Anastassia Makarieva: Science: One Bar For All

Posted: January 27, 2013 by tallbloke in Analysis, atmosphere, climate, general circulation, methodology, Philosophy
Tags: , , ,

In my opinion, this comment by Anastassia Makarieva on the interactive part of the ACS website, as well as being a strong defense of their paper, is a powerful indictment of the state of affairs in the peer review of climate science. Given the grief of rejection, the tone is remarkably restrained. Instead, there is a channelling of energy into a righteous intensity which makes this a bit of a classic.  Use the first link for the full version with all footnotes included.
anastassia_makarievaInteractive comment on “Where do winds come
from? A new theory on how water vapor
condensation influences atmospheric pressure
and dynamics” by A. M. Makarieva et al.
A. M. Makarieva et al.
Received and published: 26 April 2011

Aside from our technical response to Dr. Held1 we also wish to discuss the criteria he
uses to assess our manuscript. All theories should be subjected to similar standards
of scrutiny regardless of whether they conform to conventional thinking or not. We find
many examples to show that much of the argument against our theory and in favour
of conventional ideas appears based on misconceptions. We conclude with an appeal
concerning the wider practical importance of our ideas.

1 High bar for unconventional findings
Dr. Held starts his review with the recommendation to reject our manuscript. He explains
that a study that goes against the standard perspective or aims to overturn the
conventional wisdom has to pass a high bar.

As science students we are taught about the sins of confirmation bias – that is the need
not to allow our preconceptions and judgements to cloud our objectivity. We should not
reject ideas, or data, that fail to conform to our expectations any more readily than
those we agree with. We all agree with that as an abstract idea though it can be hard
to achieve in practice. Biases are often hard to perceive for those who hold them
especially if they are pervasive. But we should strive for objectivity – when biases are
identified we must do what we can to remove or minimise them.

Dr. Held believes our theory has to pass a high bar because of the accumulated evidence,
implicit as well as explicit, that argues against it. Dr. Held does not spell out any
evidence at all (so it all remains implicit in this case), but it is presumably a statement
of his confidence that climate modelers already have a handle on the basic principles
on how the World’s climate works. Clearly we disagree with this. If there were such
evidence the scientific approach requires that it is presented to us so that we can address
it explicitly. When it is not specified we can perhaps be forgiven for believing that
the bar has been placed infinitely high: a serious case of “confirmation bias”. Such
biases must be questioned by all of us with a training in science (regardless of views
concerning our theory).

Dr. Held concludes his review by recommending that we should avoid appealing to
authorities. He is referring to our selected quotes by Brunt and Lorenz. Normally we
would agree that such quotes are out of place in a physics paper. However, in this case
we made an explicit choice as we felt that we needed them. Our appeals to authority
are intended to counter the (familiar) arguments based on implicit evidence: that our
theory is not needed, or wrong, or the bar should be raised just because climate scien-
tists already have all the theory they need and there are no major gaps and flaws. Our
point is that climate theory is full of basic questions and doubts and many respected
authorities say so. For us a particular value of the opinions of Brunt and Lorenz lies in
the fact that they were formulated prior to the era of computer modelling. Nowadays
we find simulation is too often allowed to replace physical understanding2. Targeted
coding of numerical models can successfully simulate numerous patterns, including
many prohibited by the laws of nature3.

2 Low bar for “conventional” findings?
A higher bar for unconventional ideas automatically implies a lower bar for conventional
ones. Introducing a positive feedback – relating the height of the higher bar to the
number of studies that have passed the lower bar – in time if this continues a once
vibrating scientific community can be trapped in dogma. An objective culture might be
eroded by overconfidence in its own monolithic vision. This may be especially likely
in climate sciences due to the considerable effort invested in presenting a confident
unified front to the outside world.

Attitudes to vapor sink dynamics illustrate our concerns. Water vapor condensation is
a ubiquitous process in day-to-day weather and climate processes. Nonetheless the
nature of the associated pressure gradients has never received a theoretical investigation.
These gradients are seldom mentioned (even to discount them) in the reviews of
the general circulation theory. We find that in the microphysical studies of the phase
transitions of water vapor the physical causes of condensation are usually neglected.
E.g., in the study of Vesala et al. (1997), characterized as a clear and accurate theoretical
description of condensation by Kolb et al. (2010), it is emphasized that any
global processes that induce the conditions required for condensation to occur are not
considered and that such conditions are assumed to be predetermined. Most existing
accounts of the vapor sink, like the study of Lackmann and Yablonsky (2004), are
based on empirically fitted numerical models and do not provide or allow for a transparent
physical interpretation. These simulation studies differ in their formulations, results
and outcomes. But we see that in climate science the outputs of models of varying
degree of complexity can contradict each other without triggering any discussion or

The few studies designed to investigate the vapor sink appear from our evaluations to
be based on unphysical assumptions, like the gravity defying levitation of liquid droplets
in motionless air5 in the study of Bryan and Fritsch (2002) or atmospheric warming by
spontaneous drying in the study of Spengler et al. (2011), as endorsed by Dr. Held
in his review6. Nevertheless, because these flawed results have been “conventional”
enough to pass over the bar, they are considered benchmarks or physicality tests for
getting other ideas over the same bar.

A paradox, considering the common claims of general consensus, is that we find that
even basic questions are controversial in climate science circles. This is a result of
limited attention and investigation. These are not idle claims. For example, does condensation
in a volume of atmosphere (near-) immediately lower pressure at the surface
or must one wait until the droplets precipitate out to the surface – views among climate
scientists differ on this basic question with a majority apparently believing the latter to
be true7. Recently a modelling study (Spengler et al., 2011) was required to illustrate
that, since gas obeys a different equation of state than liquid, changes of air pressure
and fall out of droplets occur on different time scales. Indeed, as condensation occurs
the air pressure is lowered at the surface near instantaneously (at the time scale defined
by the speed of sound and height of condensation). These points are all obvious
enough when viewed from the perspective of basic physics.

