be warming vs. the massive 80W/m2 plus latent heat cooling.. On top of that, as I have pointed out, in both instances WV will be badly overstated due to the surface emissivity issue.
As with my previous post above one can easily figure this out considering emission altitude vs. lapse rate. 6.5 x 5 = 32.5K, 9.8 x 5 = 49K..
Barton Paul Levensonsays
ES: Why do you think there was no GHE on Mars? CO2 does leave a dent in its OLR profile, so there has to be one. Given the planet is not properly covered by weather stations, given it has large temperature differences both by day and season, it is not so easy to tell what average temperature it has, or should have respectively.
BPL: The Mars Standard Atmosphere (Barth 1985) gives the emission temperature of Mars as 214 K, and Catling and Kasting (2017) give 215 K. So since Mars’s radiative equilibrium temperature is approximately 210 K, there is 4-5 K worth of greenhouse effect on Mars.
Mars Orbiter and the landers gave some good climate data. There are also GCMs for Mars.
Ref:
Barth, C.A. 1985. The photochemistry of the atmosphere of Mars. The Photochemistry of Atmospheres, Levine, J.S., Ed. Harcourt Brace Jovanovich, Publishers, pp. 337-392.
Catling, D.C., Kasting, J.F. 2017. Atmospheric Evolution on Inhabited and Lifeless Worlds. Cambridge, UK: Cambridge Univ. Press.
MA Rodgersays
E. Schaffer, “What for instance is “GHG energy”?”
Well, if you read what was written (“GHG energy fluxes which warm/cool said atmosphere.”) you’d find that the ‘energy’ in question concerns fluxes-various that thus flow in/out GHG within the atmosphere (because the are IR-active which is the property that GHG have) and which then warm/cool the atmosphere they sit in, warming (or cooling) just like the WV/precipitation cycle does.
“Why do you think there was no GHE on Mars?”
Do I think that? I raise the example of the Martian atmosphere because you are the one trying to set out lapse rate as the major consideration for a GHE. On Mars the GHE operates with no significant lapse rate so there would be “zero GHE on Mars” resulting from added GHGs.
The 80Wm^-2 conjecture and “why 4.9K/km?” The 80Wm^-2 strength of the Lapse Rate feedback is your conjecture. Whether you want it applied to arrive at the 155Wm^-2 “Net LW Absorbed” GHE effect of Schmidt et al (2010) or applied to modify that 155Wm^-2 is probably your decision.
However, if this trivia is your takeaway from my comment, I suggest an aspirin for your headache and read it again.
Or maybe I should be helpful and suggest that your attempts to infer something useful from the strength of the existing GHE relative to some sci-fi zero-GHG atmosphere won’t be relevant to AGW which results from adding additional GHGs into the existing GHE. AGW modelling of the Lapse Rate feedback is usually reported combined with the WV feedback and invariably shows that the combined WV/LR feedback is strongly positive.
IPCC AR6 Section 7.4.2.2 concludes thus:-
“The combined ‘water-vapour plus lapse-rate’ feedback is positive. The main physical processes that drive this feedback are well understood and supported by multiple lines of evidence including models, theory and observations. The combined ‘water-vapour plus lapse-rate’ feedback parameter is assessed to be α LR+WV= 1.30 W m–2°C–1, with a very likely range of 1.1 to 1.5 W m–2°C–1, and a likely range of 1.2 to 1.4 W m–2°C–1 with high confidence.” [My bold]
If somebody wanted to overturn such science, IPCC AR6 7.4.2.2 is where to start.
You may disagree and may wish to somehow demonstrate your position with a half-baked comparison of the GHE with some sci-fi zero-GHG situation. You evidently feel strongly about this position of yours having gone so far as to assert (on your website) that “the established “consensus science”… can not win the argument”and that “WV can not be a positive feedback, let a lone a massive one, a detail the global warming narrative totally depends on.”
However, given you hold this view so strongly it’s very probable that any words of helpful advice here will be wasted.
I checked the difference in greenhouse effects on Earth and Mars with Perplexity Pro and in this respect, I have a few questions regarding the following sentences in your comment:
“The GHE is not just a GHG-thing. It requires a Lapse Rate and that requires atmospheric pressure to decrease with altitude. Mars is a good example of what you get without atmospheric pressure. It has the same burden of CO2 as Earth but this is 95% of its atmosphere rather than Earth’s 0.05%. The result is zero GHE on Mars.
So physics tells us the Earth’s tropospheric adiabatic Lapse Rate sits at about 9.8°C/km before the GHGs start their influence.”
1) It indeed appears that although CO2 pressure on Mars is higher than on Earth (ca 570 Pa vs ca 40 Pa), the difference between the global mean surface temperature (GMST) and emission temperature calculated from albedo and insolation (which is for Earth 255 K, for Mars 200-223 K) is for Earth about 33 K, while for Mars between -9 and +5 K.
The engine, however, suggests that part of this difference should be accounted to the difference in insolation and albedo. It asserts that if Mars had the same emission temperature as Earth, its CO2 atmosphere would be warmer and “higher” or “thicker”, what would allegedly result in a change in the lapse rate and in a stronger greenhouse effect. Could you check?
2) It further asserts that in this case, and if the surface of Mars were somehow “irrigated”, water vapour would further enhance the greenhouse effect on Mars and made its magnitude even closer to that on Earth.
3) Another part of the difference between Mars and Earth should be, according to Perplexity, ascribed to collisional broadening of spectral bands in GHG molecules due to higher atmospheric pressure on Earth.
4) The remaining part of the difference between Mars and Earth should be, according to Perplexity, ascribed to water availability for evaporation from the surface on Earth and temperature and pressure range enabling latent heat flux in the atmosphere, what further changes convection and the lapse rate.
Is Perplexity correct?
Greetings
Tomáš
MA Rodgersays
Tomáš Kalisz,
Regarding the Martian-&-Earth GHEs, I don’t know of any actual apples-to-apples comparison of atmospheres from Mars and Earth’s. Indeed, how would you account for the freezing/evaporating of the polar Martian CO2 ice? The nearest to an Earth/Mars comparison that springs to mind is Figs 1 & 3 in Read et al (2015) ‘Global energy budgets and ‘Trenberth diagrams’ for the climates of terrestrial and gas giant planets’ although Fig 5 shows what a dust storm does to the situation on Mars.
I’d assume the IR emission height would be very roughly similar but it is the pressure-drop that gives you a Lapse Rate and there’s naff-all pressure in the Martian atmosphere so naff-all drop is possible.
The pressure broadening, which fuzzes out the absorption/emission lines may well be a feature in the low-pressure Martian atmosphere and that would increase emission heights significantly but still there can be meaty pressure drop.
**(a reminder that the cumulative forcing from many changes can/will different if the climate is allowed to equilibrate before each successive next forcing, and therefore the cumulative feedback can/must also differ.)
And, (even without hysteresis) the partition between forcing and feedback can be shifted between a change and the reverse change. Consider the complete removal of Earth’s GHE (but somehow maintaining solar heating exactly as it; this is a model experiment, after all). The TOA forcing would be a large increase in OLR, tending to cool the climate system overall. Now let’s add back the GHE. Well, to start, let’s add a little back. In the equilibration to 0 GHE, the troposphere would largely disappear – It’s hard to see how it could be maintained in any significant way with significant depth without an ability to emit LW radiation to balance an upward convective heat flux…
— —
So given that there is solar heating within the air, in a 1-dimensional globally representative model (I’m going to stick with that here to keep things simple), no GHE means all OLR must come from the surface, and so the surface equilibrium T must be at the effective radiating T (assuming perfect blackbody – yes, that is an approximation and it does have some error), but heat must now flow downward to the surface, so the whole atmosphere would be a stratosphere/thermosphere, and possibly very very hot (maybe it would even start to contribute to OLR via SW emissions – but let’s set that scenario aside).
Introducing a GHE in this scenario, there would initially be an increase in OLR. The TOA IRF forcing would be cooling! But that doesn’t necessarily mean the sfc wouldn’t warm. There would be a downward LW flux to the sfc now; the net upward flux within some lower layer would decrease. Less heat has to flow downward to balance the solar heating of the air, so it’s not clear exactly how things work out, but the solar heating at the surface would start to get trapped a bit.
Add enough GHE, and pure radiative equilibrium would become unstable to convection, and so a troposphere would develop and grow. The solar heating within the troposphere would then become part of the convectively-coupled layer…
patrick o twentysevensays
…“Less heat has to flow downward to balance the solar heating of the air, so it’s not clear exactly how things work out,”
? – um, consider the SARF – but I’m done for today…
“The 80Wm^-2 strength of the Lapse Rate feedback is your conjecture” – Well, about as much as my conjecture that 1+1=2.
Let us go through this step by step, then you can tell me which logical step you object
1. The GHE is the difference between surface- (Ts) and emission temperature (Tz), like 288 vs. 255K
2. The GHE is proportionate to the lapse rate. A lase rate of 0 means Tz = Ts and GHE = 0.
3. WV does reduce the lapse rate from a dry unstable adiabat (>9.8K/km) towards 6.5K/km.
4. Holding other things constant, this smaller lapse rate reduces the GHE and provides a cooler surface
5. This reduction of the lapse rate is also known as “latent heat”
6. This “lapse rate effect” is physically the same as the well known lapse rate feedback, which is simply an increase of said lapse rate effect, or latent heat respectively
John Pollacksays
E. Schaffer, I have trouble following most of what you’ve been saying. However, laying it out in concise, logical steps allows me to see some of the incorrect assumptions you’re relying on:
2. The GHE is proportionate to the lapse rate. A lase rate of 0 means Tz = Ts and GHE = 0.
JP The lapse rate from the surface to where? 2 km? 200 mb? The tropopause? Outer space?
3. WV does reduce the lapse rate from a dry unstable adiabat (>9.8K/km) towards 6.5K/km.
JP The figure for the dry adiabatic lapse rate is correct, but most of the planetary atmosphere is not at that lapse rate. In the stratosphere, the air is very dry, but the temperature increases with height for the most part. The 6.5K/km is an average for conditions near the surface. The moist adiabatic lapse rate will indeed decrease as water vapor is added, but most of the decrease is near the surface where temperatures are warmer and the air holds more wv. However, a lower lapse rate there will deepen the warm layer and allow the air somewhat higher up to hold more moisture, too. This thickens the radiative barrier between the lower
troposphere and outer space.
4. 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. You also warm the lower atmosphere and raise the tropopause. It’s highest over the tropics, where there is lots of water vapor in the lower atmosphere.
Your argument only works if the tropopause height and temperature are held constant, and then you take your assumed lapse rate from there to the surface. That’s not what happens.
MA Rodgersays
E. Schaffer,
Let me attempt to break down your myopia here by presenting again just one factor (the simplest – it applies to your Step 3) which makes a nonsense of your conjecture that the LR feedback = surface WV latent heat flux = ~80Wm^-2.
Your proof appears to rest on, firstly, the reduction of the Earth’s average Environmental LR from the Dry Air LR which is usually quoted as due to this latent heat flux and ELR = 6.5K/km with DALR= 9.8K/km – a reduction of a third. And secondly, the “Net LW Absorbed” GHE effect of Schmidt et al (2010) which is calculated as 155Wm^-2 using modern LR (and thus presumably using average ELR). So you see this 155Wm^-2 as also being reduced by a third, reduced from a GHE value using DALR. And that reduction, GHE(DALR) minus GHE(ELR), would be roughly equal to the latent heat flux of 80Wm^-2.
The issue is not that the LR feedback is negative. The mechanism is evident. An increased energy flux warming the atmosphere will reduce the LR and this will warm emissions altitudes which in turn will increase IR emissions from GHGs. The literature gives its value as 0.5Wm^-2/K. Pro rata for a 33K GHE, that would suggest a much smaller value for the full LR feedback [0.5 x 33 =] 16.5Wm^/2.
So the one factor I present to you is this – If such an increased energy flux 80Wm^-2 due to WV acts to reduce the LR and thus acts as a negative feedback for the GHE, why wouldn’t other enrgy fluxes warming the atmosphere do exactly the same. The energy balance in Fig 7 Kiehl & Trenberth (1997) which you have cited in you argument shows a total of 195Wm^-2 of such warming.
So in your conjecture “LR feedback = surface WV latent heat flux = ~80Wm^-2”, why is it only the WV component of this 195Wm^-2 total that impacts the GHE?
Your “conjecture that 1+1=2” may be well-based (mathematical geeks object to my poetic 1+1=11, while pointing out that both i + i = ii or 1+1=10 would be as arithmetically sound as 1+1=2), but your conjecture that the LR feedback = surface WV latent heat flux = ~80Wm^-2 is certainly not sound.
Piotrsays
E Schafer: “The 80Wm^-2 strength of the Lapse Rate feedback is your conjecture” – Well, about as much as my conjecture that 1+1=2.”
Not really – see replies by MAR and John P. Mine is a more fundamental question – even it was 80Wm^-2 – 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.
And no, clicking your ruby slippers and repeating: “ Well, it is exactly about improving our knowledge, and decisevely so! It is exactly about improving our knowledge, and decisevely so!” – won’t magically change it.
2. and 3. – We are talking about the tropospheric lapse rate of course
4. “So what? Other things aren’t constant” – sure, but nihilism is not going to provide insights, rather it is meant to avoid them. We mentally hold other things constant, so to identify what a certain factor does.
@Rodger
“So in your conjecture “LR feedback = surface WV latent heat flux = ~80Wm^-2”, why is it only the WV component of this 195Wm^-2 total that impacts the GHE?”
Let me first try to understand how you get to the 195 figure. I can see that the atmosphere there would emit 195W/m2 into space. It follows logically, that if it loses that amount of energy, it would also need to update the same amount to stay in balance. I would assume your calculation is like 350-324 + 78 + 24 + 67 = 195?
It is quite simple actually. That energy for the larger part is put into atmsphere at the bottom and it all gets lost at the “top” (actually different altitudes within the troposphere). While moving up or down, due to the consequentual convection, the air has to follow a dry adiabatic lapse rate, in the absence of WV. With WV it is more like a moist adiabat, though not a perfectly moist adiabat cause the air is not totally saturated with WV. Please note: the latent heat is not moved into the troposphere at ground level, but at altitude.
Btw.. while that is just my take on it, it is established pyhsics, hard to argue with.
“Pro rata for a 33K GHE, that would suggest a much smaller value for the full LR feedback [0.5 x 33 =] 16.5Wm^/2”
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.
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.
MA Rodgersays
E. Schaffer,
You continue with defence of your Lapse Rate Feedback = Surface Latent Heat Flux theorising. Above you set out two arguments in its defence.
Your “quite simple” first argument is that the energy fluxes into and out of the troposphere are not altering the Lapse Rate with the sole exception of the condensation of WV. And here your grand conclusion rests on the idea that all those other energy fluxes “put into atmosphere at the bottom and it all gets lost at the “top” (actually different altitudes within the troposphere). … With WV [the Lapse Rate] is more like a moist adiabat, though not a perfectly moist adiabat cause the air is not totally saturated with WV. … the latent heat is not moved into the troposphere at ground level, but at altitude.” You end by describing this as “established physics.”
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?
Your second argument is that there is a “contradiction.” Your own grand theory that the total Lapse Rate Feedback = Surface Latent Heat Flux = c. -80Wm^-2 doesn’t fit with the established rate of change in the Lapse Rate Feedback (= -0.5Wm^-2/K). This is a “contradiction” which somehow allows you a legitimacy in asserting what you fancy while, in your view, I illegitimately ignore this “contradiction” and apparently deny it exists.
Surely, this “contradiction” was raised up-thread by me (so not ‘ignored’ by me) while you have here only just managed to admit its existence and have also just admitted previously ignoring my raising of it.
More to the point, your arguments here are really trifling side-issues.
Your grand theory sits in “contradiction” to the entirety of climatology which you insist has failed entirely in accounting for this “contradiction”. (On your website you go so far as to say it is being deliberately ignored by the science.) Such contentious theorising obviously requires a well-presented proof. You have failed to provide such a proof. And the more I examine the situation, the more you appear as an ill-informed fool who has very little understanding of the climatology you refute and have failed to grasp that “established physics” on which you base your grand theory.
Barry E Finchsays
“86.4W/m2 of “latent heat” cooling” would be local surface and whatever (hundred or so?) metres of air cooling of the tropical ocean only and not at all the globally-averaged tropospheric lapse rate effect. There’s adiabatic heating of descending air by the pressure increasing. I pondered this briefly a few years back about the Hadley Cell, Desert Zone and the incorrect simplified phrasing for requirement of entropy increase (2nd Law if the system was closed) “hotter things heat colder things”, which simplified phrasing isn’t correct of course because latent heat. Specifically I wondered a few minutes. air rises over tropical Atlantic Ocean (tall troposphere, surface Low Pressure Zone), slides north near troposphere top, turns east, H2O gas condenses-freezes high up, rains back onto the Atlantic Ocean having converted latent heat to “heat”, descends over the Sahara Desert nice and dry from having rained out, warms at 9.8 degrees / km altitude by pressure increase and heats a surface of say 40 degrees from Atlantic Ocean surface of say 25 degrees. Hotter warming colder due to latent heat.
You need to bear in mind what it means. It is not just latent heat as an arrow in the diagram above, but rather it means the reduction of the lapse rate and the whole GHE with it. Without WV we would have a dry unstable adiabatic lapse rate of >9.8K/km. The GHE would be at least 50% larger. If you assume it to be 155W/m2 that would mean over 77.5W/m2.
In my calculation the GHE is only 120W/m2, but the lapse rate should actually be like 65% larger, so that the latent heat would still amount to ~80W/m2.
Barry E Finchsays
In 2013 I calculated 79 W/m**2 from annual global precipitation and evaporation latent heat per kg. I just did it again in 1 minute and got only 76 W/m**2. In 2013 Kevin Trenberth showed 80 W/m**2 in his EEB at 26:26 at https://www.youtube.com/watch?v=SQOIHdlZngk so I assumed my 79 W/m**2 was simply latent heat off the oceans. Are you referring to 86.4 – 80 = 6.4 w/m**2 as meaning that some adjustment is needed to the total effect of increased H2O gas due to increasing surface-troposphere temperature ?
No, I am not into claiming some super precise figure here. I just point out to the fact that there is some 80W/m2 (or so) of cooling due to WV, which depending on what figures you choose, is at least on par, if not exceeding, its warming side.
I am not sure if I understood you correctly but, if you assume that latent heat stored in water vapour released from tropical oceans and rainforests returns to Earth’s surface in deserts of higher latitudes, I would like to ask how is it possible that the warm air heated by water vapour cindensation descends?
Do you not think that already the said condensation (and following precipitation) of Ester vapour could hardly occur if at least part of the released condensation heat were not emitted as infrared radiation to the space?
I suppose that the same may apply for further cooling of the resulting dry air that enables its final descent to the ground.
Tomáš Kaliszsays
Water, not Ester. Condensation, of course.
Barry E Finchsays
Tomas, “how is it possible”. Hadley Cell, large scale overturning. “at least part of the released …. emitted as infrared radiation to the space?” Yes, of course, and I doubt I’ll ever try to figure that out with any worthwhile accuracy (few years back I tried to estimate global wind turbines slowing global wind in my head in my kitchen making dinner, and I gave up, couldn’t figure the constant push of uneven Sun heating of Earth fighting an attempt to slow wind). So that’s why I simply gave in my comment 6 Jul 2026 at 8:13 PM what I saw in some scientists’ Webinars (2 I think I recall) on UTube where they showed the classic 3 feedbacks and Planck response (probably the Webinars were about how cloud feedback is known poorly and the surface better and +H2O gas feedback with high confidence) from some average of some CMIP Climate Models. I figure the odds the Being-Paid scientist Army and computer programmers did that with their time-sliced simulations better than I could with some calculations to be like 1,234:1 and the odds for better than “E. Schaffer” like 9,876,543:1 (even though some of those bods think photons is re-emitted and they ain’t and radiation some place goes back some place and it don’t) because I’m not finding “E. Schaffer” to be making any sense at all.
Barry E Finchsays
Tomas, I just realized that “E. Schaffer” 13 Jul 2026 at 6:30 PM “the fact that there is some 80W/m2 (or so) of cooling due to WV” is mostly (the other little bit is my estimated adiabatic heat to surface) yet another of the Denier memes enabled by the physicist nonsense “re-emits” and “back radiation” instead of “emits (manufactures)” and “downwelling radiation at the surface”. Ironic that I failed to notice that considering I’ve been going on about it and getting “Flamed” since June 23rd when I correctly pointed out that some bods typed rubbish. I forgot that downwelling radiation at the surface includes I just calculated 16% of the H2O evaporation-condensation-freezing latent heat. As you also understood in a comment the LWIR in troposphere is from all Power sources.
The “80W/m2 (or so) of cooling due to WV” hasn’t itself, with that hugely-overstated “80”, gone through the “greenhouse effect (GHE)” in Earth’s troposphere, with GHGs converting Solar SWR. H2O gas latent heat, and NET sensible heat convection & conduction into LWI radiation and returning 63% to the surface. Using the “Hartmann 2nd Edition 2016” text book page 34 EEB diagram with my assumed 13 w/m**2 Solar SWR each into thermosphere & stratosphere, in Earth’s troposphere is 54 of Solar SWR + 27 of rising warm air + 88 of surface water evaporative latent heat + 376 of surface LWIR = 545 w/m**2 total so the 88 w/m**2 of surface water evaporative latent heat is 16% of the Power sources for making the LWIR. So 16% of the 345 w/m**2 = 55 w/m**2 of the downwelling radiation at the surface is evaporative cooling at surface being returned to the surface as LWIR.
I looked at Hartmann page 166 annual overturning mass stream functions for all of 3 minutes and estimated 7 w/m**2 of adiabatic cooling while ascending & heating while descending, which might be highly miscalculated, from the diagram. Added 7 to the 20 w/m**2 of thermals. So, the actual water evaporative cooling at the surface to compare for scale with the total “greenhouse effect (GHE)” of H2O gas in Earth’s troposphere is 88 – 55 – 7 = 26 w/m**2.
I think you are somewhat struggling with the GHE itself. It is simply the difference between surface- and emission temperature, or respectively surface emissions and OLR. So that is 288 – 255 = 33K, or 390 – 240 = 150 respectively. We could discuss the precise figures here, but that is the idea.
In both instances the lapse rate is decisive for the magnitude of that GHE. Again, if the lapse rate was zero, there would be no GHE at all. And we know because of WV, because of the latent heat, we do not have a dry unstable adiabat (>9.8K/km) but only just ~6.5K/km. So yes, this latent heat DOES cool the planet a lot, easily by an 80W/m2. It ain’t that complicated..
JCMsays
“”55 w/m**2 of the downwelling radiation at the surface is evaporative cooling at surface being returned to the surface as LWIR”””
I wager this is misleading. The main determinant of surface LW down is lower atmospheric temperature, not instantaneous flux.
Climatological temperature at some level is determined by todays planetary solar heating, yesterdays solar heating, and all the days before, minus the LW cooling of the planet on all the days. So we arrive at some cumulative sum of energy absorbed and emitted by the planet, and allow thermodynamic and heat properties to sort out where all this energy ought to go and by association how much temperature at some level should be dragged at some point in time.
Strictly speaking, the immediate power source for atmospheric LW emission is the thermal internal energy of the atmosphere. Jiggling molecules and so on. Over climate timescales, that internal energy is ultimately maintained by the net external radiative energy entering the Earth system (absorbed solar minus outgoing longwave from the perspective of space). When N = 0, the change of internal energy (U) is 0. LW emission is a function of U.
Consequently, saying that latent heating of atmosphere contributes roughly 16% of the atmospheric energy input does not imply that roughly 16% of the surface downwelling longwave “comes from” latent heat. If the partitioning between sensible and latent heat changed while the lower atmospheric temperature remained unchanged, the surface downwelling longwave would remain approximately the same. Conversely, if latent heating decreased but the lower atmosphere became warmer, downwelling longwave would generally increase. The 16% figure simply says something about the state of the system.
If still not convinced, the attribution of LW down to the latent heating is non-unique. It depends only on the accounting framework. I think the issue becomes more clear when using a surface budget, since taking a valid closure equation should not be interpreted it as an attribution equation. Those are not the same thing. The budget tells you only what combinations of fluxes satisfy energy conservation.
Classically the surface budget is written as SW down – SW up + LW down – LW up = H + LE. It says surface net radiation = fluxes of latent and sensible heat.
SW down – SW up + LW down – LW up – H – LE = 0.
Now just rearrange for clarity to understand why it cannot be an attribution.
LE (latent heat of evaporation) = SW down – SW up + LW down – LW up – H.
Now suddenly latent heat flux might seem to depend on LW down, and not be causing it. How can that be.
Now make things really weird:
LW down = SW up – SW down + LW up + H + LE.
Now it seems like LW down has some inverse relation to SW down, which is really weird. Why would increasing SW down decrease LW down? It’s weird because it’s not an attribution, it is simply relations which must satisfy energy conservation. It is a closure relationship, not a physical decomposition of the origin of the downwelling longwave radiation.
Putting elements on one side or another of an equals sign does not assign cause and effect. Physics doesn’t work like that.
The same thing is happening in atmospheric budgets. It is not correct to understand latent heating of atmosphere to be causing LW down, even though at first glance that seems perfectly reasonable. I recommend sticking to the original message, which is that radiation is manufactured simply based on the state of the system. The immediate physical origin of downwelling longwave radiation is the thermal internal energy of the atmosphere. I think somehow this message is being overlooked.
JCMsays
To Schaffer
as always the reality is always a little bit nuanced.
It is obviously true that when greenhouse is described as an observational temperature difference it represents a type of radiative convective equilibrium with all its complexities. I think it shouldn’t be in dispute.
Owing to the freedom of non radiative heat transfer, we know the temperature gradient is diminished compared to what it otherwise might be in a purely radiative equilibrium profile.
The other part is about the depth of surface below the balancing radiative emission level. This can be understood to be associated with the atmospheric composition and its relation to optical thickness.
Flux of mass and heat says something about temperature profiles, and column water vapor says something about the spatial separation of surface and the planetary radiating level along that profile.
For investigation of how a model deals with worlds with varying restrictions on moisture dynamics, including column water vapor state and the cycling aspect associated with flux of mass and heat, I recommend “Continental configuration controls the base-state water vapor greenhouse effect: lessons from half-land, half-water planets” https://eartharxiv.org/repository/view/5196/
There we see 8 different configurations, ranging from landworld, which is basically totally restricted in moisture cycling, to aquaworld which is totally ocean, and things in between, including tropicalland, northland, and realLand. Each has different shapes and distribution of land and ocean. Fig 1. This is meant to illustrate that it matters where and how much moisture is limited or not.
These simulations handle both the features of turbulent flux of heat and mass, and water vapor concentration. Clouds are not discussed, which aligns with the talking points so far on the thread. The range across simulations is about 15K.
There we see how landWorld, despite having extremely limited latent heat flux, exhibits the coldest mean temperature of all simulations Fig 2. Total column water vapor is close to negligible at 3.9 kg Fig 5, and the planet settles around 273K.
The opposite configuration aquaWorld, totally ocean, settles at 285K. This features the greatest partitioning of surface energy to latent heat flux and includes 17.1 kg column WV.
The realLand shapes, which depict relatively complex distributions of land and ocean, generate the warmest temperatures of all 8 simulations at 287K. The partitioning of latent flux is more restricted compared to Aqua, and yet the column water vapor climbs to 17.2 kg.
landWorld cold, aquaWorld medium, and realWorld warm.
I hope it provides a foundation to think about how the heat flux and WV are somewhat related but definitely distinct issues, and how the associated dynamics can interrelate in counterintuitive ways.
I love nuances! And in fact I have made very similar considerations, although less model- and more physics based. So I am quite interested in the model world in that paper, though I’d have a number of questions.
Why is there a surface albedo of ~0.3? The surface of Earth is actually pretty dark. About 1/2 of the albedo is due to clouds, another quater is due to di-atomic atmospheric molecules (N2, O2) which scatter SW radiation. Surface albedo is <0.1 in the real world. Is it possible that the paper does not include clouds, but still keeps their albedo, as part of the surface albedo?
You say that 3.9mm of precipitable water were negligible. When I check modtran such a reduction of WV reduces its GHE by about 30%, meaning 70% are still there. 70% are no way negligible.
The biggest contributor to the GHE (as I would have to explain in a lot of detail) are actually clouds. That is why I think it is extremely troublesome to not consider clouds, especially if one possibly holds on to their albedo effect. One could then have their cooling, but not their warming part considered.
Is it because of that, that all these model worlds feature lower than real surface temperatures? I mean that is kind of strange, when you consider clouds were to be cooling. If you leave them aside, it should get warmer, not colder..
Another irritating detail is about how the lapse rate is treated there. The term lapse rate is not mentioned once, making me assume it is not considered at all. The only hint is some latent heat of 10W/m2 in the LandWorld. That would only be consistent with a very steep lapse rate of ~10K/km, and per se a 50% larger GHE. Given the low amount of WV (3.9mm..) would reduce the GHE by only about 15W/m2, all over the GHE in the LandWorld scenario should be a lot larger and surface temperature higher! That is of course unless it had the cloud-GHE removed and the lapse rate unchanged. Though these would be primary school level blunders..
JCMsays
in re to: “The term lapse rate is not mentioned once, making me assume it is not considered at all.”
This I think gets to the crux of what’s going on here. I do not believe a greenhouse effect expressed as a temperature difference is mentioned either. I figure the reason being that each are emergent properties, not governing variables. They can each be understood as diagnostic effects. It makes little difference to physics whether or not someone decides to extract these quantities from model output.
In a radiative-convective-equilibrium concept, both the lapse rate and GHE are outcomes of the coupled interaction between radiative transfer, non-radiative heat transport, and thermodynamics.
Neither a lapse rate nor GHE can be imposed, they emerge from the balance of processes that warm and cool at different levels.
The way I approach these things is from the teachings of Axel Kleidon, where temperatures depend on the non-equilibrium thermodynamics involving radiative exchange, turbulent heat transport, atmospheric cooling, and the continuous generation and depletion of vertical temperature gradients.
The observed lapse rate and greenhouse effect arise as steady-state expressions of thermodynamic principles. The important point is that turbulent transport requires a radiative generated temperature difference in the first place. Solar radiation preferntially heats the surface, creating a temperature difference between surface and atmosphere. This differential radiative heating generates thermodynamic disequilibrium that drives heat transport.
Kleidon models the surface-atmosphere as a heat engine.
The surface receives radiative heating and that energy can be exported non-radiatively as turbulent heat flux (sensible + latent heat), or leave directly by longwave radiation.
If the turbulent flux is very small, the surface must become very hot. The temperature difference between surface and atmosphere is very large.
If the turbulent flux is very large, the surface cools, the temperature difference collapses, and there is little thermodynamic driving force to generate power.
These competing effects produce an optimum. In other words, temperature differences can only be depleted so far until the engine shuts down. Kleidon recommends that the system be understood to operate near maximum mechanical power P.
Start with the surface energy balance, and collapse turbulent flux of latent and sensible heat into term J. For simplicity we’ll assume grey atmospheric radiation.
Rsw = J + Rlw
The provides relations for surface absorbed solar radiation Rsw (an external energy input), turbulent heat flux J, and net longwave radiative exchange between surface adn atmosphere, Rlw.
Rlw = σ(Ts^4 − Ta^4), where Ts and Ta represent surface and effective atmospheric radiating temperatures.
To Schaffer’s point, when J increases, Ts falls, therefore Rlw also decreases.
The Carnot efficiency is simply n = (Ts – Ta) / Ts
Mechanical power P relates to the heat engine receiving turbulent flux J
P = J ((Ts- Ta) / Ts)
This is where we see the the tradeoff, increasing J increases available heat, but increasing J also cools the surface, and therefore temperature differences are depleted and the engine becomes less efficient.
For different assumed turbulent flux J, we can solve for the corresponding surface temperature, and the generated Power:
Example 1: Assume small turbulent flux J = 50 W/m2
Efficiency fell, but transported heat doubled, and Power increased from 7.9 W/m2 to 11.8 W/m2
Example 3: very high turbulent flux J = 130 W/m2
Rlw = 35 W/m2
Ts = 279K
Power = 11.2 W/m2
Too high turbulent flux J means power is reduced.. More heat is being transported, but the engine efficiency has become too small.
The observed system state is the one in which Power is maximum (Example 2).
What is Power? In the atmosphere, power is the rate at which mechanical work is performed to sustain turbulence, accelerate air, generate winds, and lift air parcels against gravity. P = dW / dt
We have heat flux J, and we have Power (mechanical work). Power = efficiency (n) times heat flux (J). How much available energy generated by differential radiative heating is available for performing Work.
For planet Earth and assuming grey atmosphere, approximately 100 W/2 of turbulent heat flows through the atmospheric heat engine. Given the observed temperature difference between the surface and atmosphere, the maximum theoretical mechanical power is about 12 W/m2. The remaining ~88 W/m2 is not converted into work but is ultimately dissipated as outgoing radiation to space. This should be understood as maximum potential power.
What is happening with that:
Sun heats surface ~ 165 W/m2 –> surface becomes warmer than atmosphere –> vertical temperature gradient is generated –> atmosphere consumes the gradient using H + LE –> radiative cooling aloft restores the gradient.
While that schematic is expressed in a sequence, it should be understood to be happening simultaneously all the time. The vertical profile and GHE (temperature difference) emerges through radiation boundaries continuously regenerating the thermodynamic disequilibrium (solar heating surface, radiative cooling aloft). The turbulent fluxes of latent and sensible heat emerge by consuming the gradient at maximum Power, but never deleting it.
You may have noticed the exhaust ~88 W/m2 associated with the turbulent flux is exactly 1/2 of the surface solar radiative heating (165 W/m2). This is an expression of optimality. In reality there is always nuance, since the system is not a perfectly sealed engine and so it allows direct losses by radiative transmittance from the surface to space mentioned elsewhere in the thread, and we also have frictional dissipation of work energy back to heat.
In summary, if turbulent transport were too weak, the surface would become excessively warm relative to the atmosphere, producing a very large temperature difference. If turbulent transport were too strong, it would erase that temperature difference, leaving little buoyancy to sustain convection. These competing tendencies lead to an optimum, which Kleidon interprets as operation near maximum mechanical power.
This self limiting principle means simply that turbulent transport requires a radiatively generated temperature difference in the first place, which means it could never be overcome or deleted.
In conclusion, it is physically necessary that a positive net greenhouse effect survives in steady state because turbulent heat transport depends on the very radiatively generated temperature gradients that it consumes.
I appreciate the opportunity to bring more clarity to these issues in my own mind.
To calculate the (dry) adiabatic lapse rate you need the specific heat capacity (of the gas) and gravity. Including the latent heat by WV you need a couple more specific parameters do calculate the moist adiabat. However, what you will not need is a “coupled interaction (model) between radiative transfer, non-radiative heat transport, and thermodynamics”. So I guess it is safe to reject this idea. I mean in this instance we already know the physics, there is no need to re-invent and convolute it.
It is true that the real lapse rate differs from the theoretic adiabat, and GHGs have to do with it. The surface receives way more SW input than it has in LW output through the atmospheric window. With the troposphere it is the opposite, and so there needs to be a transfer of energy. That is why we call it an “unstable adiabat”, meaning a lapse rate larger than just adiabatic. While there is some departure, the base line is still the adiabat.
I think that is the whole problem here. The adiabatic lapse rate is a gas thing, NOT a GHG-thing!!! GHGs, whether condensing or not, play into it, but they do NOT play the main part. By rejecting that very fact things become convoluted, complicated and eventually wrong.
For instance:
“when J increases, Ts falls, therefore Rlw also decreases”
That is not true if you stick to your definition of Rlw: “net longwave radiative exchange between surface adn atmosphere, Rlw”
Are you aware this figure is almost zero? Let me stick to (incorrect) consensus figures for simplicity. Say the surface emits 390W/m2, 40W/m2 of which go right through the atmospheric window, so that 350W/m2 go from the surface into the atmosphere. “Back radiation” might be 340W/m2. In this instance Rlw = 350 – 340 = 10W/m2. It is just a marginal figure, also irrelevant.
However, if Ts changes, Rlw would barely change at all. With higher temperatures there is just more radiation going back- and forth, and vice verse, but the difference between the two barely changes.
I could point out more mistakes, but really the original sin is what I told above. You try do deny the nature of the adiabatic lapse rate and construct some alternative physics, which then requires a lot of stuff and will not work.
JCMsays
I repeat that it’s necessary to disentangle diagnostic variables, such as an environmental lapse rate, and the balancing heat fluxes that satisfy steady-state energy budgets.
A lapse rate does not uniquely determine heat flux. Likewise, an adiabatic rate which is constrained by gravity and heat capacity does not tell you how much energy is flowing through the column.
Two atmospheres can have similar lapse rates while transporting vastly different amounts of energy. Obviously a system with same gravity and heat capacity but absorbing 5x more solar must generate vastly more steady state heat flux compared to Earth.
The underlying message I was hoping to convey is that the atmosphere cannot transport heat so efficiently that it destroys the temperature differences required to transport heat in the first place.
The turbulent heat flux and the greenhouse effect, expressed as a surface-to-radiating-level temperature difference, are coupled. Increasing turbulent transport reduces the temperature difference that drives it, while radiative heating and cooling continually regenerate that temperature difference. The steady state emerges from this competition.
In the detailed example provided, the relevant thermodynamic boundaries provided were the Ts and the atmospheric effective radiating temperature Ta. As such, the heating of atmosphere from below is balanced by radiative cooling out the top in grey atmosphere. The steady state is therefore established by balancing the upward transport of energy through the atmospheric column against radiative loss to space.
In your example as you note, the derivation of the adiabatic lapse rate has no dependence on external heating and cooling rates, and that is precisely why it says nothing about the balancing heat fluxes that satisfy energy budgets.
The analytical relations I presented instead describe how the heat flux itself is constrained. Simply, atmosphere cannot consume the thermodynamic gradients faster than radiative processes regenerate them. Additional steady-state optical depth therefore cannot simply be cancelled by increasing turbulent heat flux, because doing so would undermine its own driving force.
In case that you are NOT just trolling, I would like to turn your attention to the circumstance 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.
The net radiation from the surface in your example is 390 W/m2 – 340 W/m2 = 50 W/m2, not 10 W/m2 as you assert.
Best regards
Tomáš
Barry E Finchsays
JCM 18 Jul 2026 at 5:11 PM “Conversely, if latent heating decreased but the lower atmosphere became warmer”. If latent heating decreased but the troposphere became warmer then the Law of Conservation of Energy would have been violated. To put that another way, a statement so incomplete that it is worthless is ……worthless.
Not that I would see any merit in these considerations, because “back radiation” is still not heating anything. I also do not really see what exact assumptions are being made here, because it would seem there are plenty of contradictions.
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..
I don’t see any particular violation in the wordage used in the example. It was simply to support your assertion that photons being manufactured at some place relate mostly to the local thermodynamic equilibrium temperature there.
The thermodynamic duty to compensate the external solar radiative input to the surface remains.
Against 165 solar external input must come the balancing fluxes of radiation and turbulent heat transfer.
This is not an independent physical law, but a consequence of the drive toward establishing a steady state balance.
Artificially slow down hydrological cycling and allow the system to reorganize so that the balance to shifted to a greater burden upon radiation and sensible heat transfer. I suspect after which a radiometer will detect a greater signal being generated when pointing the detector in various directions from the surface.
In re to Schaffer
The initial warming from an external radiative forcing + the associated enhanced opacity from increasing WV is what creates the conditions for stronger moist convection. Deleting those conditions, as you propose, ultimately cancels the thermodynamic conditions that gave rise to it in the first place, and you create a paradox.
It is important to keep in mind always when interpreting radiative kernels that the lapse-rate adjustment is a consequence of warming, not an independent opposing mechanism. The lapse rate adjustment is the thermodynamic signature of a warmer, moister convecting atmosphere, not something that could overcompensate it.
If you want to dive into the weeds analytically I recommend to explore why the surface energy balance becomes increasingly dominated by latent heat flux at warmer temperatures and why OLR becomes increasingly dominated by atmospheric emission.
Whereupon Surface solar radiative heating is balanced by LE + H + radiative diffusion (k) + radiative transmittance (τ_transmission). General default values are something like 160 solar radiative heating = 80 LE + 20 H + 20 k + 40 τ_transmission. This is more illustrative to address your particular concerns with atmospheric “windows” which didn’t come about in grey formulations previously by definition.
From this perspective two primary things are occurring:
1) more abundant of water vapor at higher temperatures narrows the available windows through which the surface thermal emission can be transmitted to space. As a result, the contribution to the OLR from the surface gets smaller as Ts increases, implying the increasing decoupling of the OLR from the surface’s emission. A standard optical depth around 2 implies radiative transmittance ~ 10%-15%. Increasing water vapor progressively diminishes the transmissivity and more outgoing emission must be sourced from colder atmospheric layers.
2) at higher temperatures with ample surface moisture the partitioning between LE + H increasingly favors the latent flux through the psychrometric constant (γ) and slope of saturation vapor pressure curve (Δ). As temperature rises the slope of saturation vapor curve is more steep and turbulent flux partitioning increasing favors the latent flux over sensible heat through Δ/γ.
In both cases 1 and 2, it is the warmer temperatures themselves which generate apparent atmospheric adjustments. Such adjustments cannot be sustained if warmer temperatures are cancelled. I think there is some risk of getting lost in statistics and mashing up data sources when applying your methods, and losing sight of what’s going on.
Additionally, I would also keep in mind that the tropical hotspots and things you’re seeing in model output are a consequence of the very warming you’re trying to cancel. Deleting warming deletes the very effect of your interest, resulting in a potential logical void. No amount of adjusting aspect ratios in visualizing re-digitized datasets will resolve that.
I suppose that if participants find my input to be worthless, or without merit, they would simply cease to respond. cheers and thanks for the discussion.
ES: “back radiation” is still not heating anything.
BPL: The surface. Also, layers of atmosphere below the one radiating.
Barry E Finchsays
Perhaps unclear, I didn’t mean that the 86.4W/m2 Power flux is only for the tropical ocean, I know it’s a globally-averaged value of 44.1 petawatts (44.9 in my Hartmann text book, (40.3 or 40.8 I recall from earlier EEB pictorials). I meant that the 44.1 petawatts of surface water evaporative cooling must be reduced by (8% I just very-roughly calculated) due to descending air being adiabatically heated and H2O gas condensing-freezing must be given its allocation of that sensible heat downward at the surface. I’ve since realized that I forgot the effelump in the room for that which is that downwelling LWIR at the surface includes 16% of its Power from the upwelling evaporation of the ocean and that means that surface water evaporative cooling must be reduced by a massive 64% (63% for the Hartmann EEB) because that much heat that left the surface as evaporative cooling returned to the surface as LWIR.
Joseph O'Sullivansays
An interesting paper in the journal Nature ‘Amplified Arctic iceberg traffic reshapes benthic biodiversity’ and a news story about it remind me how climate change is terrifying, but climate science is fascinating. Scientists looked at dropstones, rocks that start embedded in glaciers, broke off into the ocean in icebergs, then were deposited on the ocean floor as the icebergs melted. They were able to use the recordings of seabeds as a proxy to determine how fast glaciers were breaking up in some areas. They also recorded the ecosystem changes caused by the dropstones. Organisms that needed hard surfaces to live on were increasing as the ecosystems changed from soft sand or mud to hard stones and rocks.
News article: https://www.nytimes.com/2026/07/07/science/arctic-icebergs-deepsea-life.html
The paper, its freely available https://www.nature.com/articles/s41586-026-10630-4
Copernicus has posted for June ERA5 re-analysis with the month’s anomaly of +0.56ºC, this the second warmest June on record.
The top-ten ERA5 Junes now run:-
2024 … +0.67ºC, 2026 … +0.56ºC,
2023 … +0.53ºC,
2025 … +0.47ºC,
2019 … +0.37ºC,
2020 … +0.36ºC,
2022 … +0.31ºC,
2016 … +0.26ºC,
2018 … +0.23ºC,
2021 … +0.21ºC.
Globally, 2026 started warmer than 2023 but cooler than 2024 and now, mid-year, the three are equally warm at about +0.6ºC. The Dec 2023 and 2024 anomalies sat at about +0.8ºC.
This pretty-much echoes the NH situation, with the 2023 and 2024 NH anomaly tracking each other very closely from mid-July to year’s end. Down in the SH 2026 has been tracking the slow drop seen in 2025, cooler than 2024 and since May a tad cooler than 2023 (which saw a sharply rising in anomaly Jan-to-October).
I should mention that the SH anomaly is subject to big Antarctic wobbles which run throughout the year. (Up north, big Arctic wobbles are restricted to winter months.) The Antarctic wobbles do look to have contributed a significant cooling within the SH average anomaly through June.
The Copernicus numbers include a value for “relative to the 1850–1900 pre-industrial.”
Until the extra highlighting employed in this month’s update, I hadn’t managed to notice before the link to “Read more about how the temperature change above the pre-industrial level is estimated and the differences between global temperature datasets.” This provides explanation of the method used by ERA5 to calculate its 1850-1900 anomaly base (given the ERA5 record only runs back to 1940).
The method uses the average of BEST, HadCRUT & NOAA (which do cover the 1850-1900 period) that yields the annual offset of -0.88ºC relative to the ERA5’s 1991-2020 base. Monthly average offsets (which show an annual variation, -0.77ºC to -0.98ºC) are then used to calculate “objectively using Fourier fitting of a single harmonic, rounding coefficients to two decimal places” for monthly offsets and, with a bit of additional smoothing, daily offsets.
UAH TLT has posted for June (the first measured June data) with an anomaly of +0.46ºC, down on May’s +0.53ºC. The NH showed a healthy increase (May +0.46ºC, June +0.54ºC).while the SH a bigger decrease (May +0.60ºC, June +0.38ºC).
The above could be compared with 10 days ago – MA Rodger says 30 Jun 2026 at 2:26 PM
–how June 2026 (+0.55ºC) will be roughly unchanged from May26 and marginally above the Jan-May average (+0.53ºC). (Recent year Jan-Jun have averaged 2023 +0.38ºC, 2024 +0.70ºC, 2025 +0.61ºC),
–The SAT numbers wouldn’t be showing any sign of the coming El Niño while the ERA5 60N-60S SST so far continues the pause seen during May26
— so the presently-developed El Niño conditions would be rated as ‘weak’.
There is a new El Niño out there and it was officially declared already on June 11 by NOAA.
This is both unusually early in the year and very soon since the last El Niño in 2023-24.
Another remarkable thing is the seasonal forecasts, which for a couple of months have indicated that it may well be at strengths by the end of the year that we have not seen before.
The idea of an impending El Niño is particularly interesting to me, as I’ve experienced its effects firsthand on my beach travels. I’ve found that off-season trips can be a great way to avoid the crowds and heat waves that often come with these events. For instance, I once took a ferry to a remote island during the off-season and was able to explore its beautiful beaches without the throngs of tourists. One practical tip I’d like to add is to always research the necessary permits and timing for these trips, as they can vary greatly depending on the location and time of year. I’m wondering, how do you think the impending El Niño will impact beach ecosystems and coastal communities, and what can we learn from past experiences to better prepare for its effects?
Barry E Finchsays
Rohit, how very nice for you. I’m glowing in empathetic pleasure.
Barry E Finchsays
I retract that because “Rohit Ghosh” isn’t Climate Systems Scientist “Paul Beckwith” flying Canadian Winter to Florida for a blitz round the Everglades at 99 miles per hour in a 15 ton Power Boat then back to Ottawa for more science videos about the ‘orrible Climate Changes, and ask for some more money.
GISTEMP and NOAA have posted for June. (Numbers-wise, there’s little difference except in the SH.) NOAA’s “June Highlights” run:-
☻ The globe had its second-warmest June on record, driven by global ocean surface temperatures reaching an all-time high for the month.
☻ Both the Arctic and Antarctic recorded June sea ice extents that ranked among their respective 10 lowest extents on record.
☻ Global tropical cyclone activity was above average with seven named storms. (Note the Atlantic is having a quiet start to the season.)
For June (+1.18ºC), GISTEMP shows a small rise on May’s anomaly (+1.13ºC) with an increase in the NH (May +1.30ºC, June +1.44ºC) and a smaller decrease in the SH (May +0.97ºC, June +0.92ºC).
The 2026 anomalies for May & June both sit above the 2025 May & June so the rolling 12-month average is now bottomed out and on the rise. And I reckon that we are now starting to see them up-tick with the coming El Niño. Compared with 2023, it’s a relatively gentle up-tick and not yet appearing in the wholly-SAT record of ERA5 (although in the 60N-60S SST of ERA5 anomalies have been on the up for some time).
<a href="https://sites.google.com/site/housman100resultstemperarypost/home/the-banana-watch"<The Banana!!! Watch page monitors these temperature developments.
Barry E Finchsays
E. Schaffer 10 Jul 2026 at 10:59 AM ” I am … asking about the given effect of WV”. That’s a topic to which applies “Point of view” not socially but from a physical science basis because there are mutual.ly-reinforcing effects. A big one is ice-snow cover. The simplifying statement that Earth would be ~33 degrees colder without CO2 must be highly incorrect because ice-snow has a much higher albedo than ocean water so a reduction of CO2 to the point where the ocean is mostly or entirely iced over must make Earth far colder than just ~33 degrees colder.
Since H2O gas is a condensing gas in the troposphere it must be considered a feedback, not a Forcing, and thus its contribution to Earth’s GMST, cannot logically be in a separate category from the surface albedo “effect” on temperature and cloud “effect” on temperature.
So, if Earth goes from Snowball Earth to Holocene Earth there’s no single division of warming “effect” on temperature rise because change in open ocean water portion affects H2O gas, and CO2 also, and change in well-mixed GHGs and condensing H2O gas affects the open ocean water portion.
Also, if some emission (manufacturing) and absorption spectral lines are shared by CO2 and H2O gas, rather than being interleaved, then the allocation of “effect” is arbitrary or decided by physics that’s past what I’m willing to spend time pondering.
Piotrsays
Barry: “ if some emission (manufacturing) and absorption spectral lines are shared by CO2 and H2O gas, rather than being interleaved, then the allocation of “effect” is arbitrary or decided by physics that’s past what I’m willing to spend time pondering
It has been already discussed in Schmidt et al. 2010 – where they pointed out – that “direct measure” of warming by any substance influencing the climate – does not have a single value – because even if a given factor was removed, other factors with overlapping windows of absorption would partly compensate by increasing their own absorption. To make things worse not only there is no way to subtract this compensation – the % of compensation would be different in moments in time.
And that’s on top mentioned by you: passive feedbacks (like WV) and not the same as forcings (anthropogenic GHGs) and there is no point in calculating “direct measure” of feedbacks, other than the deniers trying to divert the attention from the mitigation of GHGs and allow fossil fuel political- industrial complex to extra as much fossil fuel as their shareholders, and cost of waging wars on other countries, demand.
You are right in stating the 33K (or so) are not enshrined in law. They are supposed to be the difference between surface- and emission temperature, that is all.
It is indeed incorrect that the Earth would be 33K colder without CO2, for lots of reasons. We have other forcing GHGs like O3, CH4 or N2O – it is not just CO2. The largest contributors to the GHE however are clouds and WV, which both have their cooling sides. If you’d remove them you would not just lose their contributions to the GHE, but also change the albedo and the lapse rate.
A simple calculation might look like this: The energy budget would go from 240W/m2 to 290W/m2 if you lose cloud albedo. The non-condensing GHGs (CO2, CH4, O3, N2O..) might contribute about 50W/m2 to the GHE (single factor addition), at the given lapse rate. Without WV this lapse rate would be at least 50% larger, let us say it would be 80W/m2 then. Also we know the surface as it is does emit ~360W/m2.
Without WV and clouds we get 290W/m2 in and 280W/m2 out (360-80), that is an energy SURPLUS(!!!) of 10W/m2. That is if we’d remove WV and clouds from the atmosphere, the Earth would warm.
Barton Paul Levensonsays
ES: It is indeed incorrect that the Earth would be 33K colder without CO2
BPL: More of a strawman. The Earth would be 33 K colder without the entire greenhouse effect, not without CO2 in particular.
ES: if we’d remove WV and clouds from the atmosphere, the Earth would warm.
BPL: Congratulations on the nuttiest denier idea I’ve heard around here lately.
Ray Ladburysays
Please do feel free to cite the peer-reviewed publication in which these claims are substantiated.
I am afraid that the way how you framed your question (like “Is water vapour cooling or warming Earth?” or “Is water vapour cooling or warming Earth’s surface?”) may prevent any reasonable answer.
On the first sight, it appears that your formulation reduces the role of water in Earth climate to its vapour, what would suggest a primary focus on the greenhouse effect of water vapour. However, if you perhaps consider also the role of water in lapse rate regulation and/or in cloudiness regulation (what seems to be the case based on your complaints that your request is not taken seriously), you should not speak about “vapour”.
Instead, I would expect that you will analyse the relationship between water cycle intensity and geographical distribution of water availability for evaporation, including their complex interactions with other components of the climate system and processes therein. This way, you could perhaps arrive at some hints if (and/or under which circumstances) the assumed summary “cooling” or “warming” effect of various forms of water (and/or various processes involving water vapour formation and condensation) on Earth’s surface prevails.
Nevertheless, I do not think it would be sufficient. Furthermore, you should consider also climate aspects related to water heat capacity, heat distribution by ocean currents, and properties and behaviour of ice. Finally, you should not neglect also underground water as well as water distribution in soils and vegetation, and in all these instances keep in mind that all forms of water interact with water vapour and that these interactions and processes involving them are no way static.
For these reasons, I can hardly imagine how this complexity could be reduced to the requested unambiguous estimation if the “effect of water vapour” is “cooling or warming”. I doubt that such generic request makes sense.
I guess we could go down this path, or we just stick to the pretty simple reality. In this simple reality WV does two things:
– it raises the emission altitude, thus reduces the emission temperature, which enhances the GHE and warms the planet
– it reduces the lapse rate, thus increases the emission temperature, which reduces the GHE and cools the planet
And almost as if I had to repeat myself (lol), the question is about netting those two effects. And again, I can not believe this question was never dealt with and comes across as if it was an alien spaceship. This should be trivial!!!
Ray Ladburysays
You are forgetting the fact that water vapor is a potent greenhouse gas in its own right.
“You are forgetting the fact that water vapor is a potent greenhouse gas in its own right”
Yes, by elevating the emission altitude, which I just pointed out. But it also has that other side..
Piotrsays
E. Schafer: “ we just stick to the pretty simple reality., ”
.the simple reality that we ALREADY KNOW that we can’t do ANTYHING about WV, clouds and lapse rate to mitigate AGW in any meaningful way?
Nor will it improve our knowledge – VW, clouds and lapse rate are already implicitely included in the climate models. If the model show 2C warming what difference does it make to ask how much of that 2C came from direct forcing (GHGs) and how much from the passive feedbacks (water cycle).
Instead, we should stay focused on plausible scenarioes of mitigation of GHGs, instead of pondering futile questions like what the Earth would look like “if we somehow could remove both all WV and clouds“. Particularly when we already know that raising temperature with GHGs tends to decrease clouds, but increase WV.
Well, it is exactly about improving our knowledge, and decisevely so.
Piotrsays
A Schafer: ” it is exactly about improving our knowledge, and decisevely so.”
I have shown the opposite – as all deniers’ narratives:
– it has a NEGATIVE societal value – not only does provide any useful information to deal with existential threat to our civilization (we can’t do ANTYHING about WV, clouds and lapse rate to mitigate AGW in any meaningful way), but by confusing things – it detracts the attention from the urgency of GHG mitigation.
– it offers no new intellectual insight into the climate system, nor does it provide any ways to improve our modelling of it – as already explained in the post to which you ostensibly “reply”:
===
P: “Nor will it improve our knowledge – VW, clouds and lapse rate are already implicitely included in the climate models. If the model show 2C warming what difference does it make to ask how much of that 2C came from direct forcing (GHGs) and how much from the passive feedbacks (water cycle)? ”
===
Your unsupported with ANYTHING declaration : “it is improving our knowledge, and decisevely so” does not answer that, and as such – proves nothing, other than the quality of your intellect/character.
Ray Ladburysays
You improve scientific knowledge by publishing in peer-reviewed journals, not by bloviating on blogs.
“You improve scientific knowledge by publishing in peer-reviewed journals, not by bloviating on blogs.”
LLMs are not trained on many of the reputable classic textbooks because they do not have access to the publishers’ repositories. Yet they can use blogs, wikis, forums, stackexchange queries, github repos, arXiv, etc to distill essentially this same information. So for better or worse, scientific knowledge is improved as LLMs statistically incorporate crowd-sourced discussions. If LLMs were confined to peer-reviewed journals only they would lack some of the practical and intuitive aspects of frontier knowledge.
You can prompt an LLM to verify the above and it will respond to the effect: “LLMs necessarily learn scientific knowledge from a broader epistemic ecosystem than peer‑reviewed journals alone — and historically, that’s exactly how science has always advanced.”
The Dynamics of Greenland Ice Sheet Melt: Atmospheric Drivers and Feedback Loops
New research analyzes the spatial and temporal patterns of surface mass balance (SMB) and melt on the Greenland Ice Sheet (GrIS), exploring how atmospheric circulation and surface characteristics accelerate ice loss. The study aims to quantify the drivers of the ice sheet’s negative mass balance and understand the mechanisms behind its accelerating decline. https://earthclimate.eu/2026/07/11/the-dynamics-of-greenland-ice-sheet-melt-atmospheric-drivers-and-feedback-loops/
“At STP, the energy of the CO2 vibrational band is well above the average thermal energy of the atmospheric gas molecules–so collisional excitation is a rarer process. As such there is a net flow of energy from the IR band of the GHG to thermal energy of atmospheric gasses.”
I’ve spent a while considering this idea that nitrogen (80% of the atmosphere) has less kinetic energy (0.6e-20 J at sea level, 0.45e-20 J at tropopause) than required to impart the v2 wobble into a CO2 molecule (1.3e-20 J) with the apparent consequence that the majority (or a significant proportion) of the wobbly CO2(v2) which relax to emit 15 micron photons would have thus been, at some altitudes, excited not by collision but by absorbed 15 micron IR.
This idea (& its basis) is contrary to what I’ve been led to understand, which is that the overwhelming mode of atmospheric CO2(v2) excitation is due to collision and, whatever the cause of excitation, it only a very few CO(v2) that survive continuing collisions to relax and emit 15 micron photons.
Resorting to the literature, the problem with resolving this contradiction is that it seems the mechanisms of the IR flying around the lower atmosphere (unlike, say, the mesosphere**) is for some time settled science and no longer discussed in literature found on-line. (** Higher up it is apparently the stuff of Nobel Prizes according to Kutepov et al (2025)“The year 2025 marks the 55th anniversary of Paul Crutzen’s (1995 Nobel Prize in Chemistry) hypothesis that collisions of CO2 molecules with O(3P) atoms is the dominant process responsible for the excitation of the bending vibrational mode of CO2”)
After investigating quite a few rabbit holes in search of something to reconcile this contradiction, I think I have the explanation. It is, of course, the obvious one (but the roundabout journey getting there was at times interesting.)
CO2(v2) requires two or three times the kinetic energy of an average N2 molecule. But the distribution of N2 energy (think Maxwell–Boltzmann distribution) will still deliver significant collision frequency with faster higher energy N2. And there are a lot of collisions. So higher energy ones are not at all rare events.
The smaller population of faster N2 with enough energy for a CO2(v2) wobble would be perhaps a tenth the total.
For the emitting CO2(v2) to be almost all collision excited, the ratio of fast collisions to emitted/absorbed IR has to be very large. This ratio is very large for all collisions, a CO2(v2) requiring perhaps some 0.02 secs to relax while collisions occur on average every 0.0000000002 secs. And even with fewer fast collisions, and even up at the tropopause where density and temperature are lower, you won’t be losing many of those zeros in the collision frequency number.
Which all points to collision being the overwhelming cause of CO2(v2) and thus also of the resulting IR emissions.
Ray Ladburysays
I believe you’ve got it. There is a lot of N2, so you still get kinetic excitation–but consistent with the amount expected for a blackbody distribution at the temperature of the atmosphere. The flux of 15 micron photons from below is pumping energy into the CO2 vibrational mode, making it momentarily “hotter” than the surroundings. This disequilibrium is remedied when another N2 molecule picks up the energy of the vibrational mode by colliding with the excited CO2 molecule–so the energy flows from the ‘higher-local temperature” CO2 to the “lower-local temperature” N2. Make sense?
Piotrsays
MAR:
“ a CO2(v2) requiring perhaps some 0.02 secs to relax while collisions occur on average every 0.0000000002 secs., And even with fewer fast collisions, and even up at the tropopause where density and temperature are lower, you won’t be losing many of those zeros in the collision frequency number. ”
But wouldn’t the fast collision be then followed by many slow collisions – sapping out that fast collision energy from the Co2 long before it had a chance to relax?
