By Andy May and Philip Mulholland
Earlier this year Philip Mulholland published the details of his Dew-Point Anchor Hypothesis or “DPAH” (Mulholland P. , 2026a). The hypothesis assumes that the Dew-Point Lifting Condensation Level (LCL), as calculated from surface conditions in the tropics, can act as an anchor for the lower and middle troposphere. This idea moves the independent variable in climate modeling from the radiative balance at the top of the atmosphere (TOA), to the cloud level inside the troposphere.
The traditional TOA anchor oversimplifies the climate since it essentially assumes that nothing that happens in the atmosphere and oceans matters. It is often claimed that atmospheric and ocean processes just “move heat around” and the only thing that really matters is radiation-in minus radiation-out at the top of the atmosphere (IPCC, 2021, Ch. 7, section 7.2) & (Held & Soden, 2000). However, this is clearly incorrect, since Earth’s oceans and atmosphere store heat for varying lengths of time as shown by the AMO and other ocean or climate oscillations. These oscillations are not simply short-term random variability as claimed by the IPCC (IPCC, 2021, Ch. 10). The AMO for example is a 60-70-year climate oscillation that correlates well with global mean surface temperature (May & Crok, 2024). Moving the climatic anchor into the climate system brings the climate oscillations into the accounting, an important conceptual shift.
The DPAH Model
The DPAH hypothesis is closely related to the moist-adiabatic theory discussed in a previous post. Both rely on the concept that rising moist air cools, causing water vapor in the air to condense, releasing latent heat that energizes the air and delays natural cooling. The speed of the rising air is closely related to surface temperature, and the LCL is related to the dewpoint and air temperature lapse rate. In essence, the dewpoint depression (dpd) is used as a proxy for the moist enthalpy of the surface atmospheric layer. Moist enthalpy is conserved for rising air parcels below the freezing point, which can be a problem as discussed later in this post.
The DPAH and the moist adiabatic theory work in the tropics, but the higher latitudes are a different environment and depend much more on advective heat carried into them from the lower latitudes by winds and ocean currents (Feldl & Merlis, 2023) and (Feldl et al., 2026). Here we will only discuss the tropics and subtropics.
The LCL height is essentially the lower cloud base. It is closely related to surface temperature and near-surface humidity. That is, the aforementioned dewpoint depression or dpd, which equals T or temperature – Td or the dewpoint temperature.
For a given relative humidity, higher surface temperatures increase the saturation vapor pressure, allowing a lifted air parcel to rise farther before reaching saturation and thus producing a higher LCL. Cooler surface temperatures result in a lower LCL. In the real tropical atmosphere, warmer surfaces are usually accompanied by higher absolute humidity (higher dewpoints), which partially offsets the temperature effect, but the net relationship still holds, and warmer surfaces tend to support higher cloud bases on average, while cooler surfaces produce lower cloud bases.
The DPAH treats the LCL (or an effective dewpoint anchor derived from surface conditions) as setting the lower boundary condition that influences the entire convective column, at least to 250 hPa (~10.6 km). This is why surface temperature (T0) and dewpoint depression (dpd) emerge as key control parameters in the Markov model.
We turned Mulholland’s scoping analysis (2026b) into a model and, at first, ran it on its own with the results shown in figure 1.
Other than using optimized initial surface values, the modeled (solid) curves for the air ascent and descent in figure 1 use no observations. The observations (dashed lines) are 10 hPa binned averages for all IGRA2 radiosondes that passed a basic QC check between 5°S and 10°N (Ascent or ITCZ) and 30°S to 20°S or 25°N to 35°N (Descent or subtropics). ITCZ stands for the Intertropical Convergence Zone or the deep tropical region where clouds and thunderstorms are nearly perpetual. The lapse rates (vertical bars in the right graph of figure 1) are emergent from the model and match observations to ~250 hPa reasonably well, but not perfectly.
In figure 1 the parameters used are identified in Table 1.
Initial surface temp (Ascent, K) | Surface temp (Descent, K) | Lapse Rate (Ascent K/km) | Lapse Rate (Descent, K/km) | Dew Point Depression (Ascent, K) | Dew Point Depression (Descent, K) |
299.98 | 297.54 | 6.18 | 6.06 | 5.37 | 4.00 |
Table 1. Optimized initial parameters for the Dew-Point Anchor Hypothesis model.
The global mean IGRA2 observed values within the ITCZ, and subtropical regions used in this study fall very close to the modeled values which is encouraging.
Nudging the model with the observations
The next step was to optimize the model with the observations and use the optimized model to help choose the best initial parameters. This was done with the R function “nlopter” and its “NLOPT_LN_BOBYQA” algorithm. It is a local derivative‑free algorithm. It refines the initial guess within physically realistic bounds but does not attempt a global search of the full parameter space. Figure 2 shows the results.
Once optimized with the global mean observations the fit is visually good. The R2 for the variables modeled versus observations, after optimization, are shown in Table 2.
R2 Temperature | R2 Dew Point Depression | R2 Lapse Rate |
99.85% | 99.89% | 79.10% |
Table 2. The R2 for the modeled values versus observations up to 250 hPa.
