On the Cloud Thermostat Hypothesis

August 22nd, 2013 by Roy W. Spencer, Ph. D.

Over 20 years ago, Ramanathan and Collins (1991 Nature, “RC91”) advanced what became known as the “Thermostat Hypothesis”, relating to how cloud formation over the tropical Pacific warm pool limits just how warm surface waters there can get. While this conceptual view was OK for the warm pool itself, they made some extrapolations which weren’t really warranted about what this meant for the larger-scale climate system, and ultimately whether “global warming” will be reduced by a resulting increase in cloud cover reflecting more sunlight back to space (negative cloud feedback).

Now, I will say that I believe that the climate system is stabilized by negative cloud feedback, even though most if not all climate models exhibit positive cloud feedback. But you have to be careful about what you use as evidence, and cloud formation over warm areas (e.g. at the end of this post at WUWT) is simply not evidence. Even climate models with strong positive cloud feedback (decreasing clouds with warming) are going to form clouds over warm areas of the oceans. That’s the way atmospheric circulation systems work.

The RC91 Thermostat Hypothesis paper, written by experts in radiation but not so much in atmospheric dynamics, was quickly attacked for neglecting the fact that, for all of that rising cloudy air over the warm pool, there has to be sinking air elsewhere which suppresses cloud formation.

In other words, just because clouds preferentially form over locations with warmer ocean waters doesn’t mean it’s evidence for negative cloud feedback (clouds increasing with global warming). Here’s a cartoon showing the basic idea:
clouds-over-warm-spots-do-not-prove-thermostat

Hartmann and Michelsen (1993 J. of Climate) and Lau et al. (1994 Geophys Res. Lett., 21, 1157-1160) were among the first papers to point this out. Much work has been done on the issue in the last 22 years with the general conclusion that, for the tropical oceans at least, warming leads to virtually no net cloud feedback, either positive or negative. The problem gets even more complicated because the temperature inversion in the above cartoon — the result of subsidence-warmed air overlying cooler ocean waters — also causes marine stratus clouds to form, which have a strong cooling effect on the climate system as a whole.

If the cloud feedback problem was that simple, it would have been solved long ago. But it ain’t that simple.


59 Responses to “On the Cloud Thermostat Hypothesis”

Toggle Trackbacks

  1. Bri says:

    Is there long term data to support ether hypothesis? I would like to see a graph of the all the feedbacks used in the models vs. the temperatures over the longest time possible. It would seem that this would eventually resolve the feedback issue.

    ßri

    • Good question, and I’ve spent years trying to answer it. The problem is complicated by (1) not being sure what the radiative forcings are that “caused” the temperature change to begin with, and (2) whether changes in ocean mixing have also changed surface temperatures.

      • buck says:

        At some point I would think Dr. Spencer’s and RSS satellite temperature trends become the long term data that validates or disproves model feedbacks.

  2. tetris says:

    So what, if anything, does the CERES data used by W.E. tell us? Is the albedo linked to more cloud cover not important?

    • I’ve been analyzing CERES data for years trying to find some metric related to feedback…it is a very difficult problem (see my comment just above this one). No, the CERES correlation plot with surface temperature W.E. showed doesn’t tell us anything about cloud feedback.

  3. Kusigrosz says:

    How important is cloud shape and, in case of anisotropic clouds, orientation relative to the Sun? Are there any long-term trends in these?

    • Clouds become highly reflective with relatively little thickness, so large areas of marine stratocumulus have a greater cooling effect than all of that cloud water gathered into a deep, narrow tower.

      Cloud reflection is definitely anisotropic, and the CERES people have to make corrections for how much sunlight is reflected as a function of angle. They get this through measurements, since part of the CERES instrument package is meant to gather statistics on how clouds reflect sunlight from different angles. I don’t know whether anyone has looked at changes over time in this, though. Probably looking for a needle in a haystack.

  4. The effect of clouds that matters most appears to be the reflectivity as seen from space.

    That inevitably affects the proportion of solar energy arriving at ToA which is then able to penetrate the atmosphere to the surface and, especially, the oceans.

    Thus my contention that changes in global cloud cover will in due course affect ocean heat content and then ultimately the temperature of the air.

    That led me to consider how those changes in total global cloud cover could occur in the first place and my conclusion was that ultimately it is solar driven via changes in stratospheric temperatures and in particular the gradient of tropopause height from equator to poles.

    A steeper equator to pole gradient gives zonal jets and less clouds whereas a shallower such gradient gives meridional jets and more clouds.

    The oceanic response would then simply cause a variable out of phase modulation of the initial solar driven change in cloud cover.

    Rather than relying on the Svensmark hypothesis I think it is simpler to suggest that zonal jets and poleward climate zones produce less cloud cover for a warming planet and meridional jets with equatorward climate zones produce more clouds for a cooling planet.

    The reason being longer lines of air mass mixing in a meridional scenario.

    For a more detailed hypothesis see here:

    http://www.newclimatemodel.com/new-climate-model/

    In so far as GHGs have any net thermal effect they would simply contribute infinitesimally to the existing natural effects caused by solar and oceanic variability.

    As far as I can see that proposition fits with recent observations and in particular the timing of recent observed climate shifts such as the one in the late 70s and the one that is becoming more apparent in the early 2000s.

    Each of those shifts was accompanied by a change in the level of solar activity. More active from cycles 20 to 23 and less active for cycle 24.

    It also fits nicely with the longer term observations which include the Little Ice Age (meridional jets) and the MWP (zonal jets).

    Someone with a little gravitas in the climatology establishment should pick up on these suggestions and run with them.

  5. Leon says:

    I try to stay open to different theories. So far I find Svensmark to be the most credible, regarding the basic climate drivers involved.
    I wonder why we don’t hear more about the NASA satellite data finding, their announcement that atmospheric CO2 greatly facilitates solar reflectivity and so is a cooling influence?

    • Leon, that NASA report had to do with the cooling effect of CO2 on the upper stratosphere, which is nothing new. Greenhouse gases cause the upper atmosphere to be colder, and the lower atmosphere to be warmer, than without those gases.

      The Slayers picked up on the NASA report as some sort of new proof that CO2 “causes cooling”, not warming. The truth is it causes both, but has opposite net effects on temperature in the upper atmosphere versus lower atmosphere.

      • Stephen Wilde says:

        “Greenhouse gases cause the upper atmosphere to be colder, and the lower atmosphere to be warmer, than without those gases.”

        I suspect that it is soon going to be found that any such effect is negligible compared to solar effects from the tropopause upward.

  6. I agree with Dr. Spencer in that greenhouse gases cool the upper atm. and warm the lower atm.

    I also believe clouds overall are a negative feedback, with the exception being large snow covered land areas such as Antartica where more cloud cover would likely lower the overall albedo in that case and cause warming.