There are deeper misunderstandings, like confusing heat and work. Any spatially nonuniform
warming of the atmosphere can be brought back to equilibrium by heat transfer
(e.g., by radiation to space) without any work performed or dynamic flow generated. In
contrast, to compensate for gas removal from an atmospheric volume, mass can be
only resupplied there by performing work on that volume by way of its compression.
This fundamental difference between pressure gradients associated with heat versus
gas removal is ignored. E.g., Spengler et al. (2011, p. 358) believe that as heating is
offset by diabatic cooling there would be analogous compensation for drying (=vapor
sink). This confusion between heat and work is perhaps a key (pervasive) misconception
about vapor sink dynamics.

In summary, despite some claims by climate modelers at GFDL8 that the vapor sink
is comprehensively included in current global circulation models, it is unclear how this
could be achieved given the poor state of either theory or empirical understanding.

3 Science: One bar for all
In our paper, the pressure gradient due to the inherent spatial inhomogeneity of the
condensation/evaporation process is derived for the first time in the scientific literature.
It is further shown that in the atmospheric context this gradient is of sufficient magnitude
to be considered as a major driver of atmospheric dynamics.

The criteria to be applied in assessing the value of new scientific propositions are
reasonably simple. Is our proposition consistent in terms of its form and logic with
basic physical principles? We say “yes”. Does it make predictions that can be tested?
Again we say “yes”. Has anyone shown any error in the mathematical or physical
reasoning? Here the answer is, as far as we can see, “no” – there are valid differences
of opinion, and no shortage of misunderstandings, but no-one has shown an error. We
thus see no valid scientific argument for a “higher bar” and none has been presented. A
study should not be rejected (i.e. given a “high bar”) because its results are surprising
or because people find the equations hard to follow or because the reviewer believes it
is wrong but can present no evidence – these are not the criteria we should accept.
The validity of any new proposition in science is ultimately tested by empirical evidence.
In climate science the situation is peculiar in that the objects of interest often exist in
a single number – e.g., the Hadley circulation – and the underlying systems cannot
be modified at the discretion of the investigator. Unlike chemists or particle physicists,
climate scientists cannot easily set up and perform experiments on many of the phenomena
they study (see, e.g., Held (2005) for a discussion). In such a situation the
first test a new theory (in our case, the proposition that atmospheric dynamics is driven
by the vapor sink) can be expected to pass is to explain the existing evidence better
(i.e., with fewer empirical parameterizations and on a unified physical basis) than the
old one (in our case, the proposition that winds are driven by heat). In our present
work and other recent papers we offer a number of examples indicating that this can
be achieved (e.g., Makarieva and Gorshkov, 2010, 2011; Makarieva et al., 2009, 2010,
2011). One cannot expect a few people to fully elaborate a new theory and analyze all
available evidence in one paper. This is seldom how science progresses. We believe
that publication of our findings in the meteorological literature will encourage other climate
students to investigate the problems, test our propositions and reach their own

We also see an opportunity to study some aspects of condensation dynamics in the
laboratory. Condensing vapor allows for a one-dimensional flow between warm and
cold liquid surfaces. Vapor arises by evaporation at the warmer surface, flows towards
the colder surface and “disappears” (i.e. condenses) there. Such a one-dimensional
motion with zero velocity at both boundaries is by definition impossible for a noncondensable
gas in which mass is conserved. There is a number of studies, both
empirical and theoretical, where such phenomena have been investigated independent
of atmospheric sciences, see a recent review by Kryukov et al. (2010). In the Earth’s
atmosphere the surfaces where condensation occurs (i.e. on moisture droplets) do not
coincide with the upper boundary surface of the circulation but are distributed within
the atmospheric column allowing for macroscopic motions in all directions. Once our
theoretical propositions are seriously analyzed and studied, this may stimulate other
investigators to attempt replicating relevant processes in the laboratory.

Why bother? There are several reasons. Here we highlight one which we consider the
most important. If we are correct that it is principally the vapor sink that determines
large-scale atmospheric dynamics, then all changes in terrestrial vegetation (and associated
changes in terrestrial vapor dynamics) threaten drastic changes in seasonal
wind patterns and, associated climates (Sheil and Murdiyarso, 2009). Such modifications
are unrelated to other aspects of climate change such as those related to
greenhouse gases etc. – but will be inherently regional and threaten many human
populations9. Those who find these ideas unlikely should pause to consider the costs
of being wrong – how certain need they be to discount the risk? We do not have a
model South American or North American continent in the laboratory to test empirically
what happens if their forests are destroyed or degraded.

While the climate community strives to impose global regulations on humanity based
on their understanding of climate, climate scientists should not be afraid of investing
in the constructive evaluation of provocative results10. Despite considerable scrutiny
from large numbers of climate scientists over the last nine months, for which we are
grateful, we have not been shown to be wrong. We request that reviewers begin to
allow for the fact that we might be right. If true our theory has implications for the lives of
many people and for the global environment. For both scientific and moral reasons we
believe our ideas urgently need concerted objective examination by climate scientists.
To make responsible predictions and elaborate strategies that are compatible with longterm
human well-being, we need to ensure that the physical principles underlying our
shared understanding of atmospheric circulation are correct. This calls for scrutiny
not just of our propositions but of all the models that are currently utilised. We call
on climate scientists to ensure good open-minded science striving for objectivity and
insight – please place all bars accordingly.