The spikes in the lapse rate curves in figure 2 are probably due to sample size differences by height in the IGRA2 data since they fall on the required measurement heights for weather balloons. The “nlopter” optimizer function was used to derive the starting parameters shown in table 1, that were used to create figure 1.
This is a local optimization that adjusts only the six surface control parameters (T₀, ELR, and dewpoint depression for ascent and descent). The Markov transition matrices are not tuned directly; they are generated deterministically from these parameters at each iteration.
All optimized parameters correspond to physically meaningful surface quantities. The optimization is constrained to realistic ranges, ensuring that the resulting parameters remain physically interpretable.
Methods and Limitations
This analysis is based on regional mean radiosonde observations from 2015-2025 between fixed latitude boundaries, one representing the ITCZ (the Ascent) and another representing the subtropics (the Descent). Thus, it is not precise, since each location is different and the ITCZ moves progressively from month to month. However, it does show that surface temperature, lapse rate, and dewpoint depression very closely predict the vertical tropospheric profile to 250 hPa, at least on average, in both the deep tropics and the subtropics (Mulholland P. , 2026c).
The optimization was only run up to 250 hPa due to the divergence of the residuals starting about there, as shown in figure 3 for air temperature. It is possible that the divergence near 250 hPa is due to the supercooled water droplet limit of -40C, that occurs around this pressure, and above which water vapor freezes directly into ice (Atkinson et al., 2016). Some of these tiny ice crystals will form cirrus clouds and some are ice crystals dropping out of the cirrus clouds as “virga” or falling streaks of ice-crystal precipitation that later sublimate back into water vapor at a lower altitude thereby absorbing latent heat and cooling the surrounding air (Fueglistaler et al., 2009) & (Jensen et al., 2013). This process typically occurs between 6 and 18 km.
The residuals plotted in figure 3 show the model breaks down near 250 hPa. On average the height is about 10.6 km. At that altitude the ITCZ temperature averages -41°C, the relative humidity 42%, and the average specific humidity or “q” is 0.18 g/kg. In the subtropics, the average mean temperature is -42°C, the relative humidity is 35%, and the specific humidity is 0.15 g/kg. So, they have significant differences, but not the dramatic differences we might expect.
The dewpoint residuals are shown in figure 4.
As with the temperature residuals, the dewpoint residuals also show that the model breaks away from observations at around 10 km or 250 hPa. We find this interesting because this is also the altitude where the AR6 climate models show their maximum divergence from observations (McKitrick & Christy, 2020).
Physical Interpretation of the ~250 hPa Limit
The optimization was deliberately limited to 250 hPa because residuals for temperature, dewpoint, and lapse rate begin to grow systematically above this level. This is not a failure of the model but rather an indication of its domain of applicability.
Below ~250 hPa, latent heat release from vapor-to-liquid condensation dominates the energetics, and the two-regime (ascent/descent) Markov process anchored by the dewpoint depression reproduces the observed profiles with high fidelity. Around –40 °C (typically near 250–300 hPa in the tropics), supercooled liquid water droplets become rare; homogeneous freezing occurs rapidly. Above this isotherm, the dominant processes shift to ice-phase microphysics: ice crystal formation, cirrus cloud development, and virga (falling ice streaks that sublimate in drier air below). This is a phase-transition boundary and a thermodynamic constraint, see more about the molar density intersection below the tropopause in (May, 2025).
Sublimation of these ice crystals absorbs latent heat, cools the surrounding air, and can increase local density. This provides a stabilizing mechanism that helps maintain the tropopause temperature inversion and the sharp change in lapse rate into the near-isothermal stratosphere. The presence of cirrus ice also affects radiation — the clouds reflect incoming solar radiation while trapping outgoing longwave radiation (Mitchell, 2002) & (Yang et al., 2013).
Because the current DPAH implementation does not explicitly model the liquid-to-ice transition or the radiative and microphysical effects of cirrus and virga, divergence above ~250 hPa (≈ 10.6 km) is physically expected. This transition layer marks the upper boundary of the convectively controlled, latent-heat-dominated regime that DPAH is designed to capture. It also coincides with the altitude where CMIP5 and CMIP6 models show their largest divergence from radiosonde observations, suggesting that inadequate treatment of ice-phase processes and their associated feedbacks may contribute to known tropical tropospheric biases in comprehensive climate models.
The IPCC Models
The AR5 and AR6 tropical models are illustrated in figure 5. AR6 discusses this mismatch on page 444 of the WG1 report. In figure 5 it is apparent that the maximum difference between the models and the observations occurs between 150 hPa and 250 hPa in both the AR5 CMIP5 model iteration and the AR6 CMIP6 iteration.
AR6 claims that over half the difference between observations and models in the tropical troposphere can be explained by the models overestimating sea surface temperatures (SSTs) in the region, but this simply raises the question: “Why do the models overestimate tropical SST?” Figure 5 compares AR6 models with forced SSTs (in blue, where model SSTs are forced to match observations) and unforced SSTs (red). As shown here, erroneous SSTs directly affect moist static energy, convection depth, and upper tropospheric warming rates. For a more complete discussion of this problem, the so-called “model hot-spot error” see here.