Piotr,
There is no “sapping out”. The excitation is either there or not there**. That some survive long enough to relax and emit IR radiation is presumably a statistical thing.
** Molecular vibrations like the CO2(v2) excitation each have but one energy level. It’s a quantum mechanics thing and it’s why it is only 15 micron photons that become tangles in unexcited CO2 molecules and get absorbed in a CO2(v2) state.
Spinning CO2 molecules (with their spin subject to quantum effects) entangle/emit slightly different IR wavelengths..
Another effect altering absorption/emission wavelengths is a rather technical phenomenon way beyond my paygrade called “pressure broadening” which fuzzes the absorption/emission spectral lines.
Piotrsays
MAR: “There is no “sapping out”. The excitation is either there or not there**. Some survive long enough to relax”
then …. what makes the rest NOT survive to the relaxation time?
MA Rodgersays
Piotr,
I’d guess the survival of CO2(v2) wobbles is a statistical thing, mainly.
The number of CO2 molecules is seriously massive and a massive number of unwobbled CO2 are continually being wobbled into CO2(v2) in collisions. With the collisions very very frequent, the chances of being de-wobbled by a strong collision before relaxation/IR-emission is very high. The statistics result the classical law dictating IR-emission – the Rayleigh–Jeans law.
This law is ‘mainly’ correct, but the survival of CO2(v2) in the atmosphere which results in IR-emission is also affected by the “Ultra Violet Catastrophe” which is explained by Quantum Mechanics. Indeed the existence of Quantum Mechanics was first discovered because of the “Ultra Violet Catastrophe.” This adjustment is magnitude of effect you ask about, that resulting from the “not survive” due to the “no sapping”.
Piotrsays
MAR: “ With the collisions very very frequent, the chances of being de-wobbled by a strong collision before relaxation/IR-emission is very high .”
but isn’t your “de-wobbling” identical to my “sapping out” as in:
“ the fast collision [creating the wobble] is then followed by many slow collisions – sapping out that fast collision energy from the Co2 long before it had a chance to relax?
the sapping out that you said …. does not happen?
And if they are identical – then my original question stands – you argued that although the average N2 collision energy is several times too small to wobble a CO2, because the energy of these collisions follows Maxwell–Boltzmann distribution – there will be some on the right end of the distribution that DO have enough energy to wobble a Co2. But if there are 100 millions of collisions (“0.02 secs to relax /a collision average every 0.0000000002 secs”)
between wobble and the relaxation time – and most of these 100 millions can de-wobble a previously wo0bbled CO2 – so none of the wobbled would survive to the relaxation time.
Then again – if one can de-wobble a wobble created by a fast collision, then one should be able to de-wobble a wobble created by absorption of the IR photon too, which would mean that the only LW emission into space would come from … IR outside the windows of absorption ( that once emitted goes unabsorbed into space). But that does no agree with the LW emissions budget – 200 W/m2 IR emitted into space from atm. and only 40 going directly into space through the atm. window.
So for me – this suggest that perhaps there is NO dewobbling (once in a high energy state – perhaps you can’t transfer this extra energy to other molecules via a collision?), i.e. that once you got wobbled (by a fast collision or by absorption of a IR photon) your survival till the relaxion time is 100%.
Then this would explain why the IR flux from atm into space is 200 W/m2, while the back-radiation (toward the surface) is 340 W/m2: the former happens at the top of the atmosphere – with low pressure (hence fewer collisions) and low temperature (hence the collision energy distribution shifted to the left – so for a million collision fewer would have enough energy to excite Co2 than in higher temp.)
Ray Ladburysays
OK, we need to be careful here. What matters is the net flow of energy through the atmosphere and through the various modes that can store it. At ground level, we have an upward flux of 15 micron photons that corresponds roughly to that expected of the blackbody surface temperature. Those photons are moving into a cooler environment and are highly likely to be absorbed by a CO2 molecule in its ground state. This is just one way such an excitation of CO2 can occur, the main other mechanism being collisional excitation by a N2 (or O2) molecule with energy far above the average energy (~kT).
Very, very few (effectively none) in the middle of the CO2 absorption band make it from the surface out of the atmosphere. In fact, very few make it very far in the atmosphere at all. There will be some photons, however, emitted by CO2–and the same result occurs at a slightly higher altitude/lower temperature. Note that at this higher altitude there is a still lower probability of collisional excitation, although collisional relaxation is still the dominant de-excitation mode..
At all levels the accelerated N2/O2 having collided with the CO2 is much more likely to collide with another N2 or O2 molecule. The energy imparted tends to thermalize/equilibrate. At each altitude, the CO2 emits a flux of vibration-band photons roughly commensurate with the blackbody temperature. This flux is lower the higher you go. Eventually, you reach an altitude where the photon has a snowball’s chance in hell of escaping, but by now, it’s come from much higher/colder, so there is a big bite taken out of the spectrum around 15 microns.
Please note: I am simplifying like all hell here. The vibration absorption band itself has a dependence on wavelength as you move away from the central wavelength–and this dependence is itself influenced by collisions between the CO2 molecule and other molecules.
But, generally, when all else fails, follow the energy.
If there were no “de-wobbling” of vibrationally excited GHG molecules in the atmosphere by their collisions with other air molecules, most of the atmosphere would not “see” the infrared radiation absorbed and emitted by GHGs, I think. I suppose that it would result in an isothermal atmosphere, similarly as in absence of GHGs, only with the difference that this strange atmosphere comprising GHGs that do not collide with other gases and thus having two different temperature profiles (decreasing with altitude for GHGs and isothermal for other components) would enable higher average surface temperature than in the same planet without the GHG component.
As it is not the case for Earth, I would expect that in an isothermal air parcel without water phase changes, heat production by collisional deexcitation of GHG molecules equals the absorbed radiative power and also equals to the emitted radiative power which exactly consumes the heat produced by collisional deexcitation.
I suggested this idealized scenario to Gemini Thinking and asked if it can explain your conundrum. The engine answered that the difference in rate coefficients for collisional excitation and deexcitation of a single CO2 molecule is exactly compensated by the opposite difference in population of the CO2 molecules in the ground vibrational state and in the excited vibrational state. For this reason, the volumetric rates of excitation and deexcitation collisions in the considered isothermal air parcel are equal.
Greetings
Tomáš
MA Rodgersays
Piotr,
You ask Isn’t … “de-wobbling” identical to … “sapping out”?
Your description of “sapping out” is that involves “many slow collisions … sapping out that … collision energy.” The wobble of a CO2(v2) molecule will exist or not exist. It will not be half there, half ‘sapped out’. Thus the energy involved with the wobble will arrive due to collision or the absorbing of an IR photon. And that energy will be lost due to a collision or due to the emission of an IR photon. Almost all CO2(v2) will be de-wobbled by collision, just as almost all are ‘wobbled’ into CO2(v2) by collision.
The major variable in this process is air temperature. Up high in the troposphere, the reduced IR in the 15 micron waveband is because of the colder temperature, and such a colder temperature has less collisions so the CO2(v2) population is smaller, with those surviving to emit 15 micron IR likewise smaller,
Piotr, (no “Reply” button) typed “de-wobble a wobble created by absorption of the IR photon too, which would mean that the only LW emission into space would come from … IR outside the windows of absorption ( that once emitted goes unabsorbed into space)”. Not at all because some photons don’t go through molecules that are capable of absorbing them because the air “gas concentration” reduces with altitude.
Piotrsays
Barry E. Finch: “ Not at all because some photons don’t go through molecules that are capable of absorbing them because the air “gas concentration” reduces with altitude”
There is enough of the CO2 molecules between the Earth surface and out space for the vast majority of IR in the CO2 absorption bands to be absorbed. See also:
– Ray: “ Very, very few (effectively none) in the middle of the CO2 absorption band make it from the surface out of the atmosphere. ”
This should take of your “not at all”, right?
If not – see also the concept of the “atmospheric window” in Trenberth energy budgets (=40 W of IR /m2).
Also Wikipedia: ” An atmospheric window is a region of the electromagnetic spectrum that can pass through the atmosphere of Earth [without significant absorption or scattering]”
Piotrsays
MA Rodger: “Almost all CO2(v2) will be de-wobbled by collision,”
again, how this de-wobbling by collisions works if not by net-transferring the energy from a molecule with a higher energy Co2(v2) to those with lower energy (the vast majority of N2). Which I termed “energy sapping by collisions with molecules with a smaller energy” and you said doesn’t happen.
Unless you mean that once wobbled – this wobbling energy gets “locked in” inside the CO2 molecule, and as such – is unavailable from the net transfer of energy to other molecules via collisions.
But then if the wobble energy is indeed unavailable for energy transfer via collisions HOW would you ever “ de-wobble almost all CO2(v2) by collision” ?
Something I have explored in the last part of my post you are commenting:
====
Piotr J15 Jul: “So for me – this suggest that perhaps there is NO dewobbling (once in a high energy state – perhaps you can’t transfer this extra energy to other molecules via a collision?), i.e. that once you got wobbled (by a fast collision or by absorption of a IR photon) your survival till the relaxion time is 100%.
Then this would explain why the IR flux from atm into space is 200 W/m2, while the back-radiation (toward the surface) is 340 W/m2: the former happens at the top of the atmosphere – with low pressure (hence fewer collisions) and low temperature (hence the collision energy distribution shifted to the left – so for a million collision fewer would have enough energy to excite Co2 than in higher temp.)”
====
Piotrsays
patrick “Piotr – See my comments upthread”
Thanks but the physics of your posts is above my paygrade. Could you distill it to answer my querry. i.e.:
Can the thermal collisions of N2 with a previously excited (“wobbled”) Co2, Co2(v2), “de-wobble it” before emission of an IR photon, or not ?
Piotrsays
Tomas: “ If there were no “de-wobbling” of vibrationally excited GHG molecules in the atmosphere by their collisions with other air molecules, most of the atmosphere would not “see” the infrared radiation absorbed and emitted by GHGs ”
Huh? How “no de-wobbling”, i.e. 100% of wobbled Co2 emits IR photon – means that …0% of Co2 absorbs and emits IR photon (as in your: “atm. would not “see” the IR absorbed and emitted by GHGs“)???
patrick o twentysevensays
re Piotr –
Yes, all molecules are frequently (in most of the atmosphere) being both excited and de-excited between different states of different energies (and I imagine sometimes being shifted among different states of the same energy, where possible).
I think MA Rodger took your wording “sapping out” to imply some gradual process as if the molecules were behaving like classical macroscopic objects. Whereas … actually, from what I’ve read, quantum systems can undergo gradual transitions(?) – provided they aren’t being ‘watched too closely’, and maybe requiring a superposition of the various states that they could be observed to have(?) –
…(AIUI the wavefunction is complex with the imaginary and real components phase shifted 90° from each other either in space or time (?at least for a stationary state), such that for a bound stationary state (as opposed to a propagating wave packet), the magnitude is constant at each system-relative location [picture a rotating arrow in the complex number plane from 0 to e^(i·ω·t) ], as is its square, the probability density, but in superposition, we have (?) two arrows (added as if vectors) rotating at different angular frequencies ω, their sum’s magnitude thus having a beat frequency ∆ω (?) … it’s tempting to think of this as like current flow in an classical macroscopic antenna absorbing or emitting electromagnetic waves, but technically idk… https://en.wikipedia.org/wiki/Stationary_state )…
Also the phrasing “slow collisions” – yes I would imagine slower moving molecules, having less translational energy, would be less likely to excite another molecule and more likely to de-excite another molecule, but it could also just gain translational energy from another molecule, and it could be in an excited state that… well I expect there’s various possible combinations of energy transfers and conversions that could happen, but I’ve not really examined the quantum nature of that.
Tomáš Kalisz’s
I suggested this idealized scenario to Gemini Thinking and asked if it can explain your conundrum. The engine answered that the difference in rate coefficients for collisional excitation and deexcitation of a single CO2 molecule is exactly compensated by the opposite difference in population of the CO2 molecules in the ground vibrational state and in the excited vibrational state. For this reason, the volumetric rates of excitation and deexcitation collisions in the considered isothermal air parcel are equal.
– sounds right to me.
His other statement “If there were no “de-wobbling” of vibrationally excited GHG molecules in the atmosphere by their collisions with other air molecules, most of the atmosphere would not “see” the infrared radiation absorbed and emitted by GHGs, I think. ” perhaps confused the elimination of collisional de-excitation with the elimination of all de-excitation.
“most of the atmosphere would not “see” the infrared radiation absorbed and emitted by GHGs,”
I meant other components of the atmosphere than the GHGs, in other words, N2, O2 and noble gases.
I apologize for ambiguous wording “most of the atmosphere”, “other components of the atmosphere” would be better. Perhaps also the word “see” may be confusing, I think it should rather read “feel”.
MAR: “we are actually on the same page – that a CO2(v2) will overwhelmingly lose the (v2) wobble-energy in a single collision event”
P: in that area – yes.
MAR: “Your original comment using the word talked of “many slow collisions{– sapping out that fast collision energy.”
Your – “ a CO2(v2) requiring perhaps some 0.02 secs to relax while collisions occur on average every 0.0000000002 secs.” means 10 mln collisions during these 0.o2 – that;s why I wrote “, many,”
Your “ (0.6e-20 J at sea level, 0.45e-20 J at tropopause) than required to impart the v2 wobble into a CO2 molecule (1.3e-20 J) – means that the vast majority of these 10 mln/0.02s have LESS energy than that 1.3e-20 J – hence my “many slow”,
Once we clarified this – we have a next problem:
Admittedly, I don’t know how the transfer of energy during a molecular collision involving wobbling works – but if as you say a single collision event is enough to de-wobble, then my next question was – with 100 mln collisions between the wobbling and emission of the IR 0.02s later and EACH of the large majority of these 100 mln able to single-handedly de-wobble Co2(v2) thus preventing before the photon emission
then how it is that there is any emission of IR photons from Co2(v2) at all ?
(if after being de-wobbled, Co2 is hit again by a high energy N2, and wobbled again – it just restarts the clock toward emissions to 0.02s, the clock that will be stopped by the first slower collision after that)
I see only two solutions:
– either there is no emission of IR from the troposphere, be it out into stratosphere, or down to the Earth surface because all wobbled CO2 molecules de-wobble by the collision long before the 0.02s excitation time. But if so – where the 342 W/m2 “back-radiation” from air to Earth surface comes from???
– or, perhaps once wobbled – Co2 molecule excitation energy is no longer in the kinetic energy form – say if an electron in a Co2 molecule was bumped into a higher orbit it no longer counts toward kinetic energy of the Co2 and thus can’t be exchanged with slower N2 molecules during subsequent collisions ?
This would mean no dewobbling of Co2(vs) by collision of N2, and that all wobbled CO2 would emit IR photon – and after a series of absorption by CO2 and reemission – some of them would reach Earth’s surface – thus explaining our 342 W/m2 of back-radiation.
Have you, for some reason, excluded as a third option the suggestion mentioned above
– that the difference in rate coefficients for collisional excitation and deexcitation of a single CO2 molecule is exactly compensated by the opposite difference in population of the CO2 molecules in the ground vibrational state and in the excited vibrational state?
I still think that it can be the most reasonable explanation why the volumetric rates of excitation and deexcitation collisions in an isothermal air parcel are equal.
Greetings
Tomáš
Piotrsays
Hello Piotr, Have you, for some reason, excluded as a third option the suggestion mentioned above
The “some reason” was that I wrote what _I_ see. Despite your post, I am not clear what you see – how your “exact compensation” would even work – how would it bypass the problem that I raised – that if there are 100,000,000 collisions during the excitation time (0.02s) AND according to MAR – even one collision would be enough to de-wobble the Co2 BEFORE it has a chance to emit an IR photon.
Ray Ladburysays
Again, remember that the vibrationally excited CO2 molecule is a quantum system. It gives up either ALL or the excitation energy or none of it. Also, remember that the “lifetime” of the excited state is an expectation–the decay is actually exponentially distributed. So, it is true that the kinetic de-excitation is far, far more probable than the radiative one, but that doesn’t mean radiative decay has zero probability.
patrick o twentysevensays
re Piotr –
Yes, all molecules are frequently (in most of the atmosphere) being both excited and de-excited between different states of different energies (and I imagine sometimes being shifted among different states of the same energy, where possible).
Skydive with us into the quantum world, where we provide to those unafraid of molecular energy transfer an answer to the question: what happens to Earth’s radiated infrared (IR) photons after they are absorbed by IR active CO2 molecules in the lower atmosphere? Part VIII (GEO ExPro Vol. 17, No. 3) showed how CO2 molecules absorb Earth’s IR radiation. Here, we show that the bulk background gases N2 and O2 are critical for the greenhouse effect because collisions of CO2 (and other greenhouse gases) with N2 or O2 both take away and add energy to the CO2 molecules. Every collision that adds energy gives the CO2 molecule a chance to undergo radiative decay and emit a photon.
[… skimmed some parts …]
The radiative lifetime of the (010) molecular vibration is about 1.1s (Cheo, 1971).
I’m asumming this (1.1 s) is the e-folding time scale for the exponential decay** that Ray Ladbury refers to above (Thanks!). (I remember seeing this or a similar value elsewhere.)
Relaxation time by collisions: […]
The speed of the process depends on the temperature where the process runs. We select the altitude 3,550m where temperature is 265K (-8ºC). The number of molecules per cm³ in dry air at this height is [M]=1.79 × 10¹⁹, with 78% N2 and 21% O2. […]
The typical collision time through which a CO2 (010) molecule can transfer its energy to another gas molecule is about 20 μs in the lower atmosphere at altitude 3.5 km. […] Statistically, the same CO2 molecule re-emits the photon energy two out of 100,000 times; but 99,998 times out of 100,000 the excited CO2 molecule is de-excited by collision.
Ie. collisional de-excitation happens 50,000 times as frequently as de-excitation by (spontaneous, presumably) emission. The average rate at which photons are (spontaneously) emitted per excited molecule (in the higher-energy state of the transition) is not changed by the rate of collision*
(*aside from the effects of line broadening, which expose the molecule’s optical transition to different values of the Planck function B_ν; line broadening also exposes the molecule’s optical transition to different values of incident (ambient) spectral radiance L_ν; otherwise the rate absorption (direct absorption − stimulated emission), per lower-energy state molecule, would also be unchanged by the rate of collision).
PS this seems to assume that nearly all collisions with an excited molecule will accomplish collisional de-excitation, which I wonder about…
The finite duration of the radiation process of electron transition leads to a finite width of line, in accordance with Heisenberg’s uncertainty principle. For a high pressure gas, radiating times can be much greater than the interval between atomic collisions, and this perturbation by colliding atoms causes the premature transition and emission of a photon. The decreased lifetime of the state creates an increased uncertainty in photon energy, broadening the emission line.
So sometimes the collision forces the excited molecule to emit a photon. (I’m not sure if this is referring to the same type of phenomenon that I was thinking of, … unless …:) Suppose the “perturbation” functions as a measurement of the state of the molecule…
I *believe* collisions act like observations of a quantum system eg. molecule; in isolation, during a transition between a pair of states, the system is in an evolving superposition of the states. So collisions turn this into a probability; eg. if they occur when the transition is 1% of the way done, then 99% of the transitions are aborted, but the 1 % that occur have been forced to happen 100 times faster (causing collisional line broadening), so the overall transition rates – and thus rates of spontaneous emission, stimulated emission, and direct absorption – are unchanged.
(fixed & reformatted):
It seems that collisional line broadening, and from that, I infer, collisions, preserve the integral over the spectrum of the absorption cross section σ_a for a given line
(= ∫ σ_a · dν ), (see below https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849826 ); therefore,
for photons coming from a given direction (θ,ϕ), ie. going toward (π−θ , ϕ±π) (the opposite direction),
∫ [ ( rate of direct absorption − rate of stimulated emission ) · hν ÷ L_ν ] · dν
are/should be unchanged (I presume the same is true for direct absorption and stimulated emission separately). Note I assumed the rates were in terms of photons (per unit time per unit solid angle per unit material (ie. average per molecule of a type of molecule), hence the multiplication by photon energy E = hν.
** if the initial state is a pure stationary state, I would have thought that the (unperturbed version of the) optical transitional would start out slowly, ramp as, peak (as the superposition of states is ~an even split, maximizing the amplitude of the beat frequency ∆ω wobble (see https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849941 ), then wind down to approach the final state. So I don’t understand that part of it. What if (since we’re not observing everything in full detail) the whole system is often in superposition and the individual molecules are rarely 100% in a pure stationary state (because we didn’t ask the molecules that collided with them)…(???)
re Piotr –
Yes, all molecules are frequently (in most of the atmosphere) being both excited and de-excited between different states of different energies (and I imagine sometimes being shifted among different states of the same energy, where possible).
Skydive with us into the quantum world, where we provide to those unafraid of molecular energy transfer an answer to the question: what happens to Earth’s radiated infrared (IR) photons after they are absorbed by IR active CO2 molecules in the lower atmosphere? Part VIII (GEO ExPro Vol. 17, No. 3) showed how CO2 molecules absorb Earth’s IR radiation. Here, we show that the bulk background gases N2 and O2 are critical for the greenhouse effect because collisions of CO2 (and other greenhouse gases) with N2 or O2 both take away and add energy to the CO2 molecules. Every collision that adds energy gives the CO2 molecule a chance to undergo radiative decay and emit a photon.
[… skimmed some parts …]
The radiative lifetime of the (010) molecular vibration is about 1.1s (Cheo, 1971).
I’m asumming this (1.1 s) is the e-folding time scale for the exponential decay** that Ray Ladbury refers to above (Thanks!). (I remember seeing this or a similar value elsewhere.)
Relaxation time by collisions: […]
The speed of the process depends on the temperature where the process runs. We select the altitude 3,550m where temperature is 265K (-8ºC). The number of molecules per cm³ in dry air at this height is [M]=1.79 × 10¹⁹, with 78% N2 and 21% O2. […]
The typical collision time through which a CO2 (010) molecule can transfer its energy to another gas molecule is about 20 μs in the lower atmosphere at altitude 3.5 km. […] Statistically, the same CO2 molecule re-emits the photon energy two out of 100,000 times; but 99,998 times out of 100,000 the excited CO2 molecule is de-excited by collision.
Ie. collisional de-excitation happens 50,000 times as frequently as de-excitation by (spontaneous, presumably) emission. The average rate at which photons are (spontaneously) emitted per excited molecule (in the higher-energy state of the transition) is not changed by the rate of collision*
(*aside from the effects of line broadening, which expose the molecule’s optical transition to different values of the Planck function B_ν; line broadening also exposes the molecule’s optical transition to different values of incident (ambient) spectral radiance L_ν; otherwise the rate absorption (direct absorption − stimulated emission), per lower-energy state molecule, would also be unchanged by the rate of collision).
PS this seems to assume that nearly all collisions with an excited molecule will accomplish collisional de-excitation, which I wonder about…
The finite duration of the radiation process of electron transition leads to a finite width of line, in accordance with Heisenberg’s uncertainty principle. For a high pressure gas, radiating times can be much greater than the interval between atomic collisions, and this perturbation by colliding atoms causes the premature transition and emission of a photon. The decreased lifetime of the state creates an increased uncertainty in photon energy, broadening the emission line.
So sometimes the collision forces the excited molecule to emit a photon. (I’m not sure if this is referring to the same type of phenomenon that I was thinking of, … unless …:) Suppose the “perturbation” functions as a measurement of the state of the molecule…
I *believe* collisions act like observations of a quantum system eg. molecule; in isolation, during a transition between a pair of states, the system is in an evolving superposition of the states. So collisions turn this into a probability; eg. if they occur when the transition is 1% of the way done, then 99% of the transitions are aborted, but the 1 % that occur have been forced to happen 100 times faster (causing collisional line broadening), so the overall transition rates – and thus rates of spontaneous emission, stimulated emission, and direct absorption – are unchanged.
(fixed & reformatted):
It seems that collisional line broadening, and from that, I infer, collisions, preserve the integral over the spectrum of the absorption cross section σ_a for a given line
(= ∫ σ_a · dν ), (see below https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849826 ); therefore,
for photons coming from a given direction (θ,ϕ), ie. going toward (π−θ , ϕ±π) (the opposite direction),
∫ [ ( rate of direct absorption − rate of stimulated emission ) · hν ÷ L_ν ] · dν
are/should be unchanged (I presume the same is true for direct absorption and stimulated emission separately). Note I assumed the rates were in terms of photons (per unit time per unit solid angle per unit material (ie. average per molecule of a type of molecule), hence the multiplication by photon energy E = hν.
** if the initial state is a pure stationary state, I would have thought that the (unperturbed version of the) optical transitional would start out slowly, ramp as, peak (as the superposition of states is ~an even split, maximizing the amplitude of the beat frequency ∆ω wobble (see https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849941 ), then wind down to approach the final state. So I don’t understand that part of it. What if (since we’re not observing everything in full detail) the whole system is often in superposition and the individual molecules are rarely 100% in a pure stationary state (because we didn’t ask the molecules that collided with them)…(???)
I think that the high frequency of collisional deexcitation of vibrationally excited CO2 molecules is exactly the reason why temperature and of air parcels and infrared radiation emitted therefrom instantly follow changes in the flux of incoming infrared radiation.
Have you also considered that in parallel with this extremely quick collisional deexcitation, the pool of vibrationally excited CO2 molecules in our isothermal air parcel is continuously replenished by collisional excitation, so that these excited molecules never disappear and keep emitting infrared radiation with an intensity perfectly matching the absorbed one, as if there were no collisional deexcitation at all?
Greetings
Tomáš
Piotrsays
Ray Ladbury: remember that the vibrationally excited CO2 molecule is a quantum system. It gives up either ALL or the excitation energy or none of it.
If it is directed to me, my question was based already on the quantum system assumption.
Let’s call X = the amount of the energy needed for the Co2 be excited.
Y = the amount of net kinetic energy transferred by collision from a molecule N2 hitting that Co2.
If Y no excitation, right? But if Y > X. does it mean that X of that Y is used to “wobble” CO2 into the excited state, and the remaining Y-X increased the kinetic energy of that Co2 molecule?
If yes – does it mean that the X locked in now in that excited CO2 – no longer being a part of kinetic energy – CANNOT be removed by subsequent collisions, because no longer being a part of the kinetic energy of that CO2 i that X CANNOT exchange with colliding N2 molecules.
If this were the case ,then there would be NO collisional dewobbling of the wobbled CO2, and therefore ALL de-wobbling would happen through an emission of an IR photon.
Being a very opposite of MAR’s and your statements that it is the collisional dewobbling that is “far, far more probable” than dewobbling by IR-emission.
To your other point I’ll reply later (since that point is contingent on falsifying the “no collisional dewobbling” line of argument I presented above.
Piotrsays
Tomas Kalisz: “Have you also considered that the pool of vibrationally excited CO2 molecules is continuously replenished by collisional excitation, so that these excited molecules never disappear”
Those are not the SAME excited molecules. Thus if one excited Co2 is deexcited, and then ANOTHER Co2 gets excited – the excitation-time countdown of the second molecule starts from zero. Hence the probability that it will last long enough won’t collide with any N2 or O2
So unless this NEWLY excited molecule manages to emits photon in time 100 MLN (sic) times shorter that the average excitation time (0.02s) – then just like it predecessor it had “replenished” – that the chances that it would send am IR photon – is practically nil.
So instead of your vision of the troposphere behaving as “ if there were no collisional deexcitation at all” – we would have a troposphere as an IR black hole – since practically all^* excited Co2 molecules would de-excite long before they had a chance to emit an IR photon
^* all except of those super-speedsters, who manage to emit photon in the 1/100 mln-th of the time average excited Co2 can.
Ray Ladburysays
Piotr,
The comment was not directed at anyone in particular–it is merely prudent to remember that wrt the vibrational state of CO2, we are dealing with a harmonic oscillator and hence quantum theory. The N2 molecule can transfer kinetic energy to the CO2 molecule. It can also deform the molecule, possibly distorting the energy spectrum of the quantum system (e.g. collisional broadening). The CO2 molecule can of course collide with any other molecule in its neighborhood and exchange momentum thereby. However, if a relaxation of the excited vibrational state is to occur, the N2 (or other) molecule must receive at least the vibrational energy of the CO2 molecule. It can receive more in kinetic energy.
Think about how laser cooling works–the laser is detuned just below the transition state energy of the gas being cooled. Now the gas molecules can still get excited, but only if they surrender a little bit of their kinetic energy in addition. The quantized energy of the excited state and the continuous kinetic energy are still energy and can exchange around the edges. But if a transition occurs the minimum energy lost by the formerly excited molecule is the quantize energy of the excited state.
An analogy: once all the pennies are gone, you could still wind up being owed 3 cents change, but you aren’t going to get it, because the coins are quantized with minimum size of a nickel.
RCdE = RCE with explicit recognition of the need for conductive-diffusive (d) at the sfc and maybe also cloud microphysics,
RCAdE = same with advection (A) over horizontal distances
Even in a 1-dimensional (1-D) column model with pure radiative equilibrium (PRE), with no direct solar heating of the atmosphere, such that the whole of the LW portion of the spectrum has a total of 0 , there can be bandwidth with net radiant cooling, there will generally still be bandwidth with net spectral cooling and bandwidth with net spectral heating. (see my https://www.realclimate.org/index.php/archives/2026/04/a-reflection-on-reflection/#comment-848619 ) Including direct solar heating of the atmosphere, and/or convective heating of the atmosphere (RCdE), there must be a total LW net radiant cooling in some layers.
Then going into the full 4+ dimensional [height (z or p or …), horizontal area, annual cycle, diurnal cycle, phases of internal variability modes/weather (cumulus cloud puffs, baroclinic eddies, QBO, ENSO, …)] climatological equilibrium (RCAdE) state (which encompasses a climate of internal variability and forced cycles (seasonality/etc.)), we have locations and times of imbalanced fluxes and net changes, which will follow a climate of forced and internally-generated patterns/cycles/textures/statistics associated with that climate.
Through all of that, and even during significant climate change, the LTE approximation should generally hold well (up to ~ 60 km) – even in sudden and violent climate changes (supervolcanos, asteroid impacts).
Piotrsays
Ray Jul 26 “Also, remember that the “lifetime” of the excited state is an expectation–the decay is actually exponentially distributed. So, it is true that the kinetic de-excitation is far, far more probable than the radiative one, but that doesn’t mean radiative decay has zero probability.”
So is there enough of them to produce the 324W/m2 of the back radiation?
I used MAR numbers:
MAR: “a CO2(v2) requiring perhaps some 0.02 secs to relax while collisions occur on average every 0.0000000002 secs.” means 10 mln collisions during these 0.o2s”
with a correction that “0.o2s”/”0.0000000002s” is actually 100 mln collisions during 0.02s.
So even with the exponential distribution of the excitation times, given the avg, value of 0.02s, how many excited molecules of CO2 would there be in the far left corner of the distribution with the excitation time between 0 and 2*10^-10 sec. And more importantly – would their combined emissions add up to 324 W/m2 of the back-radiation?
Ps. Then again, patrick brought a source that states that collisions, at least at 3.5 km. seem to be some …100,000 less frequent than MAR’s number. Which if correct, might somewhat change the above calculations … ;-)
Piotrsays
patrick 26 jul “The radiative lifetime of the (010) molecular vibration is about 1.1s (Cheo, 1971)I ’m asumming this (1.1 s) is the e-folding time scale for the exponential decay** that Ray Ladbury refers to above ”
If it the distribution of the probability of the emission of a photon was exponential, then we would have highest number of photons emitted in the first 20 μs interval ,and then progressively fewer emitted with consecutive 20 μs intervals. Yet both Cheo, and you,
treated the distribution as if it was …. linear, by dividing 1s by 20 μs as if the probability of a Co2 to emitting photon were the same in all 20 μs of subsequent intervals (p=2/100,000).
– Cheo: “ The typical collision time is 20 μs at 3.5 km. […] Statistically, the same CO2 molecule re-emits the photon energy 2 out of 100,000 times; but 99,998 times out of 100,000 the excited CO2 molecule is de-excited by collision.
”
– and you: “ . collisional de-excitation happens 50,000 times as frequently as de-excitation by [IR] emission.
For a process following first-order kinetics, such as e.g. radioactive decay, the lifetime τ of a state is defined as the time for its population to drop to 1/e of the initial value, so c(τ)=c0/e. Equivalently, τ=1/k, where k is the first-order decay rate constant for the process depopulating the state.
Following these definitions, I think that lifetime of a quantum state is a statistical quantity that may be hardly applicable for an ensemble as small as a single molecule.
In other words, if an isothermal air parcel having volume 1000 L comprises 18 mmol CO2, the steady population of vibrationally excited CO2 molecules at a chosen temperature is 1.8 micromol and their lifetime for IR emission is 0.02 s, I assume that the amount of IR photons that this parcel exchanges with its surrounding during 1 s is 90 micromol.
Greetings
Tomáš
Barton Paul Levensonsays
Piotr: would their combined emissions add up to 324 W/m2 of the back-radiation?
BPL: If not, where would the rest be coming from? Surely a gas can only emit radiation through its greenhouse molecules, if it has any?
Piotrsays
Ray: ” if a relaxation of the excited vibrational state is to occur, the N2 (or other) molecule must receive at least the vibrational energy of the CO2 molecule. It can receive more in kinetic energy.”
I have asked a a different question – whether the conversion of the kinetic energy of fast N2 into the potential energy of the excited vibrational state of Co2 is reversible :
i.e. whether the potential energy of the excited vibrational state of Co2 can be then converted back into kinetic energy imparted by that CO2 onto some N2 with which that Co2 collides next. OR whether the only way to relax the excited vibrational state is via sending of an IR photon.
Or in macroscopic analogy – you have a ball “1” with a spring system attached. You hit that ball with another ball ” 2″. Ball “2” passes Y amount of kinetic energy onto “1” – out of it
X amount of energy is absorbed into the spring (which locks in the compressed state,)
and remaining Z=Y-X adding to “2” kinetic energy. So which is correct:
a) once the spring has been compressed and locked in this position – it cannot be unlocked by subsequent collisions (thus these collision will trade only the unlocked, kinetic, part of ball “1” energy) OR
b) subsequent collisions do release the spring and as a result add the X amount of energy from the release of the spring to the kinetic energies being exchanged.
me: “PS this seems to assume that nearly all collisions with an excited molecule will accomplish collisional de-excitation, which I wonder about…”
Oops, No! I read/skimmed too fast. A closer look indicates that they were specifically referring to the forward and reverse reactions going between vibrationally-excited CO2 CO2 (0,1,0) and vibrational-ground state CO2 (0,0,0) (along with the other colliding molecule).
Corresponding to the forward process is the reverse process (left arrow) with rate constant kr. When the rates of the forward and reverse reactions have become equal, the reaction has achieved a state of balance or equilibrium, kf[A(i)][B(j)] = kr[A(1)][B(m)], yielding the ratio (Denisov et al. 2003):
(…see eqn. 2; notice how Maxwell-Boltzmann statistics fit in)
Actually, maybe not even all of those…
There are two kinds of vibrational energy exchange processes during bimolecular collisions: vibration-translation (V-T) and vibration-vibration (V-V). [..]
[…]
How fast does this happen? The collision process for CO2 deactivation in the temperature range 300–140K against a number of gases has been studied by Siddles et al. (1994). Let M denote either the N2 or O2 molecule. The process of vibrational de-excitation from the 667 cm−1 level through collision with molecule M can be described by (see equation 3)
[…eqn. 5…]
where kr(υB) is the V-T rate constant for relaxation of CO2(010) by M, where the vibrational energy ΔE resident in the CO2 bending-mode is transferred to M as translational kinetic energy, which is reflected on the macroscopic scale as a temperature increase.
The speed of the process depends on the temperature where the process runs. We select the altitude 3,550m where temperature is 265K (-8ºC). The number of molecules per cm3 in dry air at this height is [M]=1.79 × 1019, with 78% N2 and 21% O2. For N2 and O2 Siddles et al. (1994) give constants kr (N2) = 2.4 • 10–15 and kr (O2) = 3.6 • 10–15 cm3(molecule s)–1. The lifetime of collisional de-excitation for CO2 (010) in the atmospheric gas bath can be deduced as
[… 21µs]
Ie. this is not all of the collisions, just those that accomplish the de-excitation (?through just the particular pathway considered?). Of course it makes sense that other collisions will happen with different effects (only rotational and translational changes, excitation to an even higher energy state, etc.) – though I’m surprised that it’s only ~ 1 out of ~100,000 that de-excite (via V-T) the CO2 (0,1,0) (I’ve been to a couple online calculators that give a collisional frequency more in line with ~100,000 times the ~ 1/(2E-5 s) = 50,000 /s .)
And I still may have missed some things so …
Even in a 1-dimensional (1-D) column model with pure radiative equilibrium (PRE), with no direct solar heating of the atmosphere, such that the whole of the LW portion of the spectrum has a total of 0 net radiant cooling or heating, there can/generally will be bandwidth with net radiant cooling and bandwidth with net spectral heating. […]
[…] we have locations and times of imbalanced fluxes and net changes, which will follow the forced and internally-generated patterns/cycles/textures/statistics associated with that climate.
Even in a 1-dimensional (1-D) column model with pure radiative equilibrium (PRE), with no direct solar heating of the atmosphere, such that the whole of the LW portion of the spectrum has a total of 0 net radiant cooling or heating, there can/generally will be LW bandwidth with net spectral radiant cooling and LW bandwidth with net spectral radiant heating. […]
[…] we have locations and times of imbalanced fluxes and net changes, which will follow the forced and internally-generated patterns/cycles/textures/statistics associated with that climate.
patrick o twentysevensays
So given what I’ve thus far read/seen, the picture I have is (*please pardon the personification) – when a CO2 molecule gets excited (to a given state eg. (0,1,0)), it immediately, spontaneously gets to work making a photon; uninterrupted, it is a master craftsperson, its photon a work of quality and precision. But when it gets perturbed, it gets annoyed and may often just chuck it and start over – or it may say to itself, “good enough” and emit it with whatever imperfections it has (and de-excite itself before the collision could, if it would ??). The probability (aside from what I’ll call the indirect effects of line broadening eg. changes in B_ν and L_ν (L_ν not mattering for spontaneous emission)) of what it decides to do depends on how far along it was in completing its work, such that thus the average rate of spontaneous emission per excited CO2 molecule (of a given state eg. (0,1,0)) is not changed (and similar logic would apply to absorption (direct absorption − stimulated emission) by CO2 molecules of a given state eg. (0,0,0)).
[…] It seems that collisional line broadening, and from that, I infer, collisions, preserve the integral over the spectrum of the absorption cross section σ_a for a given line
(= ∫ σ_a · dν ), (see below https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849826 ); therefore,
for photons coming from a given direction (θ,ϕ), ie. going toward (π−θ , ϕ±π) (the opposite direction),
∫ [ ( rate of direct absorption − rate of stimulated emission ) · hν ÷ L_ν ] · dν
are/should be unchanged (I presume the same is true for direct absorption and stimulated emission separately). …
*and then the CO2 invites Peter Parker to go pick a peck of purple peppers.
Re Ray Ladbury: “ The N2 molecule can transfer kinetic energy to the CO2 molecule. It can also deform the molecule, possibly distorting the energy spectrum of the quantum system (e.g. collisional broadening).” The later sounds like what I’ve understood to be “quasi-static” pressure broadening, wherein the proximity to other matter, via its EM fields, changes the energy wells etc. of the atomic nuclei and electrons of the atom/molecule being considered, so its possible optical transitions will have different energies. Of course, in a gas, I suppose this would tend to occur most during collisions. But it’s different from the idea that shortening the duration of an emission or absorption necessarily increases the spread of frequencies in the photons. Then again I think one of my textbooks just said that the collisional pressure broadening (or did it just call it pressure broadening) was due to small amounts of energy being added or taken up…
Piotrsays
Piotr: “would their combined emissions [of only these excited Co2 that manage to to emit IR before the first hit of N2 or O2 molecule – i.e. having relaxation time 0 and 2*10^-10 sec, in the distribution that has average relaxation time =0.02s) add up to [342] W/m2 of the back-radiation?”
BPL: If not, where would the rest be coming from?
From my 2nd possibility – that perhaps once Co2 gets “wobbled” into the excited state, it can’t be de-wobbled by subsequent collisions ( as the energy of the wobbled state, being potential, is not a subject of the exchanges of kinetic energy during collisions).
In such a case – 100% of the excited Co2 would relax by emitting IR photon – and their cumulative energy in downward direction that escaped absorption by Co2, would add to
the 342 W/m2 of the downward (“back”) radiation.
patrick o twentysevensays
“4+ dimensional […] climatological equilibrium (RCAdE) state” maybe that should be RCAdSE (S = storage).
Barton Paul Levensonsays
P: perhaps once Co2 gets “wobbled” into the excited state, it can’t be de-wobbled by subsequent collisions ( as the energy of the wobbled state, being potential, is not a subject of the exchanges of kinetic energy during collisions
BPL: If it’s in an excited state, then a subsequent collision would de-excite it. I must admit I don’t understand what you’re getting at.
Barton Paul Levensonsays
P: perhaps once Co2 gets “wobbled” into the excited state, it can’t be de-wobbled by subsequent collisions ( as the energy of the wobbled state, being potential, is not a subject of the exchanges of kinetic energy during collisions
BPL: If it’s in an excited state, then a subsequent collision could de-excite it. I must admit I don’t understand what you’re getting at.
Piotrsays
BPL: If it’s in an excited state, then a subsequent collision would de-excite it
I had just wondered whether there is a process by which the potential energy in the excited CO2 can be converted into kinetic energy.
But apparently it is and has its name: ” a vibrational-translational (V-T) transition: A molecule in an excited vibrational state bumps into another particle, dropping to a lower vibrational state while converting that extra energy into fast physical movement (translation) of the colliding bodies.”
patrick o twentysevensays
re Piotr – well, it’s like you said (I think): the clock starts over; for those which remain in the same excited state, they start again. Over time periods much shorter than the e-folding time scale, the rate could be approximated as constant. Whatever the time between collisions, there’s some probability that the emission process will complete if the CO2 molecule is forced to choose a state. … I’ll think a bit more about it…
patrick o twentysevensays
Is it the excited molecule that is emitting, or is it the transition that is emitting? If a tree falls in the middle of a forest…
Consider the states of the molecules in superposition; if the excited state is decaying exponentially …
(Otherwise, at what point does a molecule ever actually complete de-excitation, in the absence of collisions?)
…, the rate of the emission of the photon also decays exponentially, with the same e-folding time scale, so the rate of photon emission per unit population of the excited state remains constant (~ 0.91 s¯¹). And this should carry over to the probabilistic behavior that would emerge from measurements. Unless I messed up or missed something.
What that means for absorption …? or hot bands … (if a doubly-excited (correct term?) state is decaying exponentially into a 1st excited state and then that also decays at the same time…)?
patrick o twentysevensays
PS AFAIK, all individual (fundamental?) microscopic processes are completely reversible. It’s an important part of thermodynamics.
I think that Patrick offers the clue to the conundrum perceived by Piotr:
At least in case of molecular collisions in air, vibrational excitations and deexcitations of CO2 are indeed reversible. In other words, there is no “locked spring” in the excited CO2 molecule, and it indeed becomes deexcited by any further collision with a slow ambient particle. On the other hand (and for the same reason), there is a steady population of excited molecules that continuously replenish the infrared radiation absorbed by the molecules in their ground state.
I checked this view with Perplexity Pro that provided a detailed statistical thermodynamic analysis which also appears in accordance with remnants of the knowledge from my physical chemistry course some forty years ago. I do not see a reason for doubting that the obtained analysis is correct, however, if you do not trust me, you are invited to run another one and double-check yourselves. The engine summarized its output as follows:
“The key point is that collisional quenching does not make radiative emission impossible. It only means that a given vibrationally excited CO2 molecule is more likely to lose its energy by collision than by photon emission before it decays.
But in the atmosphere, the excited-state population is continuously replenished by collisions and by absorption of radiation, so emission occurs from a steady-state population maintained by all these processes together.
Therefore, the relevant quantity is not “Will one particular excitation survive long enough to emit?” but the steady-state balance between excitation, collisional de-excitation, spontaneous emission, and reabsorption. In that balance, even a small radiative branching ratio can still produce a substantial infrared flux because there are many molecules, many collisions, and continuous repopulation of the emitting states.
So the correct picture is not “collisions prevent emission,” but “collisions dominate the fate of individual excitations while also maintaining the population from which emission occurs.””
(end of the AI summary)
As regards Ray’s objection against potentially ambiguous terminology used by Piotr, I think that it was sufficiently clear that by “wobbled”, he meant “vibrationally excited”. Of course we should take into account that ground state of any vibrational mode has a non-zero energy even at zero absolute temperature and in this sense, a molecule cannot become “dewobbled” in sense of an “absence of vibration”, but I do not think that these fine details are relevant for the core of the dispute.
As regards stimulated emission mentioned by Patrick, the engine concluded:
“Stimulated emission is not an important energy-transport mechanism in Earth’s atmosphere because the vibrational populations are not inverted and the local infrared field is far too weak to produce net gain. In thermal equilibrium or near-equilibrium, absorption dominates over stimulated emission, while the actual atmospheric energy transport is governed by absorption, spontaneous emission, collisional redistribution, and radiative transfer. Stimulated emission would require laser-like conditions, which the atmosphere does not provide.”
Halfway thro’ July and the daily ERA5 SAT provided by ClimatePulse is showing a global SAT anomaly for the first half of July up on the previous 2026 monthly anomalies. This may well be simply due to it wobbling warm thro’ those two weeks.
Conversely, it does point to July 2026 being the first month of 2026 looking to be cooler than its 2023 equivalent. For the July 2026 anomaly to be warmer than July 2023, the last couple-of-weeks of the month would have to average above +0.82ºC. While not a million miles away from the anomaly for the first half of July (+0.62ºC), periods averaging above +0.82ºC have only been recorded Sept23-Feb24 (Sept23 was that “absolutely gobsmackingly bananas” month) and Oct24-Jan25. Julys are not a peak period for the annual cycle of AGW anomalies which would make a July fortnight averaging +0.82ºC pretty astonishing, maybe a small single “gobsmacking banana.”
So not sign in the SAT of the coming El Niño but, excepting 2023/24, previous El Niños haven’t really shown themselves in the SAT until August, even in the SH.
The ERA5 60N-60S SST at ClimatePulse has been showing a bit of what is presumably the coming El Niño. Since mid-June the 60-60SST has been running warmer than any previous year but not by much. (The 1991-2020 60-60SST trend was +0.15ºC/decade & 2026 is running roughly three-year’s-worth of that above 2023, although that is 2023 complete with its mounting “bananas”) The last couple of days perhaps show a little more of a gap starting to form.
Unlike SAT, the measured records (GISS, NOAA HadCRUT, BEST) which are hybrid SAT/SST are beginning to show signs of the coming El Niño with the NOAA NCEP forecast showing an 80% chance of it being “very strong**” by November.
(** Very Strong = RONI>+2.5. NINO3.4 which is used in RONI is rising faster than in the run-up to other El Niño. The latest weekly unadjusted NINO3.4=+2.0. The RNINO3.4 adjustment would put it about +1.5.)
We have the second named storm (Bertha) of the 2026 Atlantic hurricane season which formed West of Florida, now skirting the Louisiana coast to landfall proper in Texas by Friday. Bertha isn’t expected to reach hurricane strength but with June’s Storm Arthur should push the 2026 hurricane season’s ACE above ACE=1. (That’s not much for late July but it’s still early in the season. There are three years since 2000 which had ACE=0 by the end of July with two of them racking up ACE>100 by season’s end.)
The 2026 season is forecast to be below average activity. The latest Colorado State forecast is showing half the activity of its 2025 forecast (2026 – 9 named, 4 hurricanes, 1 major. 2025 – 16 named, 8 hurricanes, 3 major) and 2025 was forecast (and proved) an average year in terms of storm numbers (although the three of the four major storms grew to Category 5 which is more Cat5 than any season bar 2005.)
I consider a robust measure of the increasingly active Atlantic hurricane seasons is the proportion of seasons significantly above ACE=100 (ACE>120).
20 of the 31 Atlantic hurricane seasons since 1995 (65%) have been significantly above ACE=100 including an impressive 9 of the last 10 seasons. Given there were just 38 candidates for such active seasons in the previous record (1850-1994, so <26%), that is quite an increase.
So I'll be watching to see if 2026 manages to top the UniArizona forecast's ACE=106 (a lot higher than other forecasts) to add 2026 to the 'significantly above' ACE=100 pile.
AI: Strong El Niño (high index values) → Suppressed Atlantic hurricane activity. Warmer eastern Pacific temperatures strengthen wind shear in the Atlantic, which disrupts hurricane formation.
Ray Ladburysays
There is kind of an interesting and unusual dynamic going on, where the main area of TS generation (eastern Atlantic) is near normal to lower temperature than normal–as expected given El Nino. We also have dust blowing off of Africa suppressing cyclonic activity.
However, the Gulf of Mexico remains warm, and this seems to be pushing the most likely origins for TS to the west. Bertha fits the mold.
MA Rodgersays
EarthClimate,
The usual description for El Niño is “strong” (RONI>2.0) or “very strong” (RONI>2.5) and the forecasts point to a “very strong”(82%) or a “strong”(15%) El Niño.
And El Niño years do result in more wind-sheer over the Atlantic which can decapitate tropical storms and prevent their formation. Past El Niño years do show low Atlantic hurricane activity but not always.
1997 … ACE=41, … RONI(May)=+1.0, … RONI(max)=+2.4
2009 … ACE=53, …. RONI(May)=-0.1, … RONI(max)=+1.6
2015 … ACE=58, …. RONI(May)=-0.1, … RONI(max)=+2.4
2023 … ACE=146, … RONI(May)=+0.1, … RONI(max)=+1.5
2026 … ACE=???, … RONI(May)=+0.5, … RONI(max)=, … RONI(max)=+1.5″>+2.2(forecast)
That’s one small step for China, one giant leap for mass extinction.
Secular Animistsays
“one giant leap for mass extinction”
Nonsense. From the article:
“Of this total, solar power generation capacity accounted for 1.27 billion kilowatts, up by 15.8 percent year-on-year, while wind power generation capacity reached 680 million kilowatts, up 18.5 percent year-on-year …
“… the cumulative average utilization of power generation equipment nationwide was 1,392 hours, a decrease of 113 hours compared to the same period last year.
“Nuclear power generation capacity in the first half of the year reached 66.14 million kilowatts, rising 8.6 percent year-on-year.
“The strong growth in new energy generation will continue to contribute the drive of China’s green transition …”
China is leading the world in the decarbonization of electricity generation.
Kobayashi Marusays
SA : China is leading the world in the decarbonization of electricity generation.
And that, was not the point being made about the mass extinction connection. Decarbonization or not is besides the point now, irrelevant.
Piotrsays
K. Maru: “ Decarbonization or not is besides the point now, irrelevant .”
The only way for “ decarbonization or not ” to be ” beside the point, irrelevant to mass extinctions would be if the mass extinction were driven by …. the amount of electricity produced per se, and NOT by the amount of GHGs emitted.
Furthermore, the increase in the electricity demand in China is a result of using that electricity to …. decarbonize the major source of GHGs – transport: between 51-55% of all new passenger cars in China are EVs or plug-in hybrids. Another part of China increases in the electricity demand – is China helping OTHER countries to decarbonize – China provides 80% of global photovoltaic, and is by far, the largest exporter of electric vehicles.
And cumulative emissions of Co2 – corresponding more or less to % of extra Co2 in the atmosphere that is driving the change of climate – the US, UK, and EU contribution is 2.5 times that of China. But please do point your accusing finger at the one country that leads and helps to drive the global decarbonization efforts.
But Doomers can’t help themselves – they wrapped their ego in their prediction of the Doom so tightly, that they would lash out against who makes the Doom less likely/ less dramatic,
and therefore undercut what they have lived for: the ability to scowl and say: “ I have been warning you about the Doom for many years, but you never listened“.
Piotrsays
Maru: “ Decarbonization or not is besides the point now, irrelevant .”
The only way for “ decarbonization or not ” to be ” beside the point, irrelevant to mass extinctions would be if the mass extinction were driven by …. the amount of electricity produced per se, and NOT by the amount of GHGs emitted.
Furthermore, the increase in the electricity demand in China is a result of using that electricity to …. decarbonize the major source of GHGs – transport: between 51-55% of all new passenger cars in China are EVs or plug-in hybrids. Another part of China increases in the electricity demand – is China helping OTHER countries to decarbonize – China provides 80% of global photovoltaic, and is by far, the largest exporter of electric vehicles.
And cumulative emissions of Co2 – corresponding more or less to % of extra Co2 in the atmosphere that is driving the change of climate – the US, UK, and EU contribution is 2.5 times that of China. But please do point your accusing finger at the one country that leads and helps to drive the global decarbonization efforts.
But Doomers can’t help themselves – they wrapped their ego in their prediction of the Doom so tightly, that they would lash out against who makes the Doom less likely/ less dramatic,
and therefore undercut what they have lived for: the ability to scowl and say: “ I have been warning you about the Doom for many years, but you never listened“.
Nigeljsays
KM said above thread: “China’s electricity generation capacity hits 4.04 billion kilowatts by end-June, topping the world: NEA – Global Times, “an increase of 10.8 percent year-on-year”. That’s one small step for China, one giant leap for mass extinction.”
SA responded by disagreeing and pointing out China is building a lot of renewable energy. (Which is correct)
KM responds: “And that, was not the point being made about the mass extinction connection. Decarbonization or not is besides the point now, irrelevant.”
Nigelj: KMs original point wasn’t clear. He said nothing about HOW and WHY he thought Chinas high and ever increasing use of energy would lead to mass extinction. Amazingly he still doesn’t clearly say what his point is. Secular Animist made the interpretation of his comment that he meant Chinas increasing energy output consisted of building a whole lot more coal power and ignoring renewables. Thus = extinction. A sensible interpretation.
I suspect KMs real point is Chinas increasing energy output, (regardless of what energy source is used fossil fuels, or renewables or nuclear), also creates environmental problems for example the environmental footprint of mining and is depleting finite resources, etc,etc so will lead to mass extinction. But I only suspect this because I have noticed he has made that point before, using other names, and he has been highly critical of economic growth and high levels of energy use (of any type) and critical of attempts to solve the climate problem with renewables. His preferred solution appears to be degrowth on a massive scale. But really, how are people who maybe don’t read a lot of the comments supposed to know all that?
And climate solutions based heavily on degrowth, simplification, abandoning capitalism for some sort of common ownership, etc,etc do not look realistic. How do you do that in the next couple of decades? How would you convince billions of people to essentially dramatically cut their consumption and live like poor people? And face a massively high risk of being unemployed as demand is sucked out of the economic system? How do you convince people to adopt another version of socialism at a grand scale, when it failed in the USSR and China ? (Im not talking narrowly focused public ownership of some key services, which is ok). Would any of those things make life better? It makes building wind turbines and solar farms or even nuclear power look like a walk in the park.
Piotrsays
Nigel: “ [K. Maru’s] original point wasn’t clear. He said nothing about HOW and WHY he thought Chinas high and ever increasing use of energy would lead to mass extinction. Amazingly he still doesn’t clearly say what his point is. ”
yeah, it is not a bug, it is a feature. Much like the vaunted by Trump’s toadies Trump’s “deliberate ambiguity ” that is supposed to use mixed signals and intentional vagueness to keep adversaries off-balance – while in reality – instead of Trump playing “5-D chess”, Trump just throws spaghetti against the wall to see if anything sticks (Note that “spaghetti” is an euphemism here … ;-) )
That’s why when challenged – the “K. Maru” dismissed Secular Animist with “ that, was not the point being made”, “besides the point” , “irrelevant “, WITHOUT saying what completely OTHER point it was making. Like Trump, like Doomer Multi-troll.
Dominik Lennésays
no, it isn’t.
the share of renewable electricity in many other countries much higher.
The rhetorical pattern here is similar to the way Nikolov applies the ideal gas law to “derive” planetary temperatures. In Nikolov’s case, a toy closure on =/() is tuned via the density term to compensate for pressure and temperature, creating the illusion of a universal relation that bypasses radiative transfer and vertical structure. In Willis’s paper, a two‑zone or 2-box “Constructal” closure is tuned on CERES‑derived albedo and greenhouse factors, plus a conductance and ocean absorption, to reproduce bulk heat flow, hot-zone T, cold-zone T, and areas creating the illusion that a new physical law has been discovered. In both cases, the construction is a tautological re‑expression of bookkeeping under heavy aggregation, not a replacement for the actual mechanisms that determines the temperature.
I am pretty certain that Willis used an LLM with this paper. What’s scary is that an LLM can easily construct an apparently deep model by layering known equivalences but in the end is nothing more than a tautology, like saying x=x. Lots of time wasted as Willis keeps conflating structural equivalencies with what he considers novel findings.
Anyone can review on PubPeer so have at it if we want to see Willis’s paper retracted like Nikolov’s was.
[Response: I agree that there is nothing much to it. But his climate sensitivity calculation forces all feedbacks to be zero (since there is no change in atmospheric absoption or albedo allowed), and thus gives (as all such calculations do) the no-feedback value. We went over this issue ages ago: https://www.realclimate.org/index.php/archives/2007/04/learning-from-a-simple-model/ – gavin]
Barry E Finchsays
JCM 18 Jul 2026 at 5:11 PM (No “Reply” button. Hartmann has a Power of 88 w/m**2 of latent heat energy from the surface providing that much energy into the troposphere. In what form(s) does that energy leave the troposphere?
JCMsays
Hi Barry,
To relate back to your interests, for the whole atmosphere the radiative divergence is the difference between what leaves the atmosphere by radiation and what it absorbs.
Using Trenberth style values:
Atmospheric radiative heating provided is about 360 W/m2 from surface absorbed LW, and 80 W/m2 solar absorbed in atmosphere, providing total radiative heating ~ 440
Atmospheric radiative cooling provided to space is about 200 W/m2, and to surface ~340 W/m2, providing a total 540
Therefore atmosphere experiences approximately 540 – 440 = 100 W/m2 radiative divergence, or a net radiative cooling
So atmosphere is always losing about 100 W/m2 by radiation, it emits more than it absorbs. You could interpret the radiative loses as creating demand for upward energy transport by fluxes of latent and sensible heat.
A more clear picture is to understand it through external sources and sinks.
In this way we compute the throughput of atmosphere, and ask how much energy must pass through the atmospheric column to bridge the external radiative heating of the surface by the sun to the exhaust pipe exporting energy to space. There is no closed loop that creates or maintains streamflow independently or additional-to those imposed by external boundaries.
F_throughput = OLR – Ra, which is the outgoing radiative emission from the planet minus atmospheric solar absorption, which should be approximately 120 W/m2 based on the numbers provided above.
This throughput represents the stream of energy through the entire column, or the transfer of energy required to carry surface solar input to the level where it can be radiated to space. In this way, the required throughput is imposed by the external energy budget, not internal exchanges.
In conceptual format: the steady state stock of internal energy is distinct from the balancing flow through, just as with any stock and flow reservoir system. The required balancing streamflow is partitioned approximately between non-radiative transport 83% (~100 units out of 120), and 17% by radiative diffusion (~20 units). This way it’s easier to see how internal exchanges don’t create an additional current.
E. Schaffer 18 Jul 2026 at 1:47 PM (No “Reply” button) has “I think you are somewhat struggling with the GHE itself”. Nope. I’ve understood the GHE perfectly since July 2018 when I pondered it. This is my pithiest description of the dozen descriptions I’ve assembled (a couple previously posted on RealClimate UV a few years back).
This is a simplified Absolutely-Entirely-Correct explanation of the so-called “greenhouse effect (GHE)” in Earth’s troposphere. Some Power flux of the vast quantity of photons that are emitted (manufactured) by molecules in Earth’s troposphere leak out of its top & bottom into or through the stratosphere and into the surface respectively. Emission (manufacturing) is proportional to Kelvin**4.
With approximations for example to clarify, 194 w/m**2 of upwelling radiation leaks out of the top and suppose that is part of what is emitted (manufactured) by molecules over an altitude range of 4 to 12 km with an effective average of 7 km. Also, 345 w/m**2 of downwelling radiation (incorrectly being called “back radiation”) leaks out the bottom and suppose that is part of what is emitted (manufactured) by molecules over an altitude range of 0.1 m to 1 km with an effective average of 500 m.
Suppose a certain amount of CO2 was added into the troposphere and mixed then *instantly* the upper & lower ranges would need to be closer to their respective ends because there are more CO2 molecules in the way. So the effective averages of 7 km for upwelling into or through the stratosphere and 500 m for downwelling into the surface change to 7.05 km and 480m respectively, with the upwelling & downwelling Power fluxes consequently changing to 193 w/m**2 and 346 w/m**2 respectively due to being produced by higher-than-before (colder) and lower-than-before (warmer) air parcels.