Acknowledgements. We thank all people who discussed our work. Our special thanks are
due to Drs. Peter Belobrov, Paulo Nobre and Stefan Emeis, whose comments submitted to this
discussion did not require a response from the authors but were greatly appreciated. Upon the
closure of this discussion, further developments with regard to our paper can be monitored at
Bryan, G. H. and Fritsch, J. M.: A benchmark simulation for moist nonhydrostatic numerical
models, Mon. Wea. Rev., 130, 2917–2928, 2002.
Makarieva, A. M. and Gorshkov, V. G.: The Biotic Pump: Condensation, atmospheric dynamics
and climate, Int. J. Water, 5, 365–385, 2010.
Makarieva, A. M. and Gorshkov, V. G.: Radial profiles of velocity and pressure for condensationinduced
hurricanes, Phys. Lett. A., 375: 1053–1058, 2011.
Makarieva, A. M., Gorshkov, V. G. and Li, B.-L.: Precipitation on land versus distance from
the ocean: Evidence for a forest pump of atmospheric moisture, Ecological Complexity, 6,
302–307, 2009.
Makarieva, A. M., Gorshkov, V. G., Li, B.-L., and Nobre, A. D.: A critique of some modern
applications of the Carnot heat engine concept: the dissipative heat engine cannot exist.
Proc. Roy. Soc. A, 466, 1893–1902, 2010.
Makarieva, A. M., Gorshkov, V. G., Nefiodov A. V.: Condensational theory of stationary tornadoes,
Phys. Lett. A., in press, doi:10.1016/j.physleta.2011.04.023, 2011.
Held, I. M.: The gap between simulation and understanding in climate modeling, Bull. Am. Met.
Soc., 86, 1609–1614, 2005.
Kolb, C. E., Cox, R. A., Abbatt, J. P. D., Ammann, M., Davis, E. J., Donaldson, D. J., Garrett,
B. C., George, C., Griffiths, P. T., Hanson, D. R., Kulmala, M., McFiggans, G., Pöschl, U.,
Riipinen, I., Rossi, M. J., Rudich, Y., Wagner, P. E., Winkler, P. M., Worsnop, D. R., and O’
Dowd, C. D.: An overview of current issues in the uptake of atmospheric trace gases by
aerosols and clouds, Atmos. Chem. Phys., 10, 10561–10605, 2010.
Kryukov, A. P., Levashov, Yu. V., and Pavlyukevich, N. V.: Condensation from a vapor-gas
mixture, Journal of Engineering Physics and Thermophysics, 83, 679–687, 2010.
Lackmann, G. M. and Yablonsky, R. M.: The importance of the precipitation mass sink in tropical
cyclones and other heavily precipitating systems, J. Atm. Sci., 61, 1674–1692, 2004.
Sheil, D. and Murdiyarso, D.: How forests attract their rain: an examination of a new hypothesis,
Bioscience 59, 341–347, 2009.
Spengler, T., Egger, J., and Garner, S. T.: How does rain affect surface pressure in a onedimensional
framework? J. Atm. Sci., 68, 347–360, 2011.
Vesala, T., Kulmala, M., Rudolf, R., Vrtala, A., and Wagner, P. E.: Models for condensational
growth and evaporation of binary aerosol particles, J. Aerosol Sci., 28, 565–598, 1997.
Interactive comment on Atmos. Chem. Phys. Discuss., 10, 24015, 2010.

  1. tallbloke says:


    If you want to help raise the profile of the talkshop and improve our chances of getting sponsorship for the September conference, please nominate us for the Bloggies awards.

    Nominations close later today.

    Thanks for your help.

  2. Nick Stokes says:

    “Given the grief of rejection,”

    It actually wasn’t rejected, despite having no favorable reviews.

    Not exactly one bar for all.

  3. tallbloke says:

    Wotcher Nick. It was recommended for rejection by peer reviewer Isaac Held, who Anastassia was responding to here.

    Do you feel she successfully defended her paper against your questions in this reply:

    Or are there outstanding issues you feel were not addressed?
    If so, what are they?

  4. Nick Stokes says:

    The revised paper was never made available for discussion, by reviewers or anyone. And no, it doesn’t answer the criticisms.

    It doubles down on one bizarre claim – that precipitation rate is proportional to humidity (eg appendix A1). Now there is always some humidity, but it isn’t always raining. This is just one claim that in Held’s terms might be given a little extra attention.

    In a technical sense the claim may have a basis, but in that sense it’s quite irrelevant. You can see this by the fact that they never say what the first order kinetic rate is. It’s very fast. What determines precipitation rate is not the molecular kinetics of phase change but the rate at which moist air becomes saturated, usually by being cooled.

  5. Konrad says:

    well of course you have got my nomination, after all the “Great Anthony Watts” has condescended to remove your link on his site from the “Transcendent Rant and way out there theory” category. I currently reserve judgement on whether this is any indication of spinal fortitude on Mr Watt’s part.

  6. tallbloke says:

    Nick: I’m interested in what the differences are between the original and revised paper. Which are the important ones from your viewpoint?

    Regarding your criticism of their humidity precip relation, surely the temporal and spatial aspects come into play here. Can you specify your objection more precisely in those terms?

    Konrad: It was more a quid pro quo thing. I removed WUWT from the ‘lukewarm cornflakes’ category; Anthony reciprocated. We can still respect each others approach, despite the differences. 😉

  7. Paul Vaughan says:

    This is interesting:

    “Condensing vapor allows for a one-dimensional flow between warm and
    cold liquid surfaces. Vapor arises by evaporation at the warmer surface, flows towards
    the colder surface and “disappears” (i.e. condenses) there. Such a one-dimensional
    motion with zero velocity at both boundaries is by definition impossible for a noncondensable gas in which mass is conserved. […] In the Earth’s
    atmosphere the surfaces where condensation occurs (i.e. on moisture droplets) do not
    coincide with the upper boundary surface of the circulation but are distributed within
    the atmospheric column allowing for macroscopic motions in all directions.”

  8. oldbrew says:

    There’s the Lifted Condensation Level and the Convective Condensation Level, brief explanations from a meteorologist here.

  9. Richard LH says:

    A possible practical experiment.