    • dallas says:

      The overall net impact of clouds would depend on what “normal” is. The average tropical ocean SST has long term secular trend with not as much sensitivity to CO2 equivalent “forcing”. According to the Oppo 2009 Indo-Pacific Warm Pool reconstruction, that warming trend started around 1700 AD and has about 100 year pseudo-cycles. There is no reason to assume that cloud feedback would be the same during a general warming or cooling mode since a thermostat has a setpoint it is trying to maintain.

      http://tinyurl.com/ljccxpf

      That is the Oppo IDPW spliced to some instrumental, if clouds are a thermostat in the tropical Pacific, we just recently reached the control range.

  7. Bri says:

    Thanks for your answer about the comparison of forcing vs. temperature. The problem is that the global warming community love’s to show us graphs of co2 vs. temperature. I understand the factors involved are complicated and often inconclusive but they have you trapped by letting them frame the argument.
    I suggest a graph of co2 vs. temperature for the last 1000 years but the absolute temperature and co2 not the anomaly. This would be an almost straight line and would reveal the hype they want use to scare more funding from the government.

    ßri

  8. RokShox says:

    On the one hand you say subsiding air suppresses cloud formation. On the other you say it stimulates marine stratus clouds. Which is it?

  9. RokShox says:

    It’s not just the direct effect of the higher tropical albedo, but the fact that a higher albedo is a marker for increased convection.

    That is, it’s not sufficient to just point at tall narrow clouds and say their reflective ability is less than that of the potentially larger area of clouds at higher latitudes that they end up suppressing.

    Those tall narrow columns are pumping an enormous amount of heat into space.

  10. RW says:

    “If the cloud feedback problem was that simple, it would have been solved long ago. But it ain’t that simple.”

    Actually, it might even be simpler.

    I would argue these two plots provide the answer to the critical cloud feedback question:

    Cloud amount vs. temperature:

    http://www.palisad.com/co2/gf/st_ca.png

    Above about OC, the net effect of clouds is to cool by reflecting more solar power away than is delayed beneath them. Below about 0C, the net effect of clouds is to warm by delaying more energy beneath them than is reflected away in total (due to snow and ice below 0C, which is roughly as reflective to solar energy as clouds are).

    Water density vs. temperature

    http://www.palisad.com/co2/gf/st_wc.png

    The data unambiguously confirms that warmer temperatures result in increased water vapor in the air as frequently claimed; however, notice how as the water vapor increases the temperature increase is not equally proportional throughout the range. Above the current global average temperature (about 288K), the water vapor content increases by a much greater amount than the temperature increases.

    Looking at this in relationship to the cloud amount vs. temperature plot (from the same data set) provides critical insight. At about the same point where the clouds start to increase again as the data goes into the tropics, the rapidly increasing water content no longer results in a further rise in temperature. The fundamental physical mechanism(s) behind this is beyond a certain temperature there is so much water being evaporated, removing so much heat from the surface (as the latent heat of evaporation), providing so much ‘fuel’ (i.e. water) for cloud formation, that the combination of cloud caused and evaporative caused cooling appears to overwhelm any increase in atmospheric opacity from increased water vapor.

    If anything, in a warmer world there would less snow and ice covered surface – not more. Hardly a case for the net effect of clouds acting to further warm rather than remain to cool on incremental global warming.

    The fundamental physics appear to be the dynamic, interrelated processes of water vapor and clouds, where clouds – whose formation is first driven by evaporated water from the surface (primarily the oceans), act as the primary control mechanism that maintains the energy balance from the forcing of the Sun. The mechanism appears to be that, on global average, increasing cloud coverage causing cooling (more solar energy is reflected than is delayed) and decreasing cloud coverage causes warming (more solar energy is absorbed than exits to space). Or that on global average, when clouds are increasing, the surface is too warm and trying to cool, and when clouds are decreasing, the surface is too cool and trying to warm. That is, these counter balancing mechanisms dynamically maintain the energy balance. In the end, it’s probably just basic thermodynamics and cloud physics.

    • RW says:

      Each small orange and brown dot is a one month average for a single grid area in a 2.5 degree slice of latitude. The larger green and blue dots are the 25 year averages for each whole 2.5 degree slice of latitude. The data is sourced from ISCCP (1983-2008).

  11. Christopher Game says:

    Very helpful post, thank you Dr Spencer.

  12. Willis Eschenbach says:

    An interesting post, Dr. Roy. However, I fear I don’t understand your criticism of my work.

    What I have shown is a positive correlation between temperature and albedo in the tropics. In other words, increasing clouds with warming. I’d never seen that mapped out before. Had you? I’d be interested in any links if someone’s done that before.

    In response, you say:

    But you have to be careful about what you use as evidence, and cloud formation over warm areas (e.g. at the end of this post at WUWT) is simply not evidence. Even climate models with strong positive cloud feedback (decreasing clouds with warming) are going to form clouds over warm areas of the oceans. That’s the way atmospheric circulation systems work.

    How is a climate model with decreasing clouds with warming going to show increasing clouds with warming? I’m not following that.

    I discussed in the same post how a combination of the cumulus cloud formation threshold and the cumulonimbus formation threshold act to control the surface temperature on an hour by hour basis. This is not “clouds form over warm oceans” as you allege.

    Instead, this is a sequence of emergent climate phenomena which act in concert to control the temperature.

    What I realized was that the action of this control system should be reflected in an overall positive correlation between albedo and temperature in the tropics. I found that … but not as “clouds form over warm areas of the oceans”. I found it as varying levels of correlation, with a whole lot of information in the map. And when you look at my map you can see that in some of the warmest areas of the ocean, like the Pacific Warm Pool, the correlation was less than in cooler areas. So I certainly didn’t show that “clouds form over warm areas of the oceans”.

    Regarding your citations that you say back up your claim:

    Hartmann and Mitchelson say that the average cloud coverage in cumulus fields is not particularly temperature dependent. My own work shows the same thing, with the exception of a slight increase in albedo in cumulonimbus versus cumulus fields.

    What I have said from the start is that the timing of the onset of the clouds is the albedo regulating mechanism, not the average cloud coverage of the cumulus field. I know that doesn’t change much. So H&M93 is not to the point.

    Lau et al. 1994 is a critique of the cirrus-cloud thermostat hypothesis. Again, not to the point. I have proposed an entirely different mechanism is at work.

    You close by saying:

    If the cloud feedback problem was that simple, it would have been solved long ago. But it ain’t that simple.

    I agree that it’s not simple feedback that keeps the temperatures in line. So you’re right, the problem is not that simple.

    Instead, what we are looking is a governor, or rather a number of governors, and subtle ones. In general the temperature is regulated through the timing and location of the appearance of emergent climate phenomena that work in a host of ways to keep the planet’s tropical surface temperature within a fairly narrow range. These phenomena include:

    Cumulus clouds forming to regulate incoming energy.

    Thunderstorms emerging from the cumulus to cool the surface and the planet in a host of ways, including driving the Hadley circulation.

    El Nino/La Nina alterations pumping warm water from the tropics to the poles, and in the process exposing the atmosphere to the cooler subsurface water.

    Thunderstorms taking a large-scale emergent form with long rows interspersed with canyons, which more efficiently moves heat aloft.