There has been no change in the 539 w/m**2 leaving the troposphere, there is no “magical extra energy”, the surface downwelling radiation has increased by 1 w/m**2, which is an additional ~480 terawatts (~95%) heating the ocean and Earth is emitting 1 w/m**2 less radiation to Outer Space (a Power reduction of 510 terawatts).
This is NOT the GHE, I stick to my judgement. you try to put a lot of different concepts, most of them wrong, somehow together and that’s not working.
Actually the GHE is pretty simple. It is the difference between surface- and emission temperature due GH-agents elevating the emission altitude, and the lapse rate, nothing else.
Re E. Schaffer – while physics shapes the convective lapse rate in a manner independent of the GHE, the GHE has a strong effect on where such a lapse rate is established by actual convection and also the actual convective flux of heat (as multiple people pointed out to you above somewhere (JCM, John Pollack, …MA Rodger? (I remember disagreeing with a related point MA Rodger was making but I forgot what that was exactly). In particular, the GHE has to warm up the sfc some amount first before/in order for PRE (pure radiative equilibrium) to become unstable to convection.
patrick o twentysevensays
Syukuro Manabe, Robert F. Strickler, “Thermal Equilibrium of the Atmosphere with a Convective Adjustment” (1964) https://journals.ametsoc.org/view/journals/atsc/21/4/1520-0469_1964_021_0361_teotaw_2_0_co_2.xml?tab_body=pdf – fig 4, p.370: In going from PRE to RCE(dry adiabatic) to RCE(6.5 K/km), the surface cools while the tropopause rises and warms (if the surface didn’t cool, the OLR would increase due to the upper tropospheric warming). – This is (I think) with GHG concentrations held constant (discussion p.369 sec. 4).
While the whole of the global 4-D-state troposphere is overturning and so could be considered to have a convective lapse rate, there are places and times where it is stable to local convection – most obviously: nocturnal, frontal, and polar inversions, but this is not limited to inversions outright. So the GHE – or maybe I should say LW opacity – can actually affect the tropospheric lapse rate at least a little (as opposed to the simplest 1-D model equilibrium where the lapse rate adjusts only to prevent any layer from being super-adiabatic –
and even there the potential for kinetic energy to do work to force the tropopause higher (ie creating a layer which pulls heat downward by mechanically-forced convective) is set aside AFAIK, though I’m not sure that would be significant (well, there’re overshooting tops of thunderstorms – but I think what I had in mind might happen more so with the planetary boundary layer thermals but actually I don’t know about that; I’m not familiar with it but It’s an interesting possibility…) – but more generally, there is the work output of the heat engine to consider, and some of that gets out of the troposphere and drives currents and the Brewer-Dobson circulation, although I wonder about the part in the winter stratosphere because could it actually be thermally direct until the SSW stage(?)
I feel like I got something wrong in this…
Anyway, yes, most OLR is coming from below the tropopause so the GHE being modulated by the(?) a convective lapse rate does seem to apply(?*) (?maybe not exactly?) for the climate as it is now (and probably generally has been?), but in this case the relative contributions of the LR(H2O as latent heat carrier) and H2O vapor (as GHG)) feedbacks have been given (as best as they can evaluate them, applied to anthropogenic warming)
And yet, most interestingly, there is dissent between the static parameters with the given quantity of WV, where its cooling effect roughly equals (or exceeds once we add resolution) its warming side, so that it must be a net cooling agent, and its role as feedback where this should be inverted.
Sure I could explain the solution of said dissent, which I happen to know, but that is only a harder pill to swollow. I think the insight that WV is rather a net cooling agent should be exiting enough for the moment.
Barton Paul Levensonsays
ES: I think the insight that WV is rather a net cooling agent should be exiting [sic] enough for the moment.
BPL: Clouds are a net cooling agent. Water vapor is a greenhouse gas.
patrick o twentysevensays
…, the GHE/LW opacity and solar heating have a strong effect on where such a lapse rate is established by actual convection and also the actual convective flux of heat …
patrick o twentysevensays
“There has been no change in the 539 w/m**2 leaving the troposphere,” – that’s true if B_ν is linear over τ through the troposphere and τ is evenly increased (ie profile is stretched by a constant factor over height), or if deviations from that cancel each other just so. But more generally, the changes in the portions F↓ and F↑ which are emitted within the from troposphere would not sum to 0. (see my https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850156 ) I think there may be a general tendency for B_ν(τ) to have convex curvature through the troposphere over a large fraction of bandwidth…(?) (this would imply
Also you earlier (June?) once said something about the effect of inversions that partly works for sufficiently opaque inversions if you specify that increasing opacity reduces the net F↓ through that layer (noting that near the boundaries of that layer, you’ll be able to see across into other layers and their B_ν values will change the F↑,↓ there, so it’s not necessarily a ↓ net F throughout the whole inversion, but anyway, the implied radiative forcing of reducing a net F↓ has a cooling effect below, but where exactly is this cooling distributed? If the stratosphere is sufficiently opaque for a portion of it to have a ↓ net F, due to an abundance of a WMGHG, then the troposphere generally should be quite opaque… (although the ↓ net F could cross the tropopause eg. ( https://www.realclimate.org/index.php/archives/2026/04/a-reflection-on-reflection/#comment-847393 … ).
Okay, I got a bit lost in the details there, sorry (and I was focused on IRF; of course going from IRF to SARF transfers (propagates/spreads**) some of the stratospheric cooling to the tropopause-level radiative forcing (total forced warming of all that’s below):
Starting from near transparency (in the part of the spectrum being considered), given a general* T decline from the sfc to TOA, adding LW opacity (increasing the concentration (eg. relative to mass of air) of absorption cross sectional area; k_{a,air} = ∑_i n_i σ_{a,i} ÷ ρ_{air} ) generally* reduces OLR and generally* reduces the net F↑ (= F↑ − F↓ ; nearer the sfc this will depend more on increasing F↓; close to TOA this will depend more on decreasing F↑) As opacity increases, there will be some uppermost layer for which IRF is a cooling effect because of the reduced flow of LW heat from below and increased concentration of emission capability (also increases capability of absorption, but @ TOA there is the sharp dropoff to ~ 0 K … etc.). OLR emitted from some upper layer increases but the decrease in the contribution from lower is greater, so OLR in total (in the part of the spectrum being considered) still decreases (generally*).
With an upper level inversion (up to TOA), greater T near TOA should enhance the cooling. OLR emitted from the warmer layer next to TOA should increase more, but the decrease in the contribution from lower is still greater, until the opacity of the inversion layer gets large enough so that the EWF(F↑-weighted) for OLR (= integral over hemisphere of directional EWF · cos(ϑ); EWF per unit mass of air = k_{a,air} · exp[−τ_{vc} / cos(ϑ)] ? did that quickly; did I get it right?; I just realized I goofed it up recently in another comment…) is sufficiently dominated by that upper level inversion.
** some of this (edited) from prior comments ( https://www.realclimate.org/index.php/archives/2023/05/cmip6-not-so-sudden-stratospheric-cooling/#comment-811772 ):
equilibrium temperature response won’t generally match instantaneous forcing (or even SARF) exactly [aside from (even without H2O, cloud, snow ice, etc. feedbacks)] – of course convective fluxes can respond (eg. convective adjustment in troposphere in a 1-D model, hence the tendency for the surface and troposphere to warm up together, aside from the lapse rate feedback’s effect), but also because when a layer warms, it’s abs. cross sections brighten, which will have a warming effect on any layer which would absorb some of that radiance; likewise a cooling layer can have a cooling effect on other layers.
also ongoing convection & advection, as well as changes in those, will redistribute the T response
Formatting fixed:
…
With an upper level inversion (up to TOA), greater T near TOA should enhance the cooling. OLR emitted from the warmer layer next to TOA should increase more, but the decrease in the contribution from lower is still greater, until the opacity of the inversion layer gets large enough so that the EWF(F↑-weighted) for OLR …
(= integral over hemisphere of directional EWF · cos(ϑ); EWF per unit mass of air = k_{a,air} · exp[−τ_{vc} / cos(ϑ)] ? did that quickly; did I get it right?; I just realized I goofed it up recently in another comment…)
… is sufficiently dominated by that upper level inversion.
…
patrick o twentysevensays
Formatting fixed (2nd try):
…
With an upper level inversion (up to TOA), greater T near TOA should enhance the cooling. OLR emitted from the warmer layer next to TOA should increase more, but the decrease in the contribution from lower is still greater, until the opacity of the inversion layer gets large enough so that the EWF(F↑-weighted) for OLR …
(= integral over hemisphere of directional EWF · cos(ϑ); EWF per unit mass of air = k_{a,air} · exp[−τ_{vc} / cos(ϑ)] ? did that quickly; did I get it right?; I just realized I goofed it up recently in another comment…)
… is sufficiently dominated by that upper level inversion.
…
patrick o twentysevensays
Where τ, τ_{vc} is measured going away from POV; ϑ is angle from vertical up or down:
EWF per unit τ (directional) = −∂/∂τ exp(−τ) = exp(−τ)
(directional) EWF per unit τ_{vc}
= −∂/∂τ_{vc} exp[−τ_{vc} / cos(ϑ)]
= (1/cos(ϑ)) · exp[−τ_{vc} / cos(ϑ)]
∂τ_{vc}/∂[vertical mass path kg/m²] = k_{a,air}(τ_{vc})
(directional) EWF per unit mass of air
= k_{a,air}(τ_{vc}) · (1/cos(ϑ)) · exp[−τ_{vc} / cos(ϑ)]
EWF(F↑-weighted, to be multiplied by B_ν) per unit mass of air
= 2π ∫_0^{π/2} k_{a,air} · [sin(ϑ)·cos(ϑ)/cos(ϑ)] · exp[−τ_{vc} / cos(ϑ)] dϑ ??
= 2π ∫_0^{π/2} k_{a,air} · sin(ϑ) · exp[−τ_{vc} / cos(ϑ)] dϑ ?? Is this right? I should try just doing this for a F and then extract the B from that and see what’s left over…
patrick o twentysevensays
For an optically very-thin layer (∆τ_{vc} ⟨⟨ 1), its internal distribution of B_ν over its τ_{vc} won’t matter much to the spectral L↓ at/through its base and L↑ at/through its top (except in directions close to horizontal), and generally*** to the spectral F↓ and F↑ at those boundaries, respectively (which don’t depend strongly on variations in L values sufficiently close to horizontal). This is because the cross-sectional area it contains is mostly visible from the boundaries of the layer (except in directions close to horizontal); they don’t get in each other’s way much, ie. don’t hide each other. So it’s the B_ν averaged over its τ_{vc} that ~≈/tends to determine it’s effect on whatever is outside the layer (except for L in directions close to horizontal); glossing over some potential complications**, if layer-averaged B_ν is colder/warmer than what’s behind it, the spectral (L,F***) ↑/↓ decreases/increases) going from its bottom/top to its top/bottom (referring to the difference in values between the two boundaries (∆L,∆F***) ↑/↓ .
**except (∆L)in directions close to horizontal; noting the L entering the layer cold be dimmer(colder) than the than the layer’s average B_ν in some directions and brighter(warmer) in others, in which case it’s hard to make a simple general statement about the sign of ∆F.
(***?only general tendency for F?)
Barry E Finchsays
E. Schaffer 18 Jul 2026 at 1:47 PM “I think you are somewhat struggling with the GHE itself” and also “So yes, this latent heat DOES cool the planet a lot, easily by an 80W/m2. It ain’t that complicated”. What the! Since it “ain’t that complicated” please explain how “latent heat DOES cool the planet” Earth and give the quantity of Earth’s ocean that condenses in Outer Space at least 100 km above the ocean surface. Thanks.
As in the depiction of the GHE above, the lapse rate is crucial for the GHE. And the latent heat does reduce it..
Barry E Finchsays
“E. Schaffer” typed “the lapse rate is crucial for the GHE. And the latent heat does reduce it.” Absolutely! Finally the “E. Schaffer” quietly drops its wildly-incorrect, hugely-overestimated 86.4W/m2 of “latent heat” cooling in its 6 Jul 2026 at 5:45 PM “some 86.4W/m2 of “latent heat” cooling (as in the NASA Earth Energy Budget), then WV was barely warming at all, or rather net cooling”. Excellent.
I have for consideration the pretty-good approximation of 88 (surface evaporation, mostly tropical ocean) minus 54 (returned to surface as LWIR, the tropospheric GHE being 38.2% of total LWIR manufacturing leakage at its top and 61.8% of total LWIR manufacturing leakage at its bottom) minus 7 (returned to surface as sensible heat) = 27 w/m**2 of Net water evaporative surface cooling, for comparison with whatever is the total GHE assigned to H2O gas by the scientists.
I am afraid you must not subtract any such numbers from latent heat..
Barry E Finchsays
I’m done with this one folks (and bloviator “JCM” for this topic). It has 86.4 w/m**2 of energy surface -> troposphere and then 100% of that PARTICULAR energy either goes to Outer Space or disappears up the tail end of the Magical Sky Unicorn.
Pete Bestsays
presently the world sits at around 1.3/1.4C above the base and warming might have accelerated to 0.3C per decade up from 0.2C due to aerosols, low level cloud formation and earths albedo absorbing more sunlight etc. So within 30 years we could be over 2C permanently and this brings in large scale climate impacts that would well impact us severely.
I looks at the remedies to our vast fossil fuel usage and just see a very slow decline at best. Technologies that dont help so much are desalination, 22.000 plants globally at present and growing, Air conditioning which again is growing at a pace but means which every energy technology is used it needs a lot of it.
The technologies helping us get off of fossil fuels are battery storage for the grid, HGV, Buses, cars and some shipping, wind and solar, geo thermal and all the renewables but CCS isnt really on the cards and nothing big is working and deployment is unlikely.
Shipping and flying isnt as yet a big option for alternative energy and these industries are growing allowing for goods and people to be shipped all over the world.
So if people had to lay their cards on the table here at RC: what would you say is the mostly likely scenario for climate change temps and impacts. I am suggesting a minimum of 2.5C personally as we have alternative technologies but politics, economics and other factors mean they cant be deployed fast enough to offset warming for another 30/40 years
They say, that low cloud cover increased and the radiation power balance feedback is negative, i.e. low clouds did have a cooling effect in the past. They did not mention aerosol effects on cloud cover, though, which is a big omission.
So, if this is true, where does CERES albedo decrease come from??
Piotrsays
Dominik: Lenne: We have a problem – we can’t explain the CERES albedo increase.
Among possibilities I see – used different satellites, different time-scale, perhaps different area of the Earth looked at. And. If as you say- they ignored the change in aerosols over time – this would confounded their results even further – although differently early in the 47-year period, when the aerosols there were likely on the increase and differently in the later part of that period when they were dropping.
And on top of that, they may have had a different way to calculate cloud feedback
– in their abstract they are perplexed themselves:
” Counterintuitively, models that best capture observed LCC climatology and trends exhibit the strongest positive abrupt-4×CO2 feedback and highest climate sensitivity. ”
when based on their results – it should have been the opposite.
It appears that the controversial results pertain to low cloud feedback in tropical region; I have not grasped if the authors somehow generalized them to the entire planet, so that it could be justified to conclude that “low cloud cover increased and the radiation power balance feedback is negative”.
Could you check their results to be sure?
Furthermore, it appears that the results are derived indirectly from a set of “cloud controlling factors” (CCF). This procedure seems to be quite complicated and thus potentially prone to bias if any of these parameters were estimated incorrectly.
Best regards
Tomáš
Barry E Finchsays
Throwing out a quick thought because it’s my 1-Trick Pony does that Paper have enough regional detail to indicate that it is or isn’t this seen at 21:38 to 23:20 at https://www.youtube.com/watch?v=agKayS6h6xA colour-coded pictorial of GMST anomaly 1976-98 to 1999-2012 and wind explanation, Pacific Ocean eastern 2/3rds cooled over the 18 years while almost everywhere else except the Southern Ocean warmed.
2014 Quote: “Atlantic warming turbocharges Pacific trade winds Date: August 3, 2014 Source: University of New South Wales. New research has found rapid warming of the Atlantic Ocean, likely caused by global warming, has turbocharged Pacific Equatorial trade winds. Currently the winds are at a level never before seen on observed records, which extend back to the 1860s. The increase in these winds has caused eastern tropical Pacific cooling…”.
Is it possible that cooling over Pacific Ocean eastern 2/3rds (I calculated ~13% of Earth’s area for my assessment of adjustment for GMST acceleration) increased its low-cloud deck so much that, combined with much land area surface warming, it gave a situation of Global average surface temperature increasing as Global average cloud cover decreased, owing to low cloud cover decreasing over land with the surface warming but overwhelming Pacific Ocean eastern 2/3rds increasing with the cooling ocean surface?
However, the above wouldn’t resolve anything if it’s simply that separate analyses of low-cloud reflectivity change give opposing results.
Kobayashi Marusays
I’ll repeat and edited view of this next month, but for now things that make me curious:
There is the A) Populist Maga-Like Conspiracist Forum and then there’s the B) RealClimate / Academic-Scientific Educated Cohort
Are they so different?
A – Refuses to look because they distrust the source (government, experts, “elites”).
B – Refuses to look because they trust the source so completely that independent verification feels unnecessary or even heretical.
A – Sees a line of text, but interprets it through a lens of conspiracy (“they’re hiding something”).
B – Sees the same line of text, but interprets it through a lens of settled authority (“this is settled, why are you even asking?”).
A – Dominated by tribal identity—any question is an attack on “us.”
B – Dominated by institutional identity—any question is an attack on “science” itself.
A – Bias = overt cynicism.
B – Bias = overt credentialism.
A – Result: utterly incapable of parsing plain language that contradicts their worldview.
B – Result: utterly incapable of parsing plain language that contradicts their worldview.
It’s the same disease, different hosts.
The common denominator is identity-protective cognition. The brain doesn’t process information neutrally—it processes it through a filter of “what keeps me safe in my tribe?”
For the Maga-illiterate forum-goer, safety means not being a “sheep.”
For the RealClimate educated commenter, safety means not being a “denier.”
Both identities are so fragile that a single piece of contradictory plain text feels like an existential threat.
But the rank refusal to even look at the material on offer in the first place? That’s the real kicker. It’s not that they look and misinterpret it — it’s that they actively avoid looking, because deep down maybe they know that seeing the raw data, follow the link and honestly review it, or the original document might force them to hold two opposing thoughts at once.
And that cognitive dissonance is too painful.
So they outsource their thinking. The populist outsources to a conspiracy laden YouTuber or the resident bully.
The academic outsources to the IPCC summary or some other “authority”, or the resident bully
Yet both call it their “research.” Neither reads the actual damn text in front of them, work out what it means, or what it is pointing to.
The saddest part?
You can’t reason someone out of a position they didn’t reason themselves into. You can only point it out
Martin Smithsays
KM: …
A – Result: utterly incapable of parsing plain language that contradicts their worldview.
B – Result: utterly incapable of parsing plain language that contradicts their worldview.
MS:
A: Changes its mind when the populist clowns at the top change.
B: Changes its mind when the science at the bottom changes.
“B: Changes its mind when the science at the bottom changes”
Some say so, some say so. I am doing a little thing here over WV, and I never asked questions I did not know the answer to already. It is more an inquiry into how it was possible to get things so badly wrong. And as expected it appears to be to the strict rejection of anything contradicting the narrative, strictly believe driven, strictly anti-science.
Perception :: Reality
Ray Ladburysays
Do feel free to submit your ideas to any decent peer-reviewed journal. We’ll discuss them when they are published. How ’bout that?
If your ideas had merit, you could easily find a co-author with institutional support. By saying you won’t publish unless someone else pays, you are admitting that your ideas aren’t worth support.
Barton Paul Levensonsays
ES, you can get a waiver from most journals if you can’t afford the page charges.
Kobayashi Marusays
MS, I prefer another approach.
Minority Group C – The outsider looking in?
Sees both A and B clearly, because they’re not emotionally invested in either tribe. Reads the actual text. Follows the links. Asks “what does this actually say?” rather than “which team does this help?”
Result: accused by A of being a secret B, and by B of being a secret A. Which only proves the point.
Nigeljsays
Minority group D. The doomers. They belong to the doomer tribe so they cant see things objectively. They are convinced society will soon collapse and that capitalism is doomed. They are convinced renewables cant work. They are convinced we will run out of resources and in the near future. They are convinced climate change will be at the very high end of the spectrum and are impervious to evidence otherwise. There’s a pattern there neatly sumarised by the term “doomers”.
Martin Smithsays
KM: Minority Group C – The outsider looking in?
MS: I think that is multiplying entities beyond necessity because it uses the fact you have defined A and B so that they are clearly intellectually equivalent — A and B make the same mistake from different ends of the belief spectrum — but that creates the impression that A and B are the same percentage of the population. I think that is wrong. I am in set B.
B – Refuses to look because they trust the source so completely that independent verification feels unnecessary or even heretical.
Well, what is heretical when I argue with an A is that A refuses to use the scientific method to validate and invalidate their argument. I know I can find the science to verify my argument, but it really should be unnecessary for me to do that independent verification because A should have already done it as I have. Then A would have long ago changed his mind.
As a B, I refuse to look because I have already looked. Suppose you are an A, and we have an argument today about the increasing GHE. I say it is increasing CO2 from burning fossil fuels; you say it is a 3560 year cycle of the orbits of Jovian planets, and you require me to go do the independent verification. I do that, and I come back and say, “Nope, it really is anthropogenic CO2.”
The next day, we have the same argument and again you tell me I am afraid to do the independent verification. Do I have to go do it again? How many times do I have to independently verify that it’s anthropogenic CO2 all the way down?
Maybe now you will say, sure, that’s an easy one, but what about this one, or what about that one? and your what-about-isms are admittedly less scientifically certain. But I will say, and rightly so, I think, that there is a scientific consensus position based on peer-reviewed science, and that is the one I have chosen in each case.
As a B, I don’t defend my beliefs; I argue for the scientific consensus position, as long as it remains the consensus position. So you can say yes, but the science might have changed since the last time you verified it, and now the consensus position is about to change. Great, then as a B, I will change my mind. But note that if you argue that way, you aren’t really an A, because an A would never have done the verification you used to prove me wrong.
As a B, I have confidence because I know my beliefs are based on science. As a B, I can even say I know my beliefs are wrong but they are always getting closer to truth as science progresses. As an A, you can’t say any of that. As an A, you don’t have the confidence that allows you to change your mind because your beliefs aren’t based on anything but the declarations of authority figures.
When you define A and B as you have, I think they are both A, and then C becomes what B should have been to start with.
Kobayashi Marusays
LOL
If you say so Martin. You’ve made some very solid points here, though I do suspect you could have had a little ChatGPT assistance, undeclared. You sneaky thing you. :-)
Martin Smithsays
KM: I do suspect you could have had a little ChatGPT assistance, undeclared.
MS: I take that as a compliment, but no, the comment was all me. I always attribute text that comes from AI.
MA Rodgersays
Kobayashi Maru,
You fail to mention C) The Skyrocketeers.
Are they so different?
C – Refuses to look because they distrust the sources (govrnment, experts, “elites”)
I would add that those you style (B) do “look” and don’t use “settled authority” to dismiss what they “see” but use the ‘scientific method.’ Such ‘method’ does not respond publicly to every proclamation of a flat-earth theory. And without itself employing the ‘scientific method,’ a flat-earth theory won’t be treated as participating in the ‘scientific method,’ although it may get a more cursory examination which may involve a level of prejudice.
The point I would make is that both (A) & (C) do not attempt to engage properly with the ‘scientific method’ which is why they get such short shrift.
Nigeljsays
Or category D, egotistical sky rockety multi identity trolls, who cant look because their heads are so far up their own backsides.
Kobayashi Marusays
Is this you trying to convince me that you are the poster boy for applying the “‘scientific method’ “?
Oh please, build a bridge and get over yourself.
MA Rodgersays
Kobayashi Maru,
You appear to be imploring me to “build a bridge.”
In this circumstance that could prove difficult.
You should know that bridges require firm foundations, usually on both sides of the divide to be bridged. I see ‘firm foundations’ for any argument based on what you term “the source” (by which I assume you mean the science generally presented in IPCC WG1 ARs) because that ‘side’ is presented within the scientific process. What I don’t see is any such ‘firm foundation’ for the other ‘side’ which, as far as both you and ‘(C) skyrocketry’ are concerned, generally remains swamp-like.
You have described your own position saying you “completely defer to Hansen and his team for accuracy” yet still manage to dismiss the views of others in your (A) & (B) hypothesis up-thread by saying “You can’t reason someone out of a position they didn’t reason themselves into.”
With such contradictory positions from you, can your position be anything other than ‘as clear as mud’ and entirely swamp-like?
It appears it is not my ‘bridge-building’ we are missing – it is the drainage ditches on your side of this divide you want bridged.
Nigeljsays
MAR, you are talking to KM who is likely the same guy who you had a discussion with on Hansens aerosols ideas except at that time he was using another internet name, possibly prieto principle.
Your point related to Hansens claims that the 2023 to 2024 temperature peak was a result of reductions in shipping aerosols, and your core argument was this is problematic because the shipping routes don’t match the areas of warming ocean very well. While I quite like Hansen, this seemed a reasonably convincing point to me. I think you provided some maps. Even if Hansen is ultimately proven right, its a fair observation that at least needs an explanation.
Multi trolls response was a huge series of attacks on you and your post, that studiously avoided any mention of the core argument you raised. You cant build bridges with people like that. They have found their leader in Hansen and wont tolerate any criticism of anything he says. Its a cult thing, a tribal thing. He hasn’t got the chops or desire to engage on the specifics and be prepared to back down from his position. He will just spray you with rhetorical s**t basically. Hes done it to me as well.
Barry E Finchsays
Nigelj (no Reply button). Why no mention of Matthew H. England et al, shoaling of the northern hemisphere oceans caused by reduced winds that apparently was tracked from region to region, matching the SST increase, caused by the El Nino if memory serves. “Cause of Extreme North Atlantic Warming in 2023 with Matthew England” at https://www.youtube.com/watch?v=39T7bW7KA18 by “Climate Chat” (Dan Miller) on Jul 17, 2025
MA Rodgersays
Nigelj,
You are likely correct as to the true identity of this particular ‘skyrocketeer’. I was myself aware of the situation but am minded not to presume when a fresh pseudonym appears in the RC comment threads. And with this particular ‘skyrocketeer’ we should also acknowledge the contributions of some anonymous AI bot which is providing significant assistance with his word salads.
ABSTRACT ‘North Atlantic Ocean circulation and temperature patterns profoundly influence global and regional climate across all timescales, from synoptic to seasonal, decadal, multidecadal, and beyond. During 2023, an extreme and near-basin-scale marine heatwave developed during Northern Hemisphere summer, peaking in July. The warming spread across virtually all regions of the North Atlantic, including the subpolar ocean, where a cooling trend over the past 50–100 years has been linked to a slowdown in the meridional overturning circulation.
Yet the mechanisms that led to this exceptional surface ocean warming remain unclear.
Here we use observationally constrained atmospheric reanalyses alongside ocean observations and model simulations to show that air–sea heat fluxes acting on an extremely shallow surface mixed layer, rather than anomalous ocean heat transport, were responsible for this extreme ocean warming event. The dominant driver is shown to be anomalously weak winds leading to strongly shoaling (shallowing) mixed layers, resulting in a rapid temperature increase in a shallow surface layer of the North Atlantic. Furthermore, solar radiation anomalies made regional-scale warming contributions in locations that approximately correspond to some of the region’s main shipping lanes, suggesting that reduced sulfate emissions could also have played a localized role. With a trend towards shallower mixed layers observed over recent decades, and projections that this will continue into the future, the severity of North Atlantic marine heatwaves is set to worsen.
Note the line “Furthermore, solar radiation anomalies made regional-scale warming contributions in locations that approximately correspond to some of the region’s main shipping lanes, suggesting that reduced sulfate emissions could also have played a localized role” which is pointing to a role for the marine aerosol reductions (but only a subsidiary one) which could assist in explaining why 2020 aerosol reductions don’t show a climatic impact until 2023. (And this is all ‘N Atlantic’ when the 2020 aerosols were a bigger event in the N Pacific.)
Thomas Fullersays
Perhaps discussing where the burden of proof lies would clarify things a bit.
Thomas Fullersays
Kobayashi Maru, you should read 23 Years on Fire by Joel Shepherd. It’s the third in a SF series. In this book Shepherd ‘introduces’ the concept of Compulsive Narrative Syndrome. It is pretty much exactly what you write in this comment.
Great comment, BTW.
Piotrsays
Thomas Fuller: “Kobayashi Maru, Compulsive Narrative Syndrome. It is pretty much exactly what you write in this comment.”
Good catch, Thomas:
“ invents or latches onto a tidy story (narrative) to make sense of reality [and] ignore any real-world data or evidence that does not fit their chosen story.
fits “Kobayashi Maru”, to a dot.
Kobayashi Marusays
Thanks Tom. That concept is real from what I’ve seen in my life. .
Piotrsays
K. Maru: “ There is the
A) Populist Maga-Like Conspiracist Forum and then there’s the
B) RealClimate / Academic-Scientific Educated Cohort
Are they so different?
More tellingly:
There are A) Deniers (General: Ken Towe, K.Woollard, Mr. KA, Victor or of the Anything but Carbon sub-type: A. Schaffer, JCM)
and there is B) a Doomer Multi-troll, currently using :Kobayashi Maru” handle
Are they so different?
A) No data nor argument can possibly change their mind
B) No data nor argument can possibly change his mind
A) The inability to change their views is tightly linked to their validation ego: if everybody else is wrong, and I, a lay person, can see what scientists can’t/are not willing to, see, then I must be really really smart/virtuous.
B) The inability to change his views is tightly linked to his validation ego: if everybody else is wrong, and I, a lay person, can see what scientists can’t/are not willing to, see, then I must be really really smart/virtuous.
A) They attack and try to discredit the mainstream climate science as corrupt/ideological for not telling them things they want to hear.
B) He attacks and tries to discredit the mainstream climate science as corrupt/ideological for not telling him things he wants to hear.
A) They cherry-pick the data and authors to appear to support their claims and discredit climate science
B) He cherry-picks the data and authors to appear to support his claims and discredit climate science
A) They sow confusion and promote distrust to science, leading to weakening or abandoning efforts to reduce the GHG emissions (“there is no problem”: “CO2 is good for us”, “it’s anything but the GHGs, stupid”)
B) He sows confusion and promotes distrust to science, leading to weakening or abandoning efforts to reduce the GHG emissions (“too late to do anything” with the implication that “it’s all or nothing” so it does not matter if we level off the emissions or ramped them out – we are doomed anyway)
A) They are either paid agents of influence, or (more likely) “useful idiots” of fossil industrial complex in general, and Russia and Saudi Arabia in particular, with the latter’s economy, ability to finance wars on others enrich its oligarchs and suppress their own society – DEPENDS on the world continuing to buy their oil and gas.
B) He is either paid agent of influence, or (more likely) a “useful idiot” of fossil industrial complex in general, and Russia and Saudi Arabia in particular, with the latter’s economy, ability to finance wars on others enrich its oligarchs and suppress their own society – DEPENDS on the world continuing to buy their oil and gas.
“Are they so different?”
=====
P.S. Right, there is a difference : the throughput – a single Multi-troll, masking it by operating under 100(s) of different names, has churned up over the many years so many thousands(?) of posts – with volume probably larger than all the RC Deniers taken together.
I agree that “Kobayashi Maru” sounds exactly like the infamous multitroll, and that it is most likely a new disguise thereof.
I do not think, however, that comparing JCM with “E. Schaffer” is appropriate. JCM does not try to invent an alternative atmospheric physics like “E. Schaffer”. Quite oppositely, I would say that his understanding to climate science is very good. From my point of view, his comments are substantive, informative and valuable and I would say that in this respect, he ranks, along with MA Rodger and John Pollack, to the best among regular Real Climate commenters.
I do not think that his criticism to widespread approach treating the role of terrestrial hydrological regimes in Earth climate regulation as “mere feedback” does entitle you to label him as a “climate science denier”.
Greetings
Tomáš
Nigeljsays
TK : I do not think that his (jcm) criticism to widespread approach treating the role of terrestrial hydrological regimes in Earth climate regulation as “mere feedback” does entitle you to label him as a “climate science denier”.
Maybe not. But maybe if you add his absurd and repeated use of the term trace gas, his frequent dismissal of climate modelling and his suggestions that scientists have ignored the water cycle does make him a denialist. If it looks like a duck, walks like a duck and quacks like a duck and keeps on doing that its probably a duck. He certainly MAKES himself look like a denialist, and that doesnt seem too clever if he isnt.
He could write one simple short paragraph clarifying his position but hasn’t to my knowledge. Like multi troll hes evasive. JCM is just a bit more knowledgeable and smarter. So he gets shoved in the denier category, rightly or wrongly, but he only has himself to blame.
Ray Ladburysays
My, but you’ve spent an amazing amount of time working out what other people must be thinking without ever really engaging or communicating with them at all! Impressive.
How unfortunate you have done so without ever actually looking at the purposes of products produced by the “authorities” you decry. Had you done so, you might have saved yourself a lot of effort and the rest of the world a lot of aggravation. But then you would have deprived yourself of that feeling of righteous indignation that you seem to so crave.
I’ll say it again. There is no “authority” for climate science. There is the evidence and the literature. The IPCC sets out to summarize that literature in a way that policy makers can digest. That is inherently a consensus building activity, because different policy makers have different experts who they trust. Bringing as many on board on the policy/politics side while not losing critical players on the science side is a balancing act. The resulting product is bound to be conservative.
In contrast to that conservative approach, we have Hansen et al. asking, “OK, just how far could that conservative position be?” These researchers are hardly bombthrowers. They are not exaggerating the potential risks–just looking at them at a different confidence level. This is precisely the sort of analysis that is needed by engineers and others trying to mitigate the risks.
There is no antagonism. There is no authority vs. opposition conflict. There is just the process of science playing itself out on a subject that is of critical public importance.
And then there are the deniers–who for reasons of lack of understanding or lack of scruples find it appropriate to try to undercut the scientific process, while rolling around on a bed of oily money with their politician friends.
Barry E Finchsays
I don’t fit in A or B for any of the 5 random stereotypes listed by the Socialite KM (the same entity that directly contradicted itself here on the fundamental essence of the “greenhouse effect” while falsely claiming to have even the foggiest clue about this simple physics).
Barton Paul Levensonsays
BEF: I don’t fit in A or B for any of the 5 random stereotypes listed by the Socialite KM
BPL: Socialite? You mean he shows up at fancy parties given by the rich?
Kobayashi Marusays
BPL: Socialite? You mean he shows up at fancy parties given by the rich?
MK: Yes! And wearing very large Gay Party Hats and High Heeled Boots.
I’m the talk of the town in all the nicest ways lol
Ray Ladburysays
OK, I think that some of the discussion of the radiative characteristics of CO2 has become too imprecise. I do not know what is meant by a “wobbly” CO2 molecule. I do not know what is meant by de-wobbling a wobbly CO2 molecule. These are not precise terms, and they could mean different things to different readers. In particular, if some are getting the idea that the CO2 molecule can lose it’s “wobbly energy” through multiple collisions rather than a single collision, then they are simply mixing quantum mechanics with classical descriptions and coming up with a nonsense slurry.
I do know what is meant by the excitation of the V2 bending mode of CO2. I know that this is a harmonic oscillation that has a given expected energy. I know what is meant by the transition (radiative or collisional) from the ground state to this excited state and by the transition (radiation or collisional) from the excited state to the ground state. I can work out the relative frequency of these transitions given the densities of CO2 and other molecules in the atmosphere, the pressure, the temperature… That is how I know that the radiative decays at any given altitude will fit nicely into the blackbody spectrum for an object at that temperature. I know there will be an near-equilibrium ratio of the radiative to collisional decay rates. I know that the energy imparted by a CO2 molecule relaxing from the V2 bending mode to an N2 (or O2) molecule will then be shared with other N2 and O2 molecules as the accelerated N2 molecule thermalizes after its initial collision with the CO2 molecule.
It is important to preserve this lack of ambiguity, because there are many factors that can introduce ambiguity to the description. For instance, collisions between a CO2 molecule and other molecules can alter the energies of the eigenstates SLIGHTLY. The V2 bending mode is still there, just with slightly different energies, and it is still a quantum system–either excited or not. If we don’t start with the unambiguous understanding of the system, the additional complications will swamp us with difficulties, and all will be lost.
Barry E Finchsays
“collisions between a CO2 molecule and other molecules can alter the energies of the eigenstates SLIGHTLY. The V2 bending mode is still there, just with slightly different energies”, Could you please clarify whether you refer to what is shown at https://www.youtube.com/watch?v=XwpjyqTOtuM (but that is described as plus rotational energies “filling in” rather than varying vibrational energies).
Barry E Finchsays
I just watched it again and hear that I misremembered from a couple years ago. Michel shows vibrational plus rotational lines initially and then states vibrational can be modified by collisions as per Ray comment.
Today, the last day of the month, and with a few days still to arrive, the ERA5 Re-Analysis at ClimatePulse is showing a July global SAT anomaly of +0.70ºC, a significant rise on previous 2026 monthly anomalies (Jan-Jun run 0.51ºC, +0.54ºC, +0.53ºC, +0.52ºC, +0.55ºC, +0.56ºC). Also a cool Antarctic through much of early July has prevented a larger rise above these earlier months.
Relative to earlier years, July 2026 is roughly matching the anomalies of July 2023 (+0.72ºC) & July 2024 (+0.68ºC) which sat as the warmest Julys on record by some margin. Even when the 2010-22 trend of +0.30ºC/decade underlying AGW is accounted-for, these years 2023, 2024 & 2026 again sit 1st, 2nd & 3rd and still well-above the throng. SAT July anomalies 2010-26 & ranking** (& adjusted for AGW trend) 2026 ,,, +0.70ºC … 2nd ,, (+0.70ºC … 3rd)
2025 ,,, +0.45ºC … 4th ,,, (+0.48ºC … 14th) 2024 ,,, +0.68ºC … 3rd … (+0.74ºC … 2nd) 2023 ,,, +0.72ºC … 1st … (+0.81ºC … 1st)
2022 ,,, +0.38ºC … 6th ,,, (+0.50ºC … 11th)
2021 ,,, +0.33ºC … 8th ,,, (+0.48ºC … 13th)
2020 ,,, +0.33ºC … 9th ,,, (+0.51ºC … 10th)
2019 ,,, +0.40ºC … 5th ,,, (+0.61ºC … 5th)
2018 ,,, +0.28ºC … 10th ,,, (+0.52ºC … 9th)
2017 ,,, +0.27ºC … 11th ,,, (+0.54ºC … 7th)
2016 ,,, +0.36ºC … 7th ,,, (+0.66ºC … 4th)
2015 ,,, +0.15ºC … 12th ,,, (+0.48ºC … 12th)
2014 ,,, +0.07ºC … 14th ,,, (+0.43ºC … 16th)
2013 ,,,. -0.02ºC … 17th ,,, (+0.37ºC … 17st)
2012 ,,, +0.04ºC … 16th ,,, (+0.46ºC … 15th)
2011 ,,, +0.14ºC … 13th ,,, (+0.59ºC … 6th)
2010 ,,, +0.05ºC … 15th ,,, (+0.53ºC … 8th)
(** 2010-14 July unadjusted anomalies are not always higher than all pre-2010 July anomalies, years which include 2009, 1998, 2005, 2002, 2007, 2006. On the full record, the rankings would run 2014 17th, =23th, 19th, =13th, 2010 18th.)
The ERA5 60N-60S SST daily anomalies at ClimatePulse has been showing some vigorous warming through July. Recording highest anomaly for time-of-year since mid-June, the margin above 2023/24 has widened since mid-July. These warming SSTs will be reflected in the conventional measured SAT/SST records.
A CarbonBrief analysis by Zeke Hausfather suggests a strongest-ever El Niño is very-likely (91%) with NINO3.4 anomalies peaking at +3.96ºC (latest weekly NINO3.4 = +2.2ºC). This peak NINO3.4 = +3.96ºC feels a bit high given the modelling shown in the NCEP.NOAA weekly ENSO Update which is presently shown peaking at RNINO3.4 = +2.2ºC. (RNINO3.4 is sitting some 0.5ºC below present NINO3.4 values.)
Yet this same CarbonBrief analysis considers the chances of 2026 becoming warmest-year-on-record is not so certain (35%) and may depend on the particular temperature record used. (Odds of a 2026 record given thus:- BEST 66%, GISTEMP 65%, HadCRUT 35%, NOAA 24%, & SATs ERA5 13%, JRA-3 9%.)
There is also a Hansen et al communique ‘Beware Media Hype. It’s Not the El Nino!’ It talks of the ERA5 60-60SST anomaly and “the gap between the two El Nino lead-in years (2026 and 2023) … there is perhaps a hint of a decrease in the gap (Fig. 9)” but this was certainly a cynical “grasping for straws” moment and since the last data show on that Fig 9 (20th July), the later data at ClimatePulse (29th July) shows the gap doubled.
The August temperature numbers will thus be interesting, the developing situation graphed out at The Banana!!! Watch.
Kobayashi Marusays
RE but this was certainly a cynical “grasping for straws” moment
No it wasn’t. But it does expose the ongoing incompetence to be found in your comments. In my view both you and Zeke et al are way out of your depth and do not know what to say anymore.
Pete Bestsays
Everyone appears to have it down as strong due to the July numbers.
Kobayashi Marusays
“Quite oppositely” is an uncommon phrasing; people usually say “quite the opposite”, “on the contrary”, or “conversely” to mean that the real truth is completely different from what was just stated.
Native speakers in LOTE would be wise to double check there understanding of what other people are saying in English. Being a non-native English speaker myself I do speak with some experience in the matter. Alien languages are tough and the consequence rough if you get it wrong.
I am not the grammar or typo police either. Call me Harry Helpful :-)
Zebra: “Do you really think what you just wrote is useful,…”
I’m sorry, Zebra. When I was an editor, they allowed us to make the assumption that our audience was literate. If you like, I can try to write something at your level–perhaps with bigger words and lots of pictures on the page?
Ray, I’m with Pascal on this. I think you should take the time to write a shorter letter, which you obviously don’t. You know, edit thyself.
I also think 99% of the population would greatly benefit from fewer words and lots of pictures in understanding and relating to science. People can be very “literate” without knowing or understanding jargon in a specialty, or even knowing that Physics Today exists, much less reading it.
My experience tells me that the first step in educating someone is listening to them and understanding “where they are at”; lecturing them at your own level isn’t going to help them move forward… usually the opposite.
https://www.realclimate.org/index.php/archives/2025/11/unforced-variations-nov-2025/#comment-842186 , https://www.realclimate.org/index.php/archives/2026/06/unforced-variations-june-2026/#comment-849365 , https://www.realclimate.org/index.php/archives/2026/04/a-reflection-on-reflection/#comment-848619 :
k_{a,air} = μ_{a,air} ÷ ρ_{air} = ∑_i ( n_i · σ_{a,i} ) ÷ ρ_{air}
n_i = number density of i
——————————————–
My understanding is that the ratios of the rates of spontaneous emission, stimulated emission, and direct absorption, are, at LTE/LEDNLIE (Local Equilibrium Distribution of Non-Latent Internal Energy), determined by T (and ν and …) via the ratio of the populations of particles in states 2 and state 1 (where E₂ – E₁ = E₂₁ ); this is easiest to show for distinguishable particles which (at LTE/LEDNLIE) fit a Maxwell-Boltzmann distribution:
Maxwell-Boltzmann distribution: population in state 2 / population in state 1 = exp[ − E₂₁ / (kT) ]
But I expect it works out more generally (eg.
https://www.realclimate.org/index.php/archives/2024/10/unforced-variations-oct-2024/#comment-826352 )
@ LTE/LEDNLIE, for the optical transitions between a pair of states1 and 2, the absorption cross section σ_a is the blackbody area that would (given a spectral radiance L_ν from a direction) absorb at the same rate as the average per particle (eg. over all CO2 molecules of a given isotopologue) rate of “total absorption” = direct absorption − stimulated emission. If you take the average per particle rate of spontaneous emission and fit it into the absorption cross section σ_a, you get the Planck function.
I *believe* collisions act like observations of a quantum system eg. molecule; in isolation, during a transition between a pair of states, the system is in an evolving superposition of the states. So collisions turn this into a probability; eg. if they occur when the transition is 1% of the way done, then 99% of the transitions are aborted, but the 1 % that occur have been forced to happen 100 times faster (causing collisional line broadening), so the overall transition rates – and thus rates of spontaneous emission, stimulated emission, and direct absorption – are unchanged.
https://en.wikipedia.org/wiki/Maxwell%E2%80%93Boltzmann_statistics (oh no, I’ve been using the wrong term!; the Maxwell-Boltzmann distribution is something else)
https://en.wikipedia.org/wiki/Fermi%E2%80%93Dirac_statistics ,
https://en.wikipedia.org/wiki/Fermi%E2%80%93Dirac_statistics#Fermi%E2%80%93Dirac_distribution
oops! “so the overall transition rates – and thus rates of spontaneous emission, stimulated emission, and direct absorption – are unchanged.” (for given L, T, etc.)
ie. it’s the ratios:
stimulated emission / direct absorption ,
and
spontaneous emission / [ (either(stimulated emission or direct absorption or their difference) / L_ν ]
which are are unchanged (for given T, etc, at any L, for photons going away from a given direction)
see https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849827 ,
&
( https://www.realclimate.org/index.php/archives/2024/10/unforced-variations-oct-2024/#comment-826352 )
https://en.wikipedia.org/wiki/Einstein_coefficients
https://en.wikipedia.org/wiki/Schwarzschild%27s_equation_for_radiative_transfer (uses different terminology, I instead of L …)
Formatting fixed: (please replace last comment)
Wait, I forgot to reiterate a corrected form of my salient point:
It seems that collisional line broadening, and from that, I infer, collisions, preserve the integral over the spectrum of the absorption cross section σ_a for a given line (= ∫σ_a · dν ), (see below https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849826 ); therefore, for photons coming from a given direction (θ,ϕ), ie. going toward (π−θ , ϕ±π) (the opposite direction),
∫[ rate of spontaneous emission · hν ÷ B_ν(ν,T) ] · dν
And
∫[ ( rate of direct absorption − rate of stimulated emission ) · hν ÷ L_ν ]
are/should be unchanged (I presume the same is true for direct absorption and stimulated emission separately). Note I assumed the rates were in terms of photons (per unit time per unit solid angle per unit material (ie. average per molecule of a type of molecule), hence the multiplication by photon energy E = hν.
I expect this also holds true for thermal Doppler broadening,
(except that time dilation would slow the rates of absorption and emission by the molecules as a function of their speed (so as some σ_a · dν is shifted away from the from the line center, some would disappear … and more would be redshifted than blueshifted), though I’d expect it would take some very very high temperatures (how high?) to make this significant.)
…
and
∫ [ ( rate of direct absorption − rate of stimulated emission ) · hν ÷ L_ν ] · dν
…
okay, *now* its fixed.
optical depths τ from O2 and N2:
https://en.wikipedia.org/wiki/Rotational%E2%80%93vibrational_spectroscopy#Homonuclear_diatomic_molecules :
“I *believe* collisions act like observations … but the 1 % that occur have been forced to happen 100 times faster (causing collisional line broadening), so the overall transition rates – and thus rates of spontaneous emission, stimulated emission, and direct absorption – are unchanged.” … Well, the math might be more complicated than that, but somehow it must work out that GHG molecules emit photons even when at LTE. – because Schwarzchild’s eqn.
dL = (B – L) dτ
works:
https://en.wikipedia.org/wiki/Schwarzschild%27s_equation_for_radiative_transfer
https://scienceopinionsfunandotherthings.wordpress.com/2025/12/09/for-asymptotic-radiances-ppia-linear-and-general-cases-wip-awaiting-final-proofread-double-check-diagrams-pending/ , https://scienceopinionsfunandotherthings.wordpress.com/2025/12/24/for-asymptotic-radiances-ppia-linear-b%cf%84/
re Ray Ladbury https://www.realclimate.org/index.php/archives/2026/06/unforced-variations-june-2026/#comment-849480
but often enough to approximately maintain an excited population of each type of GHG molecule. In fact there will even be a tiny fraction which are excited above the lowest excited vibrational state (for each mode of vibration) (I think/infer that ‘hot bands’ are significant contributors on the wings of the CO2 band). (This is in accordance with the LTE approximation.) And occasionally they will manage to emit a photon before a collision.
No, that’s wrong. For isotropic σ_a:
Net spectral radiant cooling per unit mass = 4π · ( B − L_{4π} ) · k_{a,air}
Ie.
Net spectral radiant cooling per particle (averaged for each type of particle (eg. over all CO2 molecules of a given isotopologue)) = 4π · ( B − L_{4π} ) σ_a
Ie.
it’s the difference between 4π · σ_a · the Planck function (spontaneous emission) and the 4π · σ_a · directionally averaged ambient radiance (absorption).
Ie. it depends on being able to see across a difference in T.
And the troposphere on average must experience net LW radiant cooling to balance SW heating and convergence of the convective heat flux from the sfc.
See https://www.realclimate.org/index.php/archives/2026/04/a-reflection-on-reflection/#comment-848619 :
https://www.realclimate.org/index.php/archives/2026/06/unforced-variations-june-2026/#comment-849508 :
Corrected/fixed:
“(I think/infer that ‘hot bands’ are significant contributors on the wings of the CO2 band)” – Oops! No, not quite:
https://rmets.onlinelibrary.wiley.com/doi/10.1002/qj.4485
“Radiative forcing due to carbon dioxide decomposed into its component vibrational bands†”
Keith P. Shine, Georgina E. Perry
(I think I read sec. 2 through perhaps the end of 3.1 (?) )
(background – see https://en.wikipedia.org/wiki/Rotational%E2%80%93vibrational_spectroscopy ,
CO2: https://en.wikipedia.org/wiki/Rotational%E2%80%93vibrational_spectroscopy#Polyatomic_linear_molecules ,
https://en.wikipedia.org/wiki/Rotational_spectroscopy
https://chem.libretexts.org/Bookshelves/Physical_and_Theoretical_Chemistry_Textbook_Maps/Supplemental_Modules_(Physical_and_Theoretical_Chemistry)/Spectroscopy/Vibrational_Spectroscopy/Vibrational_Modes/Combination_Bands_Overtones_and_Fermi_Resonances )
Although those ‘wing-’bands do involve transitions between pairs of states in which the lower energy state is a bending-mode (v_2) vibrationally-excited state; some even being the second such excited level** (and
thusso I’d expect the overall band (assemblage of bands) will be effectively wider/narrower at higher/lower T – and of course the same is true of each individual band because of the distribution of CO2 among rotational states …(T = 0 K should collapse the entirety of the band into just two lines (1 in R-branch and 1 in Q-branch) per isotopologue, I expect)
…– so I’d expect the effective widening (window closure) would be more/less per doubling of CO2.
**Also it turns out that the 1st excited bending-mode energy level has a degeneracy (g) of 2 – and I’m guessing, based on some logic, that the degeneracy of the n-th level of the bending-mode (where ground n = 0) will be n+1 (?); so the populations
(in (v_1, v_2 , v_3) = (x, n, z) , setting aside the Fermi-resonance states (correct phrasing?))
would then be proportional to
(n+1 [?]) exp[ −E_n / (kT) ]
(- me @ https://www.realclimate.org/index.php/archives/2024/10/unforced-variations-oct-2024/#comment-826352 )
Which means ≈ 8.17 % is the ratio of the populations in the corresponding levels. Note that this is a ratio; the fraction of CO2 molecules in v_2 = 1 will be smaller, ie.
≈ 8.17 % / [ ∑ ( 100 % + 8.17 % + … )]
Anyway, my ballpark est. of the rate of spontaneous emission from CO2 v_2 = 1 → 0 (which was… where is that comment?) would have been ~doubled if I had known this.
“Although those ‘wing-’bands do involve transitions between pairs of states in which the lower energy state is a bending-mode (v_2) vibrationally-excited state; some even being the second such excited level**”
– or state of similar energy, judging from Fig.2 (see the FF transitions)
“Which means ≈ 8.17 % is the ratio of the populations in the corresponding levels.”
– @ 300 K.
“(I think/infer that ‘hot bands’ are significant contributors on the wings of the CO2 band)” – Oops! No, not quite:
https://rmets.onlinelibrary.wiley.com/doi/10.1002/qj.4485
“Radiative forcing due to carbon dioxide decomposed into its component vibrational bands†”
Keith P. Shine, Georgina E. Perry
(I think I read sec. 2 through perhaps the end of 3.1 (?) )
(background – see https://en.wikipedia.org/wiki/Rotational%E2%80%93vibrational_spectroscopy ,
CO2: https://en.wikipedia.org/wiki/Rotational%E2%80%93vibrational_spectroscopy#Polyatomic_linear_molecules ,
https://en.wikipedia.org/wiki/Rotational_spectroscopy
https://chem.libretexts.org/Bookshelves/Physical_and_Theoretical_Chemistry_Textbook_Maps/Supplemental_Modules_(Physical_and_Theoretical_Chemistry)/Spectroscopy/Vibrational_Spectroscopy/Vibrational_Modes/Combination_Bands_Overtones_and_Fermi_Resonances )
Although those ‘wing-’bands do involve transitions between pairs of states in which the lower energy state is a bending-mode (v_2) vibrationally-excited state; some even being the second such excited level**
– or state of similar energy, judging from Fig.2 (see the FF transitions) –
(and
thusso I’d expect the overall band (assemblage of bands) will be effectively wider/narrower at higher/lower T – and of course the same is true of each individual band because of the distribution of CO2 among rotational states …(T = 0 K should collapse the entirety of the band into just two lines (1 in R-branch and 1 in Q-branch) per isotopologue, I expect)
…– so I’d expect the effective widening (window closure) would be more/less per doubling of CO2.
**Also it turns out that the 1st excited bending-mode energy level has a degeneracy (g) of 2 – and I’m guessing, based on some logic, that the degeneracy of the n-th level of the bending-mode (where ground n = 0) will be n+1 (?); so the populations
(in (v_1, v_2 , v_3) = (x, n, z) , setting aside the Fermi-resonance states (correct phrasing?))
would then be proportional to
(n+1 [?]) exp[ −E_n / (kT) ]
(- me @ https://www.realclimate.org/index.php/archives/2024/10/unforced-variations-oct-2024/#comment-826352 )
Which means ≈ 8.17 % is the ratio of the populations in the corresponding levels [@ 300 K]. Note that this is a ratio; the fraction of CO2 molecules in v_2 = 1 will be smaller, ie.
≈ 8.17 % / [ ∑ ( 100 % + 8.17 % + … )]
Anyway, my ballpark est. of the rate of spontaneous emission from CO2 v_2 = 1 → 0 (which was… where is that comment?) would have been ~doubled if I had known this.
Also,
“Why the Forcing from Carbon Dioxide Scales as the Logarithm of Its Concentration” David M. Romps, Jacob T. Seeley, Jacob P. Edman
https://journals.ametsoc.org/view/journals/clim/35/13/JCLI-D-21-0275.1.xml
sec. 5c. “Linearity of κ in pressure”:
κ = S(T) / (π·γ) · γ² / [ γ² − ( ν − ν_c)² ]
can be rewritten as:
κ = S(T) / (π·γ) · 1 / [ 1 − ( ∆ν / γ )² ]
∆ν = ν − ν_c
Their absorption coefficient κ has units of m²/mol (caption of fig. 7), which is thus the absorption cross section σ_a per molecule, multiplied by Avogadro’s number.
So the integral ∫ κ dν for the line is constant over variations in collisional line broadening (γ = line width) – at least for a Lorentz line shape – which
may beis an approximation**, but maybe there’s other p-related things going on (eg. tiny amounts of quasi-static p-broadening?).Ie, at least approximately, and except for the effects of the curvature of Planck function over ν (should generally be tiny over γ), the rates of spontaneous emission and total absorption (direct absorption − stimulated emission) should be conserved over variations in collisional line broadening.
(PS I’m using ν, the symbol for frequency, because that’s a whole lot easier to write in this space than the notation for spectroscopic wavenumber, but that’s what I’m really referring to (they are proportional)).
Note the T-dependence of the distribution of molecules of a given type among their possible states is, *I believe*, the mechanism for the T-dependence of the line strength (the quantity that is, AIUI, conserved over variations in line broadening); this includes changing the stimulated emission/direct absorption ratio. This is separate from thermal Doppler broadening.
** sec. 5c. “Linearity of κ in pressure”:
Oops!
– For a given T, and (for absorption) a given directionally averaged ambient spectral radiance L (≡L_{4π}; for isotropic σ_a), which of course can be changed by line broadening (reduced opacity at/near the center (you can see farther), & exposure to L over a larger bandwidth, including bandwidth where you can see farther).
Okay, clarification: For a type of molecules experiencing a type of line broadening (wherein the line shape (eg. Lorentz) is conserved – okay, I’m just assuming this works for…), each point ν in the spectrum maps to another point at some multiple LB of its original distance ∆ν from the line center, and the σ_a and thus κ decrease to 1/LB of the original value; since these molecules are only ‘seeing’ (absorbing) in proportion to their σ_a, they experience a reduction in optical depth and can see farther through themselves. But from the perspective of other material and at a fixed ν, reductions in optical depth near the line center occur with increased optical depth farther out, and I think that some of the effects of the later can/may dominate because nonlinearities…
________________
Related to that. With some approximating simplications (eg. the Planck function’s sensitivity to T being constant over some spectral bandwidth, opacities and other optical properties contributed by/of other materials being spectrally invariant… and ignoring variations in of line broadening (width and shape) and line strength over height (as z or p)), the same given optical depth τ contributed by a GHG eg. (an isotopologue of) CO2 …
(AFAIK, each isotopologue of CO2 only has one isotopomer (a word I only just learned within the last few hours) ; otherwise I’d have to refer to isotopomers of CO2)
(Actually let’s just assume CO2 has a set mix of its isotopologues and use a single average σ_a(ν,…) per CO2 molecule – it doesn’t but for the simplicity of wording … most is ¹²C with 2¹⁶O ; fossil C is relatively ¹³C depleted… anyway…)
… will result in the same distribution of T and therefore the Planck function B_ν (T) over the total optical depth τ (CO2 + H2O + clouds + etc.), so the resulting spectral radiances L_ν at each location (p or z, ø,λ) and for each direction (θ,ϕ) will be the same, as thus will be the spectral flux densities (F_ν↑ & F_ν↓ and F_{ν,net}↑= F_ν↑ − F_ν↓). And so…
“values at one part of the spectrum (ν₁) ‘predict’ values where (at ν₂) σ_a [or τ for the initial amount] of a GHG is halved while the concentration of that GHG is doubled” ( https://www.realclimate.org/index.php/archives/2026/04/a-reflection-on-reflection/#comment-847512 – the overlap with τ_{H2O} as well as such things(?) that disrupt a PPIA approx. (partly cloudy conditions) will disrupt/prevent/distort/mess with the stately progression of ‘isobrights’ over cos(θ), but that’ s okay for this)
Thus, with some shuffling of ν values to arrange the σ_{a,CO2} for ease of visualization, with the approx. that this σ halves going out from its peak value per spectral bandwidth ∆ν_½ (smoothly, ie. σ_a = σ_{a,max} · 2^[ −abs(ν – ν_{max}) / ∆ν_½ ] ), there is a progression of values of L_ν(θ,ϕ) (for each θ,ϕ) , F_ν{↑,↓, net↑} , and net spectral cooling NCν, at each location, which just shift outward over the spectrum by ∆ν_½ for each doubling of (an isotopologue of) CO2. New values are established and become part of that progression at the peak ν_{max}. When the ν_{max} values of F_{ν,net}↑ and NCν approach their final saturated limits (generally 0) (one value may take ‘longer’ (require more CO2) to get to that point than another), we have a set marginal bandwidth (in two margins) containing the progession of varying CO2 effects, typically containing ≈all the net radiant cooling by CO2, which shifts outward from a widening saturated bandwidth (where F_{ν,net}↑ and NCν (of all material) ≈ 0, generally), and a narrowing of the CO2 window bandwidth where the effects of CO2 are ≈0 and F_{ν,net}↑ and NCν of other materials may be significant. Then: (edited from original:
[also the sharpness of the lapse rate change at the tropopause will tend to delay the approach to saturation of net radiant warming by CO2 at/near that location (saturated limit value of 0; I’m sure it’s not paper thin – see https://www.realclimate.org/index.php/archives/2026/04/a-reflection-on-reflection/#comment-847511 last ~half);
see https://www.realclimate.org/index.php/archives/2026/04/a-reflection-on-reflection/#comment-848619
and (2020) Nadir Jeevanjee, Stephan Fueglistaler, “Simple Spectral Models for Atmospheric Radiative Cooling” https://journals.ametsoc.org/view/journals/atsc/77/2/jas-d-18-0347.1.xml (Fig. 6(c) )
&
(2020) Nadir Jeevanjee, Stephan Fueglistaler, “On the Cooling-to-Space Approximation” https://journals.ametsoc.org/view/journals/atsc/77/2/jas-d-18-0352.1.xml (Fig 7 – H, and SX term)
Note also that H (net radiant warming) by CO2 goes toward 0 (saturated limit) near the center of its 15 µm band (band of bands) except near the the tropopause and closer to TOA, and next to the sfc (another relatively sharp bend in B(τ)?; – remember that the sfc can be approx. as an infinite isothermal optical depth, as can (looking up) Space above TOA).]