    Take a large insulated box containing two plates, one hot(20c?) , one cold(1c?), and measure the steady state velocity flows in the box after some time.

    Add two water pans on top of the plates (or in a duplicate box) and measure the new steady state flows.

    This should give some idea of the magnitude of this concept I would have thought.

  10. Tim Cullen says:

    I never pay much attention to “he said this” and “she said that” stories in the glossy magazines… sometimes you have to let the dust settle before you can get a glimpse of the truth… sometimes “moral outrage” is just that… sometimes its just “public posturing”… sometimes its just “moral blackmail”… sometimes all three motivations combine into one “let me on the gravy train or your pet [cat/dog/child/project/theory] gets it”… who knows… and when it come to glossy magazines: who cares.

  11. vukcevic says:

    It’s always good when people are willing to go a little bit out on a limb..

    CERN’s scientist goes beyond the party line: dark matter and dark energy; new hypothesis.
    (click on » Look Inside )

  12. tallbloke says:

    Heh, Cullen being provocative. Under the terms of agreement between authors and Chem. Phys. Makarieva had no choice but to respond to Held’s non-criticism.

  13. Stephen Wilde says:

    Having read more of the supplementary material it does seem that the basic contention is indeed volume related because the condensing out of water vapour into more concentrated water form results in a reduction of volume of the original air parcel that then requires an inflow of air from around about and it is that redistribution of air that leads to pressure changes in the locality.

    I see that as simply the flip side of the original evaporation lower down in the vertical column (primarily but not exclusively from water at the surface).

    When evaporation occurs there is expansion as the water vapour is forced in amongst existing air molecules and so it follows that upon condensation the water vapour is removed so that the air parcel contracts and air molecules flow back to take up the new space made available.

    I also agree that evaporation effectively involves the insertion of PE into an air parcel in the form of latent heat and that condensation effectively involves the release of that PE from the air parcel.

    However that PE will then be carried by the water droplets themselves and any air molecules carried up with the water vapour as a consequence of them having been lifted from the surface to the height where condensation occurs.

    The thing is though that the air parcel without water vapour will be colder as well as more compact so
    it will be denser than surrounding parcels of air that have not yet condensed out (constantly being resupplied from below) and will begin to fall.

    It is also significant that PE does not register as heat so the locality will remain at the cold temperature set by the pressure gradient.

    In practice, the continuing upflow of vapour rich air will push that colder denser air aside and it will descend some distance away from the continuing upflow.

    Hence a circulation such as a Hadley cell.

    I certainly agree that such a process is not adequately dealt with in the climate models since it is a non radiative process that can vary in speed and / or volume.

    Indeed it is my proposition that changing atmospheric volume is sufficient to offset any thermal effects from any forcing elements other than mass, gravity and insolation.

    The net radiative outcome at top of atmosphere is merely the net thermal response from all such non radiative events in the atmosphere agreggated and netted out globally.

    My main query as regards this paper is as to whether the pressure changes are not really a result of condensation alone but more realistically a consequence of the entire water cycle whereby evaporation reduces air parcel density near the surface by injecting PE so that the air parcel rises causing a reduction of surface pressure and condensation increases air parcel density by releasing that PE into the condensed out droplets so that the air parcel becomes denser and then falls causing an increase in surface pressure at some distance away from the rising column.

    Isn’t this paper just a reaffirmation of what has always been understood as to how low and high pressure cells form and interact ?

  14. tallbloke says:

    “The physical process behind the origin of
    the condensational pressure gradient force is the loss of molar density N that occurs
    immediately upon condensation (vapor loss) rather than when the liquid precipitates
    (mass loss). In the context of condensation-induced dynamics it is natural therefore to
    speak of a vapor sink.
    In our work we consider the basic physical principles that underlie the existence of the
    vapor sink in the atmosphere. We discuss in detail the implications of the hydrostatic
    equilibrium assumption for the formulation of condensation rate. We estimate the global
    mean power of the potential energy release associated with the vapor sink. We derive
    a theoretical expression for the pressure gradient force associated with this vapor sink.
    Using the observed fundamental atmospheric constants we show that our expressions
    and the pressure gradient force they describe are relevant in atmospheric context, both in large-scale circulation patterns like Hadley cells and in hurricanes. To our
    knowledge, none of these efforts have been previously undertaken. We hope that
    our analyses will stimulate other researchers to examine and test our conclusions.”

  15. tallbloke says:

    Stephen: “Isn’t this paper just a reaffirmation of what has always been understood as to how low and high pressure cells form and interact ?”

    The purpose of the paper is to explain where the majority of the force which moves the circulation comes from, and how, and why.

    The implications are profound, because it tells us co2 variation isn’t going to have a lot of effect on general circulation, because the circulation is driven by pressure differences due to phase change rather than temperature differentials and water’s radiative properties.

  16. tallbloke says:

    Nick Stokes says:
    January 27, 2013 at 11:35 am
    The revised paper was never made available for discussion, by reviewers or anyone. And no, it doesn’t answer the criticisms.

    It doubles down on one bizarre claim – that precipitation rate is proportional to humidity (eg appendix A1).

    If Nick has been reduced to this sort of nit-pickery, he obviously doesn’t have any serious points to make. Or he would have made them on the discussion site in response to Makarieva answering his eight questions.

    Nick, Where exactly are the differences between the original manuscript and the ‘revised’ paper? No change in the equations I can see. Are you just referring to additional material given in appendices?

  17. wayne says:

    Wow, vukcevic, that article,, is close to my thoughts but repelling instead of merely being additional mass-energy density causing much faster rotation noticeable at the galactic scale.

    See: and the comments further below.

    But, I agree, this is both a bit far-out and OT here.

    [Reply] Maybe take it to Tim Cullen’s co-rotation thread? The orbit of Moons might be analogous? Or have a new thread on demand.