    Dust-devils emerging to move surface heat straight upwards.

    Cyclones (hurricanes, typhoons) forming up and cooling vast areas of the ocean by several degrees.

    The Pacific Decadal Oscillation alternately impeding and enhancing heat flows from the tropics to to poles.

    None of these, as you point out, have anything to do with simple feedback, or even complex feedback. They are independent, and in some cases independently mobile, emergent phenomena. Attempting to conceptualize the appearance of emergent phenomena as “feedback” is part of the problem with the current climate paradigm.

    Anyhow, that’s how I see it. What am I missing?

    My best to you,

    w.

    PS—I’ll cross-post this on my thread.

    • Bill Hunter says:

      Willis if I read this correctly Roy’s criticism is the problem is not as simple as you are suggesting.

      So my impression is he is not disrespecting your observation he is simply being critical of the conclusion regarding where the missing heat is.

      I thought it was interesting that your own article suggested a negative feedback to your negative feedback when you noted “how cold mornings and warm mornings affect the evolution of the temperature of the ensuing day. . . . As a result, when the day is warmer at dawn, the following morning is cooler.”

      I was totally lost there as you seemed to be describing a cycle not a net feedback and that train of thought just got dropped.

      I have been following Roy’s work in this topic I think for at least 2 years if not longer, well before his first publication on the matter. . . .well before he announced a publication would be forthcoming.

      Roy has been looking intensely at cloud feedback but is doing it on a far larger and longer term scale.

      As I recall he has noted effects that roll-out over months. Yet even Roy has to weather criticisms of how his short term analysis is way too short and a whole kit bag of other auxiliary criticisms that even run to his religious beliefs.

      So I wouldn’t take his comments as meaning you are wrong just that climate science has replies to your observations like the mid-troposphere hot spot suppressing convection and the whole enchilada of stuff that gets crammed into climate models.

      So while I enjoy reading much of your thought provoking work this does not seem to have been one of your better efforts.

      I have been trying to understand this myself for years. All I ever see are weak forcings in need of positive feedbacks to explain history. Are we just completely missing something all together? Whatever! Svensmark remains in need of quantification and its going to have to survive feedbacks as well. . . .the more negative the greater the effect is going to need to be.

      My opinion means nothing but it does seem there is more than one forcing and perhaps more than 2. Permanent feedback could well be zero.

      Ultimately its clear that argumentum ad ignorantiam is a damn poor excuse for telling folks how to live their lives.
      (obligatory political statement!)

  13. Willis Eschenbach says:

    Dr. Roy, one final comment. You start by referencing Ramanathan and Collins 91. Here’s what that paper’s abstract says:

    Observations made during the 1987 El Niño show that in the upper range of sea surface temperatures, the greenhouse effect increases with surface temperature at a rate which exceeds the rate at which radiation is being emitted from the surface. In response to this ‘super greenhouse effect’, highly reflective cirrus clouds are produced which act like a thermostat shielding the ocean from solar radiation. The regulatory effect of these cirrus clouds may limit sea surface temperatures to less than 305 K.

    Please note that the governing mechanism I propose (the daily sequential appearance of thermally regulated emergent phenomena) is totally and completely different from the feedback mechanism which they propose (an increase in cirrus clouds). As a result, R&C91 has no direct relevance to my work. Nor do the objections to R&C91 have any direct application to what I’m saying.

    Thanks,

    w.

  14. Stephen Wilde says:

    Willis asked:

    “How is a climate model with decreasing clouds with warming going to show increasing clouds with warming? I’m not following that.”

    That is a good question.

    The observations show that global cloudiness decreased during the late 20th century warming period and has been increasing since around 2000 with a cessation of warming and possibly cooling now in progress although the anomaly remains positive for the time being.

    Warming happens when there are less clouds and cooling when there are more clouds.

    That is why one needs a separate driver for the change in global cloudiness and I propose a change in the global air circulation caused by a solar effect on the gradient of tropopause height between equator and poles giving rise to changes in jet stream zonality / meridionality and the latitudinal shifting of climate zones.

    That process seems to override all else by affecting global albedo and the amount of solar energy able to penetrate the oceans.

    There is still a negative system response partly accounted for by Willis’s variant of a thermostat hypothesis involving the tropics but I aver that it is a global system response not just a tropical one and it doesn’t just involve clouds because it is the solar forced change in clouds that initiates the process in the first place.

    What one needs to consider for the negative syatem response is the additional outgoing longwave radiation from beneath the clearer skies under the expanded subtropical high pressure cells when the jets are more zonal and the climate zones more poleward.

    The truth is that the solar caused reduction in cloudiness allows both more energy into the oceans AND more energy radiating out from the surface. The increase in the speed of the hydrological cycle mops up any remaining differential but the clouds generated by the faster hydrological cycle in the tropics are not the primary negative response. The primary negative response is more outgoing radiation from clearer subtropical skies.

  15. John Olson says:

    There seems to be a problem with the diagram depicting warm air rising (and forming cumulonimbus clouds on the left) and thereby forcing cooler air to fall (on the right). I don’t follow this reasoning. Isn’t the warm air rising really mostly water vapor at 18 g/mol? Isn’t the diagram combining both micro- and macro-effects, and implying they are somehow related to one another?

    Isn’t condensation of water vapor molecules onto cool CO2 molecules and other atmospheric aerosols the equilibrating phenomenon for rising H2O? As this microscopic condensation occurs (forming what we see as clouds) the partial pressure differential is relieved and latent energy is liberated – upper tropospheric air is thus warmed in situ, not by mixing with warm air “masses” rising up from the surface.

    If a significant amount of mixing occurred during cumulonimbus formation, then those clouds would not maintain clear visual definition from a distance.

    As an analogy, consider ocean waves – the mass of water does not move in the direction of the wave: individual molecules move up and down while the wave is transmitted laterally. In our analogy for cloud formation, the water molecules move vertically to the upper troposphere while the “mass” of air molecules move laterally back and forth. There is some transfer of energy from water vapor molecules through inelastic collisions with nitrogen and oxygen molecules, but this transfer of energy is negligible in comparison with the latent heat of vaporization. This latent energy is not actually released until the water vapor condenses to form a cloud droplet.

    Furthermore, the N2 & O2 warming due to those inelastic collisions with H2O is balanced by the cooling of the surface water from evaporation. That is, the conductive exchange between surface water and surface N2 & O2 help keep surface water from cooling as much as it would have strictly due to evaporative processes, and thereby helps maintain surface temperatures on short time scales.

    Perhaps understanding suffers by extending observations of what happens over land surfaces to what happens over ocean surfaces? Land surfaces heat the surface air above them through conduction. There is relatively little H2O to evaporate, and therefore, the vertical circulation depicted in the diagram must occur (minus the cloud formation). This also happens over ocean surfaces on a global scale (not specifically annotated as such in the diagram), but does not apply to local storm systems.