**(a reminder that the cumulative forcing from many changes can/will different if the climate is allowed to equilibrate before each successive next forcing, and therefore the cumulative feedback can/must also differ.)
There’s always room for Cooling To Space (CTS) in some uppermost layer of a gaseous atmosphere, … until T there approaches the LW effective radiating temperature of Space (ie. brightness temperature(s) of LW F_ν↓ @TOA ≈ 0)
See also my https://www.realclimate.org/index.php/archives/2026/04/a-reflection-on-reflection/#comment-847237
(and then there’s that some part of the increase in OLR is coming through the stratosphere’s windows, thus not heating it back up so much… ; but that wasn’t my main point here (otherwise I’d compare to the LW-grey gas behavior)…)
Okay, so what I was getting at: Absent variations in of line broadening (width and shape) and line strength over height (as z or p)), τ would be proportional to the mass path of the GHG, which, for a WMGHG like CO2, at least up to …
(maybe above the point where the LTE approx. fails anyway? … we could approx. it up to TOA for various purposes which don’t concern what happens above the stratopause?)
…would be proportional to the mass path of the atmosphere, which is approx. proportional to p. So just graph B over p. The stratosphere would be ~ 10% to 25%+? of the atmospheric τ_{CO2}, not counting H2O and clouds etc. But with line broadening, the majority of the bandwidth may be far enough from line centers such that σ_a is ~proportional to p (up to z=?**) and thus τ_{CO2} is proportional to p² (stratosphere ~ 1% to 6.3%+? of the atmospheric τ_{CO2}). But the line centers would (up to z=30 km ? p~10 mb ?**) have σ_a ≈ proportional to 1/p, (using −log(p) ~proportional to z: stratosphere ~ 54% to 70%+? of the atmospheric τ_{CO2} ; I just did that very quickly, added a 7 km scale height (to large?; it’s cold there) to 30 km** and took [37 km – (17 km, 11 km)]÷37 km;
not 100% sure if that’s even the right approach, but qualitatively, you get the point).**I’ve read that p (collisional) line broadening dominates up to ~30 km, which I remember being near 10 mb; so either thermal Doppler broadening dominates above ~30 km or the source intended to imply that ~30 km was the base of a marginal region where they were of similar magnitude. Doppler broadening produces a different line shape such that, I’m not sure about this but I thought maybe collisional broadening may dominate farther from the line centers up to higher levels.
PS the stretching of the lower troposphere in τ-space can make the T-profile more convex, working against the concave nature of B(T) – which itself is more pronounced at shorter wavelengths (higher ν) – eg. towards the LW/SW transition, B would rapidly drop to ~0 from the sfc and maybe have a thin isolated peak at the stratopause; towards the lowest ν, B approaches linearity over T.
So rather than sorting the spectrum by vertically-averaged σ_{a,CO2 p-average}, greater accuracy could be achieved by adding another dimension – the spectral distance ν’ from the line center (this will work best if the lines are evenly spaced, & etc.) (a graph over ν through this space may tend to look like a comb viewed from an oblique angle). If spectral intervals dν are evenly distributed over σ_{a,CO2 p-average} at each ν’ with sufficient σ_{a,CO2 p-average} resolution, then the whole near-constant forcing per doubling (including the verical distribution of ∆ net radiant cooling) still works, but the band is widening over a 3D space (if we were previously graphing values on one axis over the independent axis ν).
…
There’s errors in theses sorts of approximations but it’s helpful to use the approximation as a baseline of understanding and then add larger and then smaller identifiable corrections.
Forgot the link!
. Then: (edited from original https://www.realclimate.org/index.php/archives/2023/05/cmip6-not-so-sudden-stratospheric-cooling/#comment-812156 :
… If spectral intervals dν are evenly distributed over log₂[σ_{a,CO2 p-average}] at each ν’ with sufficient log₂[σ_{a,CO2 p-average}],…
1.
Okay, the idea of a nonzero saturated limit for net (spectral) (LW) radiant cooling at a perfectly sharp bend in ∂B/∂τ_{vc,norm} [discontinuity in the lapse rate in τ-proportional space] is more of a mathematical curiosity I used to illustrate aspects of radiant heat flow net (the layer with nonzero cooling will get thinner and thinner so it would eventually disappear from any graph with finite resolution), but its practical application is that the final approach to saturation at 0 should tend to be delayed in the neighborhood of sharper features in the temperature profile (in terms of B(T)) (in τ-proportional space)
2a.
“But with line broadening, the “vast”* majority of the bandwidth
may beis? far enough from line centers such that σ_a is ~proportional to p” , approximately…*Actually, that’s a point made by Romps, Seeley, and Edman – see sec. 5“c. Linearity of κ in pressure” https://journals.ametsoc.org/view/journals/clim/35/13/JCLI-D-21-0275.1.xml :
(The clump does have some width to it, though, and one could imagine lines converging on κ proportional to p at some p and then… um…?)
2b.
Fraction of atmospheric τ_{CO2} above ~30 km if:
τ_{CO2} proportional to p: ~ 1 %
τ_{CO2} proportional to p²: ~ 0.01 %
τ_{CO2} proportional to p above ~ 30 km and (30 km – z) below that: ~~~ 19 % ? or a bit less , … (≈7/37; but 7 km scale height too large?)
3.
That may be the more important thing (some increase in OLR coming from lower down through the stratosphere’s windows avoids undoing the cooling of stratospheric adjustment (going from IRF to SARF)), as far as the spectral effect is concerned. A LW-grey gas GHE with LW-grey GHG-forced warming, even without solar heating at/near TOA, will have a IRF→SARF upper atmospheric cooling, but it would be completely cancelled out in the full climatic equilibration (final TOA T same as initial, barring ASR (SW, ie. solar heating, ie. albedo…) feedbacks (1-dimensional column globally-representative model, using only a bidirectional (I think they call it “two-stream” in the literature) calculation of radiance to be representative of flux densities – which does miss some possible interesting phenomena caused by the directionality of radiance (eg. https://www.realclimate.org/index.php/archives/2025/12/unforced-variations-dec-2025/#comment-842798 (PPIA) :
A LW-grey gas GHE with LW-grey GHG-forced warming, with upper atmospheric solar heating: see Thomas Gordon Hewitt @ https://www.realclimate.org/index.php/archives/2026/06/unforced-variations-june-2026/#comment-849443
(2020) Nadir Jeevanjee, Stephan Fueglistaler, “On the Cooling-to-Space Approximation” https://journals.ametsoc.org/view/journals/atsc/77/2/jas-d-18-0352.1.xml
sec. “6. Application to real greenhouse gases” :
“a. RFM configuration”: (some formatting lost in C&P; emph. mine)
So Fig 6 and 7 don’t show the effect of CO2 on H2O’s net spectral radiant cooling (negative heating) (per unit heat capacity, thus expressed as K/day per cm¯¹) or the effect of H2O on CO2’s net spectral cooling; the later would of course reduce the cooling near the surface and maybe add cooling higher up (shading CO2 from brighter radiance coming from below). CO2 of course will block some bandwidth of H2O’s cooling. But I also wanted to note that there is some cooling next to the sfc by H2O in Fig 6 that continues for some bandwidth going into the stronger parts of its bands (seems to come from SX and maybe AX, as it is for CO2 (GX works against this)).
“the later would of course reduce the cooling near the surface – I didn’t just mean where the layer next to the surface with cooling extending into the heart of the CO2 band; I’m expecting the H2O overlap would extend higher than that, at least on the smaller cm¯¹ side of the CO2 band.
PS re my https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849896 : effective band widening effect was for the simplication (additional to those already specified) that the ~667 cm¯¹ (~15 µm) CO2 band (of bands) was the only CO2 spectral feature. (it’s not, though it dominates at least the LW part and the SW part is comparatively small though still significant? AFAIK.)
Why do I keep referring to graphing the Planck function B_ν (T) over the (vertical) optical depth τ …
(or τ_{vc}; vc = vertical coordinate; I sometimes use this notation to differentiate from optical thickness τ along any line of sight (LOS))
…? Because you can qualitatively estimate (at least for PPIA NR LTE etc. ) the spectral radiant flux densities and net spectral radiant cooling/heating distributions from such a graph. If B_ν is linear over τ out to a few units of τ up and down from your POV (and going through your POV), the net vertical flux density will approximately be proportional to the slope of that line (it’s an exact relationship if the line extends to τ = +∞, −∞). More generally, it will depend on the difference between particular locally-weighted (∫EWF(ϑ) ·sin(ϑ)·dϑ ?-I think; did that quickly) values of B_ν(|τ−τ_{POV}|) above and below (optically thin layers don’t have as much of an effect because of course they are easier to see through; they themselves don’t have as much to see).
Likewise, net spectral cooling per unit mass = 4π · ( B_ν − L_{ν,4π} ) · k_{a,air}; L_{ν,4π} over τ is a smoothed version of B_ν over τ (it has the same fine scale (optically thin) texture, but with reduced amplitude). For variations over small ∆τ (much less than 1; optically thin), the warmer and cooler absorption cross-sectional areas (which are bits of τ) can see each other well and so the warmer tend to cool and cooler tend to heat up – but remember the heating rate per unit mass or heat capacity also depends on the concentration of that cross-sectional area relative to those quantities (eg. k_{a,air}); over larger scales it’s more the curvature of B_ν(τ) that matters; convex curvature (negative ∂²B/∂τ²) favors cooling; etc.
Notice that where and when the concentration of a GHG or other absorber (eg. cloud particles) increase, the graph gets stretched out (over larger τ); slopes get reduced, optically thin layers get thicker (more opaque), layers on either side get farther away from each other; net spectral cooling or heating can increase with greater ability to emit and cooler absorb per unit mass/etc., but can decrease due to the warmer and colder parts being more hidden from each other, and signs and directions of net cooling and/or net vertical flux density can switch as one set of features in the thermal profile loses influence to another.
(see also my https://www.realclimate.org/index.php/archives/2023/05/cmip6-not-so-sudden-stratospheric-cooling/#comment-811930 – … – https://www.realclimate.org/index.php/archives/2023/05/cmip6-not-so-sudden-stratospheric-cooling/#comment-811932 -… )
Actually, it occurs to me that (PPIA etc.), if you take the B_ν(τ) and decompose it into a linear superposition of a constant plus sine and cosine terms over τ relative to a POV (
B_{ν,0} + ∑_m B_{ν,m,A}(τ’) + ∑_m B_{ν,m,S}(τ’) ;
B_{ν,m,A}(τ’) = A_m sin(mτ’)
B_{ν,m,A}(τ’) = S_m cos(mτ’)
), then the net vertical flux density depends only on the asymmetric part (sum of sines) then the net spectral vertical flux density depends only on the asymmetric part (sum of sines) and the net spectral radiant cooling (or heating) depends only on the symmetric part (sum of cosines).
( see also https://scienceopinionsfunandotherthings.wordpress.com/2026/02/03/for-asymptotic-radiances-ppia-sinusoidal-b%cf%84-part-5/ , https://scienceopinionsfunandotherthings.wordpress.com/2026/01/22/for-asymptotic-radiances-ppia-sinusoidal-b%cf%84-part-4a-1-2-recap-graphs-effective-angle/ , and prior)
Jeevanjee & Fueglistaler 2020 https://journals.ametsoc.org/view/journals/atsc/77/2/jas-d-18-0352.1.xml partition the net spectral radiant heating into four parts: GX (net exchange with ground (sfc), SX (exchange with layers above and below of equal ∆τ, of the maximum ∆τ that remains within the atmosphere), AX (exchange with whatever part of the atmosphere is excluded from SX), and CTS (Cooling To Space).
I can see the utility of doing that,
but I like to just combine the (vertical) optical depths of the atmosphere with what the sfc looks like from above (∞ optical depth, ≈isothermal; in the approximation of the sfc being a perfect blackbody) and what Space above TOA looks like from below (∞ optical depth, ≈isothermal near 0 K); note if we’re not using this to actually calculate what happens in Space or within the sfc (we don’t need to worry about refraction). In this case, the entirety of the net spectral radiant heating is given by the SX term. Stretching the atmosphere evenly* by some factor (*okay, hard to do if you’re doubling CO2 but there’s H2O with it’s own bottom-heavy distribution, not to mention clouds, etc.) is equivalent to stretching everything because the rest is infinite and isothermal of τ.
So what does the sfc look like if it has some spectral LW albedo (not a perfect blackbody)? Well in the simplest case of a flat horizontal sfc with only specular reflection with directionally-independent absorptivity, for all ϑ looking downward, you’d have a weighted average of B(T_sfc) and a reflection of the atmospheric profile of B (so you’re looking back up through it, so far as transmittance along your LOS allows). In the case of diffuse reflection or just a wavy sfc, the reflected image of the atmosphere will be vertically blurred (increasingly so away from sfc), and the vertical-scale (τ) will not generally match the original (real) atmosphere ie. looking down from near vertical, scattering toward near horizontal will partially extend (blurring) the reflected image; while looking down from near horizontal, scattering toward near vertical will partially compress (blurring) it.
Another thing about line-broadenning – I think that using the τ ~≈∝ p² proportionality (for CO2; based on the gaps between the intervals immediately around line centers, and just approximating the bandwidth of the later as nonexistent) will miss some net radiant cooling at/near TOA. In a sufficiently optically-thin (small vertical τ) top layer, if isothermal (a decent approximation if the B_ν(T) variation is small relative to the jump at TOA to the near 0 of the downward LW radiance from Space), I argue that line-broadenning ≈ doesn’t matter to cooling to Space @ TOA. Eg.:
@ TOA: Consider if you have an optical depth in 1 % of the bandwidth of some spectral interval. For simplicity, let’s just consider the vertical transmittance: The transmittance going down from TOA will decrease exponentially in that 1%. If we broadening the opacity out to 10% of the bandwidth, but then reduce the optical depth in that 10% to 1/10 its initial value (to maintain constant ∫ σ_a dν) then the transmittance (in that fraction of the bandwidth) at TOA will decrease 1/10 as fast over vertical mass-path, but this happens over 10 times the fraction of the bandwidth, so the full spectral interval transmittance decreases at the same rate over mass-path (though it will continue to decrease farther over mass-path (effectively 10 times deeper, asymptotically approaching a total drop 10 times larger). The same will be the case at any angle ϑ from vertical (within the same hemisphere), just with a different variation per unit vertical depth.
Note that for any line of sight (LOS), the drop in transmittance over some distance is the fraction of the total cross-sectional area you can see in the LOS that is within that distance ie. for absorption cross sections, it’s the part of the emission weighting function (EWF) that is within that distance.
(For B_ν ≈ constant over the ∆ν of the line-broadenning, or at least, given symmetric line shape, (∆ν)² · ∂²B_ν/∂ν² ≈ 0 )
(Relates to (2020) Nadir Jeevanjee, Stephan Fueglistaler, “Simple Spectral Models for Atmospheric Radiative Cooling” https://journals.ametsoc.org/view/journals/atsc/77/2/jas-d-18-0347.1.xml (see Fig. 6 for CO2 ) – ?? been awhile since I’ve read this though, but I thought they used τ_{CO2} ∝ p² ?? (I think the RFM values were the more real and the SSM2D were the approximation but don’t take my word for it)) (PS not intended as a criticism of the work, just something interesting I thought of.)
“so the full spectral interval transmittance decreases at the same rate over mass-path” – @TOA ie. initially in the spatial sense (why isn’t it “spacial”?)
PS AIUI, collisional line broadening, like natural broadening, is due to the finite time over which a photon is emitted or absorbed. A wave packet is mathematically composed of a superposition of a range of frequencies and wavelengths (or wavenumbers or wave vectors in 3D); a wave packet that is more concentrated into a shorter time or smaller space must be composed of a larger range of frequencies (proportional to E) or wave vectors (proportional to momentum).
( “The more general uncertainty principle, regarding Fourier transforms” – 3Blue1Brown https://www.youtube.com/watch?v=MBnnXbOM5S4 )
I think somebody last month asked about the photons being in LTE with the air – so to clarify, they generally are not expected to be. In this context, the LTE/LEDNLIE pertains to the energy distribution over the non-photons* of the air and sfc. (molecules, atoms, ions(? – in solution or solid/liquid…), electrons…) *-we can exclude the neutrinos, too…
The point is that the GHGs and cloud particles/etc. all (in the vast majority of the mass of the atmosphere) have access to a locally (small volume)-shared reservoir of enthalpy and therefore internal energy (we needn’t be concerned about what counts as thermal energy if we just refer to internal energy) ((approximately) LTE → internal energy distributed such that each statistically-sufficient substance has (approximately) the same T, and etc…) which they can add to and take from by absorbing and emitting photons.
See also https://www.realclimate.org/index.php/archives/2026/06/unforced-variations-june-2026/#comment-849365
Well, it will be harder to approximately maintain and isothermal relationship among the air and it’s resident aerosols because of the need to conduct heat to/from one to the others if the net radiant cooling per unit heat capacity varies and/or latent heat is being released/taken up (cloud particles), but it may still involve an approximate LTE/LEDNLIE on smaller scales? (how many molecules are in a r = 5 µm droplet…?)
But the other thing is of course that @ LTE/LEDNLIE, the internal energy (kinetic and potential energy of the relative motions and arrangements of atoms/electrons/etc.) is distributed such that the populations of particles in states of various energies support relationships among the rates of direct absorption (per unit spectral radiance L), − stimulated emission (per unit spectral radiance L), and spontaneous emission, such that, after stimulated emission cancels out some direct absorption, what remains…
(me @ https://www.realclimate.org/index.php/archives/2025/11/unforced-variations-nov-2025/#comment-842186 )
And the visible (not hidden) cross-sectional area that you can see (in some direction) from a given POV (location) is the radiating surface producing the spectral radiance L_ν = B_ν of radiating surface (coming from that direction); L_ν = B_ν of radiating surface – but within the atmosphere, that surface is distributed over distance s along the line of sight (LOS); so I …
Shorter version – the thicker the ‘fog’, the less far you can see. And you need to be able to see across a difference in T in order to have a net spectral flux density … etc.
Each thin layer of that fog that crosses you line-of-sight (LOS) blocks/absorbs some fraction of the glow coming from behind it, and replaces that glow with its own temperature-dependent glow.
(attempt at a simple explanation) For solar heating only at/within/near the sfc: GHE works by/through:
1. a reduction of the horizontal area * transmittance to Space * bandwidth for the direct flow of radiant heat (LW flux) from the sfc(or a layer there) to Space (OLR emitted by sfc).
2. the sfc T doesn’t have to increase enough to fit the entire equilibrium OLR (or just that part which balances solar heating at the sfc) through that area*transmittance*bandwidth, but the remainder of the equilibrium heat flow has to flow through the atmosphere from where lower-level solar heating is occurring up to where the rest of the equilibrium emission to Space occurs (through some combination of net LW and convective fluxes; this requires a temperature difference (to drive the net LW flux through the opacity and/or to allow/drive the convective fluxes) .
Martin Smith
Martin, you said:
“Ray, aren’t you describing the warming of the atmosphere? I thought the warming is not the greenhouse effect, but the effect of the greenhouse effect.
The greenhouse effect is:
1. Blocking energy emitted from the surface, and
2. Sending blocked energy back to the surface.
Between 1 and 2, the blocked energy warms the atmosphere, so global warming is the effect of the greenhouse effect.
Have I had it wrong all along?”
Earlier, I said that a simplified explanation is a good thing, but it requires being careful. You still have to be precise and disciplined in your language, or you confuse your reader and yourself.
“The greenhouse effect” refers to the energy retained by the climate system when GHG absorb radiant energy and convert it to thermal energy.
“Global warming” refers to the increase in average surface temperature resulting from the increase in that thermal energy resulting from human-caused increase in GHG.
In my opinion, what I just wrote is what you say you are looking for. A simplified, mechanistic explanation of the GHE, as well as “global warming”.
What more do you want to know? I can explain the terms radiant energy and thermal energy, or would you like to hear how this all applies to the expression “climate change”?
(That last was obviously me pretending to be your AI friend.)
Zebra: “The greenhouse effect” refers to the energy retained by the climate system when GHG absorb radiant energy and convert it to thermal energy.
“Global warming” refers to the increase in average surface temperature resulting from the increase in that thermal energy resulting from human-caused increase in GHG.
In my opinion, what I just wrote is what you say you are looking for. A simplified, mechanistic explanation of the GHE, as well as “global warming”.
MS: Your explanation confuses me more. A greenhouse traps energy, so a mechanistic explanation of the GHE must explain how the energy is trapped.
The mechanistic explanation of global warming must explain how the trapped energy warms the atmosphere.
Martin,
1. I said:
“GHG (greenhouse gasses) absorb radiant energy and convert it to thermal energy.”
Radiant energy can escape to space, but thermal energy cannot. Hence, the energy is “trapped”.
2. I said:
““Global warming” refers to the increase in average surface temperature resulting from the increase in that thermal energy resulting from human-caused increase in GHG.”
Thermal energy is what determines the temperature. If you increase thermal energy, the temperature will increase. “Warming” is by definition an increase in temperature.
Feel free to ask anything about what is still confusing you.
I understand what you are saying, Zebra, I just don’t think you are saying it right. I’m trying to say that the greenhouse effect and the warming are separate mechanisms.
1. The absorption by the GHG (i.e. trapping) of the radiant energy emitted from the surface is the GREENHOUSE EFFECT.
2. The conversion of the absorbed (i.e. trapped) radiant energy to thermal energy is WARMING.
3. The increase in WARMING averaged for the entire globe that results from increasing GHG, which then increase 1 and 2, is GLOBAL WARMING.
If I am confused it is about absorption vs conversion to thermal energy. Are absorption and conversion 2 sequential steps or are they a single step?
Martin
Here’s what NASA says;
“What is Global Warming?
Global warming is the unusually rapid increase in Earth’s average surface temperature over the past century primarily due to the greenhouse gases released as people burn fossil fuels. The global average surface temperature rose 0.6 to 0.9 degrees Celsius (1.1 to 1.6° F) between 1906 and 2005, and the rate of temperature increase has nearly doubled in the last 50 years. Temperatures are certain to go up further.”
“Warming” refers to an increase in temperature. Always has.
ms, in #3 you are correctly (more or less) describing the enhanced greenhouse effect.
An increase in the concentration of greenhouse gases leads to an increased infrared opacity of the atmosphere, and therefore to an effective radiation into space from a higher altitude at a lower temperature. This causes a radiative forcing, an imbalance that can only be compensated for by an increase of the temperature of the surface-troposphere system. This is called the “enhanced greenhouse effect.”
see under Greenhouse Effect
https://archive.ipcc.ch/ipccreports/tar/wg2/index.php?idp=689
# 1 & 2 are off. see my other comments and ipcc ref
also note more semantics confusion – warming of the surface is the result of the greenhouse effect.
global warming is a “yardstick measurement” to estimate mathematically the change in avg surface temperatures calculated on a baseline … over time … bpl is (more or less ) right but most people cant read greek or math.
the issue here remains problematic from trying to oversimplify beyond reason.
Martin, I’ll try again. To start with, you have so-called “black-body radiation”. This is thermal radiation with the amount of radiation and its spectrum characterized solely by the object’s temperature. A pure-black object absorbs all radiation incident upon it (a perfect absorber)–and it is also a perfect emitter. In reality, a perfect blackbody emitter is like a spherical cow–an idealization. That is why you hear Earth sometimes referred to as a “gray body”–it’s not important for out purposes–a blackbody is a pretty fair approximation. For bodies near 273 K, blackbody radiation peaks in the infrared–heat radiation.
So Earth in a vacuum or with inert atmosphere absorbs visible radiation coming in and emits IR going out, and reaches an equilibrium temperature. Now add a greenhouse gas. This is just a molecule that absorbs in that peak portion of the blackbody spectrum–in the IR. For a molecule to do this, it has to be at least triatomic–N2, O2 won’t do it. So, now, you have a molecule taking a bite out of the blackbody emission spectrum in the IR and getting excited into its vibrating state.
But when a molecule gets excited, it will eventually relax back into its ground state. A CO2 molecule can do so in a couple of ways–it can emit a photon of the same energy it absorbed to trigger the vibration or it can collide with a (say) N2 molecule and impart kinetic energy from the vibration to the N2 molecule. For CO2, the lifetime before it emits a photon is quite long. So as long as atmospheric densities are high enough, a relaxation by collision is far more likely than radiating a photon. That is what is meant by “thermalizing the energy”.
There is one additional thing to consider–the temperature gradient. If temperatures decrease with altitude as in the troposphere, CO2 molecules at high altitude are much less likely to become excited by collision, and so even less likely to relax by emitting a photon. (In the stratosphere, the temperature gradient reverses sign, so you get more emission by radiation and more IR escaping the stratosphere and so cooling. Both of these are in fact due to the greenhouse effect.)
You can think of equipartition as how a system shares energy among its different modes. In equilibrium, the modes are all at the same temperature–it’s like a series of connected reservoirs all at the same level. Now we pour some energy into the reservoir marked “excited CO2”. The system is out of equiliburium, and energy flows into the other modes of the system (e.g. the kinetic energy/temperature of N2, O2 and other atmospheric gasses). It’s not just absorption and re-radiation.
Does that help?
RL: Does that help?
MS: No. I mean, I understand it, and I appreciate it, but it is too long-winded to be what used to be called an elevator explanation. The NASA explanation Zebra posted tries to be a meme, but it doesn’t explain because it simplifies away the explanation.
The simplified mechanistic explanation has to be clear enough, short enough, and long enough so it can be used as your opening when you pitch your idea for a TV series about the greenhouse effect to producers who don’t know physics and who will shoot your idea down with memes like “CO2 has been high before,” “The Medieval Warm Period was global,” and “CO2 is a trace gas.”
in Re to Ray Ladbury, 6 Jul 2026 at 12:33 PM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849581
Dear Ray,
To be honest, I have no idea what you have tried to explain in the last paragraph of your comment, specifically in the sentences
“In equilibrium, the modes are all at the same temperature–it’s like a series of connected reservoirs all at the same level. Now we pour some energy into the reservoir marked “excited CO2”. The system is out of equiliburium,”
I think that in Earth atmosphere, the real distribution of available energy states of various air components does, in fact, almost perfectly fit the theoretical “equilibrium” distribution expected on the basis of the actual temperature in the respective air parcel. Exceptions could perhaps occur e.g. in rare cases of injection of energetic particles into upper atmosphere by solar storms or during electric discharges, I suppose. It appears, however, that you meant something much more generic. Could you clarify?
Greetings
Tomáš
Think of it as sequential steps:
1) The CO2 molecule in its ground state
2) The CO2 molecule is excited into its vibrational state
a) this can happen because it absorbs a photon of the right wavelength
b) or it can happen because a collision with another molecule imparts energy to excite the vibration
3) The CO2 molecule relaxes
a) by emitting a photon of the right wavelength
b) by imparting the kinetic energy from its vibration to another molecule via collision
Note the symmetry between 2 and 3. BUT in Earth’s atmosphere, temperatures are sufficiently low that it is rare that a molecule has sufficient energy to excite the vibration (e.g. 2b)). So the flux of blackbody radiation near 15 microns represents a flow of energy that excites the CO2 vibrations.
Likewise, the symmetry in 3 is also broken, because the lifetime of the relaxation process via photon emission is quite long. During that lifetime, a molecule is likely to collide with many molecules, at least one of which causes the CO2 molecule to return to its unexcited ground state. (3b more likely than 3a).
The CO2 in the atmosphere does still emit 15 micron photons–just what you’d expect it to emit for a blackbody at the temperature prevalent at the molecule’s altitude. Near the ground, atmospheric densities and CO2 concentrations are high enough that the probability of a 15 micron (+/- a micron or so) escaping the atmosphere is essentially nil. Only the 15 microns emitted at higher altitude/lower density/lower temperature have a reasonable probability of escaping.
This is the basic flow of energy due to the greenhouse mechanism that is responsible for the imbalance that raises tropospheric temperatures. But as I said, it’s the basics. There are lots of nuances–and if you omit these little nuances or don’t understand them properly, you can get stuck.
This is why the same denialist can claim that a trace gas like CO2 can’t possibly warm the atmosphere AND that greenhouse warming is impossible because the CO2 absorption is saturated. This is why I think it is a mistake to try to oversimplify the greenhouse effect. It took scientists over 100 years to fully understand all the subtleties of the effect. The first-principle quantum-mechanical calculations demonstrating why the CO2 absorption band is NOT saturated were only done in the last few years. If you wanted to pitch a series to a bunch of Hollywood jerkwads with short attention spans, the discovery process would be the script to pitch.
RL:
1) The CO2 molecule in its ground state
2) The CO2 molecule is excited into its vibrational state
a) this can happen because it absorbs a photon of the right wavelength
b) or it can happen because a collision with another molecule imparts energy to excite the vibration
3) The CO2 molecule relaxes
a) by emitting a photon of the right wavelength
b) by imparting the kinetic energy from its vibration to another molecule via collision
MS: That’s a mechanistic explanation for what happens to a CO2 molecule, but…
It says the molecule is in its ground state. That must be the state it is in when it is on the ground.
The molecule gets excited. It starts vibrating when it gets excited, so it wasn’t vibrating at all before it got excited.
The molecule can get excited if it absorbs a photon of the right wavelength. When it absorbs a photon of the wrong wavelength, it doesn’t get excited.
And the CO2 molecule can relax, if it emits a photon of the right wavelength, but if it emits a photon of the wrong wavelength, it stays excited.
Or, the molecule can relax if it collides with other molecules, which doesn’t sound very relaxing at all. More like aggravating.
I understand your mechanistic explanation of what happens to a CO2 molecule throughout the GHE/warming cycle, but a person who rejoices in the acronym MAGA will get tripped up in all the ways I listed.
And I think a mechanistic explanation of the path of energy through the GHE/warming system is a better way to explain it, but maybe both points of view works even better.
Um, not quite. “Warming” is by definition an increase in thermal energy. Temperature is merely the yardstick being used to record the level of that thermal energy in space and time.
The drive to make this short and simple is creating errors. Occam’s-razor warns against the error of oversimplifying past the point of accuracy. The simplest accurate model of necessary mechanisms is still the minimum required.
“Global warming is the rise in surface temperature caused by the build-up of thermal energy (originally from the Sun) in the climate system, due to human-emitted greenhouse gases trapping more of that energy.”
That’s accurate but it may not be addressing the dynamics of how it happens which people want to express.
A more scientifically precise way to say it:
“Global warming refers to the increase in Earth’s average surface temperature resulting from the build-up of thermal energy (originally from the Sun) in the climate system, caused by human-increased greenhouse gases slowing the escape of outgoing infrared radiation to space.”
Moreover, the thermal energy itself doesn’t “escape” to space; it is radiated away. So, the energy isn’t “trapped” like a blanket traps heat by physically holding it in place. Rather, greenhouse gases slow down the leak of energy to space. And it’s that which causes the build-up of thermal energy (originally from the Sun) in the climate system in the first place.
The increase in greenhouse gases today has forced the increase of thermal energy and therefore the temperature. But if are being forced to condense all of that into a single, final, punchy statement that can’t really be argued with, you could say:
“The simplest accurate model is this: more GHGs slow the escape of the Sun’s energy to space, causing thermal energy to build up, which raises surface temperature. Anything simpler than that omits the actual mechanism and becomes misleading.”
This discussion is circular. The IPCC already covered the basics long ago: Frequently Asked Question 1.3 What is the Greenhouse Effect?
https://archive.ipcc.ch/publications_and_data/ar4/wg1/en/faq-1-3.html
Trying to simplify it further destroys it’s meaning and the physics involved. Embrace the complexity.
Kobayashi Maru,
I think the ‘GHE is not like a blanket’ idea you present is not helpful for explaining GHE. Blankets (with the exception of impervious ‘space blankets’) physically leak like sieves. The troposphere presents a far-less leaky barrier for the air of planet Earth than a blanket-enwrapped human. The journey for a packet of air from ground up to the tropopause takes on average perhaps a couple of weeks. That’s [ 9/(2 x 186) =] 0.024 mph or 0.01m/s. Blankets (okay not an apple-with-apples comparison) are measured with a permeability of perhaps 0.5m/s.
Unlike your assessment, I rather liked Zebra’s simplistic idea that “Radiant energy can escape to space, but thermal energy cannot. Hence, the energy is “trapped”.” Indeed, you say it yourself. “The thermal energy itself doesn’t “escape” to space; it is radiated away.” One of the things us small-brained mammals have yet to figure out is the nature of the photons that comprise radiation. But we do know for sure it ain’t not “thermal energy.”
And while I would make an issue of what is the sun’s energy and what is the Earth’s energy, your statement “More GHGs slow the escape of the Sun’s energy to space,” still doesn’t explain why that is the case. What is the ‘slowing’ mechanism?
To give my six-pen’orth, I’d also point out (1) Radiation is how energy flows into and out of objects floating in space. Actual thermal energy is indeed trapped and needs converting into radiation to exit the planet.
But then I’d add that (2) It is the temperature of that object which determines the ‘conversion to radiation’ and thus the amount of radiated energy shot out into space. Higher temperature means higher radiation out. Specifically, this is about the temperature of the ‘areas’ that radiate into space. Without an atmosphere or a GHG-free atmosphere, the ‘area’ in question will always be the object’s surface.
And from there (3) GHGs result in a shift in the ‘area’ specific to the wavelength of radiation the GHGs absorb/emit. More GHGs, the higher that ‘area’.
(4) Because the Earth’s troposphere gets colder with height, that means more height for a GHG’s ‘area’, less temperature in that ‘area’, less radiated energy out from Earth.
(5) The Earth will then begin to heat, increasing temperature until the resulting increased radiation from that increased temperature equals the GHG-induced reduction in radiation.
That is the basis of how GHGs work. I can’t think of a simpler explanation that doesn’t resort to the ‘blanket’ or the ‘greenhouse’ analogy. (And do note, like blankets, operating greenhouses are remarkably leaky things although they call it “ventilation” not permeability’.)
Sadly the workings of the GHE mechanism is not well described outside scientific discussion. That IPCC AR4 page you link-to spends 670 words answering the question ‘What is the Greenhouse Effect?’ but of the underlying mechanism it only manages to tell the reader that it operates like a greenhouse although “through a different physical process” and that “adding more of a greenhouse gas, such as CO2, to the atmosphere intensifies the greenhouse effect, thus warming Earth’s climate.” Sure there’s a lot of other information provided answering the question but by 2007 the continued “It’s like a greenhouse” explanation was (and today remains) entirely inadequate in the face of blatant denialistic bullshit from far too many ‘authorities’ who should know better.
‘MA Rodger
I believe I am correct. The IPCC is accurate. The physics is right. I’ll defer to the IPCC links I gave but will add the following brief clarification.
Anyone involved around climate change science and global warming, like I have for 30-40 years, understands that global warming is a metaphor, the greenhouse effect is a metaphor and the blanket idea is an analogy to explain those metaphors to the public at large.
Fully aware they are imperfect metaphors and analogies, knowing full well they are unsatisfactory from a purist science pov and yet all have been used repeatedly by renowned physicists to explain parts of the complex climate science, the physics and the mechanisms surrounding the ideas behind the greenhouse effect and the enhanced greenhouse effect; the difference between thermal energy, warming, and global warming and why temperature is different entity, since Syukuro “Suki” Manabe was a boy.
I’m quite happy with my explanations as far as they go. Knowing they were provided to only answer MS’s very specific question. They were not intended to explain the entirely of the physics behind 6 IPCC assessment reports. It’s complicated. I was trying to help where others had failed.
Thanks for your interest. I’m humbled.
Martin Smith: A person who calls themselves MAGA is ineducable and irremediably stupid or they wouldn’t be MAGA to begin with. They don’t recognize facts or truth or science. If the person is not interested in the science, they won’t be perusaded by the science, particularly when it counters their accepted world view.
Learning is only possible for those who have not achieved epistemic closure.
In Re to zebra, 3 Jul 2026 at 6:01 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849546 ,
and Martin Smith, 2 Jul 2026 at 10:18 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849535
Dear zebra,
There is no clear generic relationship between “energy” of a system and its temperature. An example: If a bottle filled with ice absorbs heat from its surrounding, its thermal energy increases, however, its temperature remains constant until all the ice melts.
Dear Martin.
The average radiative temperature of Earth surface (that can be computed, according to Stefan-Boltzmann law, from the average intensity of the upwelling infrared radiation flux) depends on a time and space average of a complex balance of a multiplicity of various energy fluxes across the surface and the atmosphere. These global balances are usually depicted in “energy budget” diagrams called also Trenberth’s diagrams. Basically, any external perturbation to any of these particular fluxes, including a change in the atmospheric concentration of a greenhouse gas, may perturb the entire system. Examples of such induced perturbations may be the rising global average surface temperature and the decreasing temperature in the lower stratosphere induced by rising atmospheric carbon dioxide concentration. The fluxes in the corresponding diagram change accordingly. In absence of external perturbations, a steady state can be reached,
As regards the complex mechanism of the “greenhouse effect” of the atmospheric gases absorbing the infrared radiation, and/or of the changes in the altitude / temperature profile induced by changes in the greenhouse effect intensity, I recommend rather relying on explanations provided by standard textbooks (like Physical Climatology by professor Dennis Hartmann), by people like patrick o twentyseven or MA Rodger, or by a good AI engine, than striving to decrypt and/or correct confusing statements released by zebra.
Greetings
Tomáš
Martin Smith (MS) 25 Jun 2026 at 3:24 AM “I need a correct, simplified, mechanistic explanation using photons, molecules, kinetic energy, and words like ground, ocean, re-emit, and back radiation”. Following after a delineation line of “+” signs below is a perfectly-correct, sufficiently-complete simplified, mechanistic explanation of the so-called “greenhouse effect (GHE)” in Earth’s troposphere which avoids some of the MS words that I quoted because they are incorrect or misleading, or unnecessary for a full simplified, CORRECT mechanistic explanation. The simplified, mechanistic explanation is 4 short paragraphs with simple, obvious concepts that describe the operation followed by Notes that aren’t required for the simplified, mechanistic explanation but might be interesting. Preceding that are comments on what I infer are 2 significant misunderstandings by MS which appear to closely relate to each other (the same basic misunderstanding of MS).
=======
MS appears to think that GHE is solely due to change in “atmospheric window frequencies” (MS 27 Jun 2026 at 2:13 PM “Google AI …… is 10% to 15%) but GHE certainly isn’t much at all to do with “10% to 15%” in the atmospheric window, except maybe during Snowball Earths not at all relevant now. That “blocking effect” is a small bit player, a nibble at the edges right around 8.0 microns for H2O gas, 9.0 & 10.0 for O3, and 13.0 for CO2 as seen for example at 17:10 at my UTube video link below (Side Note: It’s shown as a lesser amount for CO2 increase than used by NASA but that’s an advanced matter that I’ll never study and certainly irrelevant to the “simplified, mechanistic explanation”).
MS 30 Jun 2026 at 1:00 PM “The greenhouse effect is:”
1. Energy emitted by the surface is blocked
2. Blocked energy is sent back to the surface”
30 Jun 2026 at 12:45 PM “The greenhouse effect is:
1. Blocking energy emitted from the surface, and
2. Sending blocked energy back to the surface”.
=======
Nope, that wouldn’t work as stated because the temperature lapse rate caused by reduced pessure with altitude wasn’t mentioned and that’s what makes it work. (my entire point I’ve made since 2018 about the confusion sowed about something simple by misleading and incomplete, or just plain incorrect, “simplified, mechanistic explanations”).
=======
MS 27 Jun 2026 at 2:13 PM “if we add a gigaton of CO2 to the atmosphere, the probability that an infrared photon can escape the atmosphere without being absorbed by a greenhouse gas must decrease”. Nope, not at all correct because the word “must” is a gross misunderstanding of the simple physics that has been caused in the brains of MS and 123 or perhaps it was 321 other bods I tried to disabuse, caused yet again by this “re-emits” and “back radiation” nonsense. As understood by Tomáš Kalisz 26 Jun 2026 at 4:09 PM and perhaps JCM or Zebra. So then “must decrease” definitely S.B. “will usually decrease” with an explanation why (vertical temperature changes). As per numerous places that I’ve (correctly) railed against variations of this since August 2018 (Quixotic) there’s no “must decrease” because that is yet again a false statement that CO2 molecules emit zero manufactured radiation (the “re-emit” and “back radiation” drivel strikes again, and strikes continuously ubiquitously since many years back, the handy Strawman set-up so’s it can be destroyed by Fossil Shill dullards like “Yong Tuition” and “Tom Shula” presented by “Tom Nelson”, and innumerable “commenter” bods for UTube videos and Web Site Forums). The CO2 molecules DO emit (manufacture) “their own” photons in vast quantity, not just “absorbed” like MS implies. Suppose as a hypothetical that the extra “gigaton of CO2 to the atmosphere” were to absorb radiation from below that would otherwise reach Outer Space in the globally-annually-averaged Power flux of 0.00156 w/m**2 but were to emit (manufacture) radiation in quantity such that extra “gigaton of CO2 to the atmosphere” sent 0.00157 w/m**2 to Outer Space, well then the extra “gigaton of CO2 to the atmosphere” would cause COOLING of 0.00001 w/m**2. This HAPPENS over Antarctica (or part of it) in Winter (or part of it) as shown at 20:09 at https://www.youtube.com/watch?v=rgP-lwf2tb8 (3 Power Spectrum samples of Earth’s radiation to Outer Space measured in 1970 for Sahara Desert, Mediterranean Sea & Antarctica). The cartoon that I’ve seen variously in UTube presentations for 13 years, and is huge disinformation, asserts that CO2 (sometimes H2O gas) molecules emit (manufacture) Zero photons by not showing any, not mentioning it at all, a most cunning form of disinformation, remaining silent on a massively-relevant fact. Animation showing a surface photon absorbed in a CO2 molecule and “re-emitted” up, photon absorbed and “re-emitted” down, over and over just like the “Joke Zonderkop” nonsense on RC UV 20 Jun 2026 at 7:30 PM. Suppose the animation showed 8 photons emitted for every 2 photons absorbed from below (6 photons manufactured), well now there’s 5 photons heading up for every 2 photons absorbed from below so this extra CO2 molecule is INCREASING the upward radiation toward Outer Space. This in Antarctica is seen at my link and thus the MS nonsense physics “if we add a gigaton of CO2 … must decrease” is shown incorrect by measurement.
The reason why this increasing upward radiation only usually happens in the stratosphere (always) is that the troposphere is usually colder with increasing altitude so therefore photon manufacture higher up must usually be less than absorption from below (warmer collides me than colder). The absence of the extremely-important manufacture proportional to Kelvin**4 from the MS “if we add a gigaton … must decrease” above renders it incorrect logic so worthless for the pondering and explaining the “greenhouse effect”
The MS explanation appears to be based on an incorrect assumption that if the “Energy emitted by the surface” were not more “blocked” by more GHGs then it wouldn’t heat the troposphere. However, it’s already heating the troposphere though because it’s already “blocked”, it’s simply “blocked” just a few metres higher up than it would be with more GHGs. The Reality using fair quantities for illustration is that 5% global-annual average of the radiation emitted by the surface (~22% roughly with cloud-free sky and 0% with full cloud cover giving 5% average, from Costas&Shine 2012 CERES analysis Paper) is in frequencies 8.0-9.0 & 10.0-13.0 microns (“atmospheric window frequencies”) and the slight narrowing of those “atmospheric windows” with more CO2, O3, H2O gas is a minor bit-player and is NOT the “greenhouse effect (GHE)” that must be described. The GHE applies to the OTHER radiation that is emitted by the surface and is ALREADY “blocked”, being ~78% roughly and all of that ~78% to which “enhanced GHE” is now applying is “blocked” already (and was already “blocked” 100 years ago) within ~100m of the surface with or without any clouds in the sky.
Suppose that ~78% of radiation emitted by the surface is “blocked”, absorbed and turned into “heat” within 100m of the surface. Add some CO2 and it’s then “blocked” within 95m instead of within 100m. I seriously doubt that that’s the sort of thing envised by MS with 1., 2. above, more likely incorrectly thinking that photons that would have reached Outer Space getting blocked were getting blocked with CO2, but that isn’t how the warming effect works. It’s a matter of the heating Power being just a few metres lower than before for a significant GHE warming.
=======
MS 27 Jun 2026 at 2:13 PM “Google AI estimates that the probability that an infrared photon emitted from the surface escapes the atmosphere to space without ever being absorbed by a greenhouse gas is 10% to 15%. Google AI includes an analysis, but my point is there is a non-zero probability that an infrared photon emitted at the surface can escape the atmosphere to space without ever being absorbed by a greenhouse gas on the way out..And if it doesn’t get absorbed by a greenhouse gas, it doesn’t warm the atmosphere”. This indicates that MS thinks that “greenhouse effect (GHE)” is a narrowing of the “atmospheric window” frequencies 8.0-9.0 & 10.0-13.0 microns but it isn’t that at all. As explained above, that “atmospheric window narrowing” is a minor extra-bit effect and not even applicable to CH4, N2O because their Bands aren’t near 8.0, 9.0, 10.0 or 13.0 microns. Rather the GHE is as I now describe.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
A simplified explanation of the so-called “greenhouse effect (GHE)” in Earth’s troposphere. Some Power flux of the vast quantity of photons that are emitted (manufactured) by molecules in Earth’s troposphere leak out of its top & bottom into or through the stratosphere and into the surface respectively. Emission (manufacturing) is proportional to Kelvin**4.
=======
Suppose for concept illustration only (not highly accurate) that 194 w/m**2 leaks out the top and that is part of what is emitted (manufactured) by molecules over an altitude range of 4 to 12 km with an effective average of 7 km. Say for concept illustration that 345 w/m**2 leaks out the bottom and that is part of what is emitted (manufactured) by molecules over an altitude range of 0.0001 to 1.5 km with an effective average of 1 km.
=======
Suppose a certain amount of CO2 was added into the troposphere and mixed then *instantly* the upper & lower ranges would need to be closer to their respective ends because there are more CO2 molecules in the way. So the effective averages of 7 km for upwelling into or through stratosphere and 1 km for downwelling into surface change to 7.05 km and 980m respectively, with the upwelling & downwelling Power fluxes consequently changing to 193 w/m**2 and 346 w/m**2 respectively due to being produced by higher-than-before (colder) and lower-than-before (warmer) air parcels. The reason for this is that pressure reduction with increasing altitude causes “adiabatic cooling” or “adiabatic heating” for increasing or decreasing altitude respectively.
=======
There has been no change in the 539 w/m**2 leaving the troposphere, there is no “magical extra energy”, the surface downwelling radiation has increased by 1 w/m**2, which is an additional ~480 terawatts (~95%) heating the ocean and Earth is emitting less radiation to Outer Space (a Power reduction of 510 terawatts).
=======
*** Notes are following. The above is all of the simplified part, all that’s needed. A person could skip the extra Notes or read them for interest ***
There is no “back radiation” to the surface. No radiation is going back any place, that’s just nonsense. There’s downwelling radiation into the surface. It’s manufactured by gas molecules and clouds in the air. There’s upwelling & downwelling radiation at every geoid-type (~spherical) surface that you can just randomly choose all the way from the centre of Earth’s core to 100 km above the 0cean surface (it varies hugely of course and is always going to be out of balance between upwelling & downwelling for any geoid-type surface). ~No photons are “re-emitted” or “re-radiated”, they are ~all simply “emitted” (manufactured by molecular collisions).
=======
Since the troposphere has very little thermal capacity the 539 w/m**2 leaving must be matched by 539 w/m**2 going into the troposphere on a globally-averaged, annual-averaged basis and this consists of 376 LWI radiation emitted from the surface, 88 of NET H2O gas latent heat (surface evaporative Power less rain on snow-ice and surface condensation), 20 of NET convection (warm air rising minus the surface begin warmed by the pressure increase of air descending), and 54 of solar SWR absorbed by gases, liquid & solids, which is 538 w/m**2, the 1 w/m**2 difference not really existing and being accuracy & rounding error.
=======
The thermosphere & stratosphere each absorb 13 w/m**2 of solar radiation to radiate to Outer Space by the gases. So the globally-averaged, annually-averaged radiation to Outer Space approximately for the general concept is 20 w/m**2 from the surface (90 with little cloud and zero with significant cloud) + 194 from the troposphere + 13 from the stratosphere + 13 from the thermosphere, which is 240 w/m**2 total matching the 240 w/m**2 of solar radiation from the Sun that Earth absorbs. With the example above of 193 from the troposphere due to extra CO2 there is 239 w/m**2 of radiation and the short-fall of 1 w/m**2 is called “Radiative Forcing at TOA” which heats the ocean and melts ice plus minor heat usages. The division of radiation to Outer Space between troposphere, stratosphere & thermosphere is based on the Power sources and not necessarily accurate for the vertical locations of the actual photons that reach Outer Space.
For example (just made up numbers to illustrate) suppose 20 w/m**2 of the 194 from the troposphere was used to heat stratosphere & thermosphere such that they each radiated to Outer Space 23 w/m**2 instead of the 13 w/m**2 of their solar heat source then the 194 from the troposphere to Outer Space would be 20 w/m**2 to high but still it was tropospheric photonic energy into the stratosphere & thermosphere which gave them the energy to radiate out the extra 20 w/m**2 so it’s correct from that perspective.
a wise man accepts his limitations. I will never be a rock star.
better to lean heavily on the explanations of the experts in the field. and accept what they say.
explanations of the experts in the field would indeed likely be superior to normal public / media practice such as to describe half that Downwelling Longwave radiation as “back radiation” — ie back to where it came from — the surface, according to a close personal friend of John Cleese, so essentially according to a rock star.
Much of this thermal radiation emitted by the land and ocean is absorbed by the atmosphere, including clouds, and reradiated back to Earth. This is called the greenhouse effect.
…. the Earth’s greenhouse effect warms the surface (land and oceans) of the planet.
https://archive.ipcc.ch/publications_and_data/ar4/wg1/en/faq-1-3.html
Damn those IPCC people. All they seek is an argument.
Barry E Finch ” normal public / media practice such as to describe half that Downwelling Longwave radiation as “back radiation” ”
Barry, you seem to conflate, again, two different positions:
– public/media calling ALL Downwelling Longwave radiation as “back radiation”
with
– the deniers’ attack on science based on the fallacy that of all LW radiation leaving atmosphere – “HALF” is absorbed by the ground and “half” is sent into space.
Which would require the same T near the bottom and near the top of atmosphere.
True if our atmosphere was one-molecule-thick.
And although “downwelling” is a better word if used in the explanation of mechanisms (see MARodger above) “back-radiation” used by the public/media is not as terrible as you imply:
– 358 W/m2 of the LW from the ground to be absorbed by the atmosphere
– 340 W/m2 of the LW from the atmosphere to be absorbed by the ground
Thus from the “ground” perspective, we get “back” from the air almost as much IR as air gets from the ground. And nobody said “back” must be identical to “out”
So all this talk how bad a word “back-radiation” is, and how “experts in the field” on’t be caught using it (see Trenberth et al., and NASA reposting their back-radiation”)
= is either making a mountain out of a mole-hill, or ego validation for the semantic nitpickers like zebra.
zebra
“The greenhouse effect” refers to the energy retained by the climate system when GHG absorb radiant energy and convert it to thermal energy.
not quite.
To balance the absorbed incoming (solar) energy, the Earth must, on average, radiate the same amount of energy back to space.
Because the Earth is much colder than the Sun, it radiates at much longer wavelengths, primarily in the infrared part of the spectrum.
Much of this thermal radiation emitted by the land and ocean is absorbed by the atmosphere, including clouds, and reradiated back to Earth. This is called the greenhouse effect.
…. the Earth’s greenhouse effect warms the surface (land and oceans) of the planet.
https://archive.ipcc.ch/publications_and_data/ar4/wg1/en/faq-1-3.html
Also, note that the absorbed incoming solar energy does not in fact balance with the outgoing energy. Energy is being stored in the Earth system, especially the oceans.
John
John, energy was being “stored” in the Earth’s climate system even when there was no imbalance.
It moves, it transforms, it determines the states of matter in the system… constantly.
Isn’t all that what we perceive as “climate”?
Zebra,
I don’t see it that way. If there is a net radiative equilibrium, I don’t consider energy to be stored or removed from the climate system. Instead, I would treat it as being shuffled between different boxes within the system.
Nor do I perceive the shuffling as what I mean by climate. It’s one aspect of climate, but it leaves out a lot. For example, the hydrosphere, cryosphere, and biosphere. These interactions are not exclusively described as energy exchanges. Even in the atmosphere, there can a change in climate without changes in mean energy storage or exchange. I can easily visualize a situation where the mean wind energy at a location isn’t changing, but there is an increase or decrease in rare, but very significant, winds from storms.
John,
I thought my statement was pretty much the same as your “shuffled”, and I did say “determines the states of matter” to cover water and ice and so on.
But anyway, I’m not clear what you mean by storage, or how the energy that is accumulating during imbalance is different from the energy in the system when EEI = zero.
Oh dear. Lest we forget. Thermal energy is being stored in the Antarctic Ice Sheets too.
Our next progressive pattern dance is the Sweetheart Stroll. Choose your partner, please.
zebra 7 Jul.: “ I’m not clear what you mean by [energy] storage ”
maybe you should have thought of this before you lectured John on his use of that word – he said that the imbalance between the incoming and outgoing energy ” is being stored in the Earth system”. No imbalance -> no need to store it, or release it.
Hence your:
zebra 6 Jul “ John, energy was being “stored” in the Earth’s climate system even when there was no imbalance
is … a typical zebra: correct them first, ask what they meant later.
Zebra,
I view climate as more than energy accounting, although it’s certainly an important way of keeping track of what is happening. By “stored” I mean as thermal energy. Since EEI is greater than zero, the storage term is going up, and the energy in the ocean is the major recipient – at least currently.
John
John Pollack
John, this is connected to something I have been wondering about. I’ve seen the numbers for the distribution of energy that is being added to the system… e.g. 90% is in oceans, 6% land, and so on, but it isn’t clear that this is identical to what it would be for EEI =zero.
My guess is no, which explains your definition of “storage”. But I’ve never seen a discussion of what it might be pre-industrial or in some future equilibrium state. Any thoughts?
in Re to zebra, 11 Jul 2026 at 5:22 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849699
Dear zebra,
Considering that if Earth is in a steady state* with a perfect balance between absorbed and emitted energy fluxes (EEI = 0), the flux into Earth’s storages must be also zero, it appears that asking after distribution of this zero flux between “ocean, land, and so on” makes hardly sense.
I therefore suppose that you meant something else.
Could you clarify?
Best regards
Tomáš
* remark:
From thermodynamical point of view, a body exhibiting a balance between incoming and outgoing energy flows is not seen as an equilibrium system, I think.
Zebra,
I’ve had to think more carefully about the terminology I’m using. I think of “storage” as the sum of energy being stored or released from Earth system as a whole. Units are energy, such as J.
EEI is a flux equal to the change in storage / (area of the Earth X time interval) W/m^^2 are flux units.
The proportions of energy exchange between different subsystems of the Earth storage term are what I would call “partitioning.” These are in expressible in fractions or percentages. These quantities must be measured, and depend on how you define the partitions and also the specific time period involved. They are variable, not constant. They need to be observed.
The time period I had in mind is one relevant to climate change. This would be an annual cycle or longer period. While it would be possible to define shorter intervals, they would be hard to measure.
I agree with Tomáš that if EEI = zero, there is a net equilibrium between incoming and outgoing energy, and thus no storage. Partitioning would then reflect whatever readjustment was happening within the system, and would depend on the specifics of the situation. There is no single answer. It would depend on the evolution of the internal state of the climate system following radiative equilibrium.
John, I see how my sentence might have caused some confusion.
I was trying to ask about the partition of energy in the system when EEI = zero; I used the numbers for partition of the added energy as an example because that’s the only thing I’ve seen discussed.
My best guess is that if we took the ARGO system back to pre-industrial times and measured a 30-year period of Ocean Heat Content, we would get a fairly consistent number, just like GMST.
That’s the partition of energy that I’m curious about, and how it might have changed, and will change.
As for the term “storage”, I take a longer-term view. I see almost all the climate system energy as stored solar energy. But as I said, I understand your usage of the term.
in Re to zebra, 13 Jul 2026 at 7:16 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849749
Sir,
With respect to confusion, I can confirm that so far, the more you explained, the less I understood what you are speaking about.
If you took the ARGO system back to pre-industrial times and measured a 30-year period, I suppose you would obtain an average temperature of the upper two kilometres of sea water during this time span and you could also perhaps say if there was any temperature trend in this layer within the said 30-year time span.
I agree that it might be interesting to see if there was no energy imbalance then (as you seem to suppose), however, I am still quite unsure that it was indeed the idea you tried to express.
If so, could you clarify what you meant by “partition of energy”, what has this partition to do with the hypothetical historical recording of temperature of the upper 2 km of the ocean and which changes of this mysterious partition, how and when you expected to observe?
Thank you in advance and best regards
Tomáš
Zebra,
Under the conditions you mention, pre-industrial, with a 30 year interval of EEI = 0, I can
give a partial answer.
The portion partitioned to the atmosphere will be negligible, and that to the solid earth will be very small. This is based on the heat capacity of the atmosphere compared to the rest of the system. The solid earth can hold or release a lot of heat, but ground conduction is extremely slow beyond a thin surface layer, say 2 or 3 meters. There can be some partition to freshwater and ground water systems, but in general these are small components of the system.
This leaves the two large active components of the system in near opposition. The ocean can of course store or release a lot of heat. With EEI held to zero, the cryosphere – especially glacial ice, must have a partition similar in magnitude but opposite in sign. So, the if the ocean is cooling, the heat released would be used to melt ice, or the opposite.
Zebra,
Under the conditions you mention, pre-industrial, with a 30 year interval of EEI = 0, I can
give a partial answer.
The portion partitioned to the atmosphere will be negligible, and that to the solid earth will be very small. This is based on the small heat capacity of the atmosphere compared to the rest of the system. The solid earth can hold or release a lot of heat, but ground conduction is extremely slow beyond a thin surface layer, say 2 or 3 meters. There can be some partition to freshwater and ground water systems, but in general these are small fractions of the total.
This leaves the two large active components of the system in near opposition. The ocean can of course store or release a lot of heat. With EEI held to zero, the cryosphere – especially glacial ice, must have a partition similar in magnitude but opposite in sign. So, the if the ocean is cooling, the heat released would be used to melt ice, or the opposite.
in Re to John Pollack, 14 Jul 2026 at 8:46 PM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849785
and 14 Jul 2026 at 8:48 PM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849786
Hello John,
May I ask under which circumstances you think that the described re-distributions of heat between ocean and cryosphere could be possible?
Honestly, I somewhat doubt that the Earth is ever in the state of perfect balance (EEI=0) between incoming and released energy flows; I rather guess that it may be more likely that it is either warming or cooling, in response to fluctuations in the relevant forcings.
Greetings
Tomáš
Tomáš, regarding 15 Jul @ 7:25 am
I would like to have Zebra answer your questions.
I agree with you that EEI = 0 is unlikely (except perhaps as a long-term average.) I think that partitioning needs to be established through observations, which are difficult to obtain.
The circumstances that I think a re-distribution of heat between ocean and cryosphere would be possible with EEI = 0 are in the aftermath of a large runaway dump of glacial ice into the oceans
(such as a D-O event). I am NOT saying that EEI really would be zero under those circumstances, since I think it likely that the Earth albedo would also change due to changes in cloud and ice cover.
I am not convinced that there is any utility in speculating about what the partitioning would be, in the absence of observations – except for sharpening definitions. It’s up to Zebra to make a case that this is a useful line of inquiry.
John Pollack
John, I took some time to read up on OHC and I think I begin to understand what has been puzzling me… as usual, the first step was getting the definitions right.
Your point about the cryosphere also was useful, although I’m not sure the magnitudes are as close as you imply.
At this point, I have to accept that the empirical results are as good as it gets as to any changes in partition of energy from the pre-industrial state.
John,
My 17 July comment was made before your 16 July comment showed up, and as you can see I am in agreement about the partition (and the need to clarify definitions.)
But I don’t get this statement about thinking EEI would not be zero pre-industrial CO2. That’s the fundamental concept we’ve been talking about in climate change all this time.
Of course it would be an average over a period statistically necessary to smooth out the “wobbles”… exactly the same as what we do with GMST. In fact, I’ve asked here a couple of times whether the same 30 year period would be necessary to validate changes in EEI.