  18. Stephen Wilde says:

    “The implications are profound, because it tells us co2 variation isn’t going to have a lot of effect on general circulation, because the circulation is driven by latent heat and pressure differences rather than temperature differentials and water’s radiative properties.”

    Agreed and accepted.

    Save that temperature differentials become relevant if any forcing element tries to drive them away from the temperature profile set by gravity and pressure.

    There is room for debate as to the details and I think Ms Makarieva needs to consider the wider picture than just a reduction in volume of an air parcel when condensation occurs. The pressure initially fell at the surface when evaporation occurred in the first instance. It is at the point of evaporation that one finds the most relevant forces.

    If anything, condensation leads to a pressure rise because the smaller colder parcel of air is more dense rather than less dense.

    She is right to say that air must flow in from elsewhere to compensate for the reduced volume of the shrunken colder denser air parcel but isn’t it just replaced by more upward flowing vapour laden air from below in a circulatory process ?

    You can’t consider a vapour sink without tying it in to the vapour source.

  19. tallbloke says:

    “The pressure initially fell at the surface when evaporation occurred in the first instance.”

    It did? Why? water vapour is buoyant because it is lighter than air.

    “She is right to say that air must flow in from elsewhere to compensate for the reduced volume of the shrunken colder denser air parcel but isn’t it just replaced by more upward flowing vapour laden air from below in a circulatory process ?”

    Understanding why the flow tends to be horizontal is the trickiest part of this paper to understand. Hopefully Douglas might return to explain it better than I can.

    At it’s simplest, Makarieva et al state in 4.2 that:

    “In the real atmosphere which is effectively very thin, most
    part of the non-equilibrium pressure gradient is transferred to
    the horizontal plane via rapid hydrostatic adjustment.”

  20. tchannon says:

    A reminder that gaseous water is an unusually light gas. (gravity again, weight)

    The effect seems likely to be stronger than if it was air alone changing temperature/density.

  21. wayne says:

    This is a prime example of what is wrong in science today. Temperature “causes” nothing by the IGL, P/ρ=RT, temperature is just a manufactured balancing variable that describes the ratio of P/ρ. Both pressure and density are real, temperature is not ‘real’ though our human minds can understand changes in temperature better, much better, than getting a sense from a change in ratio of two variables. Those two variables, of which neither is sensed by humans directly, are also not easily measured without instruments.

    So, does temperature cause air to rise? Or, is it either a decrease in density or an increase of the pressure, both are in that ratio, and the ratio change is the cause of the rise in temperature?

    Makarieva et al are saying that upon condensation there is an immediate decrease in ‘n’, the number of particles in a volume, density drops. A drop in density shows up also as a balancing rise in temperature by P/ρ=RT, that is a secondary effect. But what of the pressure? That seems to occur locally over time. This decrease in ‘n’ and density over a large area also causes a drop in pressure at the surface for the gas columns no longer have as many atoms and molecules at a mean velocity to manifest as temperature drop at the surface.

    Stephen is right, he likes words, this may be viewed as a “well known” process in one respect, but in words, temperature is the preferred explanation for it would be very hard to describe it in words using P/ρ as the prime discussion variables. Makarieva et al seem to be going to that level — into the actual physical processes and away from mixed variable of temperature and I wholly agree with their approach. There are missed effects when you ignore what is the actual cause of the temperature changes we measure at an instantaneous scale. That is why this is written in math and not a stream of easily grasped words and more simple temperature equations.

    That’s how I see this at a glance.

  22. Stephen Wilde says:

    tallbloke said:

    “It did? Why? water vapour is buoyant because it is lighter than air.”

    A vertical column of air containing water vapour is less dense and less heavy than one not containing water vapour. Hence lower pressure, which is why low pressure cells tend to be invigorated over warm water where more vapour is available.

    Evaporation, even without an increase in surface temperature, leads to lower surface air pressure as it rises vertically.

    An increase in surface temperature provokes uplift and lower pressure from warmer, lighter air but the water vapour process is far more effective and doesn’t need more heat at the surface at all.

    “Understanding why the flow tends to be horizontal is the trickiest part of this paper to understand”

    At the top of the atmosphere near the tropopause that is easy. The tropopause creates a lid on convection so the only way to go is horizontally.

    At lower levels it makes sense that the rising column will interact with the air on all sides leading to horizontal exchanges of molecules and energy.

    Within the rising column itself I would expect the upward flow to be the main process for replacing and pushing aside horizontally the cooler and denser parcels of air that have had their vapour removed.

    I accept wayne’s point that Ms Makarieva is moving on from concepts expressible in words to maths which seeks to quantify but still I think her maths is going to have to cover the evaporation process as well and merge the effects of both evaporation and condensation.

    I think that the current emphasis on condensation could be misleading in isolation.

    wayne said:

    “Makarieva et al are saying that upon condensation there is an immediate decrease in ‘n’, the number of particles in a volume, density drops”

    Actually, volume decreases instantly and more molecules flow into the previous larger volume from elsewhere.

    By removing the vapour the parcel of air becomes as cold and dense as the rest of the air around about which is not part of the rising vapour rich column.

    However withIn the vertical column there is more vapour filled air coming up to replace it but she doesn’t seem to include that factor. More coming up would obviate the need for a significant impact on the surrounding pressure field other than by way of the background circulation caused by the combined evaporation / condensation cycle.

  23. wayne says:

    Understanding why the flow tends to be horizontal is the trickiest part of this paper to understand. Hopefully Douglas might return to explain it better than I can.

    My two cents, horizontal makes huge sense to me. Don’t forget the three dimensions!! When speaking of a local isotropic effect, like a collapsing volume in a cloud, there are always more effects laterally or in the horizontal plane than in the vertical axis. Four lateral directions, just two up and down, by the geometry. Also in an atmosphere, the up and down are physically bounded by both the surface and the gravitational ToA, the lateral dimensions have no real physical bounds on a spherical planet.