    Evaporation of H2O at the ocean’s surface is a direct result of solar radiation. Any localized warming under cumulonimbus clouds would be reduced by evaporative cooling, and therefore, there is very little vertical air movement (N2 & O2) associated with the formation of clouds.

    What one observes in the eye of a hurricane is significant injection of water vapor through a vertical column of air, which causes a large localized pressure drop – but ONLY because that water vapor has not condensed in the eye of the hurricane. This forms the low pressure nucleus of the hurricane.

    Am I missing something?

  16. MikeN says:

    Is Mark Cain involved with this paper. I have heard Mike Mann reference a Pacific Thermostat Hypothesis, as to why the Medieval Warm Period was not global. You got LaNina like effects in the tropics. He was asked if this sort of long term negative feedback meant that climate models ‘vastly overstate’ warming, and said he agreed with that, after first joking the questioner didn’t look like Dick Lindzen.

  17. Willis Eschenbach says:

    Bill Hunter says:
    August 23, 2013 at 7:07 PM

    Willis if I read this correctly Roy’s criticism is the problem is not as simple as you are suggesting.

    So my impression is he is not disrespecting your observation he is simply being critical of the conclusion regarding where the missing heat is.

    Sorry, but I don’t see anywhere that Dr. Roy said anything, either positive, negative, or neutral, about missing heat …

    Roy has been looking intensely at cloud feedback but is doing it on a far larger and longer term scale.

    As I recall he has noted effects that roll-out over months. Yet even Roy has to weather criticisms of how his short term analysis is way too short and a whole kit bag of other auxiliary criticisms that even run to his religious beliefs.

    So I wouldn’t take his comments as meaning you are wrong just that climate science has replies to your observations like the mid-troposphere hot spot suppressing convection and the whole enchilada of stuff that gets crammed into climate models.

    Bill, so far Roy has just claimed that “climate science has replies” to my observations, but he has not provided those replies. Unfortunately, the two studies he cited as replies are discussing a totally different phenomenon (an increase in the amount of cirrus clouds when the ocean gets very hot). They have nothing to do with my hypothesis, which is that emergent climate phenomena control the temperature, and that the control is exerted inter alia by the timing of their emergence. Where is Dr. Roy’s reply to that?

    Dr. Roy also make a curious claim. This is that:

    The problem gets even more complicated because the temperature inversion in the above cartoon — the result of subsidence-warmed air overlying cooler ocean waters — also causes marine stratus clouds to form, which have a strong cooling effect on the climate system as a whole.

    I think he just means that marine stratus form when there is a temperature inversion, not an inversion resulting from the situation in his cartoon … because thunderstorms (AFAIK) are not a cause of marine stratus, other than by way of the Hadley Cell circulation. But his meaning not clear at all, and it has little to do with my analysis in any case. And that concluded his “replies” …

    As a result, Bill, so far Dr. Roy has provided absolutely nothing to back up his claim that I’m wrong … and it seems evident from the content of the works he has cited that he doesn’t understand my hypothesis. Don’t misunderstand me, he may indeed have answers to all my questions, and I hope he does.

    So although your “impression” of how he might answer me is interesting, I think I’ll wait for Dr. Roy to get back to me with his answers on this one.

    All the best,

    w.

    • Bill Hunter says:

      With all due respect Willis, the answer Roy gave is probably sufficient. No it does not “prove” you to be wrong. You need to first offer a falsifiable hypothesis to prove you wrong.

      I am constantly suspicious of “chaotic” explanations. It seems to me saying a system is chaotic is merely a statement that the whole danged system is so complicated I can’t understand it. Its strongly circular, if I can’t understand it it must be chaotic. It has the strong odor of egomania.

      Emergent climate phenomena very clearly have local effects, but you have not made a case that those effects are extensible to the entire climate system. I pointed out in my original response that you noted how a warm morning led to a cold morning the next day and vice versa. You did not eliminate that effect in your subsequent discussion. That’s just one example.

      Yes you did wave your arms and claim in your piece on emergent climate phenomena that it was not cyclical but you have never directly dealt with the issue as near as I can tell.

      So you tell me how somebody goes about proving you wrong? Do they have to compute out all those countervailing notions you never quantified in the first place and prove a non-chaotic system exists? If so then all you are doing is preaching.

  18. Mark Bofill says:

    No desire to clutter, but I’ll say I’m greatly interested in this discussion. Looking forward to Dr. Spencer’s response.

  19. nutso fasst says:

    Central Arizona during the monsoon season is a great place to watch convection. On a clear, hot day we can watch cumulus clouds form. Some fizzle quickly, dropping their moisture as virga. Others grow up, up, and out into cumulonimbus, creating low pressure beneath them that draws hot air from the surrounding area toward the storm. Then they dump their moisture, along with cool downdrafts. Temperatures drop dramatically in minutes.

    Hot air up, cool air down. Where is the heat going? Don’t the computer models say it should be warming the upper troposphere?

    Direct measurements say it isn’t. But Robert J. Allen and Steven C. Sherwood published a paper that apparently shows that if wind is used as a proxy for temperature, it can be demonstrated mathematically that warming is occurring. Their computer model shows that climate computer models are, in fact, reasonably correct.

  20. Willis will always reach the conclusion he is correct and every one else is wrong.

  21. In a nut shell Dr. Spencer and others are correct in that clouds create a negative feedback not a positive feedback, and that conclusion is what matters in regards to how the climate will change going forward.

    Further evidence as Stephen Wilde, presents shows that above conclusion seems to be correct thus far.

    Willis is trying to hard to pinpoint this issue and not looking at the broad overall picture.

    His approach will not work in the long run.

  22. TonyB says:

    John Olson:

    “Isn’t the warm air rising really mostly water vapor at 18 g/mol? ”

    No…http://en.wikipedia.org/wiki/Vapor_pressure

    …..air at sea level, and saturated with water vapor at 20 °C, has partial pressures of about 23 mbar of water, 780 mbar of nitrogen, 210 mbar of oxygen and 9 mbar of argon.

    “Isn’t condensation of water vapor molecules onto cool CO2 molecules and other atmospheric aerosols the equilibrating phenomenon for rising H2O?”

    WV doesn’t condense onto CO2 molecules as they are not hygroscopic nuclei. Salt crystals are most common over ocean.
    Equalising? No, latent heat is released into the updraft to maintain it.

    “upper tropospheric air is thus warmed in situ, not by mixing with warm air “masses” rising up from the surface.”

    No, some is warmed by mixing with the rising updraft but most is pushed aside (adiabatic).

    “If a significant amount of mixing occurred during cumulonimbus formation, then those clouds would not maintain clear visual definition from a distance.”

    A little mixing occurs but is swamped by the rising thermal. Air being pushed aloft above the thermal is cooled and if relatively dry will achieve a lower temp at a specific level than the thermal and so increase instability (potential instability) – do some basic plots of air temp/dew pt at a level on a Tphigram/skewT and raise it aloft, Plot the Normand’s pt. The drier the air the more convective energy is availanble (CAPE).

    “…. individual molecules move up and down while the wave is transmitted laterally”
    No, that analogy does not apply.