The IPCC also cites researchers who blame the mismatch on overestimating the sensitivity of tropospheric temperature to CO2 (McKitrick & Christy, 2020) & (Po-Chedley et al., 2022). Mauritsen and Stevens used model evidence to hypothesize that cloud effects are not properly accounted for in the climate models and are more negative than currently modeled (Mauritsen & Stevens, 2015).
The tropopause transition begins at about 250 hPa in the tropics, it is also the altitude where the moist adiabatic theory (see figure 6 here) and the DPAH hypothesis begin to have problems. There is also a distinct change in slope in the relative humidity trend at about the same altitude, as shown in figure 6.
It is not clear what is happening around 250 hPa, but clearly all models, including the CMIP5 and CMIP6, the moist-adiabatic, or the DPAH models have problems in that tropospheric region. We suspect that these models are not handling the water vapor to liquid water, and then to ice transition properly. A possible name for this level is the “Graupel Boundary,” the height above which buoyant liquid cloud droplets in rapidly ascending convection towers flash-freeze into solid ice, a true phase change boundary. Graupel is the name normally given to snow or ice crystals that have become heavily rimmed by supercooled droplets, graupel normally forms between 5 and 10 km, above 10 km cirrus ice crystals dominate. This is a conceptual term and a proposed process that we may explore later as part of the Frost Point Anchor Hypothesis of cirrus cloud energetics.
The tropopause transition (roughly 250 hPa and above) is a region of ice crystal formation in rising air that releases latent heat of crystallization. The ice crystals fall under gravity into warmer air below, then sublimate back into water vapor and absorb heat. This cirrus ice cloud convection process explains the dynamic energy flow necessary to maintain the tropospheric temperature inversion.
Below 250 hPa, latent heat release from vapor-to-liquid condensation dominates. Higher up, the transition to ice-phase processes (crystallization, cirrus formation, and virga sublimation) become important, yet as far as we know this change in regime is not taken into account in any atmospheric model. This may contribute to the various model-observation mismatches discussed in this post. These mismatches are a golden opportunity to explore modelling methods that could improve models of the tropical troposphere.
Future Work: Toward a Frost Point Anchor Hypothesis
The success of the Dew-Point Anchor Hypothesis (DPAH) in regions greater than ~250 hPa suggests a natural extension into the upper troposphere where the latent heat of fusion dominates the energetics at lower pressure. We propose exploring a complementary “Frost Point Anchor Hypothesis” focused on the liquid-to-ice transition near the –40 °C isotherm. At this level, supercooled droplets rapidly freeze, releasing latent heat of crystallization, while ice crystals in cirrus clouds and virga undergo deposition and sublimation.
These ice-phase processes, combined with the radiative effects of cirrus (reflecting shortwave and trapping longwave radiation), likely anchor the temperature structure near the tropical tropopause. A future Markov or process-based model incorporating frost-point anchoring, ice microphysics, and sublimation cooling could better capture the transition from the convectively dominated troposphere to the radiatively controlled lower stratosphere.
Such an extension offers a promising pathway to reduce model-observation mismatches in the upper tropical troposphere and improve representations of cirrus feedbacks in climate simulations.
Download the bibliography here.
Thanks Andy
All the best,
Philip
Good to see Andy now on board.
I have also worked with Philip to express the underlying physics.
In the end the only things that matter in setting the surface temperature of a planet beneath an atmosphere is atmospheric mass and the force of the gravitational field which together work to balance the upward pressure gradient force and the downward force of gravity.
If that balance fails to be maintained then atmospheres are lost.
Current climate science believes that radiative imbalances can raise the surface temperature without destabilising the balance between those two forces.
It is wrong.
Greenhouse gases play a role also. Water vapor in the lower and middle troposphere and ice in the upper troposphere. In the stratosphere, CO2 and ozone matter a lot, especially when they change. These must be accounted for in addition to energy storage (or residence time), radiation-in – radiation-out, location, mass, and pressure. The story behind surface temperature is very complex.
Andy, would Aluminum Oxide from the disintegration of deorbiting satellites be a factor worth considering?
About 44,000 kg of meteors enter Earth’s atmosphere per day. or 16,000 tonnes per year. If they contain 1% aluminum by weight, that would amount to 160 tonnes per year. Most of that material vaporizes. And that is way more than the mass of satellites deorbited every year. There have been concerns expressed by certain scientists over the effects of satellites coming back into the atmosphere. I have never heard anyone (but me) even mention the fact that the meteoric mass influx is a lot bigger, let alone compare the relative effects.
Greenhouse gases introduce additional radiative imbalances but any net thermal effect at the surface is neutralised by a change in the rate of convective overturning. Water vapour phase changes make it easier for that to happen.
Your work is sound but underlying it is the ability of the system to adapt regardless of how radiative the atmospheric constituents become.
The circulation adjustments involved as a result of human emissions would be too small to measure.
A useful start on a more realistic investigation as to how the climate system on a water planet can maintain a stable temperature despite variations in the composition of an atmosphere.
A similar process occurs on any planet with atmosphere gases that can undergo a phase change from solid to liquid to gas and back again depending on temperatures.
This complements the previous idea that changes in the rate of convective overturning will adjust the rate at which energy is lost to space so as to neutralise any potential net thermal effect from changes in the radiative characteristics of an atmosphere.