    If you are speaking of strictly convection or gravity as the effect, yes, that is strictly on the vertical axis, but both or these are being discussed and the have cross effects in this paper, some vertical only and some isotropic, they are linked equations, so keep these separated as you think of the effects. Might help to watch when they jump in nomenclature from ‘d’ derivatives to ‘∂ ‘partials. The isotropic local events by themselves will always tend to manifest more as lateral effects when looked at large scales by 2/3rd or even greater due to the bounds.

  24. Stephen Wilde says:

    wayne said:

    “A drop in density shows up also as a balancing rise in temperature by P/ρ=RT”

    A rise in temperature caused by an increase in the supply of energy can cause a drop in density as the gas expands against pressure.

    But if the supply of energy stays the same then a drop in density from reducing pressure (as in rising upwards) causes the temperature to fall as well.

    The Earth as a whole receives a very stable supply of energy from the sun so as a general rule expansion of the entire atmosphere results in lower density and lower temperature.

    Within the atmosphere the supply of energy can vary from time to time and from place to place both vertically and horizontally and so we do often see a rise in temperature causing a drop in density.

    That is not what Makarieva is considering.

    She is talking about rising parcels of air which contain excess PE from earlier evaporation injecting latent heat of evaporation into the parcel.

    When it rises up the pressure gradient it cools as more and more of its KE is converted to PE.

    When condensation occurs that PE is taken away with the water droples that are produced and the volume of the parcel contracts but the contracted parcel must then adopt the same temperature as other parcels bereft of vapour that inhabit that level. In other words that parcel is left denser and colder.

    She then says that pressure at that level drops as more air moves in to occupy the reduced space.

    So I ask whether any such pressure drop would occur if the reduced space is immediately filled by more vapour rich air coming up from below.

  25. wayne says:

    “Makarieva et al are saying that upon condensation there is an immediate decrease in ‘n’, the number of particles in a volume, density drops”

    Actually, volume decreases instantly and more molecules flow into the previous larger volume from elsewhere.

    Ok Stephen, you see volume, I see density. You see PV=nRT, I see P/ρ=RT. You are speaking of a variable molar volume and I was speaking of a fixed control volume.

    Maybe we both should be more exact and explicit in the wording. 😆

  26. michael hart says:

    (I also posted this on the earlier thread, by accident.)

    I’m not going to attempt to address the details, but Makarieva and Issac Held (reviewer) appeared to agree on one aspect that occurred to me before reading his review: The issue appears, at least in part, to be a quantitative, not qualitative, disagreement.

    Held makes it clear from the first sentence that the effect “is traditionally considered to be small.” So is his main objection merely that Makarieva et. al. are exaggerating in their choice of words?

    If, as Held does, a reviewer makes use of a paper [Spengler et. al. (2011)] too new to have even been considered by Makarieva et. al., as a stick with which to beat them, then that suggests to me that perhaps their paper is not so far-out as to merit outright rejection.

    In fairness, Held does, later on, appear to make suggestion as how the issue might be resolved with data. That sounds like science.

  27. Ray C says:

    It says here ;

    “Aerosols play a critical role in the formation of clouds. Clouds form as parcels of air cool and the water vapor in them condenses, forming small liquid droplets of water. However, under normal circumstances, these droplets form only where there is some “disturbance” in the otherwise “pure” air. In general, aerosol particles provide this “disturbance”. The particles around which cloud droplets coalesce are called cloud condensation nuclei (CCN) or sometimes “cloud seeds”. Amazingly, in the absence of CCN, air containing water vapor needs to be “supersaturated” to a humidity of about 400% before droplets spontaneously form! So, in almost all circumstances, aerosols play a vital role in the formation of clouds.”

    So first Aerosols are required before precipitation and the vapour sink process can occur.
    Aerosols come in many shapes and sizes, are subject to temporal and spatial distribution and will be at different temperatures within the atmosphere.

    What happens if aerosols are in short supply? Less cloud.

    What determines the precipitation rate?
    The water on aerosols, cloud droplets, need to combine to form raindrops. Turbulent movement brings them together until they cannot resist the force of gravity. What starts the turbulent movement? The vapour sink. Condensing vapour on aerosols in horizontal layers of aerosol at a temperature zone where condensation is allowed at the dew point. Not from above because this zone has fewer, colder, aerosols, not from below because the arriving aerosol are in temperatures which are too high for condensation. So, horizontal air movement as the condensation proliferates in the “goldilocks zone” of aerosols at just the right temperature. Warm cloud development. The cooled, condensed, colliding droplets, reach raindrop size and fall, gaining mass through collision with droplets and aerosols. The resulting pressure drop, from the leaving mass of liquid water, pulls in, from sideways because that’s where condensation is easier, a supply of more aerosols.
    Wind is born!!
    My two pence worth!!says aerosols are required.

  28. tallbloke says:

    Ray C: Interesting narrative. Just one point, the mass loss from droplets falling out of an air parcel is not the main event. Its the 1000:1 density ratio between water vapour and liquid water which causes the drop in molar density which then sucks the wind in to ‘refill the air parcel’.

  29. Doug Proctor says:

    The brilliance of this response is that it doesn’t claim their ideas are “right”, but that all work should be evaluated on the same basis as any other. “Consensus”, or authority, it says, have no place in scientific studies. This, of course, is highly critical by its nature to CAGW by the IPCC narrative.

    An abbreviated version of this, reduced to its philosophical/practical aspects of determining any scientific situation. should be promoted to the MSM as the principle behind the warmist-skeptic dispute: If the uncertainties in the IPCC positions had ever been admitted, today we would be working towards what was going to happen, not what might happen. The Uncertainty Principle would be applied towards the damage done in wrongful anticipation of a problem as well as towards wrongful dismissal of a coming problem.