    “This latent energy is not actually released until the water vapor condenses to form a cloud droplet.”
    Correct.

    “Any localized warming under cumulonimbus clouds would be reduced by evaporative cooling, and therefore, there is very little vertical air movement (N2 & O2) associated with the formation of clouds.”
    Warming when in shade? The Cb will draw in air from it’s sides, where there is still solar warming. No, there is no reduction of ascent in a mature cb, dependant on the positioning of the updraft/downdraft, as the two can cancel each other or conversely +ve’ly feedback ( dependant on the upper wind strength/shear).

    “….which causes a large localized pressure drop – but ONLY because that water vapor has not condensed in the eye of the hurricane. This forms the low pressure nucleus of the hurricane.”

    Err – a hurricane has a warm descending core – that is the reason there is a “clear” eye. Inflow at the surface is anti-clockwise and switches to clockwise as the thermal winds ( blowing from warm to cold then turned right by Coriolis ) turn around as the horizontal thermal gradient changes with height.

    “Am I missing something?”

    Apparantly so – forgive me – I’m an ex UKMO Forecaster.

  23. Christopher Game says:

    Dr Spencer posts a picture of a convective circulation in the troposphere. It shows a circulation with four limbs: an expanding-ascending adiabatically work-driven temperature-lowering, a lofted radiatively cooling transverse, a contracting-descending adiabatically work-driven temperature-raising, and a near-ground conductively warming transverse limb.

    That circulation is part of the basic explanation of the tropospheric lapse rate in terms of the adiabatic gas law, as I will now describe.

    According to the second law of thermodynamics, friction and turbulence resist the circulation and would stop it but for an external driving factor that sustains it. That external driving factor is the other part of the basic explanation of the tropospheric lapse rate.

    What is the external driving factor?

    It is daytime direct sunlight heating of land or sea.

    Does their night-time cooling by radiation to space nullify this? No, the heating wins because it is driven by a high temperature source, the sun. Cooling to space is driven by a lower temperature source, the earth, and a lowest temperature sink, outer space. So the cooling lags the sun’s heating effect.

    How does the circulation work?

    Initially, at some place where the ground (land or sea) is strongly heated by the sun, an index parcel of air near the sun-heated ground is heated by conduction, so that its temperature rises and it expands. This heating is how the convective circulation is supplied with its driving energy that originated in the sun.

    The expansion of the index parcel does work on the air above it, driving it upwards against gravity, supplying it with the necessary bulk potential energy, and with kinetic energy of bulk flow. This expansion follows the ideal gas law because it is driven by conductive heating of the air.

    The air that is so driven up finds itself in a lower pressure environment. It expands laterally, doing work on the air beside it. This expansion is nearly adiabatic, without heat transfer by conduction. (Some small amount of radiative heat transfer occurs, but this effect is not as rapid as the expansive effect, and is neglected when one speaks here of “adiabatic” expansion.) It lowers the temperature of the upwards-driven air according to the adiabatic gas law. The adiabatic gas law is just the ideal gas law specialized to describe an adiabatic expansion.

    This is the expanding-ascending limb part of the basic explanation the lapse rate.

    This adiabatic expansive work is accepted by the laterally lying air as kinetic energy of bulk flow, as the lofty air moves away from the column of rising air.

    If it were not for the fact of external driving by the high-temperature sun and by low-temperature outer space, the second law of thermodynamics would tend to see the air temperature become uniform with respect to altitude, with zero lapse rate. Along with friction and turbulence, this is the main import of the second law. The sun’s heating effect outpaces the second law, as it describes passive effects on earth. This is a non-equilibrium scenario; nevertheless, what is known as local thermodynamic equilibrium describes it to good approximation and much of it can be accounted for near enough by concepts of equilibrium thermodynamics.

    Above the index parcel of ground-heated air, the upward-driven air rises and expands and lowers its temperature. It consequently dries itself by forming clouds, as noted by Dr Spencer’s post, but this is not the focus of the present note.

    The lofting air drives the previously lofted air beside it to move transversely, away from the place of original heating. Other than in a local storm, the lofted transversely-moved air goes to a place many kilometres away, being cooled by radiation to space as it goes. This radiatively-cooled laterally-moving air contracts without much conduction of heat to its surroundings. The contraction is supported by pressure from the air following it in the circulation. Driven by radiative heat loss, the contraction is approximately described by the ideal gas law, not the adiabatic gas law. (In a local storm, some of the risen air descends locally before it has had time to cool to space; this is a transient local non-linearly-contained-intensity positive-feedback mechanism for cooling the ground, because it transiently speeds up the local convective circulation.)

    The near-ground index air parcel that is initially conductively heated by the ground has a resulting increase of temperature. So it becomes less dense. It is buoyant with respect to less heated near-ground air beside it, and so it rises. That less heated near-ground air beside it moves transversely to the place under the initially heated but now rising index air parcel, and so is exposed to the original more heated ground. This is part of a cycle of convective circulation.

    Previously descended near-ground air moves transversely, replacing the initial index parcel of ground-heated and eventually rising air. The transverse movement is driven by the descent of previously-lofted and radiatively-cooled and less buoyant air. That less-buoyant air came by transverse movement, aloft, from an area of rising air, perhaps the same area as our original parcel of heated air, perhaps many kilometres away.

    That descending air is moving into an environment of higher pressure, and, without conductive heat loss, is adiabatically compressed, giving it a pressure-volume energy gain, work being done on it by the transversely and downwardly moving air following it, which also imparts its bulk gravitational potential energy, and kinetic energy of bulk flow. The adiabatic compression follows the adiabatic gas law. This raises the air temperature, though not by heat being added to it.

    This adiabatic compression is the contracting-descending limb of the basic explanation of the lapse rate.

    The two adiabatic limbs, expanding-ascending and contracting-descending, explain why the lapse rate follows the adiabatic gas law. The cycle is driven by radiative influx of solar energy.

    The energy added by the conductive ground heating of the initial index air parcel is fully lost from it within the cycle. That energy has driven the air parcel around the cycle. It has been lost by radiation to space.

    In the commonest case where there is no strong temperature inversion in the troposphere, the index air parcel has also been slightly heated by radiation from the ground, and that small amount of energy has also been radiated to space. In this commonest case, that slight transfer of energy as heat from ground to air is one-way in the thermodynamic formalism, though its mechanism is as a small residual dynamic imbalance between two component actual large oppositely sensed radiative transfers. (An increase in this slight imbalance of radiative transfers is the initiating mechanism of the effect of added carbon dioxide on the climate.)

    In the near-ground transverse limb of the circulation, the now-descended air is gradually heated by conduction from the ground. This heating causes expansion of the transversely-moving air, according to the ideal gas law, not the adiabatic gas law. The air is carried to a place of further strong ground heating, and the cycle is thereby completed. But its expansion is also imparting kinetic energy of bulk flow on the air in front of it, the index parcel of air.

    Air descending from aloft, because it was previously dried on its way aloft, does not allow cloud formation within it, as noted by Dr Spencer’s post, but not the focus of the present note.