Essentially, the phase changes increase the efficiency of internal energy transfers so that the rate of convective overturning does not need to change as much as would otherwise be necessary in the absence of phase changes.
By definition the energy entering/leaving a BB must be radiation.
The kinetic heat transfer processes of DPAH render BB impossible.
The “extra” 396 BB/333 “back”/duplicate 63 of GHE graphic TFK_bams09 and plethora of clones are trash.
The energy entering and leaving anything in space is radiative. So what?
Look at the pictures from space – the earth is a glowing blue and white body as any fool who is not colour blind can tell you.
The total energy content of the planet is in dynamic equilibrium. To scientifically demonstrate that the balance of input and output is proving to be so difficult the ultimate aim, is there a tiny fraction being retained over time or being lost over time, is elusive.
Time is the answer. “It is often claimed that atmospheric and ocean processes just “move heat around” and the only thing that really matters is radiation-in minus radiation-out at the top of the atmosphere” Time, residence time of all energy entering the system determines the total energy within the system. The GW claim is based on residence time, an increased residence time of a small spectrum if outgoing LWIR energy, causing an increase of energy within the system. A simple claim based on residence time of earths system.
Earth’s system is the land, oceans and atmospher. CAGW is obsessed with 1/1000 of the total energy. The oceans hold 999 times more energy. Variations in potent SW entering the oceans very long residence time has a much more dramatic effect on earths energy content then a much smaller weaker soectrum of LWIR exiting the atmospher.
Energy residence time is key, and top of atmosphere radiation in, radiation out, is the eventual result of energy residence time within the system. Climate science is struggling to understand minute changes in a very small bob tail wagging a very large dog.
Residence time is correct as the critical factor.
That residence time is accounted for by the time required to convert KE to PE in ascending air and PE back to KE in descending air.
If radiative material in an atmosphere tries to alter that time then convective overturning changes speed to prevent it from doing so.
Otherwise, no atmosphere.
The process is fully automatic and almost instantaneous because a change in the lapse rate slope set by atmospheric mass and gravity in one place is immediately offset by an opposite response in another place.
That explains the heat stored in planets and moons far from the sun.
The more convection is going on the longer the residence time and the higher the temperature.
It even works in the depths of space within any gaseous mass which is what leads to the formation of suns in the first place.
It is also why the temperature of the observable universe is a tad higher than absolute zero.
I’m naively curious about geothermal heat released into the oceans. Obviously it will have a longer residence than that which is released to the atmosphere, but are both sources simply constants in the global radiation balance?
Geothermal input is highly variable but counts for little given the vast volumes of cold dense water at the bottom of our oceans.Even if it were to have any effect then a tiny change in the rate of atmospheric convective overturning would quickly neutralise any potential effect on average surface temperature.
The oceans integrate solar activity, so residence time is a function of frequency.
An ideal integrator has an amplitude response that is inversely proportional to frequency, and a constant phase response of -90°. In other words, the response to a 60-year forcing cycle would be delayed by 15 years.
The integral relationship applies for periods longer than 11 years, i.e. below the Schwabe notch at the center of the plot below. For shorter periods, the sensitivity to solar forcing does not taper off with frequency, as might be expected for atmospheric and sea-surface responses.
In this plot, red is the frequency response between global temperature and sunspots. Blue is the frequency response between predicted temperature and sunspots where the prediction is a 99-year moving average of sunspot data. Coherent averaging reduces frequency resolution, but helps attenuate uncorrelated signals (noise and random weather)
Also, they are claiming they can model molecular interactions on a 25 km grid.
You are quoting the ideal black body.
Applying Kirchhoff’s law, EM and thermal are to be in balance.
I have down much work on thermal management of electronic components.
I use the test results of materials sciences and can assure you thermal energy input with thermal conduction warms the exterior surface and it emits EM.
On the ground, convection plays in.
In space vacuum limits cooling to EM radiation.
Likewise I have done much in electromagnetic environmental testing.
EM radiation related to surface temperature is real.
Ignore those bogus flat earth energy imbalance graphics.
They are not worth the electrons expended in displaying them.
I accept that the planet has an air-conditioner system. I accept that the motion of fluids both gaseous and aqueous are essential to the mechanism and that the process of energy dissipation and transfer and absorption is essentially subject to thermodynamic analysis.
I fail to see the point in dissecting the process anatomically.
The elephant in the room, for everybody, is that fluid flow whether linear or circular cannot be modelled for the simple reason that fluids flow in at least two ways, laminar and turbulent. So far, turbulence is a mystery and patches of turbulence need sample sizes which are tiny – far beyond the number crunching power of every supercomputer in existence to get answers in times measured in human lifetimes.
Can anyone explain what this man is trying to tell us in simpler terms, remembering that I have been postgraduate scientist for more than 70 years and do not need to be answered as if I am still wet behind the ears? Like, what is his point? Are the conditions he is trying to model not regions which can be sampled and analysed and not be inventions of the imagination?
Dr. Chalmers,
Thank you for the thoughtful comment and for your long career in science — that perspective is valuable.
The core goal of the Dew-Point Anchor Hypothesis (DPAH) is not simply to describe or reproduce what the measurements show (the vertical temperature, dew-point, and lapse-rate profiles from radiosondes). It is to propose why those profiles have the shape they do — particularly in the tropics.