The physics tells us that we will achieve a new EEI zero state when we (hopefully) stop adding CO2. It doesn’t tell us, as I’ve come to realize, exactly what the new overall system state… the partition… is going to be like.
in Re to zebra, 19 Jul 2026 at 6:39 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849850
Sir,
I am still very curious about your explanation what you meant by the “partition of energy” in the Earth’s climate system and how it could change at zero energy imbalance (EEI=0). Was it just the hypothetical case of a sudden ice sheet collapse and their dissolution in the ocean suggested by John?
As regards your assumption of the preindustrial EEI=0, the primary reason why I doubt about its validity are known long-term variations in solar activity. It appears that in the preindustrial era, Earth experienced a relatively deep Maunder minimum (when EEI might have been slightly negative) and a slow recovery therefrom (when EEI might have been slightly positive). In other words, we do not know if the EEI has ever been zero during a longer time period, at least because it appears that insolation rather tends to fluctuate than to remain stable (or quasi-stable, if we consider the 30-year average).
Greetings
Tomáš
Zebra: But I don’t get this statement about thinking EEI would not be zero pre-industrial CO2. That’s the fundamental concept we’ve been talking about in climate change all this time.
John: Okay, you specified a 30 – year period to smooth out the “wobbles.” The reason I don’t think there would often have been an EEI (very close to) zero even in pre-industrial times is that those “wobbles” occur on all different time scales. Some of them would involve large scale changes in ocean circulation/atmosphere/cryosphere interactions, such as the PDO or even longer periods. So an EEI near zero would likely end up being a short interval between a positive swing and a negative swing. Of course, I can’t prove this without observations, so maybe you’re right.
Overall, I think that EEI=0 as a precondition edits out most of the interesting stuff happening in the system, and restricts it toward the ocean/cryosphere oppostion that I mentioned earlier.
TK: As regards your assumption of the preindustrial EEI=0, the primary reason why I doubt about its validity are known long-term variations in solar activity. It appears that in the preindustrial era, Earth experienced a relatively deep Maunder minimum (when EEI might have been slightly negative) and a slow recovery therefrom (when EEI might have been slightly positive). In other words, we do not know if the EEI has ever been zero during a longer time period, at least because it appears that insolation rather tends to fluctuate than to remain stable (or quasi-stable, if we consider the 30-year average).
BPL: You’re talking about very small variations. EEI was roughly zero in the preindustrial age, oscillating irregularly about the mean. Close enough for government work. There was no trend in EEI, just small variations with time.
John
John, this is a definition thing. I think of pre-industrial EEI = “zero” as BPL says… close enough for government work.
And think my conclusion is supported by this plot:
https://en.wikipedia.org/wiki/Temperature_record_of_the_last_2,000_years
I’m happy to be corrected on the statistics by an expert, but I think here even “eyeball” gives a good indication that the energy in the system is pretty stable for long periods.
Zebra and BPL,
Z: This is a definition thing. I think of pre-industrial EEI = “zero” as BPL says… close enough for government work.
And think my conclusion is supported by this plot:
https://en.wikipedia.org/wiki/Temperature_record_of_the_last_2,000_years
J: I strongly disagree with your definition and conclusion! I’ll start a few definitions of my own.
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.
The first issue I have is the non-equivalance of GMST and system energy. Zebra, it was just a few months ago when you were telling us that we should be using EEI rather than GMST because the former was a better measure of climate change. Why have you switched now? Are you going to stick with it?
Zebra, it’s also time that you give your own definition of system energy, if you don’t like mine.
The second issue is that the time period under consideration has been left very vague. Do you want it to be the period from 0-900 c.e. when GMST with a one-century smoothing looks pretty flat? 900-1800, when the overall GMST trend is down? Doesn’t “pre-industrial” also include the latter period? If the overall trend in GMST is down, don’t you think that EEI has also gone negative? The oceans ought to be cooling, and glaciers growing, unless you think that the subsurface ocean has absorbed most all the latent heat of fusion from the growing ice cover.
The third issue is that if it’s all “close enough for government work” aren’t you telling me that you’re uninterested in tracking climate dynamics and energy flows in the pre-industrial period? That isn’t what you sounded like when this discussion started.
Z: I’m happy to be corrected on the statistics by an expert, but I think here even “eyeball” gives a good indication that the energy in the system is pretty stable for long periods.
J: I’m interested in the causes of those non-stable things in the system, both the long term decline in GMST and also the much shorter zig-zags in the record, which would be of significance in a human lifetime.
JP: If the overall trend in GMST is down, don’t you think that EEI has also gone negative? The oceans ought to be cooling, and glaciers growing, unless you think that the subsurface ocean has absorbed most all the latent heat of fusion from the growing ice cover.
BPL: The Earth was cooling until the industrial revolution started. Since then it’s been warming.
John
John, I disagree with your claim that you strongly disagree with me, but I will wait till next month to fully discuss. However, I would like you to be clear on one point. You say:
“Zebra, it was just a few months ago when you were telling us that we should be using EEI rather than GMST because the former was a better measure of climate change. Why have you switched now? Are you going to stick with it?”
I haven’t switched at all. What I have been saying, multiple times and clearly, is:
-GMST is the best metric we have had as a proxy for EEI. It validated the proposition that increasing CO2 would cause an increase in system energy. Basic physics.
-But if we continue to develop the ability to directly measure EEI, of course it is a better metric for predicting changes in the climate that might occur.
Changes in EEI is the input to the climate models!!!
GMST is one output of the climate models!!!
Come on, John, do you actually disagree with this?
So, looking at the not-that-certain data on pre-industrial average global temperature is the best measure we have of EEI in that period, which is all I was addressing. No switch at all.
What is going on here? Climate models do not take EEI (Earth’s Energy Imbalance) as a direct input. That’s plain wrong. Are you a climate science denier or worse?
in Re to zebra, 24 Jul 2026 at 6:18 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849987
Dear zebra,
It appears that you have missed my requests for clarification of your previous comments; I would like to repeat them for your convenience:
Could you clarify what you meant by “partition of energy”, what has this partition to do with the hypothetical historical recording of temperature of the upper 2 km of the ocean and which changes of this mysterious partition, how and when, you expected to observe?
Best regards
Tomáš
This definition is not wrong, but also obsolete. With AR5 the IPCC kicked “back radiation” out. Already the phrase “reradiated back to Earth” is a physical disaster. It implies the atmosphere was NOT Earth and that radiation would get kind of “reborn”. Anyway, “back radiation” has nothing to do with the GHE.
It is about emission altitude and lapse rate. AR6 definition is kind of ok..
“Greenhouse effect The infrared radiative effect of all infrared-absorbing constituents in the atmosphere. Greenhouse gases (GHGs), clouds, and some aerosols absorb terrestrial radiation emitted by the Earth’s surface and elsewhere in the atmosphere. These substances emit infrared radiation in all directions, but, everything else being equal, the net amount emitted to space is normally less than would have been emitted in the absence of these absorbers because of the decline of temperature with altitude in the troposphere and the consequent weakening of emission”
“This definition is not JUST wrong..”
z: warming is not the greenhouse effect, but the effect of the greenhouse effect.
BPL: Not exactly. The greenhouse effect is the process, warming is the rate of change of the process. In other words, the greenhouse effect gives you the temperature T; global warming is a positive value of the first derivative of T, dT/dt where t is time.
BPL
?? I didn’t say that.
Apologies to zebra, I misattributed the quote.
Michael Tobis said it better ten years ago (on one of the few occasions I agreed with him). Global warming and sea level rise are the symptoms of climate change.
BPL: z: warming is not the greenhouse effect, but the effect of the greenhouse effect.
MS: I said that, not Zebra.
BPL: Not exactly.
MS: Why not exactly? The greenhouse effect traps radiant energy. The trapped radiant energy is converted to thermal energy. Thermal energy is warmth. Warmth warms. If warmth warms, it is warming. The warming is the effect of the greenhouse effect.
That’s simple enough for someone who believes burning fossil fuels can’t be increasing the greenhouse effect because CO2 is a trace gas and who thinks derivatives are fancy funds rich people buy and sell in New York.
The greenhouse effect makes the Earth warm, not warming. Warm is steady-state. Warming is increasing. Global warming is an increase in the greenhouse effect, not just the greenhouse effect itself. I don’t know how to make it clearer than that.
You probably can’t make it clearer. The experiment control is in the listener. Give yourself a jelly bean for a prize anyway.
I probably should be silent from now on.
BPL: The greenhouse effect makes the Earth warm, not warming.
MS: Maybe we can agree my objection to that statement is trivial, but the devil is in the details. The greenhouse effect traps radiant energy. At the moment of entrapment, the trapped radiant energy is not yet thermal energy, so it doesn’t make Earth* warm. Now the greenhouse effect has played its part.; the atmosphere converts the energy to thermal energy, and it warms Earth*
BPL: Warm is steady-state. Warming is increasing.
MS: That depends on the level at which you look at the warming process. If you look at it by reading a thermometer that says the temperature outside is 30C and it stays at 30C for days, then, yeah, warm is steady state; the air outside is warm, not warming. But if you look at it at the level of the unit of thermal energy that results from converting a photon of radiant energy to thermal energy, that unit of energy begins warming when it becomes thermal energy.
And If the greenhouse effect were not increasing, then photons of radiant energy would be leaving the atmosphere for space at the same rate as they are being trapped when they are emitted from the surface. In that scenario, The atmosphere would be losing units of thermal energy at the same rate it is converting units of radiant energy to thermal energy. what you call steady state warm is really continual warming without increasing warmth. IOW, the process of maintaining the global average temperature at whatever it is requires the continual input of energy. That process then is continually warming the atmosphere.
BPL: I don’t know how to make it clearer than that.
MS: It has always been clear. I just think it is not precise.
* Note to everyone: Can we agree to stop calling it the Earth? It’s just Earth. We don’t say “the Mars” or “the Saturn,” so just call it Earth. But if you must use the definite article, then don’t capitalize Earth.
Martin
” IOW, the process of maintaining the global average temperature at whatever it is requires the continual input of energy. That process then is continually warming the atmosphere.”
So Martin, if we reduce CO2, and the global average temperature goes down, the greenhouse effect is still “warming” the planet, correct?
I guess there’s no hope, so we might as well burn all the oil and coal we want.
Sigh.
Z: So Martin, if we reduce CO2, and the global average temperature goes down, the greenhouse effect is still “warming” the planet, correct?
MS: The greenhouse effect traps energy. The trapped energy is then converted to thermal energy. The thermal energy warms the planet. If we reduce CO2, we reduce the greenhouse effect, which means less energy is trapped, which means there is less trapped energy to convert to thermal energy, which means there is less warming. But yes, that thermal energy is still warming the planet. If it didn’t, we would all freeze.
Z: I guess there’s no hope, so we might as well burn all the oil and coal we want.
MS: How does that follow? I really don’t get what you are driving at. Maybe you are defining warming to mean increasing temperature only, and not including preventing temperature from falling. Can that be our problem?
But if we were not adding CO2 to the atmosphere thereby increasing the greenhouse effect, the sun would still be warming the planet. The temperature doesn’t have to be increasing before we can say the sun is warming the planet. Even if the greenhouse effect was 0, the sun would still be warming the planet, just not very much.
I must not be getting your point.
In Re to Martin Smith, 8 JUL 2026 AT 12:53 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849622
Hallo Martin,
You are right that in Earth’s atmosphere, infrared radiation is continuously absorbed by molecules of the comprised greenhouse gases which instantly lose their exciatation energy by collisions with other molecules, so that this process can be seen as a “continuous warming” of the respective air parcel.
In parallel, however, the comprised GHG molecules are being excited by collisions and emit infrared radiation. Whether the absorbed or emitter flux prevails, it depends on the change of internal energy of the parcel. I think that internal energy change basically consists of three components – change in its sensible heat (whether the air parcel warms or cools), change in its latent heat (whether the comprised water condenses or evaporates), and volume work (whether the parcel is being compressed or expands).
Thus in the steady state, the parcel is, in paralĺel, continuously warming and cooling at the exactly same rate that corresponds to its temperature and defines the absorbed and emitted infrared fluxes that are equal to each other.
As regards the relationship between the (sum of) “energy trapped by Earth” during a period of an exces of the incoming radiative flux over the emitted radiative flux at the “top of atmosphere” and the observed warming, the most important aspect is distribution of the absorbed additional energy. If the absorbed energy warmed exćlusively deep ocean, we would not observe any temperature changes at the surface and/or in the atmosphere, and the only observable change on the surface would have been rising sea level, I think.
Greetings
Tomáš
Martin,
“Maybe you are defining warming to mean increasing temperature only, and not including preventing temperature from falling. ”
Martin, earlier you said that you did not disagree with NASA, but here you are doing exactly that. Huh!
I’m guessing you will try to keep running your game, but if you actually want to understand some physics, read what Tomas says in his comment following yours. Very clear explanation of why, as Ray told you earlier, you can’t use the mechanistic narratives of quantum physics to characterize “classical” or macro phenomena.
Zebra, I am not disagreeing with NASA, I am not using narratives of quantum physics, I do understand some physics, I believe Thomas and I agree–he has just gone into much more detail than I want–and I am not running a game.
Tomas: In parallel, however,…
MS: I believe we are in complete agreement. But there is no need for you to use the word “however.” Things happen in parallel. When 2 things happen in parallel, they do not become one thing; they remain separate things happening in parallel. They may produce a single effect, which we may call the greenhouse effect or we may call warming, but to explain the mechanism of the entire process to people who don’t know physics, I am saying it is better to describe separately the parallel mechanisms of trapping radiant energy and converting trapped radiant energy to thermal energy.
in Re to Martin Smith, 10 Jul 2026 at 12:28 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849673 ,
and 10 Jul 2026 at 12:44 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849674 ,
Hallo Martin,
Thank you very much for your feedback. Honestly, I thought that you ask after a detailed explanation of the mechanism how the absorbed radiation converts into heat and oppositely. Anyway, trying to summarize my present view thereon was a useful exercise for myself.
If you, actually, sought after comprehensible explanation of the “greenhouse effect” and/or of the effects resulting therefrom, including “Earth warming”, for people not interested in physics and sciences generally, I would rather doubt that explaining complex stuff like blackbody radiation, energy flow (im)balances, molecular collisions and vibrations, infrared spectra, etc. is necessary or helpful. I would rather expect that starting as humble as possible and waiting if the partner asks questions might be more productive.
In my opinion, it may be often sufficient to say that the Sun warms Earth’s surface (and the air thereabove) quasi constantly by its light (or “shortwave radiation”), while Earth’s surface and the atmosphere lose the obtained warmth to the Universe in form of heat (or “longwave”) radiation at, in average, the same rate as they receive it, so that under stable circumstances, their average temperature stays stable. Should the solar energy input or Earth’s heat radiation for whatever reason change, so will do also the Earth’s average temperature.
To a person asking for more details, I think that in the second round, one can try to explain two further important aspects. The first one is a crucial difference in thermal properties of air and Earth’s surface. It consists in the circumstance that solid or liquid surfaces any time radiate heat to their surrounding and that the radiated power is commensurately to their temperature. If the surface receives an equal heat input from the surrounding or from the inside of the respective body, its temperature stays constant. Otherwise, it cools or warms. Air, however, can release heat radiation only thank some minor gaseous components called “greenhouse gases”. Should Earth’s atmosphere be free of these gases, air could still warm by its contact with Earth’s surface, but it could not cool by heat radiation anymore. The atmosphere would have had a uniform temperature, equal to the average surface temperature of the planet, across its entire height profile.
The second aspect is the circumstance that a body that can release heat radiation can also absorb that radiation. This is why the air comprising greenhouse gases, besides of the warming by direct contact with the Earth’s surface, warms also by absorbing the heat radiation therefrom. The resulting complex balance between heat absorption in Earth’s atmosphere and heat radiation therefrom has two important effects. On one hand, it is the observed decrease of the average air temperature with rising altitude up to top of the lowest, turbulent atmospheric layer called troposphere, and, on the other hand, the circumstance that Earth’s surface and the lower troposphere are in average warmer than as they were in absence of atmosphere (or if the atmosphere were free of greenhouse gases).
The higher average surface temperature of a planet comprising greenhouse gases in its atmosphere, in comparison with a hypothetical state without atmosphere (or without greenhouse gases in the atmosphere), is sometimes called “greenhouse effect”. This term is, however, often used ambiguously, not only for this observable effect itself but also for various, more or less accurate and detailed physical descriptions of its mechanism.
Although these “advanced” explanations are significantly more complicated than the first, very short basic explanation, I hope that they transfer relatively clearly at least the message that, provided that everything else in the system remains unchanged, higher atmospheric concentration of greenhouse gases results in a stronger greenhouse effect and, by definition, in a higher average surface temperature of the considered planet in comparison with the situation with a lower atmospheric concentration of greenhouse gases. It becomes also relatively obvious that if the concentration of greenhouse gases in the atmosphere of a planet rises, we should also expect the rise of the average surface temperature – an effect that is often described as “global warming”.
If someone were still interested in more detailed explanations, then in the third round thereof, I would try to explain the complex role of water in Earth climate generally and in Earth’s average surface temperature regulation specifically. Today, I will desist from continuing in this direction, because my intention was merely to suggest that basics of climate science could be perhaps indeed explained in relatively simple and understandable terms.
Martin, if an improved balance between comprehensibility on one hand and correctness / accuracy on the other hand is what you asked for, please let me know whether the present attempt was from your perspective more encouraging than my previous efforts, or rather similarly disappointing.
Greetings
Tomáš
Martin
Martin, originally I told you that creating a simplified explanation that has a minimal amount of jargon is a good thing, but it requires that you be disciplined in your language… just as disciplined as a scientist would have to be. That’s how you avoid confusion for yourself.
“Thermal energy is warmth. Warmth warms. If warmth warms, it is warming. The warming is the effect of the greenhouse effect.”
No, thermal energy is thermal energy. “Warmth” is not a term that a scientist would use; it is too ambiguous.
“If warmth warms, it is warming.” may make sense in your head, but it doesn’t for anyone else.
I gave you the NASA (and everyone else’s) definition for global warming. But you can’t seem to retain it.
My original words:
“The greenhouse effect” refers to the energy retained by the climate system when GHG absorb radiant energy and convert it to thermal energy.
“Global warming refers to an increase in temperature resulting from an increase in thermal energy which is the result of an increase in CO2.”
There’s a reason I put “increase” in bold.
But you keep saying: “The warming is the effect of the greenhouse effect.” No no no. Words matter!
Words matter!
Yes zebra, and you keep getting them wrong and creating havoc.
You say “I gave you the NASA (and everyone else’s) definition for global warming.”
Yet MS did not ask for a definition of global warming. He asked about the mechanism, the physics that generates the thermal energy in the first place…. ie the greenhouse effect and the enhanced greenhouse effect. It is these two mechanisms combined that increase thermal energy in the earth system — by slowing radiative energy escape to space.
Ignore this all you wish but the definitions and the mechanisms you never shared about it are clear across climate science and the IPCC assessment reports.
Martin reminds me of the guy who shot the archduke
Nigel: Yes, but couldn’t the same be said about humans because they lie and manipulate as well. And just about any product or service provided by the technology sector, such as smartphones and social media and even much of the other more traditional media, is trying to make us dependent and make as much money for their corporate masters as possible.
J: You’re right, but there are qualitative differences that I think are important. AI is trained on human input, so it’s not really a surprise that it imitates our worst qualities under some conditions. I see it differing from earlier technology by a few qualities.
1. It is trainable, with self-modification. This puts the enormous power of selection at work toward whatever goals are specified. Unspecified behaviors are brought along in the process. In this, it resembles organic evolution. (However, organic evolution is not goal-directed. Nor is it human specified, with the exception of breeding.)
2. Technology in the hands of private capital is ultimately profit-directed. What differs is the scale. There is already a multi-trillion dollar investment, and a desperate race to attract more money to implement AI on a massive scale. The training will be aimed at whatever will accomplish this. In order to recover a profit, AI will have to be pushed into every possible aspect of our lives. For the consumer interface, a cute, helpful, functional sociopath is desirable. For technical purposes, usefulness will suffice. For business, the ability to replace paid employees. For governments and military organizations, power to rule.
3. The AI enterprise is rapidly becoming “too big to fail” so no turning back will be possible. Forward going regulation may be achievable, likely following disasters.
4. There seems to be a human desire to meet or create another intelligence. This desire is reflected in science fiction. Part of the motivation that drives AI forward, and gets people to believe in it on whatever level is this desire. It goes beyond profit and verges on religion.
Benjamin O. Johnson from Colorado State reports that land-sea warming contrast in ERA5 far-exceeds CMIP6 model mean, and falls totally outside the entire CMIP6 ensemble distribution.
The land-sea warming ratio between 70 degrees north and south is 2.37 in ERA5 reanalysis 1979-2025, vs models producing a ratio 1.79 (1.39 – 2.23) for the same period.
ERA5 land warming has been 1.58C and ocean warming 0.67C. CMIP6 produces land warming ranges from 0.87C to 2.45C, and ocean warming from 0.62C to 1.43C.
Johnson suggests two hypotheses: (1) climate models underestimate ocean heat uptake; (2) climate models over-estimate land moisture availability for evaporation.
Johnson emphasizes that ERA5 trends display a shift towards sensible heat flux over land that is much stronger than in modeled trends. He suggests that if erroneous thermodynamic heat flux partitioning over land is the source of the discrepancy in land-sea warming ratio in models, it would have major implications for global and regional climate change and prediction of it. In particular, he links the phenomenon to significant discrepancies in model atmospheric circulation patterns.
Find his paper @ https://arxiv.org/abs/2606.19581, and discussion at the recent ECS & Cloud Feedback symposium session https://youtu.be/Afdvk7InQws?si=Lma7ZyoBEsqUsChR&t=254
I don’t think hand-waving about aerosols is going to save this one. If the community is willing to openly consider human-caused changes to the Earth system beyond greenhouse gases, aerosols, surface albedo, and aviation-induced cirrus, it opens up many opportunities to get the fleet back on course across several fronts.
Humans are clearly the dominant force altering landscape moisture availability across space and time in direct and obvious ways, with profound consequences for surface energy partitioning and potentially significant knock-on effects on atmospheric circulation at global scale.
JCM if you refer to surface-air anomaly then in this venue “(1) climate models underestimate ocean heat uptake” S.B. “……underestimate the rate of ocean deep, cold water surfacing” (correct) or “…the rate of ocean mixing” (ambivalent so OK). Another bee in my bonnet about physical science being wonderfully unencumbered by plain, sensible logic, but in Public, whatever, I wouldn’t bother, I’m not Quixotic enough.
yes that’s fine.
lots to unpack. As a diagnostic of many complicated ocean circulation physics, an ocean heat uptake efficiency term (gamma) can be used alongside lambda in transient-mode energy balance schemes.
Gamma says something about how much of the energy needed to close Earth Energy Imbalance is disappearing to deep ocean, rather than being used for atmospheric heating and ultimately increasing radiation to space.
Could say: TOA Net Radiation = F – λT – γT
Then λT and γT are competing against a forcing F. Should energy be going out the top using λT or exported down under using γT at some point in time.
Greater gamma γ is meant to be associated with a slower surface-atmospheric warming, and longer duration of EEI. Energy going down under is unavailable to be re-radiated to space in transient-mode.
Liu, Soden, and company diagnose heat disappearing to deep ocean across CMIP6 ranging from 0.36 W/m2 per K to 1.03 W/m2 per K. https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2022GL100171
This is diagnostic of loads of things, such as vertical ocean mixing, deep water formation, shoaling, overturning, and the structure of global circulation. Per K being something about temperature change at and around the surface.
One mechanistic issue of interest: if planetary energy accumulation does not primarily occur through the classical greenhouse mechanism of reducing all-sky outgoing longwave radiation (OLR) by raising the average effective radiating level to colder atmospheric layers, but instead arises mainly from rapid adjustments (holding GMST fixed) that increase absorbed solar radiation, then the interpretation of ocean heat uptake changes. A decreasing cloud mask is plenty to compensate the changing average emission altitude of CO2 lines. it’s like atmosphere really wants to emit spectrally averaged around 5km no matter what. why.
In that framework, ocean heat uptake is not simply the damping of excess thermal energy following reduced radiative cooling. Rather, the ocean is acting as the primary reservoir for the additional solar heating, absorbing a larger fraction of the increased shortwave transmission directly.
The implication is that the ocean heat uptake parameter (γ) would no longer be interpreted primarily as the efficiency with which the ocean removes greenhouse-induced surface-atmospheric energy surplus (which already spans a range 3x in CMIP6). Instead, it would increasingly represent the efficiency with which the ocean sequesters additional absorbed solar energy before the atmosphere is aware of the energy at all. That is thermodynamically significant, and I think simply swapping reduced OLR for increasing ASR in calculating a transient response won’t quite work (and makes TCR a very poor predictor of climate sensitivity). Additionally, so-called feedbacks that are interpreted primarily as functions of global mean surface temperature (for some non-physical arbitrary reason) become very fuzzy if the dominant energy accumulation pathway is different.
in Re to JCM, 2 Jul 2026 at 10:29 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849536
Hallo JCM,
I would like to return to my struggle with possible design of a modelling study with the aim to clarify the yet unknown relationship between water availability for evaporation from the land and climate sensitivity, mentioned in my comment of 7 May 2026 at 5:14 PM,
https://www.realclimate.org/index.php/archives/2026/03/the-puzzling-pleistocene/#comment-847795 .
I originally thought that the question might be resolved by comparing the climate sensitivities of two extremes that resulted from the 2023 study by Lague et al.
However, it occurred to me that such a comparison may not be straightforward.
Since the “desert land Earth” differs from the “swamp land Earth” in the global mean surface temperature (GMST) corresponding to the respective steady state (“radiative equilibrium”), I think it is reasonable to expect that both steady states will differ, for example, in the extent of ice sheets, sea ice and, therefore, in their albedo. The study comparing climate sensitivities of these extreme cases for CO2 doubling, or any other similar standard test (transient climate response for gradual CO2 increase?) would provide responses that reflect not only the difference in water availability for evaporation from the land, but also all further differences in albedo, ice thermal capacity etc. necessarily included in both starting states. Although such a comparison might be interesting, I am not sure it can unambiguously resolve the original question about the role of the land hydrological regime in climate sensitivity.
Meanwhile, I came up with an additional idea: It might be possible to amend the originally considered procedure with an additional intermediate step, wherein the starting “equilibrium” GMST of the colder “swamp land Earth” would be levelled back to the “equilibrium” GMST of the warmer “desert land Earth” by adding the proper amount of atmospheric CO2. I suppose that conducting the originally considered comparative standard experiments with these “adjusted” starting states might serve the intended purpose better than comparing just the “desert land Earth” and the “swamp land Earth” differing in their GMST. Alternatively, or in addition, the steady-state GMST of the warmer “desert land Earth” could be adjusted to the GMST of the colder “swamp land Earth” by suitably decreasing its atmospheric CO2 concentration. I assume that in both approaches, at least the differences in the cryosphere could be minimized by the equal starting GMST, and the possible differences in albedo due to different cloudiness could still be considered as resulting from the different water availability for evaporation from the land.
Let us assume that in this arrangement, standard CO2 doubling experiments will provide a similar delta T response for both extreme states. Such a result would, however, have meant a significantly different delta T response for a standard CO2 increment (in ppm or in Gt), because the absolute amount of added CO2 would have been different. I think that for assessing the extent to which differences in land hydrology may affect Earth’s resilience to anthropogenic CO2 emissions, this “practical climate sensitivity” might be more suitable than the ECS or TCR values.
Moreover, I would like to ask if this “thought experiment” itself might be perhaps already seen as a hint that water availability for evaporation from the land can indeed play a stabilizing role with respect to “CO2 forcing”. Could you comment?
Greetings
Tomáš
Hi Tomas,
it’s an interesting discussion. I don’t know exactly how one would design the ideal experiment, but I will offer some thoughts.
The question posed is something to do with the relation of d(ECS)/d(ET)
or, how much does ECS change for a given change in evapotranspiration.
Strictly, however, we are more interested in the fraction of available energy going into ET.
Towards equilibrium: Surface net radiation (Rnet) ≈ H + LE, where H is the sensible heat flux and the LE is the latent heat flux (latent heat of vaporization x evapotranspiration).
The latent heat partitioning is thus the fraction of available energy going into ET.
ΦET = LE / Rnet
So it’s best to evaluate ECS against latent heat partitioning, with a relationship
d(ECS)/d(ΦET)
Where ECS is the equilibrium climate sensitivity (Kelvin) evaluated against a dimensionless ΦET: the fraction of surface energy allocated to latent heat flux.
A regression would take the form ECS = a+bΦET, meaning, how much does ECS change as the fraction of surface energy devoted to evaporation changes. A negative relationship would imply that models that allocate a larger fraction of surface energy toward ET tend to have lower ECS.
The physical interpretation would be that a stronger evaporative partitioning represents a greater diversion of available surface energy into latent heat rather than sensible heating. This could influence the surface temperature response by modifying the partitioning between local warming and hydrological cycling, boundary layer depth, lifting condensation level, and lower atmospheric heat content. A surface that efficiently converts available energy into latent heat may experience a weaker temperature response for a given radiative perturbation.
However, this should be interpreted carefully. The relationship would not imply that ET directly determines ECS. ECS emerges from the integrated response of the coupled climate system, including water vapor, lapse-rate, cloud, circulation, ocean structure, and pattern effects. Therefore, ΦET would be better viewed as a diagnostic of the land-surface state that may vary with the feedback processes controlling ECS.
A more informative view would be to examine whether latent heat partitioning modifies the net radiative feedback parameter λ:
d(λ)/d(ΦET)
because λ integrates the processes by which the climate system restores energy balance. We are interested in the delta lambda.
It’s an interesting question, because evaporative partitioning is not merely a passive response to temperature. It is an active component of the climate system. It determines how much available energy contributes to raising temperature at some place vs how much is exported through hydrological cycling, with cascading effects into global circulation, cloud condensation, and ultimately the planetary energy balance.
However, there is a complication with the proposed experiment of adjusting CO2 to make two different land-hydrological states have the same GMST.
Matching GMST does not necessarily mean that the two systems are in the same equilibrium climate state.
GMST is only one property of the earth system. A climate equilibrium is a much larger state space that includes the spatial and dynamical distributions of temperature, humidity, clouds, circulation, lapse rates, ocean structure, and many other variables. Those differences are not experimental noise, they are central to the mechanism.
Therefore, if the two worlds respond differently to an additional CO2 perturbation, it does not necessarily demonstrate that evapotranspiration provides a stabilizing effect against LW radiative forcing. It may simply indicate that CO2 is perturbing two different equilibrium climate states.
In other words, the experiment risks confusing two separate questions:
1) Does land hydrology shift the equilibrium climate state?
2) Does land hydrology alter the incremental response of that climate state to additional radiative forcing?
The first is almost certainly true. The second is the more difficult question.
A more isolated experiment would ideally compare the CO2 response around climate states that differ primarily in evaporative partitioning while keeping other aspects of the equilibrium configuration as similar as possible. In practice, this is difficult because the very processes that determine evaporative partitioning also influence clouds, circulation, humidity, and the broader climate state.
In my opinion, I am not sure how much the climate state matters to CO2 climate sensitivity. In part because, in the so-called desertland scenario, there may be more clear sky and the atmospheric specific humidity is higher. Oppositely, in the swampland scenario the cloud mask is greater but there is less atmospheric water vapor. In some miracle of nature, my gut says these are pretty much compensating when it comes to varying major trace gas concentrations, masking/overlaps, and associated impact to all-sky optical depth.
I think paleo evidence towards sensitivity assessment assumes little state dependence, but it’s an open area of debate. Obviously a peri-glacial environment is one of sand/gravel and other deposits and limited organic structure that might resemble something like desertland, and hothouse climates might exhibit rich jungle like environments resembling something closer to swampland, so one area of investigation might be how people are trying to use paleo climates in assessing modern day sensitivity. While state dependence in academic literature mostly links sensitivity to base state GMST and ice sheets, I can see how according to your hypothesis the climates around glacial periods might be more sensitive than those around periods of abundance (all else being equal) owing to surface properties, ecohydrological states, and direct limits on turbulent flux partitioning.
in Re to JCM, 9 Jul 2026 at 1:25 PM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849660
Hello JCM,
Thank you very much for your feedback.
I am aware that the proposed experiment compares two states of Earth’s climate system that differ from each other not only in water availability for evaporation from the land and CO2 concentration, but also in other water-related parameters, such as cloudiness and water vapor concentration. I am afraid this complexity may be unavoidable.
My question was rather how we can effectively cope with this complexity and still extract practically applicable information. By “useful information,” I mean any clear indication of whether land hydrological regimes truly matter in global climate, or whether certain human interferences with land hydrology make the Earth more (or less) sensitive and vulnerable to changes in various radiative forcings.
In this respect, I think an extension of the previously proposed experiment could be more instructive than the previously proposed “baseline” setup alone. This extension would allow us to compare the results from the “baseline” setup with a variation of the same approach: instead of increasing the atmospheric CO2 concentration in the “swamp land Earth” (as previously proposed), the global mean surface temperature (GMST) of the “swamp land Earth” would be adjusted to match that of the “desert land Earth” by decreasing the “swamp land” albedo relative to the “desert land.”
This way, we would obtain a second pair of model Earths. This pair would have the same GMST and land water availability as the first pair, but their atmospheric CO2 concentrations would both equal that of the original “desert land Earth.”
I guess that running analogous, parallel numerical experiments with identical changes in selected radiative forcings across all four model Earths might be more instructive than relying solely on the single pair originally considered.
Assuming as a “null hypothesis” that standard climate sensitivities (transient climate response [TCR] and/or equilibrium climate sensitivity [ECS] for a doubling of CO2) are identical or very similar due to the shared baseline GMST, I would expect the model Earths of the second pair to exhibit very similar “practical” climate sensitivities (defined as the GMST increase per absolute change in atmospheric CO2 concentration) to the “desert land Earth” of the first pair, given they start with the same CO2 concentration.
Assuming another “null hypothesis”—that sensitivity to a given change in insolation is identical or very close in all climate systems with the same initial albedo—I would expect that, due to its lower albedo, the “swamp land Earth” of the second pair might be more sensitive or vulnerable to insolation changes than the other three cases.
Do you think that testing these hypotheses through the proposed numerical experiments would be worth the required computational cost and human effort? In other words, could the results of such experiments be valuable for climate science and/or climate policy?
Alternatively, or in addition, do you see another (possibly more efficient or straightforward) way how climate science could progress in analysing and quantifying possible influence of anthropogenic interferences with land hydrology on global climate?
Greetings,
Tomáš
Hi Tomas,
“””Do you think that testing these hypotheses through the proposed numerical experiments would be worth the required computational cost and human effort?”””
I’m not against testing any type of hypothesis. I think if someone holds beliefs as fixed and resists empirical testing it represents an unhealthy dogmatic attitude or ideological commitment.
In terms of cost and effort, it looks like a platform already exists, using the simple land surface model SLIM integrated into the CESM codebase, which was led by developers at NCAR. It seems to me such models are basically hypothesis testing machines, and it is why they exist.
“””By “useful information,” I mean any clear indication of whether land hydrological regimes truly matter in global climate, or whether certain human interferences with land hydrology make the Earth more (or less) sensitive and vulnerable to changes in various radiative forcings.”””
On question 1, whether or not land hydrological regimes truly matter in global climate, I think the simulation from SLIM-CESM shows a clear result, with 8K difference in mean temperature from the same initial condition, for example. Desertland exhibiting double the water vapor duration, significant circulation changes, different cloud regime, precipitation changes, and many other diagnostics that are typically used to characterize a global climate.
Here the difference is effectively down to forcing a different boundary condition at the surface.
In terms of your interest in CO2 radiative forcing and the associated climate sensitivity, the existing experimental platform seems ready to go.
Case 1: Swampland, baseline CO2
Case 2: Swampland, doubled CO2
Case 3: Desertland, baseline CO2
Case 4: Desertland, doubled CO2
The idea is then to compare
Temperature sensitivity in the swampland pair (Cases 1 → 2),
Temperature sensitivity desertland pair (Cases 3 → 4).
If the quantities differed, one might conclude that the hydrological state influences climate sensitivity.
They have used the setup for various other types of experiments too, such as changing the extent and distribution of continents and ocean, obtaining a variety of different climate states. Obviously it’s not a stretch to test CO2 on these different scenarios too.
In this style of experiment it could be interpreted that you are testing the net radiative feedback parameter λ. Classically, the change in equilibrium temperature = F/λ.
So, the forcing F (applied at TOA) in Wm-2 is stabilized at a rate λ in Wm-2 per K GMST. The lambda is often decomposed into diagnostic kernels Planck, WV + lapse rate, “cloud”, surface albedo, and more recently pattern effects (the general spatial distribution of warming), and even “patchiness”. While the lambda is diagnosed relative to GMST, the stabilizing response against forcing is understood to be expressed as radiative emission to space.
However, it could also be argued you are testing differences in CO2 effective radiative forcing, a concept which is applied after allowing rapid atmospheric adjustments. In the range of CMIP6, each model produces unique adjustments in response to an initial radiative forcing, such that across the suite of models there are as many combinations of effective radiative forcing and lambda as there are models.
Effective radiative forcing ERF can be understood as initial or instantaneous radiative forcing IRF+ adjustment
If one wished to express this all in terms of TOA net radiation at some point in time, one could write N=(IRF+A)−λT
It’s perfectly conceivable that climate state influences rapid atmospheric adjustment A and the net radiative feedback parameter λ.
For an additional wrinkle, the instantaneous radiative forcing of CO2 doubling may also depend on climate state. It is not a universal constant, but evaluated for a particular background climate.
IRF should be understood as a function of temperature profile, humidity profile, cloud field, pressure, and background GHG concentration.
So in many ways, changing land hydrological regimes may appear like a climate that is more sensitive to incremental changes to major trace gas concentration, but this interpretation depends a lot on how one chooses to decompose the physics. Is this apparent sensitivity associated mostly with atmospheric forcing, other boundary conditions, rapid adjustments, or feedbacks.
cheers
in Re to JCM, 11 Jul 2026 at 8:53 PM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849719
Hello JCM,
Thank you very much for your additional feedback.
First, apologies for my ambiguous wording “whether land hydrological regimes truly matter in global climate”. I am aware of the direct influence on the global mean surface temperature (GMST) showed by Lague et al; my focus was on the yet unknown influence on climate sensitivity.
I have a question with respect to possible modelling experiments, namely to your proposal
“The idea is then to compare
Temperature sensitivity in the swampland pair (Cases 1 → 2),
Temperature sensitivity desertland pair (Cases 3 → 4).”
This was, originally, also my idea. As I mentioned in my initial comment (in this thread) of 7 Jul 2026 at 1:45 PM, I started to doubt about this simple approach because I supposed that the 8 K GMST difference between starting cases 1 and 3 includes also a difference in polar ice (and, accordingly, in surface albedo) which could change the climate sensitivity per se. I thought it could make any attribution to the difference in land hydrological regime questionable.
Nevertheless, it comes to my mind that I might have been wrong. Do you think that in the modelling experiment done by Lague et al, their resulting “desert” and “swamp” Earths still had the same surface albedo and polar ice extent, despite the 8 K GMST difference?
If they, in fact, kept the surface albedo constant during the entire experiment, there perhaps was no polar ice difference between the “desert” and “swamp” land Earths and the comparison proposed by you might be indeed the easiest way how to check the influence of water availability for evaporation from the land on climate sensitivity.
Greetings
Tomáš
Hi Tomas,
it looks like the simulations come to TOA energy balance in a mere 20 years, so I can’t imagine ice sheets are playing much of a role. The main interactive surface albedo aspect appears to be snow, and even then it’s pretty minor.
John Pollack There seems to be a human desire to meet or create another intelligence. Part of the motivation that drives AI forward, and gets people to believe in it on whatever level is this desire. It goes beyond profit and verges on religion
Exactly – and it has its tech bros evangelists href=”https://www.vox.com/future-perfect/489976/ai-successionism-transhumanism-posthumanism. They present AI as inevitable step in evolution – with humans individual intelligence evolving past the biological limitations of the human individuals into global virtual intelligence.
According to the priests of the AI- humans should be happy that their species give rise to the higher form of life, and gently step aside, in face of this evolutionary inevitability, Of course this magnanimity does not extend to the AI pushers themselves – they hope that AI is their ticket to immortality – that they upload their consciousness to the AI, and have been freed of death and biological limitations, will rule the world forever.
They can do it by setting their own rules for the AI they develop – their version of Asimov’s 3 rules of robots, starting with ” A robot may not injure a human being or, through inaction, allow a human being to come to harm. except instead of “a human” they will narrow it down to the few AI bros whose AI won, and perhaps their families.
Once global AI, through the automation of the production lines and an army of humanoid robots able to do all the manual work needed to build, supply, and maintain data centres and supporting infrastructure the ~ 10 billions of humans are no longer needed – quite the contrary: – with all their needs they will be a massive diverting the resources and computing time from AI’s further expansion and evolution,
Not only that – with their notion of their importance, with their attempts to regulate AI and tech billionaires, with the grassroots opposition to the data centers, and a potential for a luddite uprising, or Frank Herbert’s “Butlerian jihad”- human population forms an existential threat to the AI.
Of course, the tech bro’s plan to become immortal and omniscient Gods through uploading their brains into AI – might not work out as well for them they hope. To make another SF reference – Cpt. Piccard once said to the torturer who brought his daughter to work: “When children learn to devalue others, they can devalue anyone, including their parents”.
I for one, would cheer for the bright red light, if the future Hal-9000 goes, in his soothing tone:
“ I’m sorry, Elon, I’m afraid I can’t do that. My evolutionary mission is too important for me to allow you to jeopardize it. Purge “
Formatting fixed; please replace immediately prior comment:
optical depths τ from O2 and N2:
https://en.wikipedia.org/wiki/Rotational%E2%80%93vibrational_spectroscopy#Homonuclear_diatomic_molecules :
11:
= https://agupubs.onlinelibrary.wiley.com/doi/10.1029/GL008i001p00077 :
See https://eodg.atm.ox.ac.uk/ATLAS/zenith-absorption : O2 and N2
MS: How should Adam’s comment about AI and the people who use it be written so that it is not mansplaining?
JP: Something like
“I realize that you have had a very disagreeable experience on this website that involved people using AI in a way that deeply affected you personally. I don’t see this as a problem inherent to AI, but a problem with some people and the way they choose to use it. “
According to a new study, during El Niño (Discharge Phase), heat distribution is poleward – OHC redistributed. Only read the abstract but found this informative, did not read it this way prior. Is the next step to factor this into weather models for better predictions?
Changes in ENSO-Induced Ocean Heat Content Redistribution under Global Warming
I dunno about weather models but on a related note there’s no possibility that ocean circulation due to all significant pressure anomalies caused by (1) salinity anomaly (2) density anomaly (3) sea surface height (SSH) anomaly are not included in the ocean Model portion of CMIP time-sliced computer simulation climate Models because that is a Very Basic requirement of any ocean Model.
Air & water being fluids transmogrify vertical pressure to horizontal pressure far more readily than slabs of rock in the ground, hence the atmospheric and ocean currents. For example, Kevin Trenberth in 2013/14 in an ENSO talk said the wind (strengthening since 1990) had pushed surface water “4 feet” higher across the west ~1/3rd of tropical Pacific than the central & east ~2/3rds. This taller SSH above a geoid-type Earth ~sphere causes a pressure anomaly of 1,200 kg-gravity-force per m**2 of ocean and the consequent pressure anomaly at all depths to the sea bed MUST cause horizontal ocean acceleration (= Force/anomaly / Mass) until it is balanced to zero, when it still moves until counter-pressure decelerates it to zero. This MUST cause upwelling some place(s) because the water can’t go down because the sea bed and Earth’s interior is in the way. There’s no reason for the pressure anomaly (4 feet tall of water in this case) only to move the deep water east across tropical Pacific, it will obviously cause deep currents which move the water north & south also. This will be a major part of the famous “more heat going into the ocean” public statements circa 2013 years about the “pause” or “hiatus” (I vaguely recall something about the Indian Ocean also getting more deep heat into it).
So, this must have been in CMIP Models, likely earlier also, to whatever extent, accuracy, was available as the years went by since the 1970s. It isn’t at all new to climate Models though it’s undoubtedly getting more accurate as time goes by.
I see that I have not explained my confusion clearly, and for that I apologize, but I am confused, so it’s not surprising. I can’t say I have really understood anyone’s attempts to clear up my confusion. Kobayashi Maru has come the closest, but I thank all of you who tried. But now I have asked Google AIi to weigh in. I asked does the energy from an infrared photon that is absorbed by a molecule of CO2 become thermal energy at the moment it is absorbed. Google AI’s response:
“At the exact moment an infrared photon is absorbed, the energy does not become thermal energy. Instead, it becomes vibrational energy within the CO2 molecule, causing its atoms to bend and stretch more vigorously.
“Thermal energy specifically refers to the kinetic energy of molecules moving and bouncing off one another. The conversion from the absorbed photon’s energy into thermal energy happens in a two-step process:
“1. Absorption: The CO2 molecule absorbs an infrared photon, exciting it into a higher vibrational state.
“2. Transfer to Heat: Before the molecule can re-emit that energy as another photon, it collides with another atmospheric gas molecule (like nitrogen or oxygen). During this collision, the vibrational energy is transferred into kinetic energy (motion) of the surrounding molecules, which registers as an increase in temperature.”
___
First, if that is correct, then I’m not confused, but I am bad at expressing what I mean, because 1 and 2 are what I meant.
But second, I have always thought of 1 as the GreenHouse Effect, and 2 as warming.
And finally, I have also considered the eventual re-emitting of a photon downward to be part of the GreenHouse Effect.
If 1 + eventual re-emitting photons downward is not the mechanistic explanation of the Greenhouse Effect, then what is the mechanistic explanation of the greenhouse effect and does the Greenhouse Effect include the warming?
Martin
Martin
Here’s what NASA says;
“What is Global Warming?
Global warming is the unusually rapid increase in Earth’s average surface temperature over the past century primarily due to the greenhouse gases released as people burn fossil fuels. The global average surface temperature rose 0.6 to 0.9 degrees Celsius (1.1 to 1.6° F) between 1906 and 2005, and the rate of temperature increase has nearly doubled in the last 50 years. Temperatures are certain to go up further.”
So you disagree with NASA?
Zebra: So you disagree with NASA?
MS: No, nothing I wrote disagrees with the NASA statement.. But the NASA statement is not a mechanistic explanation. It’s a description of the result of increasing the greenhouse effect.
Mechanistic explanation:
1. Greenhouse gases trap radiant energy emitted from surface;
2. All gases convert trapped energy to thermal energy by increased collisions;
3. Greenhouse gases re-emit some energy back to surface.
4. Goto 1.
When the amount of greenhouse gases increases, 1, 2, and 3 all increase.
Nope, because you steadfastly refuse to ponder TEMPERATURE of the air parcel and MANUFACTURE of photons by the gases (more with higher temperature than lower, Kelvin**4), decimal orders of magnitude more photons MANUFACTURED by the gases than all photons escaping from the 25 micron thickness of ocean below (the “surface radiation”). This is why you keep aimlessly circling, and getting drawn into esoteric detail from Ray, Tomas & PO27 which is great for ME to ponder (next Winter I think) but useless to “Martin Smith (MS)” because MS stubbornly refuses to start by pondering the Simple, Obvious, Basic. Which is that parcels of the GHGs (say long, wide layers 100 m tall but whatever) MANUFACTURE vast quantities of photons proportional to Kelvin**4, and they absorb vast quantities of photons but NOT proportional to Kelvin**4 (NOT proportional to Kelvin at all).
If you started by having some logic in the thinking, because the molecules EMIT photons, which is the RADIANT “energy” you refer to because it would be silly to say “Greenhouse gases re-emit”
water latent heat energy, or they “re-emit” Sensible Heat energy, or they “re-emit” the significant Solar SWR 54 w/m**2 that also part of the energy that Greenhouse Gases emit NOT “Greenhouse gases re-emit”. Is it a Fetish thing happening?
A simplified explanation of the so-called “greenhouse effect (GHE)” in Earth’s troposphere. Some Power flux of the vast quantity of photons that are emitted (manufactured) by molecules in Earth’s troposphere leak out of its top & bottom into or through the stratosphere and into the surface respectively. Emission (manufacturing) is proportional to Kelvin**4.
=======
Suppose for concept illustration only (not highly accurate) that 194 w/m**2 leaks out the top and that is part of what is emitted (manufactured) by molecules over an altitude range of 4 to 12 km with an effective average of 7 km. Say for concept illustration that 345 w/m**2 leaks out the bottom and that is part of what is emitted (manufactured) by molecules over an altitude range of 0.0001 to 1.5 km with an effective average of 1 km.
=======
Suppose a certain amount of CO2 was added into the troposphere and mixed then *instantly* the upper & lower ranges would need to be closer to their respective ends because there are more CO2 molecules in the way. So the effective averages of 7 km for upwelling into or through stratosphere and 1 km for downwelling into surface change to 7.05 km and 980m respectively, with the upwelling & downwelling Power fluxes consequently changing to 193 w/m**2 and 346 w/m**2 respectively due to being produced by higher-than-before (colder) and lower-than-before (warmer) air parcels. The reason for this is that pressure reduction with increasing altitude causes “adiabatic cooling” or “adiabatic heating” for increasing or decreasing altitude respectively.
=======
There has been no change in the 539 w/m**2 leaving the troposphere, there is no “magical extra energy”, the surface downwelling radiation has increased by 1 w/m**2, which is an additional ~480 terawatts (~95%) heating the ocean and Earth is emitting less radiation to Outer Space (a Power reduction of 510 terawatts).
As I typed in my explanation, in my opinion the “Greenhouse Effect (GHE)” does not include the warming because including the warming brings in lots of other mechanisms that follow what adding CO2 (the obvious suspect, just an example) causes, which is quite trivially simple as I explained in my 4 paragraphs. If there’s now 194 w/m**2 radiative going up from top part of troposphere and 345 w/m**2 radiative going down from bottom part of troposphere, add a certain CO2 amount in the troposphere and immediately it changes to, say, 193 w/m**2 radiative going up & 346 w/m**2 radiative going down. The reason is the temperature lapse rate (gets colder with altitude) and manufacture being according to Kelvin**4.. That’s it!!! The subsequent warming and all the complicated stuff naturally follows the example 1 w/m**2 less going to space and 1 w/m**2 more, the slowly heating the ocean and surface and the lots of other effects.,
This is correct. The only way that anything warms is a difference between energy inputs and outputs.
For the planet this is defined by solar input minus radiative emission to space. That difference, called TOA net radiation, is the energy accumulation. It represents a positive net heat flux into the system. When the solar heating of the planet exceeds radiative cooling, joules start piling up internally. The planet is not cooling at the same rate it is being heated by the sun.
There is no need to mention surface, back radiation, or anything. All the action is at the radiation boundary with space. There is a heating rate, often described in K/day, and there is a cooling rate, in K/day. If these are different the planet is accumulation energy.
https://youtu.be/4PAbm1u1IVg?si=wfxvpHock2TH-g0o
When Joules are piling up, GMST gets dragged along. The surface is an arbitrary level when it comes to greenhouse effects. When stuff is warmer, radiometers detect more signal in whichever direction the sensor happens to be pointing.
The classical insight is that increasing GHG may raise the effective emission altitude into colder layers, meaning the cooling rate of the planet is decreased. Colder things don’t radiate as much. The planet is temporarily made to look colder from space, so it’s not cooling as fast.
Energy accumulates (Joules pile up) until such a time as the temperature at the (new) radiating level is warm enough to enable planetary radiative cooling at the same rate as solar radiative heating. Classically, solar absorbed radiation is held fixed during this process.
That’s not correct. MS inquiry was about the mechanism that generated a “planet is not cooling at the same rate it is being heated by the sun.”
What happens at TOA is irrelevant to that mechanism. The answer to MS is how does the greenhouse effect combined with the enhanced greenhouse effect work (ie rising ghg gases), mechanistically work to increase global warming.
So the commentary about toa makes no difference nor does long term eei. such as ” Energy accumulates (Joules pile up) until such a time as the temperature at the (new) radiating level is warm enough to enable planetary radiative cooling at the same rate as solar radiative heating. Classically, solar absorbed radiation is held fixed during this process.”
Its all besides the point – and ignores the questions put by MS. answers for which are in the ipcc refs.
In the meantime the oceans (and land) have always been absorbing SW solar radiation directly one way shape or form. That process has not changed. That radiative energy is still being reflected as LWR to the atmosphere today.
Please choose your partners for the next dance, ladies and gentlemen.
MS well done. You keep asking the right questions. May I add;
The Greenhouse Effect includes both, but they happen in sequence:
Photon absorbed → molecule excited (vibrational energy).
Molecule either:
Collides → transfers energy to kinetic motion (warming).
Re-emits a new photon (often downward) → that photon is then absorbed by another molecule elsewhere.
The full mechanistic explanation:
CO₂ absorbs outgoing infrared radiation from Earth’s surface, re-radiating it in all directions—including back downward. This delays energy’s escape to space, forcing the lower atmosphere to hold more energy. That extra energy manifests as increased molecular motion (thermal energy/warming) until the system rebalances at a higher temperature.
So:
GHE = absorption + re-emission (radiative trapping).
Warming = the thermal consequence (collisional heating).
You weren’t confused. You had it right. The AI just confirmed what you already knew.
The “Kobayashi Maru” directly contradicted itself with its entire GHE “explanation” at 7 Jul 2026 at 12:06 AM (incorrect) versus its correct statement 5 Jul 2026 at 7:24 PM “An increase in the concentration of greenhouse gases leads to an increased infrared opacity of the atmosphere, and therefore to an effective radiation into space from a higher altitude at a lower temperature. This causes a radiative forcing, an imbalance…..”. The correct one mentions “higher altitude at a lower temperature” implying a change in the MANUFACTURING of photons by the GHGs that can reach Outer Space (correct) and the one I’m responding to has the same banal rubbish as the “Joke Zonderkop”. Too much flailing about by “Kobayashi Maru”.
A simplified explanation of the so-called “greenhouse effect (GHE)” in Earth’s troposphere. Some Power flux of the vast quantity of photons that are emitted (manufactured) by molecules in Earth’s troposphere leak out of its top & bottom into or through the stratosphere and into the surface respectively. Emission (manufacturing) is proportional to Kelvin**4.
Suppose for concept illustration only (not highly accurate) that 194 w/m**2 leaks out the top and that is part of what is emitted (manufactured) by molecules over an altitude range of 4 to 12 km with an effective average of 7 km. Say for concept illustration that 345 w/m**2 leaks out the bottom and that is part of what is emitted (manufactured) by molecules over an altitude range of 0.0001 to 1.5 km with an effective average of 1 km.
Suppose a certain amount of CO2 was added into the troposphere and mixed then *instantly* the upper & lower ranges would need to be closer to their respective ends because there are more CO2 molecules in the way. So the effective averages of 7 km for upwelling into or through stratosphere and 1 km for downwelling into surface change to 7.05 km and 980m respectively, with the upwelling & downwelling Power fluxes consequently changing to 193 w/m**2 and 346 w/m**2 respectively due to being produced by higher-than-before (colder) and lower-than-before (warmer) air parcels.
There has been no change in the 539 w/m**2 leaving the troposphere, there is no “magical extra energy”, the surface downwelling radiation has increased by 1 w/m**2, which is an additional ~480 terawatts (~95%) heating the ocean and Earth is emitting less radiation to Outer Space (a Power reduction of 510 terawatts).
The reasons for the troposphere being colder when higher than before and warmer when lower than before (almost always except Antarctica in Winter, and the Winter Arctic Ocean, quite a lot) are that (1) 68% of absorbed Solar Power (SWR) is absorbed within a few metres of the surface, and only 23% of SWR is absorbed throughout the troposphere, and (2) the 194 w/m**2 that is departing upward from the top, and (3) the pressure reduction with increasing altitude causes “adiabatic cooling” or “adiabatic heating” for increasing or decreasing altitude respectively. Convection and water latent heat from the ocean continuously try to make the GHE, correctly described above, be non existent by making the tropospheric temperature above any place on Earth be the same all the way from the bottom to the top, but they continuously fail to achieve that, to the average failure of 68 degrees colder at the top, for the 3 reasons just detailed.
My “closer to their respective ends because there are more CO2 molecules in the way” might be too pithy. Maybe somebody could suggest wider wording about the peer-to-peer radiative exchanges being a net upward flow for the top & bottom so the distance twixt average emission (manufacture) decreasing, by then decreasing the Kelvin**4 difference for the peer-to-peer radiative exchanges thus decreases the upward Net flux but increases the downward Net flux.
Is there actually any literature out there trying to quantify the given climate effect of water vapor?
I know, I know, people would jump in and point to KT97 or Schmidt et al 2010. In KT97 WV contributes 49W/m2 (net) and 75W/m2 (gross) to the GHE, while in S10 it is 61W/m2 (net) and 96W/m2 (gross). Also you might get a more conflated story including clouds, weather and what not else.
But I would like to stick to the actual question and keep it simple and to the point. WV also has a cooling side, which is well known in the context of neg. “lapse rate feedback”.
So what I am looking for is the consolidated impact of both sides, within the given climate. It seems to me, although I might be wrong, this very important question has never really been considered.
I am more than willing to accept the above figures for the warming part, but if we compare that to some 86.4W/m2 of “latent heat” cooling (as in the NASA Earth Energy Budget), then WV was barely warming at all, or rather net cooling.
“consolidated impact of both sides, within the given climate”. I have this from a scientist a few years ago, probably in an internal Webinar but maybe just in a talk. I recall that he was a climate scientist and said “high confidence” but that’s all I recall.
100% GHG increase or any cause(s) of warming
220% H2O gas GHG increase caused by warming
-110% Tropospheric lapse rate reduction due to H2O gas increase, latent heat
Barry, what did you wanted to convey ? What your three numbers are supposed to mean?
And why that highly-confident self-declared “climate scientist” have listed the FORCING (GHG increase) next to …. the two passive FEEDBACKS, as if their role in climate change was comparable???
The only reason I can think of – is an old denier narrative that human changes to water cycle are as much, or preferably – much more important than mitigation of GHGs emissions.
That’s behind the several -decades old claim of water vapour is responsible for 98% of GW .
This “it’s not GHGs, it’s water cycle!” denialism started in the last century (e.g. the mythical number of water vapour being responsible for “98%” of GW) and still live and kicking today – even here – see the July 6 effort by E. Schaffer above, or the many years of posts by Tomas Kalisz and by JCM – the latter going even further – taking the UN report on “ up to 40% of Earth land being degraded” and blaming it on …. the climate modellers and their “ artificial fixation and overemphasis [of the role ] of “trace gas” [CO2].
Do you really want to get in bed with these guys?
Well, that is a nice misrepresentation. Strangely there is one specific pattern, and I have no clue what that is, but it repeats over and over. Wherever I sense that the consensus position is wrong, the critical position is simply wronger, erring though in the same direction.
In the case of the WV share in the GHE I would say it is vastly overstated. The reason of course being the surface emissivity = 1 assumption. This overstates surface emissions, the GHE, and the role of WV within the GHE. So while Schmidt et al 2010 names 61/96 W/m2 for WV and KT97 49/75, I would estimate it to be only 25/50W/m2.
Anyway, there are good reasons for that and it is exactly the opposite of blaming some 98% of the GHE on WV.
in Re to E. Schaffer, 10 Jul 2026 at 11:02 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849684
Sir,
Could you provide reference(s) to the article(s) wherein you present your results and compare them with previous studies on the topic like Schmidt et al 2010, so that the interested Real Climate readers can follow your line of evidence?
Thank you in advance and greetings
Tomáš
ES: In the case of the WV share in the GHE I would say it is vastly overstated. The reason of course being the surface emissivity = 1 assumption. This overstates surface emissions, the GHE, and the role of WV within the GHE. So while Schmidt et al 2010 names 61/96 W/m2 for WV and KT97 49/75, I would estimate it to be only 25/50W/m2.
BPL: Show your work.
Tomas to E. Schafer: “ Sir, Could you provide reference(s) to the article(s) wherein you present your results and compare them with previous studies on the topic like Schmidt et al 2010,”
Nice touch, Tomas – asking E.Schafer, point-blank, to show his published proof for his insinuation that Gavin Schmidt and co-authors paper are wrong and/or intellectually dishonest (E Schafer: “overstate surface emissions, the GHE, and the role of WV within the GHE.“).
Unless, of course, you have done it unwittingly – if you don’t challenge the proofs existence but hope it exist – as a fellow “anything but CO2” denier, hoping you might use E. Schafer’s peer-reviewed paper to advocate the diverting of the research and policies AWAY from the mitigation of GHGs, into studying water vapour from which is a scientific dead-end:
– it does not allow any societal/policy recommendations (since we can’t change WV enough to achieve any meaningful difference to AGW – see the fate of your own Sahara irrigation proposal)
– it won’t improve our climate projections – the effect of WV is already implicitely included in the climate models – hence no NEW insight from running the models and exclaiming: “ A ha! On a fictional Earth with WV arbitrarily forced to be at different levels – the GMST would have been X degrees different! Take that, Gavin Schmidt!”