    Dr Spencer’s post observes that the result is that cloudiness is not distributed in a simple two-dimensional Poisson random array. Instead, more or less well-defined clouds form preferentially in an area of ascending air, which they cover largely, and leave large cloud-free gaps in between, in areas of descending air.

    Dr Spencer’s post says that the exact degree of balance between the preferred cloud-forming area and the large cloud-free gaps is hard to understand or predict with desired precision. It says that a simple picture is not enough to account precisely for cloud formation and destruction as a climate feedback mechanism. Indeed, it says that no one has so far provided a precise and reliable account of it.

    This basic circulation is enhanced by transfer of internal energy due to moist processes. But the basic circulatory structure is established by the convective cycle I have just outlined.

    (I have here used the word circulation not to refer to the rigorous technical meteorological circulation that expresses the conservation of angular momentum due to the rotation of the earth. I have used the word circulation here to refer only to the convective cycle that I have outlined. Rigorous technical meteorologic terminology speaks of advection of various kinds, where I have spoken of convection in the ordinary-language sense.)

    • Good attempt Christopher.

      I covered the same ground here:

      http://www.newclimatemodel.com/the-ignoring-of-adiabatic-processes-big-mistake/

      and took it to the logical conclusion which is that adiabatic processes change as necessary to negate any thermal effects from changes in atmospheric composition.

      It is the balance of PE and KE in the vertical column which varies to achieve the required outcome.

      Although all my work is now archived at newclimatemodel.com it was originally published at climaterealists.com

      • Christopher Game says:

        Now I have looked at your link above, Stephen.

        In it I find the following paragraph.

        “Essentially, an adiabatic process is one where temperature changes can occur without addition or removal of energy and a diabatic process is one where temperature changes can only occur as a result of the addition and or removal of energy.”

        No. An adiabatic process in a closed system is one in which there is no transfer of energy as heat. The energy transfer that changes the temperature in an adiabatic process in a closed system is transfer of energy as work. In the present case, the closed system is nominated as a parcel of air that moves in the circulation. You can check this in any reliable textbook on thermodynamics.

        • Transfer of energy as work involves no addition or removal of energy. The process of work simply changes part of the energy present to heat.

          An example is the transfer of KE to or from PE and the work involved is the rising against or falling with gravity of the parcel of air.

          I thought your main post had that about right.

          Maybe a semantic issue?

          • Christopher Game says:

            I am using the long-and-well-established semantics of orthodox thermodynamics. In this context, transfer of energy as work means passage of energy from the system to its surroundings. The system loses an amount of energy and the surroundings gain the same amount.

            You are confabulating your own semantics, it seems. Your usage here seems to say that you regard ‘transfer’ as meaning ‘conversion’, but that you do not have in your mind a clear idea of what is involved or how it might happen. I think it boils down to your not understanding the basic thermodynamics and physics, but being unaware that you don’t.

            In just this very context I pointed this out to you before, and I suggested you read a textbook, but evidently you haven’t done so. Though I tried to express it quite explicitly in my present post, it seems your prejudiced and mistaken view made what I wrote slip right past you. You write KE and PE instead of kinetic energy and potential energy, as if the ideas are very familiar to you. But, rather than being familiar with the physics, you are missing an important and elementary point here.

            I am not going to spell it out for you. To understand it, you will have to read a textbook, or otherwise work it out for yourself.

        • “Convection of internal energy is a form a transport of energy but is in general not, as sometimes mistakenly supposed (a relic of the caloric theory of heat), a form of transfer of energy as heat, because convection is not in itself a microscopic motion of microscopic particles or their intermolecular potential energies, or photons; nor is it a of transfer of energy as work.”

          from here:

          http://en.wikipedia.org/wiki/Work_(thermodynamics)

          If convection is not a transfer of energy as work then Christopher’s point seems to be wrong since he said (in relation to convection):

          “The energy transfer that changes the temperature in an adiabatic process in a closed system is transfer of energy as work.”

          In fact where convection is concerned it is NOT a transfer of energy as work.

          Thus as per my point it is simply a conversion between different states of energy i.e. kinetic energy (KE) and gravitational potential energy (PE).

          • Christopher Game says:

            The inner workings of a convective circulation are outside the scope of classical equilibrium thermodynamics narrowly construed. Nevertheless, one can get a good idea of how the circulation works by considering snapshots of parts of it with the ideas of equilibrium thermodynamics. One treats the snapshots as if they represent transfers that can be described by equilibrium thermodynamics. This relies on what is usually called the approximation of local thermodynamic equilibrium. It is near enough for the present purpose.

            We are here interested in why the adiabatic gas law is invoked to describe the convective circulation in the troposphere.

            We focus on transfers of energy that involve four systems. (a) the sun-heated ground under an index parcel of air. (b) an index parcel of air, sitting on the sun-heated ground. (c) a parcel of air above that index parcel; let us call it the upper parcel. (d) a parcel of air that lies beside the upper parcel after the upper parcel has been driven upwards.

            (I) First we may consider a snapshot that covers the sun-heated ground and the index parcel of air. The sun-heated ground passes energy to the index parcel by conduction, as heat. This is not an adiabatic transfer. But in our treatment, it leaves us with a hot index parcel of air.

            (II) Next we may consider a snapshot that covers the index parcel of air and the upper parcel. The index parcel expands because of its being heated by the ground. The expansion presses upward on the upper parcel. The upper parcel is made to move upwards by this pressure. In this way the upper parcel gains kinetic energy of bulk flow and potential energy of position of centre of mass in the gravitational field of the earth; these transfers are adiabatic. Though the transfer of energy as work from the index parcel to the upper parcel is adiabatic, it is not the key adiabatic transfer for which we shall invoke the adiabatic gas law. Also, the pressure might tend to compress the upper parcel, but this is near-enough negligible in this case.

            An adiabatic transfer is just a transfer of energy as work. It does not allow transfer of matter and does not allow transfer of energy as heat.

            (III) Next we consider a snapshot that covers the upper parcel and the parcel that lies beside the upper parcel after the upper parcel has been driven upwards. In the transfer of interest in this snapshot, the upper parcel expands because it has reached an altitude of low pressure. In expanding it does work on the parcel that lies beside it. This is an adiabatic transfer, and is the transfer for which we shall invoke the adiabatic gas law. The approximation here is that the expansion is relatively rapid, so rapid that there is hardly any time for heat to be transferred by conduction or radiation during it. In this snapshot, the adiabatic work done by the upper parcel is pressure-volume work. The transferred energy is taken in by the parcel beside it as kinetic energy of bulk flow; this is an adiabatic transfer because it does not transfer matter or energy as heat. The parcel beside it is driven to move transversely, which endows it with kinetic energy of bulk flow.