The key insight is that the thermodynamics of the phase changes of Earth’s condensing volatile — water — provide an independent internal structural anchor to the tropospheric profile below the frost point. Surface temperature and near-surface dew-point depression (how close the air is to saturation) act as a physical “anchor” for the entire convective column. By using a simple probabilistic Markov model with two regimes (ascent in the ITCZ and descent in the subtropics), we test whether this bottom-up anchoring, driven by latent heat release from condensation, can explain the observed structure up to roughly 250 hPa. The fit is very strong in that layer (R² > 0.998 for temperature and dew point).
This offers a physical mechanism rooted in surface conditions, moist convection, and the energetics of water’s phase changes, rather than treating the profiles as purely radiative or top-down phenomena. Above ~250 hPa the model diverges, which points to the shift to ice-phase processes (freezing near –40 °C, cirrus formation, and virga sublimation). We plan to explore this upper-level regime through a complementary Frost Point Anchor Hypothesis (FPAH) focused on cirrus cloud dynamics, ice microphysics, and sublimation cooling.
We are not trying to simulate global turbulence or replace comprehensive models. We are exploring targeted, mechanism-based explanations for the average vertical thermodynamic structure in the tropics, tested directly against real balloon data.
We welcome constructive critique, especially from experienced scientists like yourself. If you have suggestions on better ways to test the idea or additional data sets, they would be appreciated.
Best regards,
Philip Mulholland
I agree with Philip’s answer but would add why I decided to work with him. I think, and have thought for many years, that there was something missing in the conceptual climate model, see here for what a conceptual model is:
https://andymaypetrophysicist.com/2024/02/28/climate-model-bias-1-what-is-a-model/
Essentially it is the idea being tested with the mathematical computer model. If the conceptual model is missing something important, you can tinker with the mathematical model forever and not get anywhere. This is what I think has happened in the CMIP models. The most glaring evidence this is true is the model “hot spot” made famous by John Christy, it never goes away, even when thousands of scientists spend years trying to fix it.
I thought Philip was heading in the right direction, start fixing it in the tropics where the obvious problem is and work from there. In my humble opinion, this problem goes all the way back to Manabe, Strickler, and Weatherald, and someone needs to start over. Start small and work our way through the system.
Starting over is indeed a great idea, Andy. Make sure you start with this correct definition:
“Radiation is […] energy”
And not this contradictory one:
“Radiation is a mechanism, it isn’t energy.”
Right?
Sorry Steve,
No — radiation cannot be “defined as energy.”
You can say:
Radiation carries energy.
Radiation transports energy.
Radiation produces measurable fluxes of energy.
Compare to “convection” and “conduction.”
“No — radiation cannot be “defined as energy.””
Then why did you write:
“[radiation is] normally assumed to be energy” ?
(Which I would agree with, except that there’s no “assumed” about it, in my universe)
Then you said:
“Radiation carries energy.”
You mean like a bus? Or a luminiferous aether? Where did you get that from? You know that physicists haven’t thought that way in over a hundred years, right? Here is the Wikipedia definition again:
“radiant energy is the energy of electromagnetic[1] and gravitational radiation.”
By definition and observation, “radiant energy” is a type of energy. (We’re looking at the electromagnetic variant here.) It is therefore the “capacity to do work”, propagating itself through space, as alternating electric and magnetic fields. There is no separate “transport mechanism” independent of the radiant energy. They are the same thing.
“Radiation produces measurable fluxes of energy.”
That’s an extremely disingenuous statement, and borderline fantasy. It certainly can, yes. But under what circumstances?
What’s the matter, Andy? Not going to bother to explain to us why you violated your own brand-new scientific principle before you had even finished writing it down? Most scientists would try to think of principles that would stand up to at least a few seconds of scrutiny, but not you, right, Andy May Petrophysicist?
Of course we also have our very own local genius Willis telling us that “Radiation is energy”. Was he wrong too? Will you tell him to his face that he is a liar? I’ll watch from here. It will be fun.
(And then, bizarrely, you had the gall and hypocrisy to call me “confused”. You’re a piece of work, Andy, you really are.)
“Compare to “convection” and “conduction.””
Yes, let’s do that. Under what circumstances do you think object A will transfer thermal energy via conduction to, or conduct energy to, or perform work on, object B?
Like it or not, radiation is energy in motion. The magnitude it has is the magnitude of the E field and the magnitude of the H field. End of story.
What happens prior to the radiation being created or after it is changed by gravity, other EM waves, absorption, etc. must be defined and analyzed separately.
Your word games do not contribute to understanding at all. If you want to make a point, show the associated mathematics. Those that can understand will, those who can’t may learn.
“Like it or not, radiation is energy in motion”
I never said anything different. But energy in motion is still energy, isn’t it, Jim?
“Your word games”
Sit down, twerp.
Call me all the names you want, but I am still correct. Your word games have no mathematics backing them up. Show your math when you make an assertion or all your protestations are meaningless.
“but I am still correct”
Note that I never said you were wrong. Not in this case, at least. Did you miss that? Of course you did.