    The push-back by here is notable moreover for the refusal to bow down. The younger generation is not beholden to the older after 25 years of alarm. If anything, the younger is antagonistic for all the years of having to keep their opinions to themselves. With Der Spiegel’s discussions recently of the un-discuss-able, we may be seeing the style of the next few years. A good style.

    A manifest disconnect between theory and observation eventually becomes intolerable for those whose careers and self-image demand they be objective investigators of testable fact. As long as temperatures do not rise as predicted by IPCC CO2 theory, the disconnect will exist. Let’s hope that the other, non-compromised researchers are as stalwart as these.

  30. Nick Stokes says:

    “If Nick has been reduced to this sort of nit-pickery, he obviously doesn’t have any serious points to make. Or he would have made them on the discussion site in response to Makarieva answering his eight questions.”
    I did exactly that. It’s point 5.

    It isn’t nit-pickery – it’s built into their basic mathematics. And it’s obviously wrong. It takes just one part of the kinetics determining equilibrium and ignores the back-reaction (evaporation).

    You’re right that there isn’t much difference in the equations in the text – they remain wrong. There is a new appendix, which doesn’t help.

    Can you explain Eq 34?

  31. michael hart says:

    Doug’s comment probably can’t be restated too often (but I will try).

    “The Uncertainty Principle would be applied towards the damage done in wrongful anticipation of a problem as well as towards wrongful dismissal of a coming problem.”

    Based on Chemistry alone, I know I could invent a different, new, and “plausible”, environmental “problem” for every day of the year. But that wouldn’t mean they were correct or that we should stop the world to try and prevent them. There is a universe of different ways that things can “go wrong”, and a much smaller number of ways for things to be “right”.

    I don’t need someone inventing extra hypothetical problems for me to worry about. I can do that much for myself.

  32. tallbloke says:

    Hi Nick,
    No Nick, that link is your eight questions.
    Makarieva responded here

    5. But following Eq 7 there is some new and strange physics. “Given our
    assumption that S is linear over Nv” I cannot see that this assumption has
    been discussed. But what is the basis for it – or what does it even mean?
    This comes back to the confusion about whether Nv is saturated. But the
    assumption makes no sense in unsaturated air. Does it mean linear as
    temperature varies?
    It is surprising in any case that the precipitation rate should be determined
    simply by the water vapor content.

    The assumption of linearity is based on consideration of condensation as a first-order
    reaction over saturated molar density Nv of the condensing gas4. The rate of firstorder
    reactions is directly proportional to the molar density of the reagent, with the
    proportionality constant having the dimension of inverse time: S = CNv, where C
    (dimension s􀀀1) is in the general case independent of Nv.
    This linearity is a natural theoretical assumption justified by the particular physical nature
    and stoichiometry of condensation, with gas turning to liquid. Note, for example,
    that the reverse process (evaporation) should be a zero-order reaction over Nv. Experimental
    validation of the linearity of condensation rate over Nv can be made by
    considering condensation of water with different isotopic composition 5.
    5)5E.g., Fluckiger B., Rossi M.J. ’Common Precursor Mechanism for the Heterogeneous Reaction of D2O, HCl,
    HBr, and HOBr withWater Ice in the Range 170-230 K: Mass Accommodation Coefficients on Ice’, J. Phys. Chem.
    107 (2003) 4103, see also here:

    Nick, contrary to what you claimed, Makarieva did answer you, and you should acknowledge that, before you go on to argue whether or not her answer satisfied you.

  33. Nick Stokes says:

    Well, there’s a response. But again it’s in terms of theoretical kinetics and
    1. leaves out the back reaction – evap.
    2. Doesn’t mention what the time scale actually is.

    And it doesn’t face up to the disconnect with observation. Do you think that the precipitation rate is determined by and proportional to humidity?

  34. tallbloke says:

    Section 4.2 of the paper says:

    Indeed, total
    molar density remains in hydrostatic equilibrium in the absence
    of condensation as well as in its presence. In the limit
    Nv -> N Eq. (34) gives a physically meaningful result, S=0.
    Indeed, when atmosphere consists of water vapor only and
    is in hydrostatic equilibrium, no condensation takes place.
    Condensation occurs only when water vapor distribution is
    When condensation is absent, dry air is in hydrostatic equilibrium.
    But when water vapor condenses and its distribution
    is compressed several-fold compared to the hydrostatic distribution,
    the dry air must be “stretched” compared to its
    hydrostatic distribution. Only in this case, when the nonequilibrium
    deficit of vapor in the upper atmosphere is compensated
    by the non-equilibrium excess of dry air, the moist
    air as a whole will remain in equilibrium. The distribution of
    Nd is non-equilibrium and cannot be used instead of N in the
    reference term in Eq. (34).

    It’s a dynamic system.

  35. tallbloke says:

    Nick Observes:
    “Now there is always some humidity, but it isn’t always raining.”

    True, but then, a lot of precip never reaches the ground.

    Nick asks:
    “Do you think that the precipitation rate is determined by and proportional to humidity?”

    Well, given that the hydrological cycle speeds up with a general increase in temperature, and warmer atmosphere are thus moister, it doesn’t seem to me to be such an unreasonable assumption.

    What is your argument against the idea?

  36. Nick Stokes says:

    It’s a precise quantitative assertion – S=CN_v, placed into a set of PDE’s describing atmospheric processes. It isn’t a proposition about maybe some average behaviour of the atmosphere. If it’s true, then it has to be true everywhere.

  37. tallbloke says:

    Model parameters and equation terms aren’t always true everywhere, but your mates wrote them, so you don’t complain.

    In the case of the hydrostatics, the models seem to levitate water with no work done at all.

  38. wayne says:

    “True, but then, a lot of precip never reaches the ground.”

    That’s one number I would love to find stated somewhere TB. You say ‘precip’ and I’m not sure that has to occur as big enough droplets to fall.