            In this account, many of the basic ideas are those of classical equilibrium thermodynamics. Classical equilibrium thermodynamics gains its strength because it concerns itself only with macroscopic processes. It therefore deliberately avoids explicit mention of microscopic molecular processes. There are three principal kinds of energy in classical equilibrium thermodynamics. (1) internal energy. (2) kinetic energy of bulk flow or motion, motion of the centre of gravity of the system. (3) bulk potential energy, potential energy of the whole parcel due to the position of its centre of gravity in the earth’s gravitational field.

            The transfer (I) increases the internal energy of the index parcel, and raises its temperature. This causes it to expand against the pressure of its surroundings. The transfer of energy by conduction of heat from the ground is not adiabatic.

            The transfer (II) between the index parcel and the upper parcel is adiabatic. It removes energy as pressure-volume work from the index parcel. It supplies energy to the upper parcel as kinetic energy of bulk flow or motion, and as bulk potential energy in the gravitational field.

            The transfer (III) between the upper parcel and the parcel beside it is adiabatic. It removes energy as pressure-volume work from the upper parcel and supplies it to the parcel beside it as kinetic energy of bulk flow. This pressure-volume work is the adiabatic work that is described by the adiabatic gas law, as it is of interest to us here.

            This transfer (III) is the reason why the adiabatic gas law is invoked in accounting for the convective circulation in the troposphere and it is the reason why the tropospheric temperature lapse rate is basically described by the adiabatic gas law.

            The kinetic theory of gases is not directly or explicitly invoked in the above account. This is because we are looking at a macroscopic process, and macroscopic thermodynamics is very suitable to describe it, providing the simplest accurate account. Talking of the kinetic theory of gases would just distract attention from the important macroscopic aspect of the phenomena. The kinetic theory of gases is primarily about microscopic phenomena.

            The kinetic theory of gases is an approximate theory that explains many results of thermodynamics. In contrast to thermodynamics proper, narrowly construed, the kinetic theory of gases is largely about the microscopic motions of molecules. The microscopic motions of molecules are reckoned relative to the centre of gravity of the air parcel. They have a mean kinetic energy of microscopic motion that is closely related to temperature. The kinetic energy of the microscopic molecular motions is quite distinct from kinetic energy of bulk flow. The kinetic energy of bulk flow is a macroscopic expression of the motion of the centre of energy of the air parcel.

            The thermodynamic quantity, the internal energy, is a macroscopic manifestation of the microscopic motions and interactions of the molecules. It represents two parts. One is the microscopic kinetic energy of the molecules reckoned relative to the centre of mass of the air parcel. This is the main concern of the kinetic theory of gases. The other part of the internal energy is the potential energy of interaction between the molecules. This is very hard to calculate, and is not considered in simple or approximate versions of the kinetic theory. Nevertheless, it is the reason why the kinetic theory can differ from the thermodynamic account. The kinetic theory makes approximate assumptions about models of molecular interactions. But the thermodynamic account avoids such modelling assumptions; that is one of its virtues.

  24. Dr. Roy Spencer says:
    “all of that rising cloudy air over the warm pool, there has to be sinking air elsewhere which suppresses cloud formation”

    Clouds are not the end of story. Even if relative humidity happens to be well below 100% in descending air masses, so cloud formation is suppressed everywhere, distribution of absolute humidity is still uneven. It is so, because absolute humidity of an air parcel does not depend on its immediate surroundings (both in space and time), but on its history, that is, on its temperature (pressure, elevation) the last time it has got saturated. In turbulent flows this history can be very different for adjacent parcels, moreover, shape of any parcel, originally bulky, tends to get distorted into a mesh of fine filaments with time. The net result is water vapor distribution is fractal-like even in air masses descending on average.

    There is no doubt the “greenhouse effect” is determined by IR optical thickness, not average concentration of GHGs. For well mixed greenhouse gases (like CO₂) the two definitions are interchangeable, but it is not so for not well mixed ones, e.g. for water vapor (the dominant GHG in the atmosphere). In that case average concentration only gives an upper bound for IR optical thickness provided by that particular component (to be attained at uniform distribution), with no obvious lower bound whatsoever.

    A thin metal plate may be completely opaque, but a wire fence, using the same amount of metal per unit area is pretty transparent.

    Do we have any data on how fractal dimension of water vapor distribution changes in regions of descending air with increasing well mixed GHG concentration? Or how higher moments of absolute humidity distribution are changing?

    It is crucial in determining how overall average IR optical thickness of the atmosphere is changing with increasing CO₂ concentration, still, I have not seen it discussed anywhere in the climate science literature.

    Some input or at least a pointer would be appreciated.

    • Didn’t Miskolczi find that the optical thickness stayed the same when CO2 increased?

      The reason seemed to be that other features of the atmosphere changed to negate the effect of the CO2 on optical thickness.

      He referred to water vapour amounts but I think the more direct cause would have been circulation changes. Those circulation changes would have had an effect on water vapour amounts and distribution.

      In my view the combined effects of gravity, atmospheric mass and insolation will always result in the same optical thickness regardless of composition and the mediating factor is changes in volume and circulation pursuant to the Ideal Gas Laws.

  25. Bart says:

    The problem with the studies which I have seen which purport to show overall positive feedback is that they are based on simplistic analysis of scatterplots, with assumed instantaneous response. But, phase delays can create apparent positive slopes in such plots even when the feedback is most assuredly negative. More appropriate analysis reveals that the cloud albedo feedback is decidedly negative (180 deg phase shift at low frequencies).

  26. Christopher Game says:

    Bart, that’s an interesting post. I would like some detail on this “More appropriate analysis”. How was it done?

  27. Bart says:

    Christopher, you can see the entire shooting match (literally, almost 😉 here.

    • Christopher Game says:

      Thanks, Bart. I would retain a worry that the data might be largely governed by drivers external to the variables of your analysis, temperature and cloud effect. The external drivers might introduce a quasi-periodic signal component not determined by random variations in temperature and cloud. A component that might swamp those random variations and dominate the response analysis? Perhaps you have considered that?

      • Bart says:

        Sure, that’s a question. I am basically assuming a high S/N in the relationship, and assumptions always color the analysis.

        However, this is the same data people like Dessler have used to try to claim a positive feedback, and the data clearly do not support that.

        Certainly, the way the temperature data have behaved of late does not tend to support a positive feedback.

  28. Stephen Wilde says:

    Christopher Game.

    “An adiabatic process is a process occurring without exchange of heat of a system with its environment. It is the opposite of a diabatic process”

    from here:

    http://en.wikipedia.org/wiki/Adiabatic_process

    There is no significant transfer of energy as heat from the Earth’s atmosphere to its surroundings (ground or space) when air rises and falls so I fail to see your point.

    There is only a transfer of energy between KE and PE within the vertical column so the vertical column represents a closed system and only KE registers as heat.

    Your original post appeared to appreciate that.

  29. Stephen Wilde says:

    Christopher Game said:

    “An adiabatic process in a closed system is one in which there is no transfer of energy as heat. The energy transfer that changes the temperature in an adiabatic process in a closed system is transfer of energy as work.”

    Those two sentences appear to be contradictory if by ‘transfer’ you mean passage of energy from the system to its surroundings.