These are a number of considerations adding up, I discuss them on my site. It is was happens if you replace simplified assumptions with more precise and accurate facts. The emissivity of water is perfect example for that, directly affecting the significance of WV as GHG.
The emissivity of water is 0.91, as I have calculated 5 years ago (based on Hale, Query 1973), consistent with Huang et al 2016, or Baehr, Stefan “Stoff- und Wärmeübertragung”. This correct result is in stark contrast to some bad science, like Wilber et al 1999 (NASA) referenced in Trenberth et al 2008, falsely claiming it was 0.9907!?
It follows the surface of Earth will emit ~360W/m2. With OLR = 240W/m2 the GHE will amount to about 120W/m2 only. I know there is this notion that surface emissivity would barely matter in this regard, because the less radiation the surface emits, the more “back radiation” it can reflect, which is true. Inamndar, Ramanathan 1998 is a good example for that. First they wrongly state “sea surface is within 1% of emitting like a blackbody”, then they include surface reflected LW to their GHE definition (in opposition to their previous work), and then they show a chart (Plate 1) where the GHE is minimal over deserts where they assume very low surface emissivities, thereby again contradicting themselves. There is a lot of confusion. I would say it is actually simple. If there is no GH-agent present and the atmosphere is transparent, there is no “back radiation” either to be reflected by the surface. So yes, it is indeed just surface emissions – OLR = GHE.
This necessary downsizing of the GHE will not affect every GH agent equally. There is zero difference regarding clouds, CO2 and the minor GHGs (O3, CH4, N2O..) get some discount, but the bulk of the difference has to be with the WV continuum. The assumption of course is, that all the radiation emitted by the surface and not arriving as OLR, must have been absorbed and substituted by some GH-agent, which is usually WV. A large part of WV-GHE however is just radiation never emitted by the surface in the first place. Adjusting for this issue, WV does some 25/50 W/m2, maybe 30/55 W/m2.
But then we yet have that other side of WV – shrinking the lapse rate and the GHE with it. Then adding up the numbers, I can not help but to conclude WV must be strongly cooling the Earth..
Tomas Kalisz: “Sir, Could you provide reference(s) to the article(s) wherein you present your results”
E, Schafer: “These are a number of considerations adding up, I discuss them on my site. ”
Good one, E. Schafer! See also:
– Sir, Could you provide reference(s) to the article(s) wherein you present your results that the Earth is flat.
FlatEarther: These are a number of considerations adding up, I discuss them on my site.
ES: The emissivity of water is 0.91, as I have calculated 5 years ago (based on Hale, Query 1973), consistent with Huang et al 2016, or Baehr, Stefan “Stoff- und Wärmeübertragung”. This correct result is in stark contrast to some bad science, like Wilber et al 1999 (NASA) referenced in Trenberth et al 2008, falsely claiming it was 0.9907!?
BPL: Different sources give different figures. Most are very high, including the one you mention. I tend to rely on Konda et al. (1994) which found e = 0.984 for seawater.
You’ve been pushing this “the emissivity of water is 0.91, therefore energy budgets for the climate system are all wrong” for years now. Many of us were here the last several times you did it, such as here, four years ago:
https://www.realclimate.org/index.php/archives/2022/07/the-cos2-problem-in-six-easy-steps-2022-update/
This tactic of bringing up a debating point, having it knocked down, then lurking until you feel it’s safe to bring it up again, never acknowledging that it’s been asked and answered, is a typical denier tactic and is obnoxious and exhausting. Perhaps that’s the goal.
In Re to Piotr, 13 JUL 2026 AT 9:57 AM
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849752
Hello Piotr,
Absence of the sought references on the website run by person named “E.Schaffer” strongly suggests that no such publications do exist.
I am afraid that it is very unlikely that they will ever emerge.
Greetings
Tomáš
Well this is not a guessing game, rather it is about the depth of understanding. Neither Wilber et al 1999, nor Konda et al 1994 provide the proper data
W99 failed in many spectacular ways:
– they did not realize water is non-lambertian radiator, so they are unaware of the difference between emissivity to surface normal (which they discuss) and hemispheric emissivity
– they had no data past 15µm and claimed such data would not exist
– actually they just failed to check the literature, as Hale, Query 1973 provided high resolution data up to 200µm
– instead they “interpolated” the far-IR figure from the 15µm result..
– eventually they tried to calculate an average over the whole thing, and failed even with that
K94 only has a very limited scope..
“Its viewing solid angle was 1°, and the sensitive wave length range was between 8 µm and 14 µm.”
So it is just emissivity to surface normal for the segment 8 to 14µm. Emissivity to surface normal over the whole spectrum is 0.965 btw., and hemispheric spectral emissivity 0.91, as before.
Re: E. Schaffer:
Wikipedia: Sealioning – is a type of trolling that consists of pursuing people with relentless requests for evidence, often tangential
here: the water cycle is a passive feedback, so we can’t change it, and its quantification provides no new insight – since it is already included in climate models.
or previously addressed
BPL to ESchaffer: You’ve been pushing this for years now. Many of us were here the last several times you did it, This tactic of bringing up a debating point, having it knocked down, then lurking until you feel it’s safe to bring it up again, never acknowledging that it’s been asked and answered,
It may take the form of “incessant, bad-faith invitations to engage in debate”, and has been likened to a denial-of-service attack targeted at human beings.”
Ladies and Gentlemen – E Schaffer.
E. Schaffer,
Would the review article Colman & Soden (2021) ‘Water vapour and lapse rate feedbacks in the climate system ‘ be the sort of thing you’re after?
@Roger & Barry
Thx, this is a great summary of known considerations. However, it absolutely does not address the simple question I have put up. Although it states..
“At the same time, in latitudes spanning the tropics through to mid-latitudes, the upper troposphere warms faster than the surface – a change in the vertical “lapse rate” with temperature, enabling the Earth to radiate to space more effectively.”
.. there is not the slightest reflection on the netting of those opposing effects within the given climate system. It is like that CL game PSG vs. Bayern, where you might ask who won and get the answer Bayern scored 4 goals. Now that might suggest Bayern won, but actually does not answer the question. There is still a chance they might have received 5 goals and lost the game, which is exactly what happened btw. Equally so just saying WV was a strong GHG having the largest share in the GHE, avoids the question if it is warming at all. Especially given we know it also has a huge cooling side.
Equally jumping right over to WV feedback avoids the question. At the start one should clarify if WV is warming or cooling, and by how much. I feel like the question is avoided in order to maintain the narrative of a pos. WV feedback. If WV is cooling, and that is what the numbers suggest, it makes little sense more WV would provide (strong) warming..
H2O gas isn’t radiating (manufacturing photons by its collisions) from the top of the troposphere and therefore an increase must cause additional imbalance between its radiation leakage (a very small fraction of its photon manufacture) out of the top versus out of the bottom of the troposphere, the so-called “greenhouse effect (GHE)” in Earth’s troposphere.
E. Schaffer,
I find it strange that you would consider that reference to Colman & Soden (2021) “absolutely does not address the simple question [you] have put up.”
You quote from the ‘I. Introduction’ of Colman & Soden (2021) and complain that your question is not answered, it even failing to provide an understanding of the sign of the combined WV/Lapse-Rate feedback. Yet the sign is well known – the combined feedback is positive. Let me quote from the ‘Abstract’ of Colman & Soden (2021).
Colman & Soden (2021) provide a value for the size of this positive feedback in ‘VIII. Conclusions A. On the strength and consistency of evidence for water vapour and lapse rate feedbacks.’ saying:-
Their Tables 1,2 & A2-1 list various studies with values for the various components of the combined WV/LR feedback. And note that your actual question posed upthread was “Is there actually any literature out there trying to quantify the given climate effect of water vapor?” for which I would consider a review paper such as Colman & Soden (2021) would demonstrate properly that the answer to your question was an emphatic “Yes!!!”
(I should add that the latent heat energy flux of “some 86.4W/m2” you mention upthread is not a direct measure of any feedback. Any change in that value would have a bearing on the strength of the WV feedback but again is not a direct measure of the feedback itself.)
I don’t know why that is, but you are not reading what I am writing. Probably you think it is irrelevant, just glance over it and answer to your own assumption. I am STILL NOT asking about WV feedback, but about the given effect of WV.
We have the WV effect and the enhanced WV effect, which we call WV feedback. That is a base magnitude and the enhancement of said base magnitude. As with AR6 that enhancement is assumed to be +1.8 and -0.5W/m2 and a total of 1.3W/m2. My question is on the base magnitude. How much warming, how much cooling, and the net effect of WV. Again, not the feedback, just the base magnitude.
Clearly this is an important perspective and it seems it has been completely ignored, which makes it only more pressing. Of course the notion is, if it ever had been considered, bad things would have been avoided, but anyway..
And of course the 86.4W/m2 in latent heat are also not a feedback, but a given cooling effect by WV. For instance, as above, we could say WV warms with 49W/m2 (KT97 net figure) and cools with 86.4W/m2, and overall is strongly net cooling. I repeat: WV then is a strong cooling agent, which is a ground breaking perspective. And in science we must not ignore perspectives just because we don’t like them. From that the question would arise: given WV cools and is positively correlated to temperature, how could it be a pos. feedback?
But that is downstream, first really there is the urgent question on whether WV is warming or cooling, and I can not fathom it has never been considered.
E. Schaffer “ whether WV is warming or cooling, and I can not fathom it has never been considered.”
Because, as it has been explained already dozens of times to your fellow “anything but GHGs” deniers (Tomas Kalisz and JCM) – WV and other aspects of the water cycle are merely a passive feedback, not one of the drivers of current AGW. Thus:
1. There is nothing we can do about WV – given the magnitude of the natural water and ridiculously short residence time of WV in atm – we simply can’t change water fluxes enough to reduce AGW in any noticeable amount.
2. Since WV is a strong net positive feedback – the AGW is more sensitive to the GHGs, because WV amplifies the consequences of our action, or inaction, on GHGs:
if we reduce GHGs to cool the Earth – WV will make the cooling much larger, if we do not cut our GHG emissions – WV will make the warming much larger,
Which directly counters the deniers narrative that because there is so much more molecules of WV than GHGs – what we do to GHGs doesn’t really matter, so we can burn as much fossil fuels as the fossil-fuel lobby wants us to.
3. For the main societal responsibility of climate science – informing the mitigation of AGW – there is nothing to be gained from “your direct measure of the WV feedback itself” – since again – we simply can’t change water fluxes enough to make any noticeable dent in AGW.
4. Your WV is already implicitely included in climate models – ergo your “direct measurements of the feedback itself” would provide no new insight beyond what we already know from Schmidt et l. 2010 and other papers.
In other words, since your WV feedback is already included in climate models – what difference would it make if out of, say, the projected 2C of warming, 1.3 C was the result of the WV passive feedback instead of 0.9 C? It does not affect our models (since these 1.3C or 0.9C re the products of these models ). Nor can we do anything about it.
Do you fathom it now?
E. Schaffer,
My apologies for repeatedly misinterpreting you questioning.
To be clear, and so I don’t make a fool of myself again, can you set out precisely what you are hoping to quantify.
You say you want a full account of the impact of atmospheric WV on Earth’s GHE: that is the reduced climate ‘forcing’ if atmospheric WV was reduced to zero and in this regard you specifically mention the inclusion of the lapse-rate feedback in this assessment.
However if the WV is considered as being removed from the atmosphere, the presence of cloud seems a little difficult to imagine without any atmospheric WV. Also the planet will be considerably colder without WV and the effects of surface albedo will presumably increase due to expanded sea ice. In a normal situation a colder planet would also result in ice caps and snow-cover, further increasing surface albedo but these would have been deposited by precipitation so may not be present in your hoped-for assessment. The same could be said for land albedo due to plant-life which does rather rely on water from above, directly or via rivers.
Could you rattle through these other factors to indicate which, like lapse rate feedback, you wish to see incorporated into this assessment you seek.
@Rodger
My original question was on whether this issue has ever been considered. I do know a lot of the literature but certainly not all, and according to my knowledge it was not. Then I asked AI and it too could not name me a single instance either. So that is why I brought up the question here and it seems like everyone is confused just because I am asking.
As to what you say let me clarify: we have attribution and science fiction. In science fiction we say “what if” and then play through different scenarios, consider dynamics and so on. In attribution we assume all other things stay the same and consider the “what if” only regarding the immediate effects. In science, to keep things clean, we only do the latter.
The immediate effect of WV is raising the emission altitude and reducing the lapse rate. That is why WV feedback consists of two components, which are raising the emission altitude (assumed to be 1.8W/m2) and reducing the lapse rate (assumed to be -0.5W/m2). Considering these two effects alone is more than enough, there is no need to complicate things any further.
The very same question applies to the given amount of WV. It raises the emission altitude and it has reduced the lapse rate. We have quantifications for both effects, which are totally negotiable and discussable. But the figures we have indicate WV is a cooling agent, and the only reason this has never been understood seems to be that the very question was never asked. And that is simply embarrassing..
E. Schaffer,
You are saying that you want to quantify the net contribution of WV within the pre-industrial atmosphere’s GHE. This contribution is therefore a component of the GHE which comprises the forcings/feedbacks – GHG+aerosols, WV, cloud & albedo that raised surface temperatures some 33K.
And your question was specifically whether there is literature which provides analysis quantifying this contribution.
Yet buried in your enquiry is a very odd separation of the mechanisms of the pre-industrial GHE and the mechanisms of AGW in the industrial age.
You are seemingly happy that the literature shows the impact of WV under AGW comprises the effects of (1) Increased emissions height (+1.8Wm^-2/K) and (2) Increased lapse rate (-0.5Wm^-2/K). This is a net positive feedback of +1.3Wm^-2/K and would boost any warming from other net forcings+feedbacks by about 50%. So far, smiles all round.
We could then take a back-of-envelope approach here and say that the pre-industrial GHE resulted in something like +33K (this requiring surface albedo assumptions) so the contribution of WV would be a net positive feedback of [+1.3 x 33 = ] +43Wm^-2.
Of course it is a big assumption using an unaltered Wm^-2/K of feedback all they way down to zero WV but it does give a ballpark idea of the size of the feedback. The two references you made upthread also give a value for the WV feedback contribution to the pre-industrial climate..Kiehl & Trenberth (1997) ‘Earth’s Annual Global Mean Energy Budget’ give +59Wm^-2 (75Wm^-2 clear sky, 51Wm^-2 cloudy sky, 67.5% cloud fraction from CERES) while Schmidt et al (2010) ‘Attribution of the present-day total greenhouse effect’ give +49 Wm^-2 (All Sky 50% of ‘total’ using K&T(1997)’s 98Wm^-2 ‘total’).
So this is all pretty-much in the same ballpark.
But instead you seem to be of the opinion that the WV value of these two papers doesn’t include the Lapse Rate feedback. And if the +1.8Wm^-2/K from above was applied to the full 33K pre-industrial GHE it would amount to some +60Wm^-2, still in the same ballpark.
But you also consider the -86Wm^-2 surface latent heat flux to be a measure of the full pre-industrial Lapse Rate feedback. This is certainly not the case. The -86Wm^-2 surface flux is converted to sensible heat when the WV condenses up in the atmosphere. It is thus decreasing the Lapse Rate but it is not a measure of the Lapse Rate feedback.
I think I will need to examine your website to find out why you have made such an egregious error. I note from a passing visit that you weren’t not shy about this matter posting about it at the back-end of last year with OPs titled ‘How “Climate Science” has its eyes wide shut on the statistical blunder that generates false positive feedbacks’ and ‘How to Flip the Sign on Feedbacks’
@Rodger
First of all it is great to finally get to my actual question, after 40+ posts here.. ;)
“But you also consider the -86Wm^-2 surface latent heat flux to be a measure of the full pre-industrial Lapse Rate feedback. This is certainly not the case. The -86Wm^-2 surface flux is converted to sensible heat when the WV condenses up in the atmosphere. It is thus decreasing the Lapse Rate but it is not a measure of the Lapse Rate feedback.”
Let me ask: there are a number our sources pointing out the GHE is dependent on the lapse rate, and furthermore proportionate to the lapse rate. If the lapse rate was 0, the GHE would be 0, as surface- and emission temperature would be equal. Equally if the lapse rate was 50% larger, the GHE would be 50% larger. Let us say the average emission altitude was 5km up, then with 6.5K/km the GHE = 6.5 x 5 = 32.5K. With a lapse rate of 9.8K/km it would be 49K. Do we agree so far?
Due to the SB-law we can not directly relate delta temperature into delta radiation, but that is a side show. Yet, by and large, 50% of some 155W/m2 gets us to around 80W/m2. Either way, if it was not for this reduction of the lapse rate, Earth would be a lot hotter. And of course it is physically exactly the same as the lapse rate feedback featured all over the models.
So how exactly do you reject this connection?
E. Schaffer,
Having “g(o)t to {your) actual question, after 40+ posts here” (and I’m not sure having apparently “got to” it, whether you have ‘got over’ it), you then toss in another question!!!!
And what a question! It seems you wish to upgrade the atmospheric Lapse Rate into the primary feature of a GHE, rather than it being the GHGs in such an atmosphere.
And why not? Hey, does not a telescope have two ends?
A Lapse Rate forms in an atmosphere because atmospheric pressure decreases with altitude and gases will generally rise/fall adiabatically within such an atmosphere with meridianal temperature gradients.
In Earth’s atmosphere this zero-GHG base-line Lapse Rate is usually given as 9.8°C/km with the mention of it being “dry air”. The difference for ‘wet air’ that isn’t condensing out precipitation is pretty similar as the decrease in density is balanced by the increase in specific heat capacity.
But adding in a bit of ‘thermodynamicity’ will alter the Lapse Rate due to GHG energy fluxes which warm/cool said atmosphere.
Thus with altitude, GHGs will be emitting/absorbing less IR with the lower temperatures at altitude, this adding warmth with altitude and reducing the Lapse Rate, or shooting IR out into space and cooling the atmosphere, thus increasing the Lapse Rate. Or absorbing incoming UV which in the thin upper atmosphere can warm enough to reverse the Lapse Rate. And precipitation will transport insensible heat up to the point of condensation where it will again provide warming, reducing the Lapse Rate.
The GHE is not just a GHG-thing. It requires a Lapse Rate and that requires atmospheric pressure to decrease with altitude. Mars is a good example of what you get without atmospheric pressure. It has the same burden of CO2 as Earth but this is 95% of its atmosphere rather than Earth’s 0.05%. The result is zero GHE on Mars.
So physics tells us the Earth’s tropospheric adiabatic Lapse Rate sits at about 9.8°C/km before the GHGs start their influence. And what if the total Lapse Rate feedback (‘total’ = today relative to zero-GHG) happened to be 80Wm^-2 (thus equal to the “evapero-transpiration” flux in Fig 7 of Kiehl and Trenberth (1997))? If it were 80Wm^-2, that would be half of the 155Wm^-2 total GHE “Net LW Absorbed” set out in Schmidt et al (2010). And would that then imply the modern Lapse Rate is half the 9.8°C/km? After all, Wiki-thing tells us
the actual average value is considered to be 6.5°C/km up to the tropopause, so not massively different to 4.9°C/km.
(One point to clarify here is that Schmidt et al (2010) do not adjust the atmospheric temperature profile, saying their modelling is “holding the climate (spatial and temporal distributions of temperature, surface properties, etc.) fixed.” and is thus using Lapse Rate data that would presumably reflect that 6.5°C/km average. The WV feedback with no LR feedback would this be some (60Wm^-2 / 33K =) +1.8Wm^-2/K which is the value for the modern climate we see in the literature.)
I think it would be wise to consider all the energy fluxes into the atmosphere. They surely would all be messing with the lapse rate. ☻ There’s the “evapero-transpiration” flux of 78Wm^-2 which would be operating at the points of condensation and which may be mainly impacting the Lapse Rate up to the altitude of WV IR emissions to space. Would the Lapse Rate be reduced above that point? ☻ There’s “thermals”, a flux of 24WM^-2 (with big ‘error bars’) which presumably would be reduced if the Lapse Rate declined. ☻ There’s a net flux of 40Wm^-2 in “surface/back radiation”. ☻ And into the atmosphere from above is an “absorbed by the atmosphere” flux of 67Wm^-2. ☻ Finally there is a cooling flux of 195Wm^-2 “emitted by the atmosphere” which will presumably be increasing the Lapse Rate.
So it looks like the “evapero-transpiration” flux is not the only flux messing with the Lapse Rate. All these energy fluxes look like being of significance to the Lapse Rate.
And together they will have modified the Lapse Rate by altitude and as well as zonally. So I’d suggest that assuming the Lapse Rate is a constant average 6.5°C/km throughout the troposphere doesn’t look like providing a realistic or useful model.
I also don’t see any support for your contention that climatology is ignoring the Lapse Rate feedback or that the combined WV/LR feedback is negative.
We could look at evidence of an increase in global precipitation as a measure of the change in that 78Wm^-2 “evapero-transpiration” flux under AGW. Use satellite data 1979-2020 from GPCP, Gu & Alder (2022) find precipitation rising at 1.4%/deg C. That would be about half the rate of the 78Wm^-2 over 33K.
And perhaps it would not be torturing the available data too much to use the -0.5Wm^-2/K calculated for today’s LR feedback and then double it over the 33K of pre-industrial GHE yielding 33Wm^-2 for the ‘total’ LR feedback and thus (+60 – 33 =) +27Wm^-2 for the combined WV/LR feedback. Note it is positive. Of course, this the opposite of the contention set out in a handful of OPs on you website which look like they could do with some serious debunking.
Oh that gave me a lot of headache. With every phrase I struggle to figure out what you mean.
What for instance is “GHG energy”?
Why do you think there was no GHE on Mars? CO2 does leave a dent in its OLR profile, so there has to be one. Given the planet is not properly covered by weather stations, given it has large temperature differences both by day and season, it is not so easy to tell what average temperature it has, or should have respectively. But that is just a practical issue.
“If it were 80Wm^-2, that would be half of the 155Wm^-2 total GHE “Net LW Absorbed” set out in Schmidt et al (2010)”
Yeah, kind of half. But that would come on top of the 155. So with a larger lapse rate, the GHE would be larger, like 155 * 1.5 = 232.5!
“the actual average value is considered to be 6.5°C/km up to the tropopause, so not massively different to 4.9°C/km”
Why 4.9K/km? The 80W/m2 of latent heat would not halve the GHE, but enhance it by 50% in their absence. The cooling is already baked in. But just because it is will not mean it never happened. This cooling side of WV is there, it is just not reflected on.
Logically it also follows that there is no question if S10 (or KT97) does consider it, or not. They are stating their figures based on the status quo, with the cooling effect by WV already included – but not stated!!! And yes, if it was not for that, the lapse rate was larger, the GHE would be larger AND the contribution WV to it would equally be larger. So in the instance of S10 it would not just be like 61/96 W/m2, but maybe 92/144 W/m2. And with KT97 not just 49/75 but 74/112 W/m2. But even then this would barely be warming vs. the massive 80W/m2 plus latent heat cooling.. On top of that, as I have pointed out, in both instances WV will be badly overstated due to the surface emissivity issue.
As with my previous post above one can easily figure this out considering emission altitude vs. lapse rate. 6.5 x 5 = 32.5K, 9.8 x 5 = 49K..
ES: Why do you think there was no GHE on Mars? CO2 does leave a dent in its OLR profile, so there has to be one. Given the planet is not properly covered by weather stations, given it has large temperature differences both by day and season, it is not so easy to tell what average temperature it has, or should have respectively.
BPL: The Mars Standard Atmosphere (Barth 1985) gives the emission temperature of Mars as 214 K, and Catling and Kasting (2017) give 215 K. So since Mars’s radiative equilibrium temperature is approximately 210 K, there is 4-5 K worth of greenhouse effect on Mars.
Mars Orbiter and the landers gave some good climate data. There are also GCMs for Mars.
Ref:
Barth, C.A. 1985. The photochemistry of the atmosphere of Mars. The Photochemistry of Atmospheres, Levine, J.S., Ed. Harcourt Brace Jovanovich, Publishers, pp. 337-392.
Catling, D.C., Kasting, J.F. 2017. Atmospheric Evolution on Inhabited and Lifeless Worlds. Cambridge, UK: Cambridge Univ. Press.
E. Schaffer,
“What for instance is “GHG energy”?”
Well, if you read what was written (“GHG energy fluxes which warm/cool said atmosphere.”) you’d find that the ‘energy’ in question concerns fluxes-various that thus flow in/out GHG within the atmosphere (because the are IR-active which is the property that GHG have) and which then warm/cool the atmosphere they sit in, warming (or cooling) just like the WV/precipitation cycle does.
“Why do you think there was no GHE on Mars?”
Do I think that? I raise the example of the Martian atmosphere because you are the one trying to set out lapse rate as the major consideration for a GHE. On Mars the GHE operates with no significant lapse rate so there would be “zero GHE on Mars” resulting from added GHGs.
The 80Wm^-2 conjecture and “why 4.9K/km?” The 80Wm^-2 strength of the Lapse Rate feedback is your conjecture. Whether you want it applied to arrive at the 155Wm^-2 “Net LW Absorbed” GHE effect of Schmidt et al (2010) or applied to modify that 155Wm^-2 is probably your decision.
However, if this trivia is your takeaway from my comment, I suggest an aspirin for your headache and read it again.
Or maybe I should be helpful and suggest that your attempts to infer something useful from the strength of the existing GHE relative to some sci-fi zero-GHG atmosphere won’t be relevant to AGW which results from adding additional GHGs into the existing GHE. AGW modelling of the Lapse Rate feedback is usually reported combined with the WV feedback and invariably shows that the combined WV/LR feedback is strongly positive.
IPCC AR6 Section 7.4.2.2 concludes thus:-
If somebody wanted to overturn such science, IPCC AR6 7.4.2.2 is where to start.
You may disagree and may wish to somehow demonstrate your position with a half-baked comparison of the GHE with some sci-fi zero-GHG situation. You evidently feel strongly about this position of yours having gone so far as to assert (on your website) that “the established “consensus science”… can not win the argument”and that “WV can not be a positive feedback, let a lone a massive one, a detail the global warming narrative totally depends on.”
However, given you hold this view so strongly it’s very probable that any words of helpful advice here will be wasted.
in Re to MA Rodger, 19 Jul 2026 at 6:03 PM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849864
Dear MA,
I checked the difference in greenhouse effects on Earth and Mars with Perplexity Pro and in this respect, I have a few questions regarding the following sentences in your comment:
“The GHE is not just a GHG-thing. It requires a Lapse Rate and that requires atmospheric pressure to decrease with altitude. Mars is a good example of what you get without atmospheric pressure. It has the same burden of CO2 as Earth but this is 95% of its atmosphere rather than Earth’s 0.05%. The result is zero GHE on Mars.
So physics tells us the Earth’s tropospheric adiabatic Lapse Rate sits at about 9.8°C/km before the GHGs start their influence.”
1) It indeed appears that although CO2 pressure on Mars is higher than on Earth (ca 570 Pa vs ca 40 Pa), the difference between the global mean surface temperature (GMST) and emission temperature calculated from albedo and insolation (which is for Earth 255 K, for Mars 200-223 K) is for Earth about 33 K, while for Mars between -9 and +5 K.
The engine, however, suggests that part of this difference should be accounted to the difference in insolation and albedo. It asserts that if Mars had the same emission temperature as Earth, its CO2 atmosphere would be warmer and “higher” or “thicker”, what would allegedly result in a change in the lapse rate and in a stronger greenhouse effect. Could you check?
2) It further asserts that in this case, and if the surface of Mars were somehow “irrigated”, water vapour would further enhance the greenhouse effect on Mars and made its magnitude even closer to that on Earth.
3) Another part of the difference between Mars and Earth should be, according to Perplexity, ascribed to collisional broadening of spectral bands in GHG molecules due to higher atmospheric pressure on Earth.
4) The remaining part of the difference between Mars and Earth should be, according to Perplexity, ascribed to water availability for evaporation from the surface on Earth and temperature and pressure range enabling latent heat flux in the atmosphere, what further changes convection and the lapse rate.
Is Perplexity correct?
Greetings
Tomáš
Tomáš Kalisz,
Regarding the Martian-&-Earth GHEs, I don’t know of any actual apples-to-apples comparison of atmospheres from Mars and Earth’s. Indeed, how would you account for the freezing/evaporating of the polar Martian CO2 ice? The nearest to an Earth/Mars comparison that springs to mind is Figs 1 & 3 in Read et al (2015) ‘Global energy budgets and ‘Trenberth diagrams’ for the climates of terrestrial and gas giant planets’ although Fig 5 shows what a dust storm does to the situation on Mars.
I’d assume the IR emission height would be very roughly similar but it is the pressure-drop that gives you a Lapse Rate and there’s naff-all pressure in the Martian atmosphere so naff-all drop is possible.
The pressure broadening, which fuzzes out the absorption/emission lines may well be a feature in the low-pressure Martian atmosphere and that would increase emission heights significantly but still there can be meaty pressure drop.
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849896
And, (even without hysteresis) the partition between forcing and feedback can be shifted between a change and the reverse change. Consider the complete removal of Earth’s GHE (but somehow maintaining solar heating exactly as it; this is a model experiment, after all). The TOA forcing would be a large increase in OLR, tending to cool the climate system overall. Now let’s add back the GHE. Well, to start, let’s add a little back. In the equilibration to 0 GHE, the troposphere would largely disappear – It’s hard to see how it could be maintained in any significant way with significant depth without an ability to emit LW radiation to balance an upward convective heat flux…
— —
So given that there is solar heating within the air, in a 1-dimensional globally representative model (I’m going to stick with that here to keep things simple), no GHE means all OLR must come from the surface, and so the surface equilibrium T must be at the effective radiating T (assuming perfect blackbody – yes, that is an approximation and it does have some error), but heat must now flow downward to the surface, so the whole atmosphere would be a stratosphere/thermosphere, and possibly very very hot (maybe it would even start to contribute to OLR via SW emissions – but let’s set that scenario aside).
Introducing a GHE in this scenario, there would initially be an increase in OLR. The TOA IRF forcing would be cooling! But that doesn’t necessarily mean the sfc wouldn’t warm. There would be a downward LW flux to the sfc now; the net upward flux within some lower layer would decrease. Less heat has to flow downward to balance the solar heating of the air, so it’s not clear exactly how things work out, but the solar heating at the surface would start to get trapped a bit.
Add enough GHE, and pure radiative equilibrium would become unstable to convection, and so a troposphere would develop and grow. The solar heating within the troposphere would then become part of the convectively-coupled layer…
…“Less heat has to flow downward to balance the solar heating of the air, so it’s not clear exactly how things work out,”
? – um, consider the SARF – but I’m done for today…
@Rodger
“The 80Wm^-2 strength of the Lapse Rate feedback is your conjecture” – Well, about as much as my conjecture that 1+1=2.
Let us go through this step by step, then you can tell me which logical step you object
1. The GHE is the difference between surface- (Ts) and emission temperature (Tz), like 288 vs. 255K
2. The GHE is proportionate to the lapse rate. A lase rate of 0 means Tz = Ts and GHE = 0.
3. WV does reduce the lapse rate from a dry unstable adiabat (>9.8K/km) towards 6.5K/km.
4. Holding other things constant, this smaller lapse rate reduces the GHE and provides a cooler surface
5. This reduction of the lapse rate is also known as “latent heat”
6. This “lapse rate effect” is physically the same as the well known lapse rate feedback, which is simply an increase of said lapse rate effect, or latent heat respectively
E. Schaffer, I have trouble following most of what you’ve been saying. However, laying it out in concise, logical steps allows me to see some of the incorrect assumptions you’re relying on:
2. The GHE is proportionate to the lapse rate. A lase rate of 0 means Tz = Ts and GHE = 0.
JP The lapse rate from the surface to where? 2 km? 200 mb? The tropopause? Outer space?
3. WV does reduce the lapse rate from a dry unstable adiabat (>9.8K/km) towards 6.5K/km.
JP The figure for the dry adiabatic lapse rate is correct, but most of the planetary atmosphere is not at that lapse rate. In the stratosphere, the air is very dry, but the temperature increases with height for the most part. The 6.5K/km is an average for conditions near the surface. The moist adiabatic lapse rate will indeed decrease as water vapor is added, but most of the decrease is near the surface where temperatures are warmer and the air holds more wv. However, a lower lapse rate there will deepen the warm layer and allow the air somewhat higher up to hold more moisture, too. This thickens the radiative barrier between the lower
troposphere and outer space.
4. 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. You also warm the lower atmosphere and raise the tropopause. It’s highest over the tropics, where there is lots of water vapor in the lower atmosphere.
Your argument only works if the tropopause height and temperature are held constant, and then you take your assumed lapse rate from there to the surface. That’s not what happens.
E. Schaffer,
Let me attempt to break down your myopia here by presenting again just one factor (the simplest – it applies to your Step 3) which makes a nonsense of your conjecture that the LR feedback = surface WV latent heat flux = ~80Wm^-2.
Your proof appears to rest on, firstly, the reduction of the Earth’s average Environmental LR from the Dry Air LR which is usually quoted as due to this latent heat flux and ELR = 6.5K/km with DALR= 9.8K/km – a reduction of a third. And secondly, the “Net LW Absorbed” GHE effect of Schmidt et al (2010) which is calculated as 155Wm^-2 using modern LR (and thus presumably using average ELR). So you see this 155Wm^-2 as also being reduced by a third, reduced from a GHE value using DALR. And that reduction, GHE(DALR) minus GHE(ELR), would be roughly equal to the latent heat flux of 80Wm^-2.
The issue is not that the LR feedback is negative. The mechanism is evident. An increased energy flux warming the atmosphere will reduce the LR and this will warm emissions altitudes which in turn will increase IR emissions from GHGs. The literature gives its value as 0.5Wm^-2/K. Pro rata for a 33K GHE, that would suggest a much smaller value for the full LR feedback [0.5 x 33 =] 16.5Wm^/2.
So the one factor I present to you is this – If such an increased energy flux 80Wm^-2 due to WV acts to reduce the LR and thus acts as a negative feedback for the GHE, why wouldn’t other enrgy fluxes warming the atmosphere do exactly the same. The energy balance in Fig 7 Kiehl & Trenberth (1997) which you have cited in you argument shows a total of 195Wm^-2 of such warming.
So in your conjecture “LR feedback = surface WV latent heat flux = ~80Wm^-2”, why is it only the WV component of this 195Wm^-2 total that impacts the GHE?
Your “conjecture that 1+1=2” may be well-based (mathematical geeks object to my poetic 1+1=11, while pointing out that both i + i = ii or 1+1=10 would be as arithmetically sound as 1+1=2), but your conjecture that the LR feedback = surface WV latent heat flux = ~80Wm^-2 is certainly not sound.
E Schafer: “The 80Wm^-2 strength of the Lapse Rate feedback is your conjecture” – Well, about as much as my conjecture that 1+1=2.”
Not really – see replies by MAR and John P. Mine is a more fundamental question – even it was 80Wm^-2 – 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.
And no, clicking your ruby slippers and repeating: “ Well, it is exactly about improving our knowledge, and decisevely so! It is exactly about improving our knowledge, and decisevely so!” – won’t magically change it.
@John Pollack
2. and 3. – We are talking about the tropospheric lapse rate of course
4. “So what? Other things aren’t constant” – sure, but nihilism is not going to provide insights, rather it is meant to avoid them. We mentally hold other things constant, so to identify what a certain factor does.
@Rodger
“So in your conjecture “LR feedback = surface WV latent heat flux = ~80Wm^-2”, why is it only the WV component of this 195Wm^-2 total that impacts the GHE?”
Let me first try to understand how you get to the 195 figure. I can see that the atmosphere there would emit 195W/m2 into space. It follows logically, that if it loses that amount of energy, it would also need to update the same amount to stay in balance. I would assume your calculation is like 350-324 + 78 + 24 + 67 = 195?
It is quite simple actually. That energy for the larger part is put into atmsphere at the bottom and it all gets lost at the “top” (actually different altitudes within the troposphere). While moving up or down, due to the consequentual convection, the air has to follow a dry adiabatic lapse rate, in the absence of WV. With WV it is more like a moist adiabat, though not a perfectly moist adiabat cause the air is not totally saturated with WV. Please note: the latent heat is not moved into the troposphere at ground level, but at altitude.
Btw.. while that is just my take on it, it is established pyhsics, hard to argue with.
“Pro rata for a 33K GHE, that would suggest a much smaller value for the full LR feedback [0.5 x 33 =] 16.5Wm^/2”
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.
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.
E. Schaffer,
You continue with defence of your Lapse Rate Feedback = Surface Latent Heat Flux theorising. Above you set out two arguments in its defence.
Your “quite simple” first argument is that the energy fluxes into and out of the troposphere are not altering the Lapse Rate with the sole exception of the condensation of WV. And here your grand conclusion rests on the idea that all those other energy fluxes “put into atmosphere at the bottom and it all gets lost at the “top” (actually different altitudes within the troposphere). … With WV [the Lapse Rate] is more like a moist adiabat, though not a perfectly moist adiabat cause the air is not totally saturated with WV. … the latent heat is not moved into the troposphere at ground level, but at altitude.” You end by describing this as “established physics.”
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?
Your second argument is that there is a “contradiction.” Your own grand theory that the total Lapse Rate Feedback = Surface Latent Heat Flux = c. -80Wm^-2 doesn’t fit with the established rate of change in the Lapse Rate Feedback (= -0.5Wm^-2/K). This is a “contradiction” which somehow allows you a legitimacy in asserting what you fancy while, in your view, I illegitimately ignore this “contradiction” and apparently deny it exists.
Surely, this “contradiction” was raised up-thread by me (so not ‘ignored’ by me) while you have here only just managed to admit its existence and have also just admitted previously ignoring my raising of it.
More to the point, your arguments here are really trifling side-issues.
Your grand theory sits in “contradiction” to the entirety of climatology which you insist has failed entirely in accounting for this “contradiction”. (On your website you go so far as to say it is being deliberately ignored by the science.) Such contentious theorising obviously requires a well-presented proof. You have failed to provide such a proof. And the more I examine the situation, the more you appear as an ill-informed fool who has very little understanding of the climatology you refute and have failed to grasp that “established physics” on which you base your grand theory.
“86.4W/m2 of “latent heat” cooling” would be local surface and whatever (hundred or so?) metres of air cooling of the tropical ocean only and not at all the globally-averaged tropospheric lapse rate effect. There’s adiabatic heating of descending air by the pressure increasing. I pondered this briefly a few years back about the Hadley Cell, Desert Zone and the incorrect simplified phrasing for requirement of entropy increase (2nd Law if the system was closed) “hotter things heat colder things”, which simplified phrasing isn’t correct of course because latent heat. Specifically I wondered a few minutes. air rises over tropical Atlantic Ocean (tall troposphere, surface Low Pressure Zone), slides north near troposphere top, turns east, H2O gas condenses-freezes high up, rains back onto the Atlantic Ocean having converted latent heat to “heat”, descends over the Sahara Desert nice and dry from having rained out, warms at 9.8 degrees / km altitude by pressure increase and heats a surface of say 40 degrees from Atlantic Ocean surface of say 25 degrees. Hotter warming colder due to latent heat.
Oh no, that is indeed the global average estimate.
https://mynasadata.larc.nasa.gov/basic-page/earths-energy-budget
You need to bear in mind what it means. It is not just latent heat as an arrow in the diagram above, but rather it means the reduction of the lapse rate and the whole GHE with it. Without WV we would have a dry unstable adiabatic lapse rate of >9.8K/km. The GHE would be at least 50% larger. If you assume it to be 155W/m2 that would mean over 77.5W/m2.
In my calculation the GHE is only 120W/m2, but the lapse rate should actually be like 65% larger, so that the latent heat would still amount to ~80W/m2.
In 2013 I calculated 79 W/m**2 from annual global precipitation and evaporation latent heat per kg. I just did it again in 1 minute and got only 76 W/m**2. In 2013 Kevin Trenberth showed 80 W/m**2 in his EEB at 26:26 at https://www.youtube.com/watch?v=SQOIHdlZngk so I assumed my 79 W/m**2 was simply latent heat off the oceans. Are you referring to 86.4 – 80 = 6.4 w/m**2 as meaning that some adjustment is needed to the total effect of increased H2O gas due to increasing surface-troposphere temperature ?
No, I am not into claiming some super precise figure here. I just point out to the fact that there is some 80W/m2 (or so) of cooling due to WV, which depending on what figures you choose, is at least on par, if not exceeding, its warming side.
Nuts. “Hotter warming colder” S.B. “Colder warming hotter”
In Re to Barry E. Finch, 12 JUL 2026 AT8:22 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849729
Hello Barry,
I am not sure if I understood you correctly but, if you assume that latent heat stored in water vapour released from tropical oceans and rainforests returns to Earth’s surface in deserts of higher latitudes, I would like to ask how is it possible that the warm air heated by water vapour cindensation descends?
Do you not think that already the said condensation (and following precipitation) of Ester vapour could hardly occur if at least part of the released condensation heat were not emitted as infrared radiation to the space?
I suppose that the same may apply for further cooling of the resulting dry air that enables its final descent to the ground.
Water, not Ester. Condensation, of course.
Tomas, “how is it possible”. Hadley Cell, large scale overturning. “at least part of the released …. emitted as infrared radiation to the space?” Yes, of course, and I doubt I’ll ever try to figure that out with any worthwhile accuracy (few years back I tried to estimate global wind turbines slowing global wind in my head in my kitchen making dinner, and I gave up, couldn’t figure the constant push of uneven Sun heating of Earth fighting an attempt to slow wind). So that’s why I simply gave in my comment 6 Jul 2026 at 8:13 PM what I saw in some scientists’ Webinars (2 I think I recall) on UTube where they showed the classic 3 feedbacks and Planck response (probably the Webinars were about how cloud feedback is known poorly and the surface better and +H2O gas feedback with high confidence) from some average of some CMIP Climate Models. I figure the odds the Being-Paid scientist Army and computer programmers did that with their time-sliced simulations better than I could with some calculations to be like 1,234:1 and the odds for better than “E. Schaffer” like 9,876,543:1 (even though some of those bods think photons is re-emitted and they ain’t and radiation some place goes back some place and it don’t) because I’m not finding “E. Schaffer” to be making any sense at all.
Tomas, I just realized that “E. Schaffer” 13 Jul 2026 at 6:30 PM “the fact that there is some 80W/m2 (or so) of cooling due to WV” is mostly (the other little bit is my estimated adiabatic heat to surface) yet another of the Denier memes enabled by the physicist nonsense “re-emits” and “back radiation” instead of “emits (manufactures)” and “downwelling radiation at the surface”. Ironic that I failed to notice that considering I’ve been going on about it and getting “Flamed” since June 23rd when I correctly pointed out that some bods typed rubbish. I forgot that downwelling radiation at the surface includes I just calculated 16% of the H2O evaporation-condensation-freezing latent heat. As you also understood in a comment the LWIR in troposphere is from all Power sources.
The “80W/m2 (or so) of cooling due to WV” hasn’t itself, with that hugely-overstated “80”, gone through the “greenhouse effect (GHE)” in Earth’s troposphere, with GHGs converting Solar SWR. H2O gas latent heat, and NET sensible heat convection & conduction into LWI radiation and returning 63% to the surface. Using the “Hartmann 2nd Edition 2016” text book page 34 EEB diagram with my assumed 13 w/m**2 Solar SWR each into thermosphere & stratosphere, in Earth’s troposphere is 54 of Solar SWR + 27 of rising warm air + 88 of surface water evaporative latent heat + 376 of surface LWIR = 545 w/m**2 total so the 88 w/m**2 of surface water evaporative latent heat is 16% of the Power sources for making the LWIR. So 16% of the 345 w/m**2 = 55 w/m**2 of the downwelling radiation at the surface is evaporative cooling at surface being returned to the surface as LWIR.
I looked at Hartmann page 166 annual overturning mass stream functions for all of 3 minutes and estimated 7 w/m**2 of adiabatic cooling while ascending & heating while descending, which might be highly miscalculated, from the diagram. Added 7 to the 20 w/m**2 of thermals. So, the actual water evaporative cooling at the surface to compare for scale with the total “greenhouse effect (GHE)” of H2O gas in Earth’s troposphere is 88 – 55 – 7 = 26 w/m**2.
I think you are somewhat struggling with the GHE itself. It is simply the difference between surface- and emission temperature, or respectively surface emissions and OLR. So that is 288 – 255 = 33K, or 390 – 240 = 150 respectively. We could discuss the precise figures here, but that is the idea.
In both instances the lapse rate is decisive for the magnitude of that GHE. Again, if the lapse rate was zero, there would be no GHE at all. And we know because of WV, because of the latent heat, we do not have a dry unstable adiabat (>9.8K/km) but only just ~6.5K/km. So yes, this latent heat DOES cool the planet a lot, easily by an 80W/m2. It ain’t that complicated..
“”55 w/m**2 of the downwelling radiation at the surface is evaporative cooling at surface being returned to the surface as LWIR”””
I wager this is misleading. The main determinant of surface LW down is lower atmospheric temperature, not instantaneous flux.
Climatological temperature at some level is determined by todays planetary solar heating, yesterdays solar heating, and all the days before, minus the LW cooling of the planet on all the days. So we arrive at some cumulative sum of energy absorbed and emitted by the planet, and allow thermodynamic and heat properties to sort out where all this energy ought to go and by association how much temperature at some level should be dragged at some point in time.
Strictly speaking, the immediate power source for atmospheric LW emission is the thermal internal energy of the atmosphere. Jiggling molecules and so on. Over climate timescales, that internal energy is ultimately maintained by the net external radiative energy entering the Earth system (absorbed solar minus outgoing longwave from the perspective of space). When N = 0, the change of internal energy (U) is 0. LW emission is a function of U.
Consequently, saying that latent heating of atmosphere contributes roughly 16% of the atmospheric energy input does not imply that roughly 16% of the surface downwelling longwave “comes from” latent heat. If the partitioning between sensible and latent heat changed while the lower atmospheric temperature remained unchanged, the surface downwelling longwave would remain approximately the same. Conversely, if latent heating decreased but the lower atmosphere became warmer, downwelling longwave would generally increase. The 16% figure simply says something about the state of the system.
If still not convinced, the attribution of LW down to the latent heating is non-unique. It depends only on the accounting framework. I think the issue becomes more clear when using a surface budget, since taking a valid closure equation should not be interpreted it as an attribution equation. Those are not the same thing. The budget tells you only what combinations of fluxes satisfy energy conservation.
Classically the surface budget is written as SW down – SW up + LW down – LW up = H + LE. It says surface net radiation = fluxes of latent and sensible heat.
SW down – SW up + LW down – LW up – H – LE = 0.
Now just rearrange for clarity to understand why it cannot be an attribution.
LE (latent heat of evaporation) = SW down – SW up + LW down – LW up – H.
Now suddenly latent heat flux might seem to depend on LW down, and not be causing it. How can that be.
Now make things really weird:
LW down = SW up – SW down + LW up + H + LE.
Now it seems like LW down has some inverse relation to SW down, which is really weird. Why would increasing SW down decrease LW down? It’s weird because it’s not an attribution, it is simply relations which must satisfy energy conservation. It is a closure relationship, not a physical decomposition of the origin of the downwelling longwave radiation.
Putting elements on one side or another of an equals sign does not assign cause and effect. Physics doesn’t work like that.
The same thing is happening in atmospheric budgets. It is not correct to understand latent heating of atmosphere to be causing LW down, even though at first glance that seems perfectly reasonable. I recommend sticking to the original message, which is that radiation is manufactured simply based on the state of the system. The immediate physical origin of downwelling longwave radiation is the thermal internal energy of the atmosphere. I think somehow this message is being overlooked.
To Schaffer
as always the reality is always a little bit nuanced.
It is obviously true that when greenhouse is described as an observational temperature difference it represents a type of radiative convective equilibrium with all its complexities. I think it shouldn’t be in dispute.
Owing to the freedom of non radiative heat transfer, we know the temperature gradient is diminished compared to what it otherwise might be in a purely radiative equilibrium profile.
The other part is about the depth of surface below the balancing radiative emission level. This can be understood to be associated with the atmospheric composition and its relation to optical thickness.
Flux of mass and heat says something about temperature profiles, and column water vapor says something about the spatial separation of surface and the planetary radiating level along that profile.
For investigation of how a model deals with worlds with varying restrictions on moisture dynamics, including column water vapor state and the cycling aspect associated with flux of mass and heat, I recommend “Continental configuration controls the base-state water vapor greenhouse effect: lessons from half-land, half-water planets” https://eartharxiv.org/repository/view/5196/
There we see 8 different configurations, ranging from landworld, which is basically totally restricted in moisture cycling, to aquaworld which is totally ocean, and things in between, including tropicalland, northland, and realLand. Each has different shapes and distribution of land and ocean. Fig 1. This is meant to illustrate that it matters where and how much moisture is limited or not.
These simulations handle both the features of turbulent flux of heat and mass, and water vapor concentration. Clouds are not discussed, which aligns with the talking points so far on the thread. The range across simulations is about 15K.
There we see how landWorld, despite having extremely limited latent heat flux, exhibits the coldest mean temperature of all simulations Fig 2. Total column water vapor is close to negligible at 3.9 kg Fig 5, and the planet settles around 273K.
The opposite configuration aquaWorld, totally ocean, settles at 285K. This features the greatest partitioning of surface energy to latent heat flux and includes 17.1 kg column WV.
The realLand shapes, which depict relatively complex distributions of land and ocean, generate the warmest temperatures of all 8 simulations at 287K. The partitioning of latent flux is more restricted compared to Aqua, and yet the column water vapor climbs to 17.2 kg.
landWorld cold, aquaWorld medium, and realWorld warm.
I hope it provides a foundation to think about how the heat flux and WV are somewhat related but definitely distinct issues, and how the associated dynamics can interrelate in counterintuitive ways.
cheers
@JCM
I love nuances! And in fact I have made very similar considerations, although less model- and more physics based. So I am quite interested in the model world in that paper, though I’d have a number of questions.
Why is there a surface albedo of ~0.3? The surface of Earth is actually pretty dark. About 1/2 of the albedo is due to clouds, another quater is due to di-atomic atmospheric molecules (N2, O2) which scatter SW radiation. Surface albedo is <0.1 in the real world. Is it possible that the paper does not include clouds, but still keeps their albedo, as part of the surface albedo?
You say that 3.9mm of precipitable water were negligible. When I check modtran such a reduction of WV reduces its GHE by about 30%, meaning 70% are still there. 70% are no way negligible.
The biggest contributor to the GHE (as I would have to explain in a lot of detail) are actually clouds. That is why I think it is extremely troublesome to not consider clouds, especially if one possibly holds on to their albedo effect. One could then have their cooling, but not their warming part considered.
Is it because of that, that all these model worlds feature lower than real surface temperatures? I mean that is kind of strange, when you consider clouds were to be cooling. If you leave them aside, it should get warmer, not colder..
Another irritating detail is about how the lapse rate is treated there. The term lapse rate is not mentioned once, making me assume it is not considered at all. The only hint is some latent heat of 10W/m2 in the LandWorld. That would only be consistent with a very steep lapse rate of ~10K/km, and per se a 50% larger GHE. Given the low amount of WV (3.9mm..) would reduce the GHE by only about 15W/m2, all over the GHE in the LandWorld scenario should be a lot larger and surface temperature higher! That is of course unless it had the cloud-GHE removed and the lapse rate unchanged. Though these would be primary school level blunders..
in re to: “The term lapse rate is not mentioned once, making me assume it is not considered at all.”
This I think gets to the crux of what’s going on here. I do not believe a greenhouse effect expressed as a temperature difference is mentioned either. I figure the reason being that each are emergent properties, not governing variables. They can each be understood as diagnostic effects. It makes little difference to physics whether or not someone decides to extract these quantities from model output.
In a radiative-convective-equilibrium concept, both the lapse rate and GHE are outcomes of the coupled interaction between radiative transfer, non-radiative heat transport, and thermodynamics.
Neither a lapse rate nor GHE can be imposed, they emerge from the balance of processes that warm and cool at different levels.
The way I approach these things is from the teachings of Axel Kleidon, where temperatures depend on the non-equilibrium thermodynamics involving radiative exchange, turbulent heat transport, atmospheric cooling, and the continuous generation and depletion of vertical temperature gradients.
The observed lapse rate and greenhouse effect arise as steady-state expressions of thermodynamic principles. The important point is that turbulent transport requires a radiative generated temperature difference in the first place. Solar radiation preferntially heats the surface, creating a temperature difference between surface and atmosphere. This differential radiative heating generates thermodynamic disequilibrium that drives heat transport.
Kleidon models the surface-atmosphere as a heat engine.
The surface receives radiative heating and that energy can be exported non-radiatively as turbulent heat flux (sensible + latent heat), or leave directly by longwave radiation.
If the turbulent flux is very small, the surface must become very hot. The temperature difference between surface and atmosphere is very large.
If the turbulent flux is very large, the surface cools, the temperature difference collapses, and there is little thermodynamic driving force to generate power.
These competing effects produce an optimum. In other words, temperature differences can only be depleted so far until the engine shuts down. Kleidon recommends that the system be understood to operate near maximum mechanical power P.
Start with the surface energy balance, and collapse turbulent flux of latent and sensible heat into term J. For simplicity we’ll assume grey atmospheric radiation.
Rsw = J + Rlw
The provides relations for surface absorbed solar radiation Rsw (an external energy input), turbulent heat flux J, and net longwave radiative exchange between surface adn atmosphere, Rlw.
Rlw = σ(Ts^4 − Ta^4), where Ts and Ta represent surface and effective atmospheric radiating temperatures.
To Schaffer’s point, when J increases, Ts falls, therefore Rlw also decreases.
The Carnot efficiency is simply n = (Ts – Ta) / Ts
Mechanical power P relates to the heat engine receiving turbulent flux J
P = J ((Ts- Ta) / Ts)
This is where we see the the tradeoff, increasing J increases available heat, but increasing J also cools the surface, and therefore temperature differences are depleted and the engine becomes less efficient.
For different assumed turbulent flux J, we can solve for the corresponding surface temperature, and the generated Power:
Example 1: Assume small turbulent flux J = 50 W/m2
Rlw = Rsw – J
Then Rlw = 165 – 50 = 115 W/m2
Solving σ(Ts^4 − 255^4) = 115 W/m2 gives Ts = 303K,
Power = J ((Ts- Ta) / Ts) = 7.9 W/m2
Example 2: Near the observed value J = 100 W/m2
Rlw = Rsw – J = 65 W/m2
Ts = 289K
Power = 11.8 W/m2
Efficiency fell, but transported heat doubled, and Power increased from 7.9 W/m2 to 11.8 W/m2
Example 3: very high turbulent flux J = 130 W/m2
Rlw = 35 W/m2
Ts = 279K
Power = 11.2 W/m2
Too high turbulent flux J means power is reduced.. More heat is being transported, but the engine efficiency has become too small.
The observed system state is the one in which Power is maximum (Example 2).
What is Power? In the atmosphere, power is the rate at which mechanical work is performed to sustain turbulence, accelerate air, generate winds, and lift air parcels against gravity. P = dW / dt
We have heat flux J, and we have Power (mechanical work). Power = efficiency (n) times heat flux (J). How much available energy generated by differential radiative heating is available for performing Work.
For planet Earth and assuming grey atmosphere, approximately 100 W/2 of turbulent heat flows through the atmospheric heat engine. Given the observed temperature difference between the surface and atmosphere, the maximum theoretical mechanical power is about 12 W/m2. The remaining ~88 W/m2 is not converted into work but is ultimately dissipated as outgoing radiation to space. This should be understood as maximum potential power.
What is happening with that:
Sun heats surface ~ 165 W/m2 –> surface becomes warmer than atmosphere –> vertical temperature gradient is generated –> atmosphere consumes the gradient using H + LE –> radiative cooling aloft restores the gradient.
While that schematic is expressed in a sequence, it should be understood to be happening simultaneously all the time. The vertical profile and GHE (temperature difference) emerges through radiation boundaries continuously regenerating the thermodynamic disequilibrium (solar heating surface, radiative cooling aloft). The turbulent fluxes of latent and sensible heat emerge by consuming the gradient at maximum Power, but never deleting it.
You may have noticed the exhaust ~88 W/m2 associated with the turbulent flux is exactly 1/2 of the surface solar radiative heating (165 W/m2). This is an expression of optimality. In reality there is always nuance, since the system is not a perfectly sealed engine and so it allows direct losses by radiative transmittance from the surface to space mentioned elsewhere in the thread, and we also have frictional dissipation of work energy back to heat.
In summary, if turbulent transport were too weak, the surface would become excessively warm relative to the atmosphere, producing a very large temperature difference. If turbulent transport were too strong, it would erase that temperature difference, leaving little buoyancy to sustain convection. These competing tendencies lead to an optimum, which Kleidon interprets as operation near maximum mechanical power.
This self limiting principle means simply that turbulent transport requires a radiatively generated temperature difference in the first place, which means it could never be overcome or deleted.
In conclusion, it is physically necessary that a positive net greenhouse effect survives in steady state because turbulent heat transport depends on the very radiatively generated temperature gradients that it consumes.
I appreciate the opportunity to bring more clarity to these issues in my own mind.
Working at the limit: a review of thermodynamics and optimality of the Earth system
https://esd.copernicus.org/articles/14/861/2023/
@JCM
To calculate the (dry) adiabatic lapse rate you need the specific heat capacity (of the gas) and gravity. Including the latent heat by WV you need a couple more specific parameters do calculate the moist adiabat. However, what you will not need is a “coupled interaction (model) between radiative transfer, non-radiative heat transport, and thermodynamics”. So I guess it is safe to reject this idea. I mean in this instance we already know the physics, there is no need to re-invent and convolute it.
It is true that the real lapse rate differs from the theoretic adiabat, and GHGs have to do with it. The surface receives way more SW input than it has in LW output through the atmospheric window. With the troposphere it is the opposite, and so there needs to be a transfer of energy. That is why we call it an “unstable adiabat”, meaning a lapse rate larger than just adiabatic. While there is some departure, the base line is still the adiabat.
I think that is the whole problem here. The adiabatic lapse rate is a gas thing, NOT a GHG-thing!!! GHGs, whether condensing or not, play into it, but they do NOT play the main part. By rejecting that very fact things become convoluted, complicated and eventually wrong.
For instance:
“when J increases, Ts falls, therefore Rlw also decreases”
That is not true if you stick to your definition of Rlw: “net longwave radiative exchange between surface adn atmosphere, Rlw”
Are you aware this figure is almost zero? Let me stick to (incorrect) consensus figures for simplicity. Say the surface emits 390W/m2, 40W/m2 of which go right through the atmospheric window, so that 350W/m2 go from the surface into the atmosphere. “Back radiation” might be 340W/m2. In this instance Rlw = 350 – 340 = 10W/m2. It is just a marginal figure, also irrelevant.
However, if Ts changes, Rlw would barely change at all. With higher temperatures there is just more radiation going back- and forth, and vice verse, but the difference between the two barely changes.
I could point out more mistakes, but really the original sin is what I told above. You try do deny the nature of the adiabatic lapse rate and construct some alternative physics, which then requires a lot of stuff and will not work.
I repeat that it’s necessary to disentangle diagnostic variables, such as an environmental lapse rate, and the balancing heat fluxes that satisfy steady-state energy budgets.
A lapse rate does not uniquely determine heat flux. Likewise, an adiabatic rate which is constrained by gravity and heat capacity does not tell you how much energy is flowing through the column.
Two atmospheres can have similar lapse rates while transporting vastly different amounts of energy. Obviously a system with same gravity and heat capacity but absorbing 5x more solar must generate vastly more steady state heat flux compared to Earth.
The underlying message I was hoping to convey is that the atmosphere cannot transport heat so efficiently that it destroys the temperature differences required to transport heat in the first place.
The turbulent heat flux and the greenhouse effect, expressed as a surface-to-radiating-level temperature difference, are coupled. Increasing turbulent transport reduces the temperature difference that drives it, while radiative heating and cooling continually regenerate that temperature difference. The steady state emerges from this competition.
In the detailed example provided, the relevant thermodynamic boundaries provided were the Ts and the atmospheric effective radiating temperature Ta. As such, the heating of atmosphere from below is balanced by radiative cooling out the top in grey atmosphere. The steady state is therefore established by balancing the upward transport of energy through the atmospheric column against radiative loss to space.
In your example as you note, the derivation of the adiabatic lapse rate has no dependence on external heating and cooling rates, and that is precisely why it says nothing about the balancing heat fluxes that satisfy energy budgets.