You contradicted Andy, though. So one of you is obviously lying. Which one, do you think? (I already told you the answer, so this should be easy.)
“Your word games”
Sit down, twerp. Science is not for illiterate undereducated engineers with room temperature IQs who can’t tell the difference between “scientific definitions” and “word games”.
“all your protestations”
What protestations? I told you you were right.
“are meaningless.”
The only meaningless thing around here is everything you wrote after you told us that “semantics are irrelevant”, or in other words, that you are functionally illiterate. By choice.
There are basic problems with terminology.
It gets worse when specific definitions are hijacked, repurposed, and subject to social/common language context derived definitions.
Context matters.
Yes, radiation is a mechanism.
There are both thermal and electromagnetic radiation mechanisms.
Radiation is also high energy particles, renown for ionizing effects.
Electromagnetic radiation is both the movement of energy and the energy itself.
Infrared radiation is not heat.
Heat is not energy.
Heat is the flow of thermal energy via kinetic interactions.
Unless one sticks to using concise definitions, the signal to noise ratio in language communications is awful. It gets worse when one does not clearly define context.
Hear, hear!
“There are basic problems with terminology.”
Indeed.
“Electromagnetic radiation is both the movement of energy and the energy itself.”
And this is one of them. Remember that the “movement of energy” is not the same as “work”, or therefore “power”.
“Heat is the flow of thermal energy via kinetic interactions.”
Or radiant ones.
“Unless one sticks to using concise definitions, the signal to noise ratio in language communications is awful.”
Absolutely. Especially when people use the same word to mean two different things, and don’t specify which one they’re talking about. Scientists try to avoid such ambiguities. Not always successfully!
Dr. Chalmers,
As a retired professional geoscientist with over 50 years’ experience in seismic interpretation and subsurface modelling, I fully understand the healthy skepticism toward over-reliance on modelling, especially when underlying concepts are pushed beyond their proper domain.
I believe overreach has occurred in climate science, where the radiative paradigm — legitimate for studying boundary conditions at the top of the atmosphere — has been extended to explain the internal generation of the environmental lapse rate solely through atmospheric thermal radiant opacity. This overlooks fundamental meteorology.
The dry adiabatic lapse rate can be derived from first principles using the hydrostatic equation, planetary gravity, and mean molecular weight. The environmental lapse rate, in turn, is strongly shaped by the latent heat released and absorbed during the phase changes of water — Earth’s dominant condensing volatile. These internal thermodynamic processes provide a physically grounded anchor for the tropospheric temperature profile.
In our Dew-Point Anchor Hypothesis (DPAH) work, we are seeking to restore the primacy of meteorology in explaining the internal energetics of the atmosphere. Radiative physics, while important, is not a “one size fits all” explanator for the most complex physical system in geoscience. Our Markov model tests whether surface conditions and moist convective processes can account for the observed vertical structure in the tropics up to ~250 hPa. The strong empirical fit supports this bottom-up view.
We openly acknowledge the model’s limits near the tropopause, where ice-phase processes dominate, and we plan to extend the approach with a Frost Point Anchor Hypothesis focused on cirrus dynamics.
Constructive critique from experienced scientists like yourself is most welcome.
Best regards,
Philip Mulholland
“I believe overreach has occurred in climate science, where the radiative paradigm — legitimate for studying boundary conditions at the top of the atmosphere — has been extended to explain the internal generation of the environmental lapse rate solely through atmospheric thermal radiant opacity. This overlooks fundamental meteorology.”
Well put, sir. In 1938, Callendar proposed an attribution of station warming to the rising concentration of CO2. From the meteorological discipline of the time, Simpson and Brunt explained concisely why the radiative paradigm does not work to support such a finding. More here. Modern modeling of the general circulation, e.g. ERA5, confirms their sense of the issue, in my view.
https://wattsupwiththat.com/2026/03/15/open-thread-181/#comment-4174555
Good to see you have managed to get this article up on WUWT. Good work and flows nicely.
This is what I have been stating for a few years. Understand ice forming processes on land, on water and in the atmosphere and you begin to understand the climate.
It is the solidification of water vapour at high altitude and low pressure during convective overshoot that creates the surface temperature regulation around 30C.
I would be looking for radiosonde outliers when surface temperature is above 30C. to see how well that fits the model. This is the condition that controls overshoot caused by more moisture below the LFC than needed to just achieve saturation above 10km. The ice that forms above 14km in the thin air is micron size and has very low terminal velocity; meaning it is persistent. It absorbs and emits OLR at very low temperature but increase in reflected shortwave is greater than reduced OLR so it cools the surface.
But it is now widely known that the open ocean cannot sustain a temperature above 30C. But it is even more widely known that the GAST is increasing. So the question is why?
Hi Rick,
Good question and one needs to keep an open mind and consider the possibilities. It could be, in part, or entirely:
And there are other possibilities that I can’t think of, I’m sure. But we will get nowhere, just looking a one possibility and ignoring the others.
Something that all the climate models are blind to is the long and gradual upward trend in annual ToA solar energy over the tropics. The common belief is that precession of perihelion moves so slowly that it could not possibly be the cause of the observed temperature rise.