    Either it is adiabatic with no such transfer or it is diabatic with such transfer.

    Yet you say adiabatic in both situations don’t you?

    The reality is that any work done in the process of air rising against gravity or falling with gravity only converts energy to and fro between KE and PE and does not transfer energy in or out of the closed adiabatic system.

    Meanwhile solar energy gets a free pass straight through so as to balance the energy in with energy out at ToA.

  30. Christopher Game says:

    Responding to the posts of Stephen Wilde of 12:02PM and of 12:16 PM on August 28, 2013.

    Dear Stephen, I wrote just above that I am not going to spell it out for you. Again I suggest you read a reliable textbook or work it out for yourself. The Wikipedia is not a reliable textbook. In this case it didn’t help you work it out for yourself. You still persist in feeling that rhetoric or speciously logical manipulation of word strings can take the place of physical understanding.

    • You simply don’t understand my point otherwise you would not categorise my words in such a way.

      There is nothing in basic thermodynamics that contradicts the scenario that I suggested. As far as I can see basic thermodynamics and established science support it.

      If you have a sound point then you should set it out for the benefit of other readers.

      • Christopher Game says:

        Dear Stephen, I have set it out above http://www.drroyspencer.com/2013/08/on-the-cloud-thermostat-hypothesis/#comment-86819 at perhaps already excessive length in the usual language of physics and thermodynamics. I do not think I would benefit other readers by writing even more. The problem here is that you do not understand, and at present have no intention of learning to understand, the basic ideas of thermodynamics as it is relevant here. As the inventor of a wonderful theory of climate, you feel it is beneath your dignity to learn basic thermodynamics from textbooks. When I correct you, instead of taking advantage of what I write by learning from it, you habitually reject it in favour of your homespun doctrines and confabulated semantics. So in explaining things to you it is of no avail that I write in the usual language of the subject; I am not about to adopt instead your homespun doctrines. You say that I don’t understand your “point”. The way I see it, in the light of basic thermodynamics, important “points” that you make are more or less nonsensical if read as they stand. It is mostly easy to see just how and why your thinking goes wrong, but it does so in many ways and places; the best course is for you climb down from your high horse and try to learn the basics from a textbook, rather than try to get me to pick up each of your errors one by one.

    • Kristian says:

      An air parcel at the surface receives energy from the surface as heat (originally by conduction). This means its internal energy increases. It is warmed and expands, rising as a result, gradually spending the energy supplied to it from the surface in the process … by doing work on its surrounding air masses. When the original extra amount of energy supplied from the surface is spent, the air parcel ideally stops rising – it has delivered its gained surplus energy to the atmosphere at large. On average, the atmosphere now contains more energy (is slightly warmer) than it was before the surface heated that parcel of air. This surplus will ideally build up during the day, simply because of the eternal lag between incoming solar energy and the cooling mechanisms of the atmosphere. During the night, the atmosphere/surface restores the energy balance before the next day begins.

      Stephen, the parcel of air does NOT contain as much energy at the top of the tropospheric column as it did when it was originally heated by the surface. This is NOT simply a matter of conversion from KE to PE. The parcel of air has lost its surplus energy by doing work on its surroundings.

      I’m quite sure this is what Christopher Game is trying to tell you …

      • Kristian,

        My understanding is that the energy content of a molecule at the top of the atmosphere is the same as the energy content at the bottom of the atmosphere.

        One must consider the ENTIRE atmospheric column, not just the troposphere.

        At the bottom it is all KE and at the top it is nearly all PE.

        It is not quite all PE at top of atmosphere because the temperature of space is not absolute zero so there will still be some KE.

        However for a molecule in contact with the surface there is no PE at all.

        The adiabatic process therefore is a closed system or as nearly so as makes no difference and so the work done in rising and falling only converts energy to and fro between KE and PE.

        If there were leakage of energy into or out of the adiabatic process then either the atmosphere would expand indefinitely or it would collapse to the surface.

        • Stephen Wilde says:

          Actually there is still some PE when a molecule is at a surface because it is still some way from the centre of the gravitational field.

          It is that remaining PE pressing on the surface that gives rise to surface pressure.

  31. Peter Garrone says:

    Perhaps the thermostat is in the upper/thermosphere/ stratosphere. The incremental ratio of surface heat transfered to temperature difference is probably quite large, because of the physical mechanism of convection. Once a certain temperature differential is reached, convection turns on, and that clamps the surface temperature.

    If a GHG such as CO2 is added, then the average height at which the heat is radiated directly to outer space would increase, and increase surface temperature. However there is a negative feedback because the extra GHG molecules help cool the stratosphere as they enhance radiative heat loss given external sources of heat, UV captured by ozone and convection from below.

    So whatever the cloud feedback is, perhaps it is overshadowed by some more significant feedback effect.

    • “If a GHG such as CO2 is added, then the average height at which the heat is radiated directly to outer space would increase, and increase surface temperature”

      It wouldn’t need to increase surface temperature if it causes a change in the lapse rate gradient along with the increase in height.

      Instead, one gets the same surface temperature and the change in the lapse rate away from that set by gravity and atmospheric mass causes more convection which increases the average height and adjusts the lapse rate slope instead of raising surface temperature.

      The higher point at which radiation is then radiated direct to space thus becomes warmer than it was before not colder (as proposed by AGW theory) and the rate of energy loss to space is faster which offsets the thermal effects of the GHGs.

      The effective radiating height would only be colder than before if that height could change without an expansion of the atmosphere but that is nonsense since additional energy in an atmosphere whether KE or PE always results in expansion.

      Expansion always results in more PE holding the atmosphere higher off the surface so the increase in PE (which is not heat) offsets the attempt to increase KE (which is heat).

      All that is in accordance with the Ideal Gas Laws.

      • Peter Garrone says:

        Looking at the aerological diagrams released by our national bureau of meteorology, it is obvious that lapse rate can be assumed to be constant for a small change in temperature at any particular operating point. Just google aerological diagram and look at the dashed background lines on the graph.

        Adding co2 will not make the planet cooler. It will warm it. How much depends on the feedback. I am suggesting that perhaps there is negative feedback from a cooling stratosphere, rather than cloud feedback, which is difficult to pin down.

  32. Kristian said:

    “it has delivered its gained surplus energy to the atmosphere at large.”

    I think that is the error.

    The parcel does not give up its energy to the atmosphere at large. The process of rising causes KE to convert to PE which results in cooling until it reaches the correct temperature for its height.

    There is leakage to adjoining air parcels by conduction which causes it to eventually stop rising and equilibriate with its surroundings but the other air parcels are also part of the adiabatic system so that doesn’t count as a transfer of energy out of the system.

    If one limits the issue to the troposphere it is actually the temperature inversion at the tropopause that stops the continuing rise and not conduction to adjoining parcels.

    For energy to be transferred out of the system means either back to the surface or out to space but that doesn’t happen as a result of the work done during uplift and descent.

    All that work done only converts energy to and fro between KE and PE with uplift always equalling descent over time for the atmosphere as a whole.