The analytical relations I presented instead describe how the heat flux itself is constrained. Simply, atmosphere cannot consume the thermodynamic gradients faster than radiative processes regenerate them. Additional steady-state optical depth therefore cannot simply be cancelled by increasing turbulent heat flux, because doing so would undermine its own driving force.
cheers
in Re to “E.Schaffer”, 22 Jul 2026 at 10:08 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849939
Sir,
In case that you are NOT just trolling, I would like to turn your attention to the circumstance 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.
The net radiation from the surface in your example is 390 W/m2 – 340 W/m2 = 50 W/m2, not 10 W/m2 as you assert.
Best regards
Tomáš
JCM 18 Jul 2026 at 5:11 PM “Conversely, if latent heating decreased but the lower atmosphere became warmer”. If latent heating decreased but the troposphere became warmer then the Law of Conservation of Energy would have been violated. To put that another way, a statement so incomplete that it is worthless is ……worthless.
Not that I would see any merit in these considerations, because “back radiation” is still not heating anything. I also do not really see what exact assumptions are being made here, because it would seem there are plenty of contradictions.
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..
in re: to Barry
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850101
I don’t see any particular violation in the wordage used in the example. It was simply to support your assertion that photons being manufactured at some place relate mostly to the local thermodynamic equilibrium temperature there.
The thermodynamic duty to compensate the external solar radiative input to the surface remains.
Against 165 solar external input must come the balancing fluxes of radiation and turbulent heat transfer.
This is not an independent physical law, but a consequence of the drive toward establishing a steady state balance.
Artificially slow down hydrological cycling and allow the system to reorganize so that the balance to shifted to a greater burden upon radiation and sensible heat transfer. I suspect after which a radiometer will detect a greater signal being generated when pointing the detector in various directions from the surface.
In re to Schaffer
The initial warming from an external radiative forcing + the associated enhanced opacity from increasing WV is what creates the conditions for stronger moist convection. Deleting those conditions, as you propose, ultimately cancels the thermodynamic conditions that gave rise to it in the first place, and you create a paradox.
It is important to keep in mind always when interpreting radiative kernels that the lapse-rate adjustment is a consequence of warming, not an independent opposing mechanism. The lapse rate adjustment is the thermodynamic signature of a warmer, moister convecting atmosphere, not something that could overcompensate it.
If you want to dive into the weeds analytically I recommend to explore why the surface energy balance becomes increasingly dominated by latent heat flux at warmer temperatures and why OLR becomes increasingly dominated by atmospheric emission.
Whereupon Surface solar radiative heating is balanced by LE + H + radiative diffusion (k) + radiative transmittance (τ_transmission). General default values are something like 160 solar radiative heating = 80 LE + 20 H + 20 k + 40 τ_transmission. This is more illustrative to address your particular concerns with atmospheric “windows” which didn’t come about in grey formulations previously by definition.
From this perspective two primary things are occurring:
1) more abundant of water vapor at higher temperatures narrows the available windows through which the surface thermal emission can be transmitted to space. As a result, the contribution to the OLR from the surface gets smaller as Ts increases, implying the increasing decoupling of the OLR from the surface’s emission. A standard optical depth around 2 implies radiative transmittance ~ 10%-15%. Increasing water vapor progressively diminishes the transmissivity and more outgoing emission must be sourced from colder atmospheric layers.
2) at higher temperatures with ample surface moisture the partitioning between LE + H increasingly favors the latent flux through the psychrometric constant (γ) and slope of saturation vapor pressure curve (Δ). As temperature rises the slope of saturation vapor curve is more steep and turbulent flux partitioning increasing favors the latent flux over sensible heat through Δ/γ.
In both cases 1 and 2, it is the warmer temperatures themselves which generate apparent atmospheric adjustments. Such adjustments cannot be sustained if warmer temperatures are cancelled. I think there is some risk of getting lost in statistics and mashing up data sources when applying your methods, and losing sight of what’s going on.
Additionally, I would also keep in mind that the tropical hotspots and things you’re seeing in model output are a consequence of the very warming you’re trying to cancel. Deleting warming deletes the very effect of your interest, resulting in a potential logical void. No amount of adjusting aspect ratios in visualizing re-digitized datasets will resolve that.
I suppose that if participants find my input to be worthless, or without merit, they would simply cease to respond. cheers and thanks for the discussion.
more contrasting perspectives on such matters:
The Quasi-Linear Relation between Planetary Outgoing Longwave Radiation and Surface Temperature: A Climate Footprint of Radiative and Nonradiative Processes
https://journals.ametsoc.org/view/journals/atsc/80/9/JAS-D-22-0261.1.xml
Earth’s outgoing longwave radiation linear due to H2O greenhouse effect
https://www.pnas.org/doi/full/10.1073/pnas.1809868115
ES: “back radiation” is still not heating anything.
BPL: The surface. Also, layers of atmosphere below the one radiating.
Perhaps unclear, I didn’t mean that the 86.4W/m2 Power flux is only for the tropical ocean, I know it’s a globally-averaged value of 44.1 petawatts (44.9 in my Hartmann text book, (40.3 or 40.8 I recall from earlier EEB pictorials). I meant that the 44.1 petawatts of surface water evaporative cooling must be reduced by (8% I just very-roughly calculated) due to descending air being adiabatically heated and H2O gas condensing-freezing must be given its allocation of that sensible heat downward at the surface. I’ve since realized that I forgot the effelump in the room for that which is that downwelling LWIR at the surface includes 16% of its Power from the upwelling evaporation of the ocean and that means that surface water evaporative cooling must be reduced by a massive 64% (63% for the Hartmann EEB) because that much heat that left the surface as evaporative cooling returned to the surface as LWIR.
An interesting paper in the journal Nature ‘Amplified Arctic iceberg traffic reshapes benthic biodiversity’ and a news story about it remind me how climate change is terrifying, but climate science is fascinating. Scientists looked at dropstones, rocks that start embedded in glaciers, broke off into the ocean in icebergs, then were deposited on the ocean floor as the icebergs melted. They were able to use the recordings of seabeds as a proxy to determine how fast glaciers were breaking up in some areas. They also recorded the ecosystem changes caused by the dropstones. Organisms that needed hard surfaces to live on were increasing as the ecosystems changed from soft sand or mud to hard stones and rocks.
News article:
https://www.nytimes.com/2026/07/07/science/arctic-icebergs-deepsea-life.html
The paper, its freely available
https://www.nature.com/articles/s41586-026-10630-4
Copernicus has posted for June ERA5 re-analysis with the month’s anomaly of +0.56ºC, this the second warmest June on record.
The top-ten ERA5 Junes now run:-
2024 … +0.67ºC,
2026 … +0.56ºC,
2023 … +0.53ºC,
2025 … +0.47ºC,
2019 … +0.37ºC,
2020 … +0.36ºC,
2022 … +0.31ºC,
2016 … +0.26ºC,
2018 … +0.23ºC,
2021 … +0.21ºC.
Globally, 2026 started warmer than 2023 but cooler than 2024 and now, mid-year, the three are equally warm at about +0.6ºC. The Dec 2023 and 2024 anomalies sat at about +0.8ºC.
This pretty-much echoes the NH situation, with the 2023 and 2024 NH anomaly tracking each other very closely from mid-July to year’s end. Down in the SH 2026 has been tracking the slow drop seen in 2025, cooler than 2024 and since May a tad cooler than 2023 (which saw a sharply rising in anomaly Jan-to-October).
I should mention that the SH anomaly is subject to big Antarctic wobbles which run throughout the year. (Up north, big Arctic wobbles are restricted to winter months.) The Antarctic wobbles do look to have contributed a significant cooling within the SH average anomaly through June.
The Copernicus numbers include a value for “relative to the 1850–1900 pre-industrial.”
Until the extra highlighting employed in this month’s update, I hadn’t managed to notice before the link to “Read more about how the temperature change above the pre-industrial level is estimated and the differences between global temperature datasets.” This provides explanation of the method used by ERA5 to calculate its 1850-1900 anomaly base (given the ERA5 record only runs back to 1940).
The method uses the average of BEST, HadCRUT & NOAA (which do cover the 1850-1900 period) that yields the annual offset of -0.88ºC relative to the ERA5’s 1991-2020 base. Monthly average offsets (which show an annual variation, -0.77ºC to -0.98ºC) are then used to calculate “objectively using Fourier fitting of a single harmonic, rounding coefficients to two decimal places” for monthly offsets and, with a bit of additional smoothing, daily offsets.
UAH TLT has posted for June (the first measured June data) with an anomaly of +0.46ºC, down on May’s +0.53ºC. The NH showed a healthy increase (May +0.46ºC, June +0.54ºC).while the SH a bigger decrease (May +0.60ºC, June +0.38ºC).
The July NOAA ENSO: Recent Evolution is posted. The RONI numbers show AMJ 2026 at +0.5, this a little behind the 1997 and 2015 El Niños, although the Central Pacific OHC pretty-much tracking 1997-98 and well above 2015-16 etc.
The above could be compared with 10 days ago – MA Rodger says 30 Jun 2026 at 2:26 PM
–how June 2026 (+0.55ºC) will be roughly unchanged from May26 and marginally above the Jan-May average (+0.53ºC). (Recent year Jan-Jun have averaged 2023 +0.38ºC, 2024 +0.70ºC, 2025 +0.61ºC),
–The SAT numbers wouldn’t be showing any sign of the coming El Niño while the ERA5 60N-60S SST so far continues the pause seen during May26
— so the presently-developed El Niño conditions would be rated as ‘weak’.
–a 2-in-3 chance of ‘very strong’ El Niño conditions (RONI>2.5) by November.
https://www.realclimate.org/index.php/archives/2026/06/unforced-variations-june-2026/#comment-849502
June 2026 was the hottest June recorded for western Europe and the second warmest globally.
It saw near-record temperatures driven by the highest sea surface temperatures (SSTs) on record for the month,
https://climate.copernicus.eu/copernicus-record-heatwave-brings-hottest-june-western-europe-during-second-warmest-june-globally
June 2026 was the second-warmest June in analyses of global weather data going back to 1850
a 95% chance that 2026 will rank among the four warmest years on record.
NOAA expects a very strong “super” El Niño event
https://yaleclimateconnections.org/2026/07/june-2026-earths-2nd-hottest-june-on-record/
and
https://climate.copernicus.eu/surface-air-temperature-june-2026
and
There is a new El Niño out there and it was officially declared already on June 11 by NOAA.
This is both unusually early in the year and very soon since the last El Niño in 2023-24.
Another remarkable thing is the seasonal forecasts, which for a couple of months have indicated that it may well be at strengths by the end of the year that we have not seen before.
All these three aspects combined seem to make this El Niño different to the previous ones.
https://www.realclimate.org/index.php/archives/2026/07/this-new-el-nino-is-different/
The idea of an impending El Niño is particularly interesting to me, as I’ve experienced its effects firsthand on my beach travels. I’ve found that off-season trips can be a great way to avoid the crowds and heat waves that often come with these events. For instance, I once took a ferry to a remote island during the off-season and was able to explore its beautiful beaches without the throngs of tourists. One practical tip I’d like to add is to always research the necessary permits and timing for these trips, as they can vary greatly depending on the location and time of year. I’m wondering, how do you think the impending El Niño will impact beach ecosystems and coastal communities, and what can we learn from past experiences to better prepare for its effects?
Rohit, how very nice for you. I’m glowing in empathetic pleasure.
I retract that because “Rohit Ghosh” isn’t Climate Systems Scientist “Paul Beckwith” flying Canadian Winter to Florida for a blitz round the Everglades at 99 miles per hour in a 15 ton Power Boat then back to Ottawa for more science videos about the ‘orrible Climate Changes, and ask for some more money.
GISTEMP and NOAA have posted for June. (Numbers-wise, there’s little difference except in the SH.) NOAA’s “June Highlights” run:-
☻ The globe had its second-warmest June on record, driven by global ocean surface temperatures reaching an all-time high for the month.
☻ Both the Arctic and Antarctic recorded June sea ice extents that ranked among their respective 10 lowest extents on record.
☻ Global tropical cyclone activity was above average with seven named storms. (Note the Atlantic is having a quiet start to the season.)
For June (+1.18ºC), GISTEMP shows a small rise on May’s anomaly (+1.13ºC) with an increase in the NH (May +1.30ºC, June +1.44ºC) and a smaller decrease in the SH (May +0.97ºC, June +0.92ºC).
The 2026 anomalies for May & June both sit above the 2025 May & June so the rolling 12-month average is now bottomed out and on the rise. And I reckon that we are now starting to see them up-tick with the coming El Niño. Compared with 2023, it’s a relatively gentle up-tick and not yet appearing in the wholly-SAT record of ERA5 (although in the 60N-60S SST of ERA5 anomalies have been on the up for some time).
<a href="https://sites.google.com/site/housman100resultstemperarypost/home/the-banana-watch"<The Banana!!! Watch page monitors these temperature developments.
E. Schaffer 10 Jul 2026 at 10:59 AM ” I am … asking about the given effect of WV”. That’s a topic to which applies “Point of view” not socially but from a physical science basis because there are mutual.ly-reinforcing effects. A big one is ice-snow cover. The simplifying statement that Earth would be ~33 degrees colder without CO2 must be highly incorrect because ice-snow has a much higher albedo than ocean water so a reduction of CO2 to the point where the ocean is mostly or entirely iced over must make Earth far colder than just ~33 degrees colder.
Since H2O gas is a condensing gas in the troposphere it must be considered a feedback, not a Forcing, and thus its contribution to Earth’s GMST, cannot logically be in a separate category from the surface albedo “effect” on temperature and cloud “effect” on temperature.
So, if Earth goes from Snowball Earth to Holocene Earth there’s no single division of warming “effect” on temperature rise because change in open ocean water portion affects H2O gas, and CO2 also, and change in well-mixed GHGs and condensing H2O gas affects the open ocean water portion.
Also, if some emission (manufacturing) and absorption spectral lines are shared by CO2 and H2O gas, rather than being interleaved, then the allocation of “effect” is arbitrary or decided by physics that’s past what I’m willing to spend time pondering.
Barry: “ if some emission (manufacturing) and absorption spectral lines are shared by CO2 and H2O gas, rather than being interleaved, then the allocation of “effect” is arbitrary or decided by physics that’s past what I’m willing to spend time pondering
It has been already discussed in Schmidt et al. 2010 – where they pointed out – that “direct measure” of warming by any substance influencing the climate – does not have a single value – because even if a given factor was removed, other factors with overlapping windows of absorption would partly compensate by increasing their own absorption. To make things worse not only there is no way to subtract this compensation – the % of compensation would be different in moments in time.
And that’s on top mentioned by you: passive feedbacks (like WV) and not the same as forcings (anthropogenic GHGs) and there is no point in calculating “direct measure” of feedbacks, other than the deniers trying to divert the attention from the mitigation of GHGs and allow fossil fuel political- industrial complex to extra as much fossil fuel as their shareholders, and cost of waging wars on other countries, demand.
You are right in stating the 33K (or so) are not enshrined in law. They are supposed to be the difference between surface- and emission temperature, that is all.
It is indeed incorrect that the Earth would be 33K colder without CO2, for lots of reasons. We have other forcing GHGs like O3, CH4 or N2O – it is not just CO2. The largest contributors to the GHE however are clouds and WV, which both have their cooling sides. If you’d remove them you would not just lose their contributions to the GHE, but also change the albedo and the lapse rate.
A simple calculation might look like this: The energy budget would go from 240W/m2 to 290W/m2 if you lose cloud albedo. The non-condensing GHGs (CO2, CH4, O3, N2O..) might contribute about 50W/m2 to the GHE (single factor addition), at the given lapse rate. Without WV this lapse rate would be at least 50% larger, let us say it would be 80W/m2 then. Also we know the surface as it is does emit ~360W/m2.
Without WV and clouds we get 290W/m2 in and 280W/m2 out (360-80), that is an energy SURPLUS(!!!) of 10W/m2. That is if we’d remove WV and clouds from the atmosphere, the Earth would warm.
ES: It is indeed incorrect that the Earth would be 33K colder without CO2
BPL: More of a strawman. The Earth would be 33 K colder without the entire greenhouse effect, not without CO2 in particular.
ES: if we’d remove WV and clouds from the atmosphere, the Earth would warm.
BPL: Congratulations on the nuttiest denier idea I’ve heard around here lately.
Please do feel free to cite the peer-reviewed publication in which these claims are substantiated.
in Re to E. Schaffer, 10 Jul 2026 at 10:59 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849683
Sir,
I am afraid that the way how you framed your question (like “Is water vapour cooling or warming Earth?” or “Is water vapour cooling or warming Earth’s surface?”) may prevent any reasonable answer.
On the first sight, it appears that your formulation reduces the role of water in Earth climate to its vapour, what would suggest a primary focus on the greenhouse effect of water vapour. However, if you perhaps consider also the role of water in lapse rate regulation and/or in cloudiness regulation (what seems to be the case based on your complaints that your request is not taken seriously), you should not speak about “vapour”.
Instead, I would expect that you will analyse the relationship between water cycle intensity and geographical distribution of water availability for evaporation, including their complex interactions with other components of the climate system and processes therein. This way, you could perhaps arrive at some hints if (and/or under which circumstances) the assumed summary “cooling” or “warming” effect of various forms of water (and/or various processes involving water vapour formation and condensation) on Earth’s surface prevails.
Nevertheless, I do not think it would be sufficient. Furthermore, you should consider also climate aspects related to water heat capacity, heat distribution by ocean currents, and properties and behaviour of ice. Finally, you should not neglect also underground water as well as water distribution in soils and vegetation, and in all these instances keep in mind that all forms of water interact with water vapour and that these interactions and processes involving them are no way static.
For these reasons, I can hardly imagine how this complexity could be reduced to the requested unambiguous estimation if the “effect of water vapour” is “cooling or warming”. I doubt that such generic request makes sense.
Best regards
Tomáš
Yes, that sounds indeed very complicated.
I guess we could go down this path, or we just stick to the pretty simple reality. In this simple reality WV does two things:
– it raises the emission altitude, thus reduces the emission temperature, which enhances the GHE and warms the planet
– it reduces the lapse rate, thus increases the emission temperature, which reduces the GHE and cools the planet
And almost as if I had to repeat myself (lol), the question is about netting those two effects. And again, I can not believe this question was never dealt with and comes across as if it was an alien spaceship. This should be trivial!!!
You are forgetting the fact that water vapor is a potent greenhouse gas in its own right.
“You are forgetting the fact that water vapor is a potent greenhouse gas in its own right”
Yes, by elevating the emission altitude, which I just pointed out. But it also has that other side..
E. Schafer: “ we just stick to the pretty simple reality., ”
.the simple reality that we ALREADY KNOW that we can’t do ANTYHING about WV, clouds and lapse rate to mitigate AGW in any meaningful way?
Nor will it improve our knowledge – VW, clouds and lapse rate are already implicitely included in the climate models. If the model show 2C warming what difference does it make to ask how much of that 2C came from direct forcing (GHGs) and how much from the passive feedbacks (water cycle).
Instead, we should stay focused on plausible scenarioes of mitigation of GHGs, instead of pondering futile questions like what the Earth would look like “if we somehow could remove both all WV and clouds“. Particularly when we already know that raising temperature with GHGs tends to decrease clouds, but increase WV.
Well, it is exactly about improving our knowledge, and decisevely so.
A Schafer: ” it is exactly about improving our knowledge, and decisevely so.”
I have shown the opposite – as all deniers’ narratives:
– it has a NEGATIVE societal value – not only does provide any useful information to deal with existential threat to our civilization (we can’t do ANTYHING about WV, clouds and lapse rate to mitigate AGW in any meaningful way), but by confusing things – it detracts the attention from the urgency of GHG mitigation.
– it offers no new intellectual insight into the climate system, nor does it provide any ways to improve our modelling of it – as already explained in the post to which you ostensibly “reply”:
===
P: “Nor will it improve our knowledge – VW, clouds and lapse rate are already implicitely included in the climate models. If the model show 2C warming what difference does it make to ask how much of that 2C came from direct forcing (GHGs) and how much from the passive feedbacks (water cycle)? ”
===
Your unsupported with ANYTHING declaration : “it is improving our knowledge, and decisevely so” does not answer that, and as such – proves nothing, other than the quality of your intellect/character.
You improve scientific knowledge by publishing in peer-reviewed journals, not by bloviating on blogs.
“You improve scientific knowledge by publishing in peer-reviewed journals, not by bloviating on blogs.”
LLMs are not trained on many of the reputable classic textbooks because they do not have access to the publishers’ repositories. Yet they can use blogs, wikis, forums, stackexchange queries, github repos, arXiv, etc to distill essentially this same information. So for better or worse, scientific knowledge is improved as LLMs statistically incorporate crowd-sourced discussions. If LLMs were confined to peer-reviewed journals only they would lack some of the practical and intuitive aspects of frontier knowledge.
You can prompt an LLM to verify the above and it will respond to the effect: “LLMs necessarily learn scientific knowledge from a broader epistemic ecosystem than peer‑reviewed journals alone — and historically, that’s exactly how science has always advanced.”
Welcome to the new world.
The Dynamics of Greenland Ice Sheet Melt: Atmospheric Drivers and Feedback Loops
Video
Ray Ladbury,
Picking up on your comments from the June UV thread saying:-
I’ve spent a while considering this idea that nitrogen (80% of the atmosphere) has less kinetic energy (0.6e-20 J at sea level, 0.45e-20 J at tropopause) than required to impart the v2 wobble into a CO2 molecule (1.3e-20 J) with the apparent consequence that the majority (or a significant proportion) of the wobbly CO2(v2) which relax to emit 15 micron photons would have thus been, at some altitudes, excited not by collision but by absorbed 15 micron IR.
This idea (& its basis) is contrary to what I’ve been led to understand, which is that the overwhelming mode of atmospheric CO2(v2) excitation is due to collision and, whatever the cause of excitation, it only a very few CO(v2) that survive continuing collisions to relax and emit 15 micron photons.
Resorting to the literature, the problem with resolving this contradiction is that it seems the mechanisms of the IR flying around the lower atmosphere (unlike, say, the mesosphere**) is for some time settled science and no longer discussed in literature found on-line. (** Higher up it is apparently the stuff of Nobel Prizes according to Kutepov et al (2025)“The year 2025 marks the 55th anniversary of Paul Crutzen’s (1995 Nobel Prize in Chemistry) hypothesis that collisions of CO2 molecules with O(3P) atoms is the dominant process responsible for the excitation of the bending vibrational mode of CO2”)
After investigating quite a few rabbit holes in search of something to reconcile this contradiction, I think I have the explanation. It is, of course, the obvious one (but the roundabout journey getting there was at times interesting.)
CO2(v2) requires two or three times the kinetic energy of an average N2 molecule. But the distribution of N2 energy (think Maxwell–Boltzmann distribution) will still deliver significant collision frequency with faster higher energy N2. And there are a lot of collisions. So higher energy ones are not at all rare events.
The smaller population of faster N2 with enough energy for a CO2(v2) wobble would be perhaps a tenth the total.
For the emitting CO2(v2) to be almost all collision excited, the ratio of fast collisions to emitted/absorbed IR has to be very large. This ratio is very large for all collisions, a CO2(v2) requiring perhaps some 0.02 secs to relax while collisions occur on average every 0.0000000002 secs. And even with fewer fast collisions, and even up at the tropopause where density and temperature are lower, you won’t be losing many of those zeros in the collision frequency number.
Which all points to collision being the overwhelming cause of CO2(v2) and thus also of the resulting IR emissions.
I believe you’ve got it. There is a lot of N2, so you still get kinetic excitation–but consistent with the amount expected for a blackbody distribution at the temperature of the atmosphere. The flux of 15 micron photons from below is pumping energy into the CO2 vibrational mode, making it momentarily “hotter” than the surroundings. This disequilibrium is remedied when another N2 molecule picks up the energy of the vibrational mode by colliding with the excited CO2 molecule–so the energy flows from the ‘higher-local temperature” CO2 to the “lower-local temperature” N2. Make sense?
MAR:
“ a CO2(v2) requiring perhaps some 0.02 secs to relax while collisions occur on average every 0.0000000002 secs., And even with fewer fast collisions, and even up at the tropopause where density and temperature are lower, you won’t be losing many of those zeros in the collision frequency number. ”
But wouldn’t the fast collision be then followed by many slow collisions – sapping out that fast collision energy from the Co2 long before it had a chance to relax?
Piotr,
There is no “sapping out”. The excitation is either there or not there**. That some survive long enough to relax and emit IR radiation is presumably a statistical thing.
** Molecular vibrations like the CO2(v2) excitation each have but one energy level. It’s a quantum mechanics thing and it’s why it is only 15 micron photons that become tangles in unexcited CO2 molecules and get absorbed in a CO2(v2) state.
Spinning CO2 molecules (with their spin subject to quantum effects) entangle/emit slightly different IR wavelengths..
Another effect altering absorption/emission wavelengths is a rather technical phenomenon way beyond my paygrade called “pressure broadening” which fuzzes the absorption/emission spectral lines.
MAR: “There is no “sapping out”. The excitation is either there or not there**. Some survive long enough to relax”
then …. what makes the rest NOT survive to the relaxation time?
Piotr,
I’d guess the survival of CO2(v2) wobbles is a statistical thing, mainly.
The number of CO2 molecules is seriously massive and a massive number of unwobbled CO2 are continually being wobbled into CO2(v2) in collisions. With the collisions very very frequent, the chances of being de-wobbled by a strong collision before relaxation/IR-emission is very high. The statistics result the classical law dictating IR-emission – the Rayleigh–Jeans law.
This law is ‘mainly’ correct, but the survival of CO2(v2) in the atmosphere which results in IR-emission is also affected by the “Ultra Violet Catastrophe” which is explained by Quantum Mechanics. Indeed the existence of Quantum Mechanics was first discovered because of the “Ultra Violet Catastrophe.” This adjustment is magnitude of effect you ask about, that resulting from the “not survive” due to the “no sapping”.
MAR: “ With the collisions very very frequent, the chances of being de-wobbled by a strong collision before relaxation/IR-emission is very high .”
but isn’t your “de-wobbling” identical to my “sapping out” as in:
“ the fast collision [creating the wobble] is then followed by many slow collisions – sapping out that fast collision energy from the Co2 long before it had a chance to relax?
the sapping out that you said …. does not happen?
And if they are identical – then my original question stands – you argued that although the average N2 collision energy is several times too small to wobble a CO2, because the energy of these collisions follows Maxwell–Boltzmann distribution – there will be some on the right end of the distribution that DO have enough energy to wobble a Co2. But if there are 100 millions of collisions (“0.02 secs to relax /a collision average every 0.0000000002 secs”)
between wobble and the relaxation time – and most of these 100 millions can de-wobble a previously wo0bbled CO2 – so none of the wobbled would survive to the relaxation time.
Then again – if one can de-wobble a wobble created by a fast collision, then one should be able to de-wobble a wobble created by absorption of the IR photon too, which would mean that the only LW emission into space would come from … IR outside the windows of absorption ( that once emitted goes unabsorbed into space). But that does no agree with the LW emissions budget – 200 W/m2 IR emitted into space from atm. and only 40 going directly into space through the atm. window.
So for me – this suggest that perhaps there is NO dewobbling (once in a high energy state – perhaps you can’t transfer this extra energy to other molecules via a collision?), i.e. that once you got wobbled (by a fast collision or by absorption of a IR photon) your survival till the relaxion time is 100%.
Then this would explain why the IR flux from atm into space is 200 W/m2, while the back-radiation (toward the surface) is 340 W/m2: the former happens at the top of the atmosphere – with low pressure (hence fewer collisions) and low temperature (hence the collision energy distribution shifted to the left – so for a million collision fewer would have enough energy to excite Co2 than in higher temp.)
OK, we need to be careful here. What matters is the net flow of energy through the atmosphere and through the various modes that can store it. At ground level, we have an upward flux of 15 micron photons that corresponds roughly to that expected of the blackbody surface temperature. Those photons are moving into a cooler environment and are highly likely to be absorbed by a CO2 molecule in its ground state. This is just one way such an excitation of CO2 can occur, the main other mechanism being collisional excitation by a N2 (or O2) molecule with energy far above the average energy (~kT).
Very, very few (effectively none) in the middle of the CO2 absorption band make it from the surface out of the atmosphere. In fact, very few make it very far in the atmosphere at all. There will be some photons, however, emitted by CO2–and the same result occurs at a slightly higher altitude/lower temperature. Note that at this higher altitude there is a still lower probability of collisional excitation, although collisional relaxation is still the dominant de-excitation mode..
At all levels the accelerated N2/O2 having collided with the CO2 is much more likely to collide with another N2 or O2 molecule. The energy imparted tends to thermalize/equilibrate. At each altitude, the CO2 emits a flux of vibration-band photons roughly commensurate with the blackbody temperature. This flux is lower the higher you go. Eventually, you reach an altitude where the photon has a snowball’s chance in hell of escaping, but by now, it’s come from much higher/colder, so there is a big bite taken out of the spectrum around 15 microns.
Please note: I am simplifying like all hell here. The vibration absorption band itself has a dependence on wavelength as you move away from the central wavelength–and this dependence is itself influenced by collisions between the CO2 molecule and other molecules.
But, generally, when all else fails, follow the energy.
in Re to Piotr, 15 Jul 2026 at 8:55 PM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849798
Hello Piotr,
If there were no “de-wobbling” of vibrationally excited GHG molecules in the atmosphere by their collisions with other air molecules, most of the atmosphere would not “see” the infrared radiation absorbed and emitted by GHGs, I think. I suppose that it would result in an isothermal atmosphere, similarly as in absence of GHGs, only with the difference that this strange atmosphere comprising GHGs that do not collide with other gases and thus having two different temperature profiles (decreasing with altitude for GHGs and isothermal for other components) would enable higher average surface temperature than in the same planet without the GHG component.
As it is not the case for Earth, I would expect that in an isothermal air parcel without water phase changes, heat production by collisional deexcitation of GHG molecules equals the absorbed radiative power and also equals to the emitted radiative power which exactly consumes the heat produced by collisional deexcitation.
I suggested this idealized scenario to Gemini Thinking and asked if it can explain your conundrum. The engine answered that the difference in rate coefficients for collisional excitation and deexcitation of a single CO2 molecule is exactly compensated by the opposite difference in population of the CO2 molecules in the ground vibrational state and in the excited vibrational state. For this reason, the volumetric rates of excitation and deexcitation collisions in the considered isothermal air parcel are equal.
Greetings
Tomáš
Piotr,
You ask Isn’t … “de-wobbling” identical to … “sapping out”?
Your description of “sapping out” is that involves “many slow collisions … sapping out that … collision energy.” The wobble of a CO2(v2) molecule will exist or not exist. It will not be half there, half ‘sapped out’. Thus the energy involved with the wobble will arrive due to collision or the absorbing of an IR photon. And that energy will be lost due to a collision or due to the emission of an IR photon. Almost all CO2(v2) will be de-wobbled by collision, just as almost all are ‘wobbled’ into CO2(v2) by collision.
The major variable in this process is air temperature. Up high in the troposphere, the reduced IR in the 15 micron waveband is because of the colder temperature, and such a colder temperature has less collisions so the CO2(v2) population is smaller, with those surviving to emit 15 micron IR likewise smaller,
re Piotr – See my comments upthread https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849537 ,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849541 … … …
Piotr, (no “Reply” button) typed “de-wobble a wobble created by absorption of the IR photon too, which would mean that the only LW emission into space would come from … IR outside the windows of absorption ( that once emitted goes unabsorbed into space)”. Not at all because some photons don’t go through molecules that are capable of absorbing them because the air “gas concentration” reduces with altitude.
Barry E. Finch: “ Not at all because some photons don’t go through molecules that are capable of absorbing them because the air “gas concentration” reduces with altitude”
There is enough of the CO2 molecules between the Earth surface and out space for the vast majority of IR in the CO2 absorption bands to be absorbed. See also:
– Ray: “ Very, very few (effectively none) in the middle of the CO2 absorption band make it from the surface out of the atmosphere. ”
This should take of your “not at all”, right?
If not – see also the concept of the “atmospheric window” in Trenberth energy budgets (=40 W of IR /m2).
Also Wikipedia: ” An atmospheric window is a region of the electromagnetic spectrum that can pass through the atmosphere of Earth [without significant absorption or scattering]”
MA Rodger: “Almost all CO2(v2) will be de-wobbled by collision,”
again, how this de-wobbling by collisions works if not by net-transferring the energy from a molecule with a higher energy Co2(v2) to those with lower energy (the vast majority of N2). Which I termed “energy sapping by collisions with molecules with a smaller energy” and you said doesn’t happen.
Unless you mean that once wobbled – this wobbling energy gets “locked in” inside the CO2 molecule, and as such – is unavailable from the net transfer of energy to other molecules via collisions.
But then if the wobble energy is indeed unavailable for energy transfer via collisions HOW would you ever “ de-wobble almost all CO2(v2) by collision” ?
Something I have explored in the last part of my post you are commenting:
====
Piotr J15 Jul: “So for me – this suggest that perhaps there is NO dewobbling (once in a high energy state – perhaps you can’t transfer this extra energy to other molecules via a collision?), i.e. that once you got wobbled (by a fast collision or by absorption of a IR photon) your survival till the relaxion time is 100%.
Then this would explain why the IR flux from atm into space is 200 W/m2, while the back-radiation (toward the surface) is 340 W/m2: the former happens at the top of the atmosphere – with low pressure (hence fewer collisions) and low temperature (hence the collision energy distribution shifted to the left – so for a million collision fewer would have enough energy to excite Co2 than in higher temp.)”
====
patrick “Piotr – See my comments upthread”
Thanks but the physics of your posts is above my paygrade. Could you distill it to answer my querry. i.e.:
Can the thermal collisions of N2 with a previously excited (“wobbled”) Co2, Co2(v2), “de-wobble it” before emission of an IR photon, or not ?
Tomas: “ If there were no “de-wobbling” of vibrationally excited GHG molecules in the atmosphere by their collisions with other air molecules, most of the atmosphere would not “see” the infrared radiation absorbed and emitted by GHGs ”
Huh? How “no de-wobbling”, i.e. 100% of wobbled Co2 emits IR photon – means that …0% of Co2 absorbs and emits IR photon (as in your: “atm. would not “see” the IR absorbed and emitted by GHGs“)???
re Piotr –
Yes, all molecules are frequently (in most of the atmosphere) being both excited and de-excited between different states of different energies (and I imagine sometimes being shifted among different states of the same energy, where possible).
I think MA Rodger took your wording “sapping out” to imply some gradual process as if the molecules were behaving like classical macroscopic objects. Whereas … actually, from what I’ve read, quantum systems can undergo gradual transitions(?) – provided they aren’t being ‘watched too closely’, and maybe requiring a superposition of the various states that they could be observed to have(?) –
…(AIUI the wavefunction is complex with the imaginary and real components phase shifted 90° from each other either in space or time (?at least for a stationary state), such that for a bound stationary state (as opposed to a propagating wave packet), the magnitude is constant at each system-relative location [picture a rotating arrow in the complex number plane from 0 to e^(i·ω·t) ], as is its square, the probability density, but in superposition, we have (?) two arrows (added as if vectors) rotating at different angular frequencies ω, their sum’s magnitude thus having a beat frequency ∆ω (?) … it’s tempting to think of this as like current flow in an classical macroscopic antenna absorbing or emitting electromagnetic waves, but technically idk… https://en.wikipedia.org/wiki/Stationary_state )…
Also the phrasing “slow collisions” – yes I would imagine slower moving molecules, having less translational energy, would be less likely to excite another molecule and more likely to de-excite another molecule, but it could also just gain translational energy from another molecule, and it could be in an excited state that… well I expect there’s various possible combinations of energy transfers and conversions that could happen, but I’ve not really examined the quantum nature of that.
Tomáš Kalisz’s
– sounds right to me.
His other statement “If there were no “de-wobbling” of vibrationally excited GHG molecules in the atmosphere by their collisions with other air molecules, most of the atmosphere would not “see” the infrared radiation absorbed and emitted by GHGs, I think. ” perhaps confused the elimination of collisional de-excitation with the elimination of all de-excitation.
in Re to Piotr, 20 Jul 2026 at 2:43 PM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849891
Hello Piotr,
By my sentence
“most of the atmosphere would not “see” the infrared radiation absorbed and emitted by GHGs,”
I meant other components of the atmosphere than the GHGs, in other words, N2, O2 and noble gases.
I apologize for ambiguous wording “most of the atmosphere”, “other components of the atmosphere” would be better. Perhaps also the word “see” may be confusing, I think it should rather read “feel”.
Greetings
Tomáš
Piotr,
My confusion in interpreting your “sapping out the collision energy” is (or perhaps ‘was’) both because of the dictionary meaning of the word ‘sapping’ and because your original comment using the word talked of “many slow collisions – sapping out that fast collision energy.”
But you seem to be saying that we are actually on the same page – that a CO2(v2) will overwhelmingly lose the (v2) wobble-energy in a single collision event, the energy being transferred into the (less exotic) thermal energy of the atmosphere. Hurrah!!!
MAR: “we are actually on the same page – that a CO2(v2) will overwhelmingly lose the (v2) wobble-energy in a single collision event”
P: in that area – yes.
MAR: “Your original comment using the word talked of “many slow collisions{– sapping out that fast collision energy.”
Your – “ a CO2(v2) requiring perhaps some 0.02 secs to relax while collisions occur on average every 0.0000000002 secs.” means 10 mln collisions during these 0.o2 – that;s why I wrote “, many,”
Your “ (0.6e-20 J at sea level, 0.45e-20 J at tropopause) than required to impart the v2 wobble into a CO2 molecule (1.3e-20 J) – means that the vast majority of these 10 mln/0.02s have LESS energy than that 1.3e-20 J – hence my “many slow”,
Once we clarified this – we have a next problem:
Admittedly, I don’t know how the transfer of energy during a molecular collision involving wobbling works – but if as you say a single collision event is enough to de-wobble, then my next question was – with 100 mln collisions between the wobbling and emission of the IR 0.02s later and EACH of the large majority of these 100 mln able to single-handedly de-wobble Co2(v2) thus preventing before the photon emission
then how it is that there is any emission of IR photons from Co2(v2) at all ?
(if after being de-wobbled, Co2 is hit again by a high energy N2, and wobbled again – it just restarts the clock toward emissions to 0.02s, the clock that will be stopped by the first slower collision after that)
I see only two solutions:
– either there is no emission of IR from the troposphere, be it out into stratosphere, or down to the Earth surface because all wobbled CO2 molecules de-wobble by the collision long before the 0.02s excitation time. But if so – where the 342 W/m2 “back-radiation” from air to Earth surface comes from???
– or, perhaps once wobbled – Co2 molecule excitation energy is no longer in the kinetic energy form – say if an electron in a Co2 molecule was bumped into a higher orbit it no longer counts toward kinetic energy of the Co2 and thus can’t be exchanged with slower N2 molecules during subsequent collisions ?
This would mean no dewobbling of Co2(vs) by collision of N2, and that all wobbled CO2 would emit IR photon – and after a series of absorption by CO2 and reemission – some of them would reach Earth’s surface – thus explaining our 342 W/m2 of back-radiation.
In Re to Piotr, 23 Jul 2026 at 6:57 PM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849973
Hello Piotr,
Have you, for some reason, excluded as a third option the suggestion mentioned above
– that the difference in rate coefficients for collisional excitation and deexcitation of a single CO2 molecule is exactly compensated by the opposite difference in population of the CO2 molecules in the ground vibrational state and in the excited vibrational state?
I still think that it can be the most reasonable explanation why the volumetric rates of excitation and deexcitation collisions in an isothermal air parcel are equal.
Greetings
Tomáš
Hello Piotr, Have you, for some reason, excluded as a third option the suggestion mentioned above
The “some reason” was that I wrote what _I_ see. Despite your post, I am not clear what you see – how your “exact compensation” would even work – how would it bypass the problem that I raised – that if there are 100,000,000 collisions during the excitation time (0.02s) AND according to MAR – even one collision would be enough to de-wobble the Co2 BEFORE it has a chance to emit an IR photon.
Again, remember that the vibrationally excited CO2 molecule is a quantum system. It gives up either ALL or the excitation energy or none of it. Also, remember that the “lifetime” of the excited state is an expectation–the decay is actually exponentially distributed. So, it is true that the kinetic de-excitation is far, far more probable than the radiative one, but that doesn’t mean radiative decay has zero probability.
– by collisions, I meant.
———-
“Recent Advances in Climate Change Research: Part IX – How Carbon Dioxide Emits IR Photons” – Lasse Amundsen, Martin Landrø
https://geoexpro.com/recent-advances-in-climate-change-research-part-ix-how-carbon-dioxide-emits-ir-photons/ (very detailed)
I’m asumming this (1.1 s) is the e-folding time scale for the exponential decay** that Ray Ladbury refers to above (Thanks!). (I remember seeing this or a similar value elsewhere.)
Ie. collisional de-excitation happens 50,000 times as frequently as de-excitation by (spontaneous, presumably) emission. The average rate at which photons are (spontaneously) emitted per excited molecule (in the higher-energy state of the transition) is not changed by the rate of collision*
(*aside from the effects of line broadening, which expose the molecule’s optical transition to different values of the Planck function B_ν; line broadening also exposes the molecule’s optical transition to different values of incident (ambient) spectral radiance L_ν; otherwise the rate absorption (direct absorption − stimulated emission), per lower-energy state molecule, would also be unchanged by the rate of collision).
PS this seems to assume that nearly all collisions with an excited molecule will accomplish collisional de-excitation, which I wonder about…
https://sciencedemonstrations.fas.harvard.edu/presentations/collisional-broadening :
So sometimes the collision forces the excited molecule to emit a photon. (I’m not sure if this is referring to the same type of phenomenon that I was thinking of, … unless …:) Suppose the “perturbation” functions as a measurement of the state of the molecule…
Last paragraph of my https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849537 , with subsequent corrections and clarifications (the original ≈works if the Planck function B_ν and incident (ambient) spectral radiance L_ν can be approx. as constant over a sufficient bandwidth for the amount of line broadening:
subsequent corrections and clarifications:
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849942 ,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849951 ,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849952
** if the initial state is a pure stationary state, I would have thought that the (unperturbed version of the) optical transitional would start out slowly, ramp as, peak (as the superposition of states is ~an even split, maximizing the amplitude of the beat frequency ∆ω wobble (see https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849941 ), then wind down to approach the final state. So I don’t understand that part of it. What if (since we’re not observing everything in full detail) the whole system is often in superposition and the individual molecules are rarely 100% in a pure stationary state (because we didn’t ask the molecules that collided with them)…(???)
But I did get the sense that it is as Ray Ladbury says from another source, which I will link to when I track it down.
Other links: https://home.strw.leidenuniv.nl/~emr/stralingsprocessen/les12.pdf
– by collisions, I meant.
———-
“Recent Advances in Climate Change Research: Part IX – How Carbon Dioxide Emits IR Photons” – Lasse Amundsen, Martin Landrø
https://geoexpro.com/recent-advances-in-climate-change-research-part-ix-how-carbon-dioxide-emits-ir-photons/ (very detailed)
I’m asumming this (1.1 s) is the e-folding time scale for the exponential decay** that Ray Ladbury refers to above (Thanks!). (I remember seeing this or a similar value elsewhere.)
Ie. collisional de-excitation happens 50,000 times as frequently as de-excitation by (spontaneous, presumably) emission. The average rate at which photons are (spontaneously) emitted per excited molecule (in the higher-energy state of the transition) is not changed by the rate of collision*
(*aside from the effects of line broadening, which expose the molecule’s optical transition to different values of the Planck function B_ν; line broadening also exposes the molecule’s optical transition to different values of incident (ambient) spectral radiance L_ν; otherwise the rate absorption (direct absorption − stimulated emission), per lower-energy state molecule, would also be unchanged by the rate of collision).
PS this seems to assume that nearly all collisions with an excited molecule will accomplish collisional de-excitation, which I wonder about…
https://sciencedemonstrations.fas.harvard.edu/presentations/collisional-broadening :
So sometimes the collision forces the excited molecule to emit a photon. (I’m not sure if this is referring to the same type of phenomenon that I was thinking of, … unless …:) Suppose the “perturbation” functions as a measurement of the state of the molecule…
Last paragraph of my https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849537 , with subsequent corrections and clarifications (the original ≈works if the Planck function B_ν and incident (ambient) spectral radiance L_ν can be approx. as constant over a sufficient bandwidth for the amount of line broadening:
subsequent corrections and clarifications:
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849942 ,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849951 ,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849952
** if the initial state is a pure stationary state, I would have thought that the (unperturbed version of the) optical transitional would start out slowly, ramp as, peak (as the superposition of states is ~an even split, maximizing the amplitude of the beat frequency ∆ω wobble (see https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849941 ), then wind down to approach the final state. So I don’t understand that part of it. What if (since we’re not observing everything in full detail) the whole system is often in superposition and the individual molecules are rarely 100% in a pure stationary state (because we didn’t ask the molecules that collided with them)…(???)
But I did get the sense that it is as Ray Ladbury says from another source, which I will link to when I track it down.
Other links: https://home.strw.leidenuniv.nl/~emr/stralingsprocessen/les12.pdf
If my most recent comment here was repeated rapidly three times, please delete two of them. I had a glitch on my end; sorry.
Also, I only posted this comment: https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849837
Because its near doppelganger hadn’t yet shown up even when comments I made after it were shown; the moment I posted, I saw the original there and realized there was no need, so https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849837 can also be deleted.
In Re to Piotr, 25 Jul 2026 at 5:23 PM ,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850041
Hello Piotr,
I think that the high frequency of collisional deexcitation of vibrationally excited CO2 molecules is exactly the reason why temperature and of air parcels and infrared radiation emitted therefrom instantly follow changes in the flux of incoming infrared radiation.
Have you also considered that in parallel with this extremely quick collisional deexcitation, the pool of vibrationally excited CO2 molecules in our isothermal air parcel is continuously replenished by collisional excitation, so that these excited molecules never disappear and keep emitting infrared radiation with an intensity perfectly matching the absorbed one, as if there were no collisional deexcitation at all?
Greetings
Tomáš
Ray Ladbury: remember that the vibrationally excited CO2 molecule is a quantum system. It gives up either ALL or the excitation energy or none of it.
If it is directed to me, my question was based already on the quantum system assumption.
Let’s call X = the amount of the energy needed for the Co2 be excited.
Y = the amount of net kinetic energy transferred by collision from a molecule N2 hitting that Co2.
If Y no excitation, right? But if Y > X. does it mean that X of that Y is used to “wobble” CO2 into the excited state, and the remaining Y-X increased the kinetic energy of that Co2 molecule?
If yes – does it mean that the X locked in now in that excited CO2 – no longer being a part of kinetic energy – CANNOT be removed by subsequent collisions, because no longer being a part of the kinetic energy of that CO2 i that X CANNOT exchange with colliding N2 molecules.
If this were the case ,then there would be NO collisional dewobbling of the wobbled CO2, and therefore ALL de-wobbling would happen through an emission of an IR photon.
Being a very opposite of MAR’s and your statements that it is the collisional dewobbling that is “far, far more probable” than dewobbling by IR-emission.
To your other point I’ll reply later (since that point is contingent on falsifying the “no collisional dewobbling” line of argument I presented above.
Tomas Kalisz: “Have you also considered that the pool of vibrationally excited CO2 molecules is continuously replenished by collisional excitation, so that these excited molecules never disappear”
Those are not the SAME excited molecules. Thus if one excited Co2 is deexcited, and then ANOTHER Co2 gets excited – the excitation-time countdown of the second molecule starts from zero. Hence the probability that it will last long enough won’t collide with any N2 or O2
So unless this NEWLY excited molecule manages to emits photon in time 100 MLN (sic) times shorter that the average excitation time (0.02s) – then just like it predecessor it had “replenished” – that the chances that it would send am IR photon – is practically nil.
So instead of your vision of the troposphere behaving as “ if there were no collisional deexcitation at all” – we would have a troposphere as an IR black hole – since practically all^* excited Co2 molecules would de-excite long before they had a chance to emit an IR photon
^* all except of those super-speedsters, who manage to emit photon in the 1/100 mln-th of the time average excited Co2 can.
Piotr,
The comment was not directed at anyone in particular–it is merely prudent to remember that wrt the vibrational state of CO2, we are dealing with a harmonic oscillator and hence quantum theory. The N2 molecule can transfer kinetic energy to the CO2 molecule. It can also deform the molecule, possibly distorting the energy spectrum of the quantum system (e.g. collisional broadening). The CO2 molecule can of course collide with any other molecule in its neighborhood and exchange momentum thereby. However, if a relaxation of the excited vibrational state is to occur, the N2 (or other) molecule must receive at least the vibrational energy of the CO2 molecule. It can receive more in kinetic energy.
Think about how laser cooling works–the laser is detuned just below the transition state energy of the gas being cooled. Now the gas molecules can still get excited, but only if they surrender a little bit of their kinetic energy in addition. The quantized energy of the excited state and the continuous kinetic energy are still energy and can exchange around the edges. But if a transition occurs the minimum energy lost by the formerly excited molecule is the quantize energy of the excited state.
An analogy: once all the pennies are gone, you could still wind up being owed 3 cents change, but you aren’t going to get it, because the coins are quantized with minimum size of a nickel.
re Tomáš Kalisz – this might be for you:
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849829 – https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849838
and re “ and keep emitting infrared radiation with an intensity perfectly matching the absorbed one” – that’s not generally expected to happen:
pure radiative equilibrium = PRE
&
radiative-convective-equilibrium = RCE
(after Jeevanjee & Fueglistaler 2020 https://journals.ametsoc.org/view/journals/atsc/77/2/jas-d-18-0352.1.xml )
RCdE = RCE with explicit recognition of the need for conductive-diffusive (d) at the sfc and maybe also cloud microphysics,
RCAdE = same with advection (A) over horizontal distances
Even in a 1-dimensional (1-D) column model with pure radiative equilibrium (PRE), with no direct solar heating of the atmosphere, such that the whole of the LW portion of the spectrum has a total of 0 , there can be bandwidth with net radiant cooling, there will generally still be bandwidth with net spectral cooling and bandwidth with net spectral heating. (see my https://www.realclimate.org/index.php/archives/2026/04/a-reflection-on-reflection/#comment-848619 ) Including direct solar heating of the atmosphere, and/or convective heating of the atmosphere (RCdE), there must be a total LW net radiant cooling in some layers.
Then going into the full 4+ dimensional [height (z or p or …), horizontal area, annual cycle, diurnal cycle, phases of internal variability modes/weather (cumulus cloud puffs, baroclinic eddies, QBO, ENSO, …)] climatological equilibrium (RCAdE) state (which encompasses a climate of internal variability and forced cycles (seasonality/etc.)), we have locations and times of imbalanced fluxes and net changes, which will follow a climate of forced and internally-generated patterns/cycles/textures/statistics associated with that climate.
Through all of that, and even during significant climate change, the LTE approximation should generally hold well (up to ~ 60 km) – even in sudden and violent climate changes (supervolcanos, asteroid impacts).
Ray Jul 26 “Also, remember that the “lifetime” of the excited state is an expectation–the decay is actually exponentially distributed. So, it is true that the kinetic de-excitation is far, far more probable than the radiative one, but that doesn’t mean radiative decay has zero probability.”
So is there enough of them to produce the 324W/m2 of the back radiation?
I used MAR numbers:
MAR: “a CO2(v2) requiring perhaps some 0.02 secs to relax while collisions occur on average every 0.0000000002 secs.” means 10 mln collisions during these 0.o2s”
with a correction that “0.o2s”/”0.0000000002s” is actually 100 mln collisions during 0.02s.
So even with the exponential distribution of the excitation times, given the avg, value of 0.02s, how many excited molecules of CO2 would there be in the far left corner of the distribution with the excitation time between 0 and 2*10^-10 sec. And more importantly – would their combined emissions add up to 324 W/m2 of the back-radiation?
Ps. Then again, patrick brought a source that states that collisions, at least at 3.5 km. seem to be some …100,000 less frequent than MAR’s number. Which if correct, might somewhat change the above calculations … ;-)
patrick 26 jul “The radiative lifetime of the (010) molecular vibration is about 1.1s (Cheo, 1971)I ’m asumming this (1.1 s) is the e-folding time scale for the exponential decay** that Ray Ladbury refers to above ”
If it the distribution of the probability of the emission of a photon was exponential, then we would have highest number of photons emitted in the first 20 μs interval ,and then progressively fewer emitted with consecutive 20 μs intervals. Yet both Cheo, and you,
treated the distribution as if it was …. linear, by dividing 1s by 20 μs as if the probability of a Co2 to emitting photon were the same in all 20 μs of subsequent intervals (p=2/100,000).
– Cheo: “ The typical collision time is 20 μs at 3.5 km. […] Statistically, the same CO2 molecule re-emits the photon energy 2 out of 100,000 times; but 99,998 times out of 100,000 the excited CO2 molecule is de-excited by collision.
”
– and you: “ . collisional de-excitation happens 50,000 times as frequently as de-excitation by [IR] emission.
in Re to Piotr, 27 Jul 2026 at 11:41 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850099
Hello Piotr,
For a process following first-order kinetics, such as e.g. radioactive decay, the lifetime τ of a state is defined as the time for its population to drop to 1/e of the initial value, so c(τ)=c0/e. Equivalently, τ=1/k, where k is the first-order decay rate constant for the process depopulating the state.
Following these definitions, I think that lifetime of a quantum state is a statistical quantity that may be hardly applicable for an ensemble as small as a single molecule.
In other words, if an isothermal air parcel having volume 1000 L comprises 18 mmol CO2, the steady population of vibrationally excited CO2 molecules at a chosen temperature is 1.8 micromol and their lifetime for IR emission is 0.02 s, I assume that the amount of IR photons that this parcel exchanges with its surrounding during 1 s is 90 micromol.
Greetings
Tomáš
Piotr: would their combined emissions add up to 324 W/m2 of the back-radiation?
BPL: If not, where would the rest be coming from? Surely a gas can only emit radiation through its greenhouse molecules, if it has any?
Ray: ” if a relaxation of the excited vibrational state is to occur, the N2 (or other) molecule must receive at least the vibrational energy of the CO2 molecule. It can receive more in kinetic energy.”
I have asked a a different question – whether the conversion of the kinetic energy of fast N2 into the potential energy of the excited vibrational state of Co2 is reversible :
i.e. whether the potential energy of the excited vibrational state of Co2 can be then converted back into kinetic energy imparted by that CO2 onto some N2 with which that Co2 collides next. OR whether the only way to relax the excited vibrational state is via sending of an IR photon.
Or in macroscopic analogy – you have a ball “1” with a spring system attached. You hit that ball with another ball ” 2″. Ball “2” passes Y amount of kinetic energy onto “1” – out of it
X amount of energy is absorbed into the spring (which locks in the compressed state,)
and remaining Z=Y-X adding to “2” kinetic energy. So which is correct:
a) once the spring has been compressed and locked in this position – it cannot be unlocked by subsequent collisions (thus these collision will trade only the unlocked, kinetic, part of ball “1” energy) OR
b) subsequent collisions do release the spring and as a result add the X amount of energy from the release of the spring to the kinetic energies being exchanged.
re my https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850060
“Recent Advances in Climate Change Research: Part IX – How Carbon Dioxide Emits IR Photons” – Lasse Amundsen, Martin Landrø
https://geoexpro.com/recent-advances-in-climate-change-research-part-ix-how-carbon-dioxide-emits-ir-photons/
me: “PS this seems to assume that nearly all collisions with an excited molecule will accomplish collisional de-excitation, which I wonder about…”
Oops, No! I read/skimmed too fast. A closer look indicates that they were specifically referring to the forward and reverse reactions going between vibrationally-excited CO2 CO2 (0,1,0) and vibrational-ground state CO2 (0,0,0) (along with the other colliding molecule).
(…see eqn. 2; notice how Maxwell-Boltzmann statistics fit in)
Actually, maybe not even all of those…
Ie. this is not all of the collisions, just those that accomplish the de-excitation (?through just the particular pathway considered?). Of course it makes sense that other collisions will happen with different effects (only rotational and translational changes, excitation to an even higher energy state, etc.) – though I’m surprised that it’s only ~ 1 out of ~100,000 that de-excite (via V-T) the CO2 (0,1,0) (I’ve been to a couple online calculators that give a collisional frequency more in line with ~100,000 times the ~ 1/(2E-5 s) = 50,000 /s .)
And I still may have missed some things so …
my “re Tomáš Kalisz” https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850105 2 sections fixed:
Even in a 1-dimensional (1-D) column model with pure radiative equilibrium (PRE), with no direct solar heating of the atmosphere, such that the whole of the LW portion of the spectrum has a total of 0 net radiant cooling or heating, there can/generally will be bandwidth with net radiant cooling and bandwidth with net spectral heating. […]
[…] we have locations and times of imbalanced fluxes and net changes, which will follow the forced and internally-generated patterns/cycles/textures/statistics associated with that climate.
my “re Tomáš Kalisz” https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850105 2 sections fixed: (2nd try; please replace 1st)
Even in a 1-dimensional (1-D) column model with pure radiative equilibrium (PRE), with no direct solar heating of the atmosphere, such that the whole of the LW portion of the spectrum has a total of 0 net radiant cooling or heating, there can/generally will be LW bandwidth with net spectral radiant cooling and LW bandwidth with net spectral radiant heating. […]
[…] we have locations and times of imbalanced fluxes and net changes, which will follow the forced and internally-generated patterns/cycles/textures/statistics associated with that climate.
So given what I’ve thus far read/seen, the picture I have is (*please pardon the personification) – when a CO2 molecule gets excited (to a given state eg. (0,1,0)), it immediately, spontaneously gets to work making a photon; uninterrupted, it is a master craftsperson, its photon a work of quality and precision. But when it gets perturbed, it gets annoyed and may often just chuck it and start over – or it may say to itself, “good enough” and emit it with whatever imperfections it has (and de-excite itself before the collision could, if it would ??). The probability (aside from what I’ll call the indirect effects of line broadening eg. changes in B_ν and L_ν (L_ν not mattering for spontaneous emission)) of what it decides to do depends on how far along it was in completing its work, such that thus the average rate of spontaneous emission per excited CO2 molecule (of a given state eg. (0,1,0)) is not changed (and similar logic would apply to absorption (direct absorption − stimulated emission) by CO2 molecules of a given state eg. (0,0,0)).
I could be wrong but it fits.
me @ https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850060 :
*and then the CO2 invites Peter Parker to go pick a peck of purple peppers.
Re Ray Ladbury: “ The N2 molecule can transfer kinetic energy to the CO2 molecule. It can also deform the molecule, possibly distorting the energy spectrum of the quantum system (e.g. collisional broadening).” The later sounds like what I’ve understood to be “quasi-static” pressure broadening, wherein the proximity to other matter, via its EM fields, changes the energy wells etc. of the atomic nuclei and electrons of the atom/molecule being considered, so its possible optical transitions will have different energies. Of course, in a gas, I suppose this would tend to occur most during collisions. But it’s different from the idea that shortening the duration of an emission or absorption necessarily increases the spread of frequencies in the photons. Then again I think one of my textbooks just said that the collisional pressure broadening (or did it just call it pressure broadening) was due to small amounts of energy being added or taken up…
Piotr: “would their combined emissions [of only these excited Co2 that manage to to emit IR before the first hit of N2 or O2 molecule – i.e. having relaxation time 0 and 2*10^-10 sec, in the distribution that has average relaxation time =0.02s) add up to [342] W/m2 of the back-radiation?”
BPL: If not, where would the rest be coming from?
From my 2nd possibility – that perhaps once Co2 gets “wobbled” into the excited state, it can’t be de-wobbled by subsequent collisions ( as the energy of the wobbled state, being potential, is not a subject of the exchanges of kinetic energy during collisions).
In such a case – 100% of the excited Co2 would relax by emitting IR photon – and their cumulative energy in downward direction that escaped absorption by Co2, would add to
the 342 W/m2 of the downward (“back”) radiation.
“4+ dimensional […] climatological equilibrium (RCAdE) state” maybe that should be RCAdSE (S = storage).
P: perhaps once Co2 gets “wobbled” into the excited state, it can’t be de-wobbled by subsequent collisions ( as the energy of the wobbled state, being potential, is not a subject of the exchanges of kinetic energy during collisions
BPL: If it’s in an excited state, then a subsequent collision would de-excite it. I must admit I don’t understand what you’re getting at.
P: perhaps once Co2 gets “wobbled” into the excited state, it can’t be de-wobbled by subsequent collisions ( as the energy of the wobbled state, being potential, is not a subject of the exchanges of kinetic energy during collisions
BPL: If it’s in an excited state, then a subsequent collision could de-excite it. I must admit I don’t understand what you’re getting at.
BPL: If it’s in an excited state, then a subsequent collision would de-excite it
I had just wondered whether there is a process by which the potential energy in the excited CO2 can be converted into kinetic energy.
