by Anastassia Makarieva, Victor Gorshkov, Douglas Sheil, Antonio Nobre, Larry Li
It’s official: our controversial paper has been published. After a burst of intense attention (some of you may remember discussions at Climate Etc., the Air Vent and the Blackboard), followed by nearly two years of waiting, our paper describing a new mechanism driving atmospheric motion has been published in Atmospheric Chemistry and Physics.
It’s been an epic process – most papers get published (or rejected) in less than a tenth of that time. The paper is accompanied by an unusual Editor Comment (p. 1054) stating that in the paper we have presented a view on atmospheric dynamics that is both “completely new” and “highly controversial”. They accept that we have made a case to be answered: they clarify that “the handling editor (and the executive committee) are not convinced that the new view presented in the controversial paper is wrong.” That’s not exactly an endorsement but it is progress.
We have not been simply waiting the last two years. Arguments and ideas have matured. We want to give you an update. In Section 1 of this post we discuss the novelty of our propositions. In Section 2 we address three of the most common objections. In doing so, we draw on all the recent work by our group thus providing an updated view on our theory.
What is new?
We have described a new and significant source of potential energy governing atmospheric motion. Previously, the only such recognised energy source was the buoyancy associated with temperature gradients.
Unlike the buoyancy mechanism, that applies to both liquids and gases, our new mechanism applies only to gases. Water vapor condenses and disappears from the gas phase when moist air ascends and cools. For this reason the water vapor pressure declines with height much faster than the other (non-condensable) atmospheric gases. As a result the exponential scale height hv of water vapor is markedly smaller than the scale height h of the air as a whole, hv << h. What are the implications of these two different scales for atmospheric dynamics?
In hydrostatic equilibrium the vertical pressure gradient force -∂p/∂z balances the gas weight in a unit atmospheric volume –ρg: -∂p/∂z – ρg = 0, where p is air pressure, ρ (kg m−3) is air density, and g is acceleration of gravity. In the absence of condensation in a circulating atmosphere the relative partial pressure γi ≡ pi/p of the non-condensable atmospheric gases, including the unsaturated water vapor, is independent of height. In hydrostatic equilibrium for such gases we have:
where pi and Ni (mol m−3) are partial pressure and molar density of the i-th gas, respectively, R = 8.3 J mol−1K−1 is the universal gas constant, and N and M are the molar density and mean molar mass of air as a whole.
This relationship determines that in hydrostatic equilibrium any work –w∂pi/∂z performed by the vertical partial pressure gradient per unit time per unit atmospheric volume is compensated exactly by the work –wγiρg performed by the force of gravity that acts on a corresponding molar share γi of the air mass (here w is vertical velocity). In other words, all work performed by the non-condensable gases as they ascend and expand is fully spent on elevating their respective molar shares of total air mass in the gravitational field. Nothing is left to generate kinetic energy.
By contrast, if we consider the saturated water vapor, condensation means that we have
That is, the work of the partial pressure gradient of water vapor greatly exceeds what is needed to overcome gravity. The main physical statement behind our new view is that this net remaining power q (W m−3)
is available to generate kinetic energy and drive the Earth’s atmospheric dynamics. Roughly speaking it is the power that remains after the water vapor has “lifted itself”. The value of q represents the volume-specific power of the “motor” that drives the atmospheric circulation.
The formation of strong vertical winds is directly inhibited by the atmosphere’s condition of hydrostatic equilibrium. For that reason the dynamic power of condensation is mostly translated into the power of horizontal pressure gradients and winds:
Here v = u + w is air velocity, u is horizontal and w is vertical air velocity, and ∇p is the pressure gradient, all measured on the circulation largest spatial scale. The kinetic energy generated by horizontal pressure gradients dissipates in smaller-scale eddies and ultimately converts to heat.
By integrating (3) over height z and noting that wN = wp/(RT) is independent of z to the accuracy of γ (e.g., see Appendix here), we obtain a relationship indicating that the driving power Q per unit area is proportional to precipitation P (mol m−2 s−1):
where T is the mean temperature in the air column, and P ≡ wNγ(0) = wNv(0) is the upwelling flux of water vapor (mol m−2 s−1), which, in the stationary state and assuming complete condensation, is equal to the downward flux of precipitating water. As discussed in our paper, this equation is exact for a horizontally isothermal atmosphere. In the general case it may be imprecise by about 10% (Makarieva, Gorshkov, 2010).
We can now compare our theory with observations. First, we note that the mean global power of atmospheric circulation estimated from (5) is about 4 W m−2, which is in close agreement with the best observational estimates. We note that this is the first and only theoretical estimate of the power of global circulation currently available. We return to this in the next section.
We then observe that the above physical relationships apply to circulation phenomena characterized by notably different spatial and temporal scales. For a steady-state global-scale pattern (e.g., Hadley cells) the mean value of precipitation P is determined by solar power I. About one third of solar power is spent on evaporation. Given that evaporation and precipitation must be nearly equal (in a steady state) we can see that P ∼ I/Lv, where Lv (J mol−1) is the heat of vaporization. In a much smaller and short-lived circulation system, like a hurricane or tornado, that moves as a whole with velocity V, precipitation within the circulation area is determined by the flux of water vapor imported. It thus depends on the height hh and radius r of the circulation, velocity V and the ambient amount of water wapor: P ∼ (hh/r)VNv (here Nv is the mean ambient molar density of water vapor).
As we can see, the physical determinants of precipitation are very different. E.g., hurricane P can be several orders of magnitude higher than the mean global P. The horizontal scale of Hadley cells is several times larger than that of hurricanes. Despite such different scales, physical determinants of condensation intensity and drastically varying P values theoretical estimates (3) and (5) successfully describe the Hadley cell as well as much more compact and transient circulation phenomena (see (Makarieva, Gorshkov, 2011) and (Makarieva et al., 2011) for details). Our approach thus provides a unified physical explanation to atmospheric circulation phenomena previously considered unrelated.
The controversy
Thanks to help from blog readers, those who visited the ACPD site and many others who we have communicated with, our paper has received considerable feedback. Some were supportive and many were critical. Some have accepted that the physical mechanism is valid, though some (such as JC) question its magnitude and some are certain it is incorrect (but cannot find the error). Setting aside these specific issues, most of the more general critical comments can be classified as variations on, and combinations of, three basic statements:
1. Current weather and climate models (a) are already based on physical laws and (b) satisfactorily reproduce observed patterns and behaviour. By inference, it is unlikely that they miss any major processes.
2. You should produce a working model more effective than current models.
3. Current models are comprehensive: your effect is already there.
Let’s consider these claims one by one.
Models and physical laws
The physical laws behind all existing atmospheric circulation models are Newton’s second law, conservation of mass, the ideal gas law and the first law of thermodynamics. Here the first law of thermodynamics is assigned the role