You will see that CERES shows a significant upward trend for its existence this century but that is mostly due to solar activity. The underlying trend has been increasing since 1255AD. It took till about 1600AD before the climate started to respond to the increasing tropical sunlight.
2024 has 4ZJ more sunlight over the tropics than 1255. Enough to increase atmospheric water by 6mm over the minimum in 1255. Half of that was trend and half was solar activity peak of SC25 in 2024.
So precession causes both seasonal shifts in solar power and annual shifts in available energy across latitudes. The common wisdom is that energy input for a given latitude is constant over an annual cycle. That is not the case and easily verified by looking at the ToA CERES data.
So more annual solar energy over the tropics is driving more evaporation in the tropics. The atmosphere is more energetic and then able to advect more heat poleward.
Seasonal solar variation in the SH is reducing but increasing in the NH. Current warmth across Europe is just a hint of things to come until the Med goes into regular monsoon and northern Africa builds more biomass.
“But it is even more widely known that the GAST is increasing.”
It is no warmer today than in the 1930’s. In fact, it is cooler today than in the 1930’s.
Assuming warming is not supported by the written record.
The AMO is cyclical and has a warm phase and a cool phase. So does the Earth’s atmosphere. You are referring to one phase when you say it is warming.
I like your theories about the Sun and how things change as the positions of Earth and the Sun change, but your theory needs to explain the cyclical movement of the Earth’s temperatures. You seem to be ignoring this with the claim the GAST is increasing. History says it increases and decreases.
I understand that people need a standard for comparison and currently that is GAST (and others).
However, GAST is only an exercise in arithmetic and has no physical meaning.
Averaging 0 C and 20 C yields 10 C
Averaging -20 C and +40 C yields 10 C.
They are not physically the same.
100%! An average of a data distribution is *NOT* a data point. It is a statistical descriptor that is sometimes useful in describing the data distribution. It becomes less useful when the distribution is skewed, multi-modal, etc. It also becomes less useful when the individual data components can’t be used to form a larger, physical system. When this is the case you can calculate an average mathematically but it is decidedly non-physical.
And look at the variance in your two examples. That is an indication of how not stating variance is lying by omission. In the first, the range variance is ±10°C while in the second it is ±30°C. That gives an anomaly of say 0.01 ±30°C. I’m burning up!
The best proxies for global surface temperature are the satellite measurements of the lower troposphere. On average, they exhibit an upward trend for almost 50 years now. A warmer atmosphere above a warmer surface – on average.
The CET is the longest standing instrument record and has trended upward for each century since it began measurements – 1700, 1800 and 1900. And that trend is continuing in to the 21st century.
The high temperature of the 1930s was essentially limited to North America and usually explained by land clearing.
Solar activity is significant in decade scale variation and is mostly responsible for the strong rise in ToA sunlight over the tropics this century but that variation sits on an upward trend of much longer duration driven primarily by precession of perihelion.
Sea level is rising and that trend goes back at least 300 years. About half the rise is due to ocean heat retention.
The temperature trend throughout the holocene has followed precession of perihelion.
Dew point, wet bulb, relative humidity & grains of water – psychrometric properties of humid air as exemplified in the kinetic turmoil of the atmos or an industrial wet cooling tower pluming at your local powerplant.
Relative Humidity = water grains actual compared to imaginary grains at saturation.
7,000 grains per pound. 1.0 Btu/lb F sensible, 1,000 Btu/lb latent. Air, 0.24 Btu/lb F
The Btu interplay between RH & DB as seen on the psychrometric graphic shows the RH/DB acting as a diurnal thermal surge tank.
So, what?
Well, a 1.0 emissivity BB system can’t include any kinetic processes.
TFK_bams09 emissivity is 63 real/396 imaginary = 0.16.
And no BB means no imaginary “extra” 396 W/m^2, no imaginary 333 W/m^2 “back” radiation, no imaginary duplicate 63 W/m^2 that you hide under your desk pad and no GHE.
Wrong.
A 1.0 emissivity BB system has to first account for kinetic processes before determining how much EM is emitted.
There is a reason why the expression grey body came into play.
No where in this universe is a true black body.
That is an ideal model constructed to help, repeat help, explain the physical reality.
Had to find this.
Andy and Philip, nice article. It deals with physical parameters that have been missing in climate science for far too long.
My own investigation started recently after becoming convinced that radiative input, insolation, can’t just be easily nor correctly translated into atmospheric temperature. The surface plays a large part in translating short wave energy into longer wavelength infrared wavelengths. This isn’t a simple process where one can derive a simple function or even a constant value to apply to the magnitudes.
I especially liked the comment:
and from Dr.
Climate science and modelers have been beguiled with statistical treatments of climate phenomena and what might be based on time. Alice in Wonderland comes to mind. This theory deals with actual physical data and proposes an hypothesis that can result in a mathematical relationship to explain what is occurring.
The only thing I see missing is a proposed treatment of enthalpy, although that is inherent in the phase changes taking place in H2O. This is probably ok for the atmosphere over oceans but over land, temperature alone will not suffice. There is too much variation in enthalpy.
Thank you both for your work that will drive climate science out of the ditch and back onto the road.