But apparently it is and has its name: ” a vibrational-translational (V-T) transition: A molecule in an excited vibrational state bumps into another particle, dropping to a lower vibrational state while converting that extra energy into fast physical movement (translation) of the colliding bodies.”
re Piotr – well, it’s like you said (I think): the clock starts over; for those which remain in the same excited state, they start again. Over time periods much shorter than the e-folding time scale, the rate could be approximated as constant. Whatever the time between collisions, there’s some probability that the emission process will complete if the CO2 molecule is forced to choose a state. … I’ll think a bit more about it…
Is it the excited molecule that is emitting, or is it the transition that is emitting? If a tree falls in the middle of a forest…
Consider the states of the molecules in superposition; if the excited state is decaying exponentially …
(Otherwise, at what point does a molecule ever actually complete de-excitation, in the absence of collisions?)
…, the rate of the emission of the photon also decays exponentially, with the same e-folding time scale, so the rate of photon emission per unit population of the excited state remains constant (~ 0.91 s¯¹). And this should carry over to the probabilistic behavior that would emerge from measurements. Unless I messed up or missed something.
What that means for absorption …? or hot bands … (if a doubly-excited (correct term?) state is decaying exponentially into a 1st excited state and then that also decays at the same time…)?
PS AFAIK, all individual (fundamental?) microscopic processes are completely reversible. It’s an important part of thermodynamics.
in Re to patrick o twentyseven,
29 Jul 2026 at 11:59 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850179
and 29 Jul 2026 at 12:04 PM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850180 ,
to Ray Ladbury, 30 Jul 2026 at 6:42 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850205
and to Piotr,
28 Jul 2026 at 9:54 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850143
and 26 Jul 2026 at 6:16 PM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850067
Dear Sirs,
I think that Patrick offers the clue to the conundrum perceived by Piotr:
At least in case of molecular collisions in air, vibrational excitations and deexcitations of CO2 are indeed reversible. In other words, there is no “locked spring” in the excited CO2 molecule, and it indeed becomes deexcited by any further collision with a slow ambient particle. On the other hand (and for the same reason), there is a steady population of excited molecules that continuously replenish the infrared radiation absorbed by the molecules in their ground state.
I checked this view with Perplexity Pro that provided a detailed statistical thermodynamic analysis which also appears in accordance with remnants of the knowledge from my physical chemistry course some forty years ago. I do not see a reason for doubting that the obtained analysis is correct, however, if you do not trust me, you are invited to run another one and double-check yourselves. The engine summarized its output as follows:
“The key point is that collisional quenching does not make radiative emission impossible. It only means that a given vibrationally excited CO2 molecule is more likely to lose its energy by collision than by photon emission before it decays.
But in the atmosphere, the excited-state population is continuously replenished by collisions and by absorption of radiation, so emission occurs from a steady-state population maintained by all these processes together.
Therefore, the relevant quantity is not “Will one particular excitation survive long enough to emit?” but the steady-state balance between excitation, collisional de-excitation, spontaneous emission, and reabsorption. In that balance, even a small radiative branching ratio can still produce a substantial infrared flux because there are many molecules, many collisions, and continuous repopulation of the emitting states.
So the correct picture is not “collisions prevent emission,” but “collisions dominate the fate of individual excitations while also maintaining the population from which emission occurs.””
(end of the AI summary)
As regards Ray’s objection against potentially ambiguous terminology used by Piotr, I think that it was sufficiently clear that by “wobbled”, he meant “vibrationally excited”. Of course we should take into account that ground state of any vibrational mode has a non-zero energy even at zero absolute temperature and in this sense, a molecule cannot become “dewobbled” in sense of an “absence of vibration”, but I do not think that these fine details are relevant for the core of the dispute.
As regards stimulated emission mentioned by Patrick, the engine concluded:
“Stimulated emission is not an important energy-transport mechanism in Earth’s atmosphere because the vibrational populations are not inverted and the local infrared field is far too weak to produce net gain. In thermal equilibrium or near-equilibrium, absorption dominates over stimulated emission, while the actual atmospheric energy transport is governed by absorption, spontaneous emission, collisional redistribution, and radiative transfer. Stimulated emission would require laser-like conditions, which the atmosphere does not provide.”
Greetings
Tomáš
Breakthrough in Measuring Methane Emissions
Halfway thro’ July and the daily ERA5 SAT provided by ClimatePulse is showing a global SAT anomaly for the first half of July up on the previous 2026 monthly anomalies. This may well be simply due to it wobbling warm thro’ those two weeks.
Conversely, it does point to July 2026 being the first month of 2026 looking to be cooler than its 2023 equivalent. For the July 2026 anomaly to be warmer than July 2023, the last couple-of-weeks of the month would have to average above +0.82ºC. While not a million miles away from the anomaly for the first half of July (+0.62ºC), periods averaging above +0.82ºC have only been recorded Sept23-Feb24 (Sept23 was that “absolutely gobsmackingly bananas” month) and Oct24-Jan25. Julys are not a peak period for the annual cycle of AGW anomalies which would make a July fortnight averaging +0.82ºC pretty astonishing, maybe a small single “gobsmacking banana.”
So not sign in the SAT of the coming El Niño but, excepting 2023/24, previous El Niños haven’t really shown themselves in the SAT until August, even in the SH.
The ERA5 60N-60S SST at ClimatePulse has been showing a bit of what is presumably the coming El Niño. Since mid-June the 60-60SST has been running warmer than any previous year but not by much. (The 1991-2020 60-60SST trend was +0.15ºC/decade & 2026 is running roughly three-year’s-worth of that above 2023, although that is 2023 complete with its mounting “bananas”) The last couple of days perhaps show a little more of a gap starting to form.
Unlike SAT, the measured records (GISS, NOAA HadCRUT, BEST) which are hybrid SAT/SST are beginning to show signs of the coming El Niño with the NOAA NCEP forecast showing an 80% chance of it being “very strong**” by November.
(** Very Strong = RONI>+2.5. NINO3.4 which is used in RONI is rising faster than in the run-up to other El Niño. The latest weekly unadjusted NINO3.4=+2.0. The RNINO3.4 adjustment would put it about +1.5.)
We have the second named storm (Bertha) of the 2026 Atlantic hurricane season which formed West of Florida, now skirting the Louisiana coast to landfall proper in Texas by Friday. Bertha isn’t expected to reach hurricane strength but with June’s Storm Arthur should push the 2026 hurricane season’s ACE above ACE=1. (That’s not much for late July but it’s still early in the season. There are three years since 2000 which had ACE=0 by the end of July with two of them racking up ACE>100 by season’s end.)
The 2026 season is forecast to be below average activity. The latest Colorado State forecast is showing half the activity of its 2025 forecast (2026 – 9 named, 4 hurricanes, 1 major. 2025 – 16 named, 8 hurricanes, 3 major) and 2025 was forecast (and proved) an average year in terms of storm numbers (although the three of the four major storms grew to Category 5 which is more Cat5 than any season bar 2005.)
I consider a robust measure of the increasingly active Atlantic hurricane seasons is the proportion of seasons significantly above ACE=100 (ACE>120).
20 of the 31 Atlantic hurricane seasons since 1995 (65%) have been significantly above ACE=100 including an impressive 9 of the last 10 seasons. Given there were just 38 candidates for such active seasons in the previous record (1850-1994, so <26%), that is quite an increase.
So I'll be watching to see if 2026 manages to top the UniArizona forecast's ACE=106 (a lot higher than other forecasts) to add 2026 to the 'significantly above' ACE=100 pile.
But how, when accounting for the “Super” El Nino?
There is kind of an interesting and unusual dynamic going on, where the main area of TS generation (eastern Atlantic) is near normal to lower temperature than normal–as expected given El Nino. We also have dust blowing off of Africa suppressing cyclonic activity.
However, the Gulf of Mexico remains warm, and this seems to be pushing the most likely origins for TS to the west. Bertha fits the mold.
EarthClimate,
The usual description for El Niño is “strong” (RONI>2.0) or “very strong” (RONI>2.5) and the forecasts point to a “very strong”(82%) or a “strong”(15%) El Niño.
And El Niño years do result in more wind-sheer over the Atlantic which can decapitate tropical storms and prevent their formation. Past El Niño years do show low Atlantic hurricane activity but not always.
1997 … ACE=41, … RONI(May)=+1.0, … RONI(max)=+2.4
2009 … ACE=53, …. RONI(May)=-0.1, … RONI(max)=+1.6
2015 … ACE=58, …. RONI(May)=-0.1, … RONI(max)=+2.4
2023 … ACE=146, … RONI(May)=+0.1, … RONI(max)=+1.5
2026 … ACE=???, … RONI(May)=+0.5, … RONI(max)=, … RONI(max)=+1.5″>+2.2(forecast)
Thank you both for the insightful response.
China’s electricity generation capacity hits 4.04 billion kilowatts by end-June, topping the world: NEA – Global Times, “an increase of 10.8 percent year-on-year”.
https://www.globaltimes.cn/page/202607/1366545.shtml
That’s one small step for China, one giant leap for mass extinction.
“one giant leap for mass extinction”
Nonsense. From the article:
“Of this total, solar power generation capacity accounted for 1.27 billion kilowatts, up by 15.8 percent year-on-year, while wind power generation capacity reached 680 million kilowatts, up 18.5 percent year-on-year …
“… the cumulative average utilization of power generation equipment nationwide was 1,392 hours, a decrease of 113 hours compared to the same period last year.
“Nuclear power generation capacity in the first half of the year reached 66.14 million kilowatts, rising 8.6 percent year-on-year.
“The strong growth in new energy generation will continue to contribute the drive of China’s green transition …”
China is leading the world in the decarbonization of electricity generation.
SA : China is leading the world in the decarbonization of electricity generation.
And that, was not the point being made about the mass extinction connection. Decarbonization or not is besides the point now, irrelevant.
K. Maru: “ Decarbonization or not is besides the point now, irrelevant .”
The only way for “ decarbonization or not ” to be ” beside the point, irrelevant to mass extinctions would be if the mass extinction were driven by …. the amount of electricity produced per se, and NOT by the amount of GHGs emitted.
Furthermore, the increase in the electricity demand in China is a result of using that electricity to …. decarbonize the major source of GHGs – transport: between 51-55% of all new passenger cars in China are EVs or plug-in hybrids. Another part of China increases in the electricity demand – is China helping OTHER countries to decarbonize – China provides 80% of global photovoltaic, and is by far, the largest exporter of electric vehicles.
And cumulative emissions of Co2 – corresponding more or less to % of extra Co2 in the atmosphere that is driving the change of climate – the US, UK, and EU contribution is 2.5 times that of China. But please do point your accusing finger at the one country that leads and helps to drive the global decarbonization efforts.
But Doomers can’t help themselves – they wrapped their ego in their prediction of the Doom so tightly, that they would lash out against who makes the Doom less likely/ less dramatic,
and therefore undercut what they have lived for: the ability to scowl and say: “ I have been warning you about the Doom for many years, but you never listened“.
Maru: “ Decarbonization or not is besides the point now, irrelevant .”
The only way for “ decarbonization or not ” to be ” beside the point, irrelevant to mass extinctions would be if the mass extinction were driven by …. the amount of electricity produced per se, and NOT by the amount of GHGs emitted.
Furthermore, the increase in the electricity demand in China is a result of using that electricity to …. decarbonize the major source of GHGs – transport: between 51-55% of all new passenger cars in China are EVs or plug-in hybrids. Another part of China increases in the electricity demand – is China helping OTHER countries to decarbonize – China provides 80% of global photovoltaic, and is by far, the largest exporter of electric vehicles.
And cumulative emissions of Co2 – corresponding more or less to % of extra Co2 in the atmosphere that is driving the change of climate – the US, UK, and EU contribution is 2.5 times that of China. But please do point your accusing finger at the one country that leads and helps to drive the global decarbonization efforts.
But Doomers can’t help themselves – they wrapped their ego in their prediction of the Doom so tightly, that they would lash out against who makes the Doom less likely/ less dramatic,
and therefore undercut what they have lived for: the ability to scowl and say: “ I have been warning you about the Doom for many years, but you never listened“.
KM said above thread: “China’s electricity generation capacity hits 4.04 billion kilowatts by end-June, topping the world: NEA – Global Times, “an increase of 10.8 percent year-on-year”. That’s one small step for China, one giant leap for mass extinction.”
SA responded by disagreeing and pointing out China is building a lot of renewable energy. (Which is correct)
KM responds: “And that, was not the point being made about the mass extinction connection. Decarbonization or not is besides the point now, irrelevant.”
Nigelj: KMs original point wasn’t clear. He said nothing about HOW and WHY he thought Chinas high and ever increasing use of energy would lead to mass extinction. Amazingly he still doesn’t clearly say what his point is. Secular Animist made the interpretation of his comment that he meant Chinas increasing energy output consisted of building a whole lot more coal power and ignoring renewables. Thus = extinction. A sensible interpretation.
I suspect KMs real point is Chinas increasing energy output, (regardless of what energy source is used fossil fuels, or renewables or nuclear), also creates environmental problems for example the environmental footprint of mining and is depleting finite resources, etc,etc so will lead to mass extinction. But I only suspect this because I have noticed he has made that point before, using other names, and he has been highly critical of economic growth and high levels of energy use (of any type) and critical of attempts to solve the climate problem with renewables. His preferred solution appears to be degrowth on a massive scale. But really, how are people who maybe don’t read a lot of the comments supposed to know all that?
And climate solutions based heavily on degrowth, simplification, abandoning capitalism for some sort of common ownership, etc,etc do not look realistic. How do you do that in the next couple of decades? How would you convince billions of people to essentially dramatically cut their consumption and live like poor people? And face a massively high risk of being unemployed as demand is sucked out of the economic system? How do you convince people to adopt another version of socialism at a grand scale, when it failed in the USSR and China ? (Im not talking narrowly focused public ownership of some key services, which is ok). Would any of those things make life better? It makes building wind turbines and solar farms or even nuclear power look like a walk in the park.
Nigel: “ [K. Maru’s] original point wasn’t clear. He said nothing about HOW and WHY he thought Chinas high and ever increasing use of energy would lead to mass extinction. Amazingly he still doesn’t clearly say what his point is. ”
yeah, it is not a bug, it is a feature. Much like the vaunted by Trump’s toadies Trump’s “deliberate ambiguity ” that is supposed to use mixed signals and intentional vagueness to keep adversaries off-balance – while in reality – instead of Trump playing “5-D chess”, Trump just throws spaghetti against the wall to see if anything sticks (Note that “spaghetti” is an euphemism here … ;-) )
That’s why when challenged – the “K. Maru” dismissed Secular Animist with “ that, was not the point being made”, “besides the point” , “irrelevant “, WITHOUT saying what completely OTHER point it was making. Like Trump, like Doomer Multi-troll.
no, it isn’t.
the share of renewable electricity in many other countries much higher.
Willis Eschenbach has successfully use Ned Nikolov-style tricks in getting a paper published.. I have a PubPeer response here: https://pubpeer.com/publications/9705D8058ED4686D92253056ED7F23
The rhetorical pattern here is similar to the way Nikolov applies the ideal gas law to “derive” planetary temperatures. In Nikolov’s case, a toy closure on =/() is tuned via the density term to compensate for pressure and temperature, creating the illusion of a universal relation that bypasses radiative transfer and vertical structure. In Willis’s paper, a two‑zone or 2-box “Constructal” closure is tuned on CERES‑derived albedo and greenhouse factors, plus a conductance and ocean absorption, to reproduce bulk heat flow, hot-zone T, cold-zone T, and areas creating the illusion that a new physical law has been discovered. In both cases, the construction is a tautological re‑expression of bookkeeping under heavy aggregation, not a replacement for the actual mechanisms that determines the temperature.
I am pretty certain that Willis used an LLM with this paper. What’s scary is that an LLM can easily construct an apparently deep model by layering known equivalences but in the end is nothing more than a tautology, like saying x=x. Lots of time wasted as Willis keeps conflating structural equivalencies with what he considers novel findings.
Anyone can review on PubPeer so have at it if we want to see Willis’s paper retracted like Nikolov’s was.
[Response: I agree that there is nothing much to it. But his climate sensitivity calculation forces all feedbacks to be zero (since there is no change in atmospheric absoption or albedo allowed), and thus gives (as all such calculations do) the no-feedback value. We went over this issue ages ago: https://www.realclimate.org/index.php/archives/2007/04/learning-from-a-simple-model/ – gavin]
JCM 18 Jul 2026 at 5:11 PM (No “Reply” button. Hartmann has a Power of 88 w/m**2 of latent heat energy from the surface providing that much energy into the troposphere. In what form(s) does that energy leave the troposphere?
Hi Barry,
To relate back to your interests, for the whole atmosphere the radiative divergence is the difference between what leaves the atmosphere by radiation and what it absorbs.
Using Trenberth style values:
Atmospheric radiative heating provided is about 360 W/m2 from surface absorbed LW, and 80 W/m2 solar absorbed in atmosphere, providing total radiative heating ~ 440
Atmospheric radiative cooling provided to space is about 200 W/m2, and to surface ~340 W/m2, providing a total 540
Therefore atmosphere experiences approximately 540 – 440 = 100 W/m2 radiative divergence, or a net radiative cooling
So atmosphere is always losing about 100 W/m2 by radiation, it emits more than it absorbs. You could interpret the radiative loses as creating demand for upward energy transport by fluxes of latent and sensible heat.
A more clear picture is to understand it through external sources and sinks.
In this way we compute the throughput of atmosphere, and ask how much energy must pass through the atmospheric column to bridge the external radiative heating of the surface by the sun to the exhaust pipe exporting energy to space. There is no closed loop that creates or maintains streamflow independently or additional-to those imposed by external boundaries.
F_throughput = OLR – Ra, which is the outgoing radiative emission from the planet minus atmospheric solar absorption, which should be approximately 120 W/m2 based on the numbers provided above.
This throughput represents the stream of energy through the entire column, or the transfer of energy required to carry surface solar input to the level where it can be radiated to space. In this way, the required throughput is imposed by the external energy budget, not internal exchanges.
In conceptual format: the steady state stock of internal energy is distinct from the balancing flow through, just as with any stock and flow reservoir system. The required balancing streamflow is partitioned approximately between non-radiative transport 83% (~100 units out of 120), and 17% by radiative diffusion (~20 units). This way it’s easier to see how internal exchanges don’t create an additional current.
https://link.springer.com/article/10.1007/bf00210625 The global heat balance: heat transports in the atmosphere and ocean
https://journals.ametsoc.org/view/journals/atsc/78/7/JAS-D-20-0290.1.xml Using “Heat Tagging” to Understand the Remote Influence of Atmospheric Diabatic Heating through Long-Range Transport
E. Schaffer 18 Jul 2026 at 1:47 PM (No “Reply” button) has “I think you are somewhat struggling with the GHE itself”. Nope. I’ve understood the GHE perfectly since July 2018 when I pondered it. This is my pithiest description of the dozen descriptions I’ve assembled (a couple previously posted on RealClimate UV a few years back).
This is a simplified Absolutely-Entirely-Correct explanation of the so-called “greenhouse effect (GHE)” in Earth’s troposphere. Some Power flux of the vast quantity of photons that are emitted (manufactured) by molecules in Earth’s troposphere leak out of its top & bottom into or through the stratosphere and into the surface respectively. Emission (manufacturing) is proportional to Kelvin**4.
With approximations for example to clarify, 194 w/m**2 of upwelling radiation leaks out of the top and suppose that is part of what is emitted (manufactured) by molecules over an altitude range of 4 to 12 km with an effective average of 7 km. Also, 345 w/m**2 of downwelling radiation (incorrectly being called “back radiation”) leaks out the bottom and suppose that is part of what is emitted (manufactured) by molecules over an altitude range of 0.1 m to 1 km with an effective average of 500 m.
Suppose a certain amount of CO2 was added into the troposphere and mixed then *instantly* the upper & lower ranges would need to be closer to their respective ends because there are more CO2 molecules in the way. So the effective averages of 7 km for upwelling into or through the stratosphere and 500 m for downwelling into the surface change to 7.05 km and 480m respectively, with the upwelling & downwelling Power fluxes consequently changing to 193 w/m**2 and 346 w/m**2 respectively due to being produced by higher-than-before (colder) and lower-than-before (warmer) air parcels.
There has been no change in the 539 w/m**2 leaving the troposphere, there is no “magical extra energy”, the surface downwelling radiation has increased by 1 w/m**2, which is an additional ~480 terawatts (~95%) heating the ocean and Earth is emitting 1 w/m**2 less radiation to Outer Space (a Power reduction of 510 terawatts).
This is NOT the GHE, I stick to my judgement. you try to put a lot of different concepts, most of them wrong, somehow together and that’s not working.
Actually the GHE is pretty simple. It is the difference between surface- and emission temperature due GH-agents elevating the emission altitude, and the lapse rate, nothing else.
https://en.wikipedia.org/wiki/Greenhouse_effect#/media/File:Greenhouse_Effect_Overview.svg
Re E. Schaffer – while physics shapes the convective lapse rate in a manner independent of the GHE, the GHE has a strong effect on where such a lapse rate is established by actual convection and also the actual convective flux of heat (as multiple people pointed out to you above somewhere (JCM, John Pollack, …MA Rodger? (I remember disagreeing with a related point MA Rodger was making but I forgot what that was exactly). In particular, the GHE has to warm up the sfc some amount first before/in order for PRE (pure radiative equilibrium) to become unstable to convection.
Syukuro Manabe, Robert F. Strickler, “Thermal Equilibrium of the Atmosphere with a Convective Adjustment” (1964)
https://journals.ametsoc.org/view/journals/atsc/21/4/1520-0469_1964_021_0361_teotaw_2_0_co_2.xml?tab_body=pdf – fig 4, p.370: In going from PRE to RCE(dry adiabatic) to RCE(6.5 K/km), the surface cools while the tropopause rises and warms (if the surface didn’t cool, the OLR would increase due to the upper tropospheric warming). – This is (I think) with GHG concentrations held constant (discussion p.369 sec. 4).
While the whole of the global 4-D-state troposphere is overturning and so could be considered to have a convective lapse rate, there are places and times where it is stable to local convection – most obviously: nocturnal, frontal, and polar inversions, but this is not limited to inversions outright. So the GHE – or maybe I should say LW opacity – can actually affect the tropospheric lapse rate at least a little (as opposed to the simplest 1-D model equilibrium where the lapse rate adjusts only to prevent any layer from being super-adiabatic –
and even there the potential for kinetic energy to do work to force the tropopause higher (ie creating a layer which pulls heat downward by mechanically-forced convective) is set aside AFAIK, though I’m not sure that would be significant (well, there’re overshooting tops of thunderstorms – but I think what I had in mind might happen more so with the planetary boundary layer thermals but actually I don’t know about that; I’m not familiar with it but It’s an interesting possibility…) – but more generally, there is the work output of the heat engine to consider, and some of that gets out of the troposphere and drives currents and the Brewer-Dobson circulation, although I wonder about the part in the winter stratosphere because could it actually be thermally direct until the SSW stage(?)
I feel like I got something wrong in this…
Anyway, yes, most OLR is coming from below the tropopause so the GHE being modulated by
the(?) a convective lapse rate does seem to apply(?*) (?maybe not exactly?) for the climate as it is now (and probably generally has been?), but in this case the relative contributions of the LR(H2O as latent heat carrier) and H2O vapor (as GHG)) feedbacks have been given (as best as they can evaluate them, applied to anthropogenic warming)And yet, most interestingly, there is dissent between the static parameters with the given quantity of WV, where its cooling effect roughly equals (or exceeds once we add resolution) its warming side, so that it must be a net cooling agent, and its role as feedback where this should be inverted.
Sure I could explain the solution of said dissent, which I happen to know, but that is only a harder pill to swollow. I think the insight that WV is rather a net cooling agent should be exiting enough for the moment.
ES: I think the insight that WV is rather a net cooling agent should be exiting [sic] enough for the moment.
BPL: Clouds are a net cooling agent. Water vapor is a greenhouse gas.
…, the GHE/LW opacity and solar heating have a strong effect on where such a lapse rate is established by actual convection and also the actual convective flux of heat …
“There has been no change in the 539 w/m**2 leaving the troposphere,” – that’s true if B_ν is linear over τ through the troposphere and τ is evenly increased (ie profile is stretched by a constant factor over height), or if deviations from that cancel each other just so. But more generally, the changes in the portions F↓ and F↑ which are emitted within the from troposphere would not sum to 0. (see my https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850156 ) I think there may be a general tendency for B_ν(τ) to have convex curvature through the troposphere over a large fraction of bandwidth…(?) (this would imply
Also you earlier (June?) once said something about the effect of inversions that partly works for sufficiently opaque inversions if you specify that increasing opacity reduces the net F↓ through that layer (noting that near the boundaries of that layer, you’ll be able to see across into other layers and their B_ν values will change the F↑,↓ there, so it’s not necessarily a ↓ net F throughout the whole inversion, but anyway, the implied radiative forcing of reducing a net F↓ has a cooling effect below, but where exactly is this cooling distributed? If the stratosphere is sufficiently opaque for a portion of it to have a ↓ net F, due to an abundance of a WMGHG, then the troposphere generally should be quite opaque… (although the ↓ net F could cross the tropopause eg. ( https://www.realclimate.org/index.php/archives/2026/04/a-reflection-on-reflection/#comment-847393 … ).
But also consider the effective band widening effect (https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-849896 or simpler: https://www.realclimate.org/index.php/archives/2023/05/cmip6-not-so-sudden-stratospheric-cooling/#comment-812156 ); adding a little opacity to a nearly transparent layer will have a rather different effect… (granted there’s typically a significant H2O τ even in clear skies in the atmospheric window)…
Okay, I got a bit lost in the details there, sorry (and I was focused on IRF; of course going from IRF to SARF transfers (propagates/spreads**) some of the stratospheric cooling to the tropopause-level radiative forcing (total forced warming of all that’s below):
Starting from near transparency (in the part of the spectrum being considered), given a general* T decline from the sfc to TOA, adding LW opacity (increasing the concentration (eg. relative to mass of air) of absorption cross sectional area; k_{a,air} = ∑_i n_i σ_{a,i} ÷ ρ_{air} ) generally* reduces OLR and generally* reduces the net F↑ (= F↑ − F↓ ; nearer the sfc this will depend more on increasing F↓; close to TOA this will depend more on decreasing F↑) As opacity increases, there will be some uppermost layer for which IRF is a cooling effect because of the reduced flow of LW heat from below and increased concentration of emission capability (also increases capability of absorption, but @ TOA there is the sharp dropoff to ~ 0 K … etc.). OLR emitted from some upper layer increases but the decrease in the contribution from lower is greater, so OLR in total (in the part of the spectrum being considered) still decreases (generally*).
With an upper level inversion (up to TOA), greater T near TOA should enhance the cooling. OLR emitted from the warmer layer next to TOA should increase more, but the decrease in the contribution from lower is still greater, until the opacity of the inversion layer gets large enough so that the EWF(F↑-weighted) for OLR (= integral over hemisphere of directional EWF · cos(ϑ); EWF per unit mass of air = k_{a,air} · exp[−τ_{vc} / cos(ϑ)] ? did that quickly; did I get it right?; I just realized I goofed it up recently in another comment…) is sufficiently dominated by that upper level inversion.
see also https://www.realclimate.org/index.php/archives/2026/04/a-reflection-on-reflection/#comment-847393
** some of this (edited) from prior comments ( https://www.realclimate.org/index.php/archives/2023/05/cmip6-not-so-sudden-stratospheric-cooling/#comment-811772 ):
equilibrium temperature response won’t generally match instantaneous forcing (or even SARF) exactly [aside from (even without H2O, cloud, snow ice, etc. feedbacks)] – of course convective fluxes can respond (eg. convective adjustment in troposphere in a 1-D model, hence the tendency for the surface and troposphere to warm up together, aside from the lapse rate feedback’s effect), but also because when a layer warms, it’s abs. cross sections brighten, which will have a warming effect on any layer which would absorb some of that radiance; likewise a cooling layer can have a cooling effect on other layers.
also ongoing convection & advection, as well as changes in those, will redistribute the T response
see also https://www.realclimate.org/index.php/archives/2024/01/unforced-variations-jan-2024/#comment-818215
Formatting fixed:
…
With an upper level inversion (up to TOA), greater T near TOA should enhance the cooling. OLR emitted from the warmer layer next to TOA should increase more, but the decrease in the contribution from lower is still greater, until the opacity of the inversion layer gets large enough so that the EWF(F↑-weighted) for OLR …
(= integral over hemisphere of directional EWF · cos(ϑ); EWF per unit mass of air = k_{a,air} · exp[−τ_{vc} / cos(ϑ)] ? did that quickly; did I get it right?; I just realized I goofed it up recently in another comment…)
… is sufficiently dominated by that upper level inversion.
…
Formatting fixed (2nd try):
…
With an upper level inversion (up to TOA), greater T near TOA should enhance the cooling. OLR emitted from the warmer layer next to TOA should increase more, but the decrease in the contribution from lower is still greater, until the opacity of the inversion layer gets large enough so that the EWF(F↑-weighted) for OLR …
(= integral over hemisphere of directional EWF · cos(ϑ); EWF per unit mass of air = k_{a,air} · exp[−τ_{vc} / cos(ϑ)] ? did that quickly; did I get it right?; I just realized I goofed it up recently in another comment…)
… is sufficiently dominated by that upper level inversion.
…
Where τ, τ_{vc} is measured going away from POV; ϑ is angle from vertical up or down:
EWF per unit τ (directional) = −∂/∂τ exp(−τ) = exp(−τ)
(directional) EWF per unit τ_{vc}
= −∂/∂τ_{vc} exp[−τ_{vc} / cos(ϑ)]
= (1/cos(ϑ)) · exp[−τ_{vc} / cos(ϑ)]
∂τ_{vc}/∂[vertical mass path kg/m²] = k_{a,air}(τ_{vc})
(directional) EWF per unit mass of air
= k_{a,air}(τ_{vc}) · (1/cos(ϑ)) · exp[−τ_{vc} / cos(ϑ)]
EWF(F↑-weighted, to be multiplied by B_ν) per unit mass of air
= 2π ∫_0^{π/2} k_{a,air} · [sin(ϑ)·cos(ϑ)/cos(ϑ)] · exp[−τ_{vc} / cos(ϑ)] dϑ ??
= 2π ∫_0^{π/2} k_{a,air} · sin(ϑ) · exp[−τ_{vc} / cos(ϑ)] dϑ ?? Is this right? I should try just doing this for a F and then extract the B from that and see what’s left over…
For an optically very-thin layer (∆τ_{vc} ⟨⟨ 1), its internal distribution of B_ν over its τ_{vc} won’t matter much to the spectral L↓ at/through its base and L↑ at/through its top (except in directions close to horizontal), and generally*** to the spectral F↓ and F↑ at those boundaries, respectively (which don’t depend strongly on variations in L values sufficiently close to horizontal). This is because the cross-sectional area it contains is mostly visible from the boundaries of the layer (except in directions close to horizontal); they don’t get in each other’s way much, ie. don’t hide each other. So it’s the B_ν averaged over its τ_{vc} that ~≈/tends to determine it’s effect on whatever is outside the layer (except for L in directions close to horizontal); glossing over some potential complications**, if layer-averaged B_ν is colder/warmer than what’s behind it, the spectral (L,F***) ↑/↓ decreases/increases) going from its bottom/top to its top/bottom (referring to the difference in values between the two boundaries (∆L,∆F***) ↑/↓ .
**except (∆L)in directions close to horizontal; noting the L entering the layer cold be dimmer(colder) than the than the layer’s average B_ν in some directions and brighter(warmer) in others, in which case it’s hard to make a simple general statement about the sign of ∆F.
(***?only general tendency for F?)
E. Schaffer 18 Jul 2026 at 1:47 PM “I think you are somewhat struggling with the GHE itself” and also “So yes, this latent heat DOES cool the planet a lot, easily by an 80W/m2. It ain’t that complicated”. What the! Since it “ain’t that complicated” please explain how “latent heat DOES cool the planet” Earth and give the quantity of Earth’s ocean that condenses in Outer Space at least 100 km above the ocean surface. Thanks.
As in the depiction of the GHE above, the lapse rate is crucial for the GHE. And the latent heat does reduce it..
“E. Schaffer” typed “the lapse rate is crucial for the GHE. And the latent heat does reduce it.” Absolutely! Finally the “E. Schaffer” quietly drops its wildly-incorrect, hugely-overestimated 86.4W/m2 of “latent heat” cooling in its 6 Jul 2026 at 5:45 PM “some 86.4W/m2 of “latent heat” cooling (as in the NASA Earth Energy Budget), then WV was barely warming at all, or rather net cooling”. Excellent.
I have for consideration the pretty-good approximation of 88 (surface evaporation, mostly tropical ocean) minus 54 (returned to surface as LWIR, the tropospheric GHE being 38.2% of total LWIR manufacturing leakage at its top and 61.8% of total LWIR manufacturing leakage at its bottom) minus 7 (returned to surface as sensible heat) = 27 w/m**2 of Net water evaporative surface cooling, for comparison with whatever is the total GHE assigned to H2O gas by the scientists.
I am afraid you must not subtract any such numbers from latent heat..
I’m done with this one folks (and bloviator “JCM” for this topic). It has 86.4 w/m**2 of energy surface -> troposphere and then 100% of that PARTICULAR energy either goes to Outer Space or disappears up the tail end of the Magical Sky Unicorn.
presently the world sits at around 1.3/1.4C above the base and warming might have accelerated to 0.3C per decade up from 0.2C due to aerosols, low level cloud formation and earths albedo absorbing more sunlight etc. So within 30 years we could be over 2C permanently and this brings in large scale climate impacts that would well impact us severely.
I looks at the remedies to our vast fossil fuel usage and just see a very slow decline at best. Technologies that dont help so much are desalination, 22.000 plants globally at present and growing, Air conditioning which again is growing at a pace but means which every energy technology is used it needs a lot of it.
The technologies helping us get off of fossil fuels are battery storage for the grid, HGV, Buses, cars and some shipping, wind and solar, geo thermal and all the renewables but CCS isnt really on the cards and nothing big is working and deployment is unlikely.
Shipping and flying isnt as yet a big option for alternative energy and these industries are growing allowing for goods and people to be shipped all over the world.
So if people had to lay their cards on the table here at RC: what would you say is the mostly likely scenario for climate change temps and impacts. I am suggesting a minimum of 2.5C personally as we have alternative technologies but politics, economics and other factors mean they cant be deployed fast enough to offset warming for another 30/40 years
We have a problem – we can’t explain the CERES albedo increase.
At least if the result of the following paper is true:
A Reconciled Satellite Record Reveals a Negative Low Cloud Feedback Over the Past 47 Years – Cesana – 2026 – Geophysical Research Letters – Wiley Online Library
https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2026GL124158
They say, that low cloud cover increased and the radiation power balance feedback is negative, i.e. low clouds did have a cooling effect in the past. They did not mention aerosol effects on cloud cover, though, which is a big omission.
So, if this is true, where does CERES albedo decrease come from??
Dominik: Lenne: We have a problem – we can’t explain the CERES albedo increase.
Among possibilities I see – used different satellites, different time-scale, perhaps different area of the Earth looked at. And. If as you say- they ignored the change in aerosols over time – this would confounded their results even further – although differently early in the 47-year period, when the aerosols there were likely on the increase and differently in the later part of that period when they were dropping.
And on top of that, they may have had a different way to calculate cloud feedback
– in their abstract they are perplexed themselves:
” Counterintuitively, models that best capture observed LCC climatology and trends exhibit the strongest positive abrupt-4×CO2 feedback and highest climate sensitivity. ”
when based on their results – it should have been the opposite.
in Re to Dominik Lenné, 28 Jul 2026 at 5:13 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850131
Dear Sir,
It appears that the controversial results pertain to low cloud feedback in tropical region; I have not grasped if the authors somehow generalized them to the entire planet, so that it could be justified to conclude that “low cloud cover increased and the radiation power balance feedback is negative”.
Could you check their results to be sure?
Furthermore, it appears that the results are derived indirectly from a set of “cloud controlling factors” (CCF). This procedure seems to be quite complicated and thus potentially prone to bias if any of these parameters were estimated incorrectly.
Best regards
Tomáš
Throwing out a quick thought because it’s my 1-Trick Pony does that Paper have enough regional detail to indicate that it is or isn’t this seen at 21:38 to 23:20 at https://www.youtube.com/watch?v=agKayS6h6xA colour-coded pictorial of GMST anomaly 1976-98 to 1999-2012 and wind explanation, Pacific Ocean eastern 2/3rds cooled over the 18 years while almost everywhere else except the Southern Ocean warmed.
2014 Quote: “Atlantic warming turbocharges Pacific trade winds Date: August 3, 2014 Source: University of New South Wales. New research has found rapid warming of the Atlantic Ocean, likely caused by global warming, has turbocharged Pacific Equatorial trade winds. Currently the winds are at a level never before seen on observed records, which extend back to the 1860s. The increase in these winds has caused eastern tropical Pacific cooling…”.
Is it possible that cooling over Pacific Ocean eastern 2/3rds (I calculated ~13% of Earth’s area for my assessment of adjustment for GMST acceleration) increased its low-cloud deck so much that, combined with much land area surface warming, it gave a situation of Global average surface temperature increasing as Global average cloud cover decreased, owing to low cloud cover decreasing over land with the surface warming but overwhelming Pacific Ocean eastern 2/3rds increasing with the cooling ocean surface?
However, the above wouldn’t resolve anything if it’s simply that separate analyses of low-cloud reflectivity change give opposing results.
I’ll repeat and edited view of this next month, but for now things that make me curious:
There is the A) Populist Maga-Like Conspiracist Forum and then there’s the B) RealClimate / Academic-Scientific Educated Cohort
Are they so different?
A – Refuses to look because they distrust the source (government, experts, “elites”).
B – Refuses to look because they trust the source so completely that independent verification feels unnecessary or even heretical.
A – Sees a line of text, but interprets it through a lens of conspiracy (“they’re hiding something”).
B – Sees the same line of text, but interprets it through a lens of settled authority (“this is settled, why are you even asking?”).
A – Dominated by tribal identity—any question is an attack on “us.”
B – Dominated by institutional identity—any question is an attack on “science” itself.
A – Bias = overt cynicism.
B – Bias = overt credentialism.
A – Result: utterly incapable of parsing plain language that contradicts their worldview.
B – Result: utterly incapable of parsing plain language that contradicts their worldview.
It’s the same disease, different hosts.
The common denominator is identity-protective cognition. The brain doesn’t process information neutrally—it processes it through a filter of “what keeps me safe in my tribe?”
For the Maga-illiterate forum-goer, safety means not being a “sheep.”
For the RealClimate educated commenter, safety means not being a “denier.”
Both identities are so fragile that a single piece of contradictory plain text feels like an existential threat.
But the rank refusal to even look at the material on offer in the first place? That’s the real kicker. It’s not that they look and misinterpret it — it’s that they actively avoid looking, because deep down maybe they know that seeing the raw data, follow the link and honestly review it, or the original document might force them to hold two opposing thoughts at once.
And that cognitive dissonance is too painful.
So they outsource their thinking. The populist outsources to a conspiracy laden YouTuber or the resident bully.
The academic outsources to the IPCC summary or some other “authority”, or the resident bully
Yet both call it their “research.” Neither reads the actual damn text in front of them, work out what it means, or what it is pointing to.
The saddest part?
You can’t reason someone out of a position they didn’t reason themselves into. You can only point it out
KM: …
A – Result: utterly incapable of parsing plain language that contradicts their worldview.
B – Result: utterly incapable of parsing plain language that contradicts their worldview.
MS:
A: Changes its mind when the populist clowns at the top change.
B: Changes its mind when the science at the bottom changes.
“B: Changes its mind when the science at the bottom changes”
Some say so, some say so. I am doing a little thing here over WV, and I never asked questions I did not know the answer to already. It is more an inquiry into how it was possible to get things so badly wrong. And as expected it appears to be to the strict rejection of anything contradicting the narrative, strictly believe driven, strictly anti-science.
Perception :: Reality
Do feel free to submit your ideas to any decent peer-reviewed journal. We’ll discuss them when they are published. How ’bout that?
Sure – would you cover the expenses?
If your ideas had merit, you could easily find a co-author with institutional support. By saying you won’t publish unless someone else pays, you are admitting that your ideas aren’t worth support.
ES, you can get a waiver from most journals if you can’t afford the page charges.
MS, I prefer another approach.
Minority Group C – The outsider looking in?
Sees both A and B clearly, because they’re not emotionally invested in either tribe. Reads the actual text. Follows the links. Asks “what does this actually say?” rather than “which team does this help?”
Result: accused by A of being a secret B, and by B of being a secret A. Which only proves the point.
Minority group D. The doomers. They belong to the doomer tribe so they cant see things objectively. They are convinced society will soon collapse and that capitalism is doomed. They are convinced renewables cant work. They are convinced we will run out of resources and in the near future. They are convinced climate change will be at the very high end of the spectrum and are impervious to evidence otherwise. There’s a pattern there neatly sumarised by the term “doomers”.
KM: Minority Group C – The outsider looking in?
MS: I think that is multiplying entities beyond necessity because it uses the fact you have defined A and B so that they are clearly intellectually equivalent — A and B make the same mistake from different ends of the belief spectrum — but that creates the impression that A and B are the same percentage of the population. I think that is wrong. I am in set B.
B – Refuses to look because they trust the source so completely that independent verification feels unnecessary or even heretical.
Well, what is heretical when I argue with an A is that A refuses to use the scientific method to validate and invalidate their argument. I know I can find the science to verify my argument, but it really should be unnecessary for me to do that independent verification because A should have already done it as I have. Then A would have long ago changed his mind.
As a B, I refuse to look because I have already looked. Suppose you are an A, and we have an argument today about the increasing GHE. I say it is increasing CO2 from burning fossil fuels; you say it is a 3560 year cycle of the orbits of Jovian planets, and you require me to go do the independent verification. I do that, and I come back and say, “Nope, it really is anthropogenic CO2.”
The next day, we have the same argument and again you tell me I am afraid to do the independent verification. Do I have to go do it again? How many times do I have to independently verify that it’s anthropogenic CO2 all the way down?
Maybe now you will say, sure, that’s an easy one, but what about this one, or what about that one? and your what-about-isms are admittedly less scientifically certain. But I will say, and rightly so, I think, that there is a scientific consensus position based on peer-reviewed science, and that is the one I have chosen in each case.
As a B, I don’t defend my beliefs; I argue for the scientific consensus position, as long as it remains the consensus position. So you can say yes, but the science might have changed since the last time you verified it, and now the consensus position is about to change. Great, then as a B, I will change my mind. But note that if you argue that way, you aren’t really an A, because an A would never have done the verification you used to prove me wrong.
As a B, I have confidence because I know my beliefs are based on science. As a B, I can even say I know my beliefs are wrong but they are always getting closer to truth as science progresses. As an A, you can’t say any of that. As an A, you don’t have the confidence that allows you to change your mind because your beliefs aren’t based on anything but the declarations of authority figures.
When you define A and B as you have, I think they are both A, and then C becomes what B should have been to start with.
LOL
If you say so Martin. You’ve made some very solid points here, though I do suspect you could have had a little ChatGPT assistance, undeclared. You sneaky thing you. :-)
KM: I do suspect you could have had a little ChatGPT assistance, undeclared.
MS: I take that as a compliment, but no, the comment was all me. I always attribute text that comes from AI.
Kobayashi Maru,
You fail to mention C) The Skyrocketeers.
Are they so different?
C – Refuses to look because they distrust the sources (govrnment, experts, “elites”)
I would add that those you style (B) do “look” and don’t use “settled authority” to dismiss what they “see” but use the ‘scientific method.’ Such ‘method’ does not respond publicly to every proclamation of a flat-earth theory. And without itself employing the ‘scientific method,’ a flat-earth theory won’t be treated as participating in the ‘scientific method,’ although it may get a more cursory examination which may involve a level of prejudice.
The point I would make is that both (A) & (C) do not attempt to engage properly with the ‘scientific method’ which is why they get such short shrift.
Or category D, egotistical sky rockety multi identity trolls, who cant look because their heads are so far up their own backsides.
Is this you trying to convince me that you are the poster boy for applying the “‘scientific method’ “?
Oh please, build a bridge and get over yourself.
Kobayashi Maru,
You appear to be imploring me to “build a bridge.”
In this circumstance that could prove difficult.
You should know that bridges require firm foundations, usually on both sides of the divide to be bridged. I see ‘firm foundations’ for any argument based on what you term “the source” (by which I assume you mean the science generally presented in IPCC WG1 ARs) because that ‘side’ is presented within the scientific process. What I don’t see is any such ‘firm foundation’ for the other ‘side’ which, as far as both you and ‘(C) skyrocketry’ are concerned, generally remains swamp-like.
You have described your own position saying you “completely defer to Hansen and his team for accuracy” yet still manage to dismiss the views of others in your (A) & (B) hypothesis up-thread by saying “You can’t reason someone out of a position they didn’t reason themselves into.”
With such contradictory positions from you, can your position be anything other than ‘as clear as mud’ and entirely swamp-like?
It appears it is not my ‘bridge-building’ we are missing – it is the drainage ditches on your side of this divide you want bridged.
MAR, you are talking to KM who is likely the same guy who you had a discussion with on Hansens aerosols ideas except at that time he was using another internet name, possibly prieto principle.
Your point related to Hansens claims that the 2023 to 2024 temperature peak was a result of reductions in shipping aerosols, and your core argument was this is problematic because the shipping routes don’t match the areas of warming ocean very well. While I quite like Hansen, this seemed a reasonably convincing point to me. I think you provided some maps. Even if Hansen is ultimately proven right, its a fair observation that at least needs an explanation.
Multi trolls response was a huge series of attacks on you and your post, that studiously avoided any mention of the core argument you raised. You cant build bridges with people like that. They have found their leader in Hansen and wont tolerate any criticism of anything he says. Its a cult thing, a tribal thing. He hasn’t got the chops or desire to engage on the specifics and be prepared to back down from his position. He will just spray you with rhetorical s**t basically. Hes done it to me as well.
Nigelj (no Reply button). Why no mention of Matthew H. England et al, shoaling of the northern hemisphere oceans caused by reduced winds that apparently was tracked from region to region, matching the SST increase, caused by the El Nino if memory serves. “Cause of Extreme North Atlantic Warming in 2023 with Matthew England” at https://www.youtube.com/watch?v=39T7bW7KA18 by “Climate Chat” (Dan Miller) on Jul 17, 2025
Nigelj,
You are likely correct as to the true identity of this particular ‘skyrocketeer’. I was myself aware of the situation but am minded not to presume when a fresh pseudonym appears in the RC comment threads. And with this particular ‘skyrocketeer’ we should also acknowledge the contributions of some anonymous AI bot which is providing significant assistance with his word salads.
Barry E Finch,
The paper behind your YouTube is England et al (2025) ‘Drivers of the extreme North Atlantic marine heatwave during 2023’ (co-authored by one of our hosts) which rather passed my by. The citations of England et al (2025) would be worth a read to see what others are saying about the paper’s findings which run:-
Note the line “Furthermore, solar radiation anomalies made regional-scale warming contributions in locations that approximately correspond to some of the region’s main shipping lanes, suggesting that reduced sulfate emissions could also have played a localized role” which is pointing to a role for the marine aerosol reductions (but only a subsidiary one) which could assist in explaining why 2020 aerosol reductions don’t show a climatic impact until 2023. (And this is all ‘N Atlantic’ when the 2020 aerosols were a bigger event in the N Pacific.)
Perhaps discussing where the burden of proof lies would clarify things a bit.
Kobayashi Maru, you should read 23 Years on Fire by Joel Shepherd. It’s the third in a SF series. In this book Shepherd ‘introduces’ the concept of Compulsive Narrative Syndrome. It is pretty much exactly what you write in this comment.
Great comment, BTW.
Thomas Fuller: “Kobayashi Maru, Compulsive Narrative Syndrome. It is pretty much exactly what you write in this comment.”
Good catch, Thomas:
“ invents or latches onto a tidy story (narrative) to make sense of reality [and] ignore any real-world data or evidence that does not fit their chosen story.
fits “Kobayashi Maru”, to a dot.
Thanks Tom. That concept is real from what I’ve seen in my life. .
K. Maru: “ There is the
A) Populist Maga-Like Conspiracist Forum and then there’s the
B) RealClimate / Academic-Scientific Educated Cohort
Are they so different?
More tellingly:
There are A) Deniers (General: Ken Towe, K.Woollard, Mr. KA, Victor or of the Anything but Carbon sub-type: A. Schaffer, JCM)
and there is B) a Doomer Multi-troll, currently using :Kobayashi Maru” handle
Are they so different?
A) No data nor argument can possibly change their mind
B) No data nor argument can possibly change his mind
A) The inability to change their views is tightly linked to their validation ego: if everybody else is wrong, and I, a lay person, can see what scientists can’t/are not willing to, see, then I must be really really smart/virtuous.
B) The inability to change his views is tightly linked to his validation ego: if everybody else is wrong, and I, a lay person, can see what scientists can’t/are not willing to, see, then I must be really really smart/virtuous.
A) They attack and try to discredit the mainstream climate science as corrupt/ideological for not telling them things they want to hear.
B) He attacks and tries to discredit the mainstream climate science as corrupt/ideological for not telling him things he wants to hear.
A) They cherry-pick the data and authors to appear to support their claims and discredit climate science
B) He cherry-picks the data and authors to appear to support his claims and discredit climate science
A) They sow confusion and promote distrust to science, leading to weakening or abandoning efforts to reduce the GHG emissions (“there is no problem”: “CO2 is good for us”, “it’s anything but the GHGs, stupid”)
B) He sows confusion and promotes distrust to science, leading to weakening or abandoning efforts to reduce the GHG emissions (“too late to do anything” with the implication that “it’s all or nothing” so it does not matter if we level off the emissions or ramped them out – we are doomed anyway)
A) They are either paid agents of influence, or (more likely) “useful idiots” of fossil industrial complex in general, and Russia and Saudi Arabia in particular, with the latter’s economy, ability to finance wars on others enrich its oligarchs and suppress their own society – DEPENDS on the world continuing to buy their oil and gas.
B) He is either paid agent of influence, or (more likely) a “useful idiot” of fossil industrial complex in general, and Russia and Saudi Arabia in particular, with the latter’s economy, ability to finance wars on others enrich its oligarchs and suppress their own society – DEPENDS on the world continuing to buy their oil and gas.
“Are they so different?”
=====
P.S. Right, there is a difference : the throughput – a single Multi-troll, masking it by operating under 100(s) of different names, has churned up over the many years so many thousands(?) of posts – with volume probably larger than all the RC Deniers taken together.
in Re to Piotr, 29 Jul 2026 at 11:44 AM,
https://www.realclimate.org/index.php/archives/2026/07/unforced-variations-july-2026/#comment-850178
Hello Piotr,
I agree that “Kobayashi Maru” sounds exactly like the infamous multitroll, and that it is most likely a new disguise thereof.
I do not think, however, that comparing JCM with “E. Schaffer” is appropriate. JCM does not try to invent an alternative atmospheric physics like “E. Schaffer”. Quite oppositely, I would say that his understanding to climate science is very good. From my point of view, his comments are substantive, informative and valuable and I would say that in this respect, he ranks, along with MA Rodger and John Pollack, to the best among regular Real Climate commenters.
I do not think that his criticism to widespread approach treating the role of terrestrial hydrological regimes in Earth climate regulation as “mere feedback” does entitle you to label him as a “climate science denier”.
Greetings
Tomáš
TK : I do not think that his (jcm) criticism to widespread approach treating the role of terrestrial hydrological regimes in Earth climate regulation as “mere feedback” does entitle you to label him as a “climate science denier”.
Maybe not. But maybe if you add his absurd and repeated use of the term trace gas, his frequent dismissal of climate modelling and his suggestions that scientists have ignored the water cycle does make him a denialist. If it looks like a duck, walks like a duck and quacks like a duck and keeps on doing that its probably a duck. He certainly MAKES himself look like a denialist, and that doesnt seem too clever if he isnt.
He could write one simple short paragraph clarifying his position but hasn’t to my knowledge. Like multi troll hes evasive. JCM is just a bit more knowledgeable and smarter. So he gets shoved in the denier category, rightly or wrongly, but he only has himself to blame.
My, but you’ve spent an amazing amount of time working out what other people must be thinking without ever really engaging or communicating with them at all! Impressive.
How unfortunate you have done so without ever actually looking at the purposes of products produced by the “authorities” you decry. Had you done so, you might have saved yourself a lot of effort and the rest of the world a lot of aggravation. But then you would have deprived yourself of that feeling of righteous indignation that you seem to so crave.
I’ll say it again. There is no “authority” for climate science. There is the evidence and the literature. The IPCC sets out to summarize that literature in a way that policy makers can digest. That is inherently a consensus building activity, because different policy makers have different experts who they trust. Bringing as many on board on the policy/politics side while not losing critical players on the science side is a balancing act. The resulting product is bound to be conservative.
In contrast to that conservative approach, we have Hansen et al. asking, “OK, just how far could that conservative position be?” These researchers are hardly bombthrowers. They are not exaggerating the potential risks–just looking at them at a different confidence level. This is precisely the sort of analysis that is needed by engineers and others trying to mitigate the risks.
There is no antagonism. There is no authority vs. opposition conflict. There is just the process of science playing itself out on a subject that is of critical public importance.
And then there are the deniers–who for reasons of lack of understanding or lack of scruples find it appropriate to try to undercut the scientific process, while rolling around on a bed of oily money with their politician friends.
I don’t fit in A or B for any of the 5 random stereotypes listed by the Socialite KM (the same entity that directly contradicted itself here on the fundamental essence of the “greenhouse effect” while falsely claiming to have even the foggiest clue about this simple physics).
BEF: I don’t fit in A or B for any of the 5 random stereotypes listed by the Socialite KM
BPL: Socialite? You mean he shows up at fancy parties given by the rich?
BPL: Socialite? You mean he shows up at fancy parties given by the rich?
MK: Yes! And wearing very large Gay Party Hats and High Heeled Boots.
I’m the talk of the town in all the nicest ways lol
OK, I think that some of the discussion of the radiative characteristics of CO2 has become too imprecise. I do not know what is meant by a “wobbly” CO2 molecule. I do not know what is meant by de-wobbling a wobbly CO2 molecule. These are not precise terms, and they could mean different things to different readers. In particular, if some are getting the idea that the CO2 molecule can lose it’s “wobbly energy” through multiple collisions rather than a single collision, then they are simply mixing quantum mechanics with classical descriptions and coming up with a nonsense slurry.
I do know what is meant by the excitation of the V2 bending mode of CO2. I know that this is a harmonic oscillation that has a given expected energy. I know what is meant by the transition (radiative or collisional) from the ground state to this excited state and by the transition (radiation or collisional) from the excited state to the ground state. I can work out the relative frequency of these transitions given the densities of CO2 and other molecules in the atmosphere, the pressure, the temperature… That is how I know that the radiative decays at any given altitude will fit nicely into the blackbody spectrum for an object at that temperature. I know there will be an near-equilibrium ratio of the radiative to collisional decay rates. I know that the energy imparted by a CO2 molecule relaxing from the V2 bending mode to an N2 (or O2) molecule will then be shared with other N2 and O2 molecules as the accelerated N2 molecule thermalizes after its initial collision with the CO2 molecule.
It is important to preserve this lack of ambiguity, because there are many factors that can introduce ambiguity to the description. For instance, collisions between a CO2 molecule and other molecules can alter the energies of the eigenstates SLIGHTLY. The V2 bending mode is still there, just with slightly different energies, and it is still a quantum system–either excited or not. If we don’t start with the unambiguous understanding of the system, the additional complications will swamp us with difficulties, and all will be lost.
“collisions between a CO2 molecule and other molecules can alter the energies of the eigenstates SLIGHTLY. The V2 bending mode is still there, just with slightly different energies”, Could you please clarify whether you refer to what is shown at https://www.youtube.com/watch?v=XwpjyqTOtuM (but that is described as plus rotational energies “filling in” rather than varying vibrational energies).
I just watched it again and hear that I misremembered from a couple years ago. Michel shows vibrational plus rotational lines initially and then states vibrational can be modified by collisions as per Ray comment.
Today, the last day of the month, and with a few days still to arrive, the ERA5 Re-Analysis at ClimatePulse is showing a July global SAT anomaly of +0.70ºC, a significant rise on previous 2026 monthly anomalies (Jan-Jun run 0.51ºC, +0.54ºC, +0.53ºC, +0.52ºC, +0.55ºC, +0.56ºC). Also a cool Antarctic through much of early July has prevented a larger rise above these earlier months.
Relative to earlier years, July 2026 is roughly matching the anomalies of July 2023 (+0.72ºC) & July 2024 (+0.68ºC) which sat as the warmest Julys on record by some margin. Even when the 2010-22 trend of +0.30ºC/decade underlying AGW is accounted-for, these years 2023, 2024 & 2026 again sit 1st, 2nd & 3rd and still well-above the throng.
SAT July anomalies 2010-26 & ranking** (& adjusted for AGW trend)
2026 ,,, +0.70ºC … 2nd ,, (+0.70ºC … 3rd)
2025 ,,, +0.45ºC … 4th ,,, (+0.48ºC … 14th)
2024 ,,, +0.68ºC … 3rd … (+0.74ºC … 2nd)
2023 ,,, +0.72ºC … 1st … (+0.81ºC … 1st)
2022 ,,, +0.38ºC … 6th ,,, (+0.50ºC … 11th)
2021 ,,, +0.33ºC … 8th ,,, (+0.48ºC … 13th)
2020 ,,, +0.33ºC … 9th ,,, (+0.51ºC … 10th)
2019 ,,, +0.40ºC … 5th ,,, (+0.61ºC … 5th)
2018 ,,, +0.28ºC … 10th ,,, (+0.52ºC … 9th)
2017 ,,, +0.27ºC … 11th ,,, (+0.54ºC … 7th)
2016 ,,, +0.36ºC … 7th ,,, (+0.66ºC … 4th)
2015 ,,, +0.15ºC … 12th ,,, (+0.48ºC … 12th)
2014 ,,, +0.07ºC … 14th ,,, (+0.43ºC … 16th)
2013 ,,,. -0.02ºC … 17th ,,, (+0.37ºC … 17st)
2012 ,,, +0.04ºC … 16th ,,, (+0.46ºC … 15th)
2011 ,,, +0.14ºC … 13th ,,, (+0.59ºC … 6th)
2010 ,,, +0.05ºC … 15th ,,, (+0.53ºC … 8th)
(** 2010-14 July unadjusted anomalies are not always higher than all pre-2010 July anomalies, years which include 2009, 1998, 2005, 2002, 2007, 2006. On the full record, the rankings would run 2014 17th, =23th, 19th, =13th, 2010 18th.)
The ERA5 60N-60S SST daily anomalies at ClimatePulse has been showing some vigorous warming through July. Recording highest anomaly for time-of-year since mid-June, the margin above 2023/24 has widened since mid-July. These warming SSTs will be reflected in the conventional measured SAT/SST records.
A CarbonBrief analysis by Zeke Hausfather suggests a strongest-ever El Niño is very-likely (91%) with NINO3.4 anomalies peaking at +3.96ºC (latest weekly NINO3.4 = +2.2ºC). This peak NINO3.4 = +3.96ºC feels a bit high given the modelling shown in the NCEP.NOAA weekly ENSO Update which is presently shown peaking at RNINO3.4 = +2.2ºC. (RNINO3.4 is sitting some 0.5ºC below present NINO3.4 values.)
Yet this same CarbonBrief analysis considers the chances of 2026 becoming warmest-year-on-record is not so certain (35%) and may depend on the particular temperature record used. (Odds of a 2026 record given thus:- BEST 66%, GISTEMP 65%, HadCRUT 35%, NOAA 24%, & SATs ERA5 13%, JRA-3 9%.)
There is also a Hansen et al communique ‘Beware Media Hype. It’s Not the El Nino!’ It talks of the ERA5 60-60SST anomaly and “the gap between the two El Nino lead-in years (2026 and 2023) … there is perhaps a hint of a decrease in the gap (Fig. 9)” but this was certainly a cynical “grasping for straws” moment and since the last data show on that Fig 9 (20th July), the later data at ClimatePulse (29th July) shows the gap doubled.
The August temperature numbers will thus be interesting, the developing situation graphed out at The Banana!!! Watch.
RE but this was certainly a cynical “grasping for straws” moment
No it wasn’t. But it does expose the ongoing incompetence to be found in your comments. In my view both you and Zeke et al are way out of your depth and do not know what to say anymore.
Everyone appears to have it down as strong due to the July numbers.
“Quite oppositely” is an uncommon phrasing; people usually say “quite the opposite”, “on the contrary”, or “conversely” to mean that the real truth is completely different from what was just stated.
Native speakers in LOTE would be wise to double check there understanding of what other people are saying in English. Being a non-native English speaker myself I do speak with some experience in the matter. Alien languages are tough and the consequence rough if you get it wrong.
I am not the grammar or typo police either. Call me Harry Helpful :-)