    I looked out the window a few days back and the visibility must have been less than two miles and there were no clouds at all, but the sun was out full force, so there was convection. The sky near the horizon up to 45 degrees was milky white-blue. I don’t think all evaporation/condensation is even at the visible droplet scale but more on a limited number of molecules clumped together, say 100 or 10,000 or 1,000,000, evaporating were warm near the surface and condensing again high where it is cool, just cycling, driven by the convection.

    Have you ever come across that topic before? I haven’t.

  39. tallbloke says:

    Wayne: I think Judy Curry discussed it. It’s a surprisingly high figure. From memory, something like 90% of condensate doesn’t fall to ground but is re-evaporated.

  40. wayne says:

    Thanks tallbloke.

    Just by saying ‘a large percentage’, that fills in the question as to how so much energy gets from here, low near the surface, to there, high in the atmosphere. My calculations were missing a large portion there and that just might complete it. I’ll search her site a bit more to fill in particulars.

    Now, all of the GCM’s are accounting for such realities aren’t they? Any of them? 😉 Is this what is meant by saying that they don’t handle clouds very well or at all?

  41. tallbloke says:

    Also from 4.2

    “in the absence of hydrostatic adjustment,
    the dry air distribution is not affected by condensation and
    remains in equilibrium both in horizontal and vertical dimensions.
    The non-equilibrium gradient of total air pressure remains
    located in the vertical dimension and is not translated
    onto horizontal dimension. Such a situation could take place
    in an atmosphere that would be much higher than it is wide.
    In the real atmosphere which is effectively very thin, most
    part of the non-equilibrium pressure gradient is transferred to
    the horizontal plane via rapid hydrostatic adjustment.”

  42. wayne says:

    Lacking a theory on condensation rate, one is unable to check the results of numerical simulations against robust physical estimates. Simulations are checked by comparison to other simulations, such that the errors have the opportunity to perpetuate from model to model. However, in less complicated (compared, e.g., to GCMs) models like that of BF02 the inconsistencies can be explicitly traced relatively easily.

    In the BF02 model to compare the moist and dry simulations, a convectively neutral temperature gradient is specified in the initial state of the dry simulation. In this state the dry atmosphere should be in stable static equilibrium: all velocities are zero. This is intended to be compared to a moist simulation. The initial static equilibrium state of the moist atmosphere is chosen such that the liquid drops levitate in the motionless air (BF02, p. 2921).

    Right TB, there are the models. Those might be the statements that caught your eye, they should. Seems it always takes reading two or three times at this depth to get all of what is being said but it sounds so close.

    Right… no convection… what planet are Bryan et al speaking of? But… they need an initial condition for the model, so… just assume zero and a motionless atmosphere.

    You know, maybe I should just re-read ALL of Anastassia’s papers and discussions now that I understand better what is being said. I even understand most of the equations this time through compared to two years ago. Much of what I have been harping on seems to parallel so much in these texts, some of my statements very well have come from discussing with her on wuwt two years ago and I just don’t realize it, though I tend to word each topic a bit differently. Sure don’t want to seem as if I’m trying to take the fire away from anyone. If they have it all worked out, discussed and published, great, then I just have to read it and spread their findings, much easier that way.

  43. tallbloke says:

    Wayne: No way! There is a huge value in you’re independent progress towards a fuller understanding of the thermodynamics of our atmosphere.

  44. suricat says:


    Surely this is all about the ‘policy’ of ‘convention’.

    Many years ago (or should that be ‘Once upon a time’? 🙂 ), my ‘lab work’ was criticised for not ‘following convention’.

    Whilst measuring the parameters of a thermistor during a B Tec 4 ‘lab experiment’ I was told to expect a ‘linear current response’, but discovered a current ‘glitch’ at one interval. Thus I repeated the experiment (without enough time to finish it) and drew a graph that included the ‘glitch’.

    The tutor drew me to one side and quietly said: “Ray, you should’ve finished the experiment and drawn the graph as a ‘straight line’, then added a comment that ‘all of the measurements’ didn’t fall inside the ‘straight line graph'”.

    All I said was, “Sorry, but the measurements didn’t result in a ‘straight line’. I thought I was in error, so I repeated the experiment up to that point!”.

    He later admitted that ‘other processes’ were happening at that point of the graph that were outside of his tutelage remit.

    What can you do? 😦 (that’s a rhetorical question)

    Best regards, Ray.

  45. wayne says:

    TB, don’t worry ;), not going anywhere fast, that’s for sure! Just tied down through tomorrow at least. I was just saying “If they have it all worked out” and bet we won’t know that for quite some time. Thought it a good time to take another good look at all of those equations in their papers once again. Trouble is, it’s just a long, slow slog.

  46. Richard says:

    I still think that it should be possible to setup a practical experiment to demonstrate this effect.

    A large sealed box containg dry air with two plates, one warm (20c) and one cold (2c) placed in the corners/edges will have a steady state air motion profile which can be measured.

    If two water pans are added on top of the plates to this setup then the water will transfer from the warm pan to the cold pan via water vapour evaporation and condensation in addition to any dry air motions.

    The air/vapour movement profiles will be different to those with just dry air I suspect.

    The magnitude of this difference should give a practical value to the magnitude of the effect I believe.

    [Reply] Sounds promising. I always encourage empirical experiment. TB

  47. Richard says:

    At the present it is just a thought experiment (I do not have access to instruments that can remotely measure air movements).

    However, even as a thought experiment, it appears to demonstrate the Makarieva effect.

    I am not sure how conventional physics explains any difference in flows. There will be more work done in the system (as evaporation/condesation is a large energy transfer mechanism) and that additional energy will be visible in the flows.

    Although the cold water surface in the pan is not exactly the same as condensing out into air I cannot see how the effect would change other than the overall surface tensions (energy) in the systems would differ.

  48. Brian H says:

    One bar for all? Are your standards really as low as climate science’s?
    For shame.