One of my favorite Pat Frank quotes:
“Statistics is no substitute for Physics”
Thanks Jim
There are many factors that affect air pressure and rainfall. The moon pulls the Earth’s atmosphere towards it and a bulge occurs in the atmosphere under the moon.
https://www.washington.edu/news/2016/01/29/phases-of-the-moon-affect-amount-of-rainfall/
Also the oceans. It is called tides.
The moon creates turbulence in the atmosphere and ocean.
Exactly, but not just the moon.
Lots of planets out there.
It is not if. It is merely a matter of degree.
This article and the comments are what real science is all about.
Thanks to all. Very interesting!
It most certainly beats the socks off political flame warrior commentary.
“higher latitudes are a different environment and depend much more on advective heat carried into them from the lower latitudes by winds and ocean currents”
Is it possible planetary geometry plays into this?
At higher latitudes the solar angle of incidents deviates from orthogonal and, as such, has a high natural reflectivity?
The result is less solar energy warming the surface.
Also, at higher latitudes, due to spherical geometry, the atmospheric optical depth is deeper.
Is is possible that is also a factor?
Path loss never seems to get included in any measurement discussion. E.g. measuring temperature using microwave frequencies which suffer from path loss. If you don’t know the path loss then you can’t figure the originating temperature. It just becomes a guess. Once again, modeling water vapor as a parameter (e.g. clouds, etc) raises its ugly head.
I should add that optical depth and path loss are related.
I elected to pose those as questions, not statements.
I believe you understood my approach and perhaps why that choice.
Cheers.
The rotation of the earth pushes a wave of atmospheric expansion and contraction due to differential solar heating. This pumping of the atmosphere once every 24 hours is doing massive work on a planetary scale. Is it accounted for?
“… the only thing that really matters is radiation-in minus radiation-out at the top of the atmosphere (IPCC, 2021, Ch. 7, section 7.2) & (Held & Soden, 2000). However, this is clearly incorrect, …”
Well, duh!!!
As I have demonstrated repeatedly GHE thinks the surface radiates “extra” energy as a 16 C BB.
Just flat wrong.
That reminds me of a point I made here in 2010.
Those types of budgets show latent heat (PE) going up within rising air but no PE coming back down as kinetic energy (heat/KE) in falling columns of air.
Instead, they use the concept of downward IR to make up for that omission.
In reality there is no net positive warming effect from downward IR because the atmosphere already warms as one goes down along the lapse rate slope. Instead, KE is released from PE within descending columns of air.
As I recall, it was Trenberth who made that error initially.
The budget balances without downward IR when one makes the appropriate adjustment.
Please throw out that bogus graphic.
There are many more things wrong with it than you point out.
Is “surface” temperature actual ground or 1.5 m air temp?
Sun heats ground, ground heats air.
Latent moisture moderates diurnal swing.
I believe the above article by Andy May and Philip Mulholland offers some key insights to one of the many feedback processes occurring in Earth’s atmosphere, hydrosphere and cryosphere—more specifically with respect to its hydrologic cycle—that “somewhat stabilize” Earth’s climate on timescales of tens to many thousands of years (understanding that in the last million of so years, Earth has cycled between glacial and interglacial conditions at an average frequency of about one cycle per 100,000 years).
However, I disagree with this summary statement made as the last sentence in the article’s second paragraph:
“Moving the climatic anchor into the climate system brings the climate oscillations into the accounting, an important conceptual shift.”
There really is no conceptual shift in the “accounting” for climate oscillations.
The accounting for climate oscillations in terms of energy flows entering, leaving and within the Earth-considered-as-a-system-bound-to-orbit-the-Sun and, in turn to orbit the center of the Milky Way galaxy, has always AVERAGED climate oscillations over various time periods, even when only considering TOA incoming-vs-outgoing radiation exchange.
This averaging in energy accounting has ranged from diurnal weather oscillations, to annual seasonal oscillations, to weather oscillations (weakly) coupled to the “11 year average” (aka Schwabe) sunspot cycles, to multiple El Nino/La Nina events, to AMO and PDO climate oscillations, to Gleissberg and DeVries solar radiation oscillations, and to various Milankovitch cycles and their resonances, with oscillation periods varying from tens to hundreds of millennia duration.
This “averaging over time” (aka “smoothing of oscillations”) is necessary to properly account for (a) the enormous heat capacity of Earth’s oceans, and (b) the transition time delays associated with the phase changes of water (solid<–>liquid and liquid<–>vapor) that involve relatively massive amounts of energy exchange without a corresponding change in the temperature of the mass of water involved.
Some of what you say is obvious. Variation over time is important. And it must be observed over a long enough period to identify all low-frequency components.
The problem is that “averaging over time” hides the variance. That’s one of the MAJOR failings of climate science, always assuming that Variance always cancels out and can therefore be ignored.
Proper accounting of the heat flows, in/out, doesn’t truly require averaging for proper accounting. Integrating the factors over time results in an extensive value that can be used. Averaging of intensive values results in a non-physical output. E.g. integrating temperature (intensive) over time results in degree-days (extensive). Averaging temperature is just plain non-physical.
Whatever is done in the future, it must include balancing joules in/out over a period long enough to identify low frequency components. It’s those low frequency components that are IMPORTANT in being able to identify the negative feedback paths so important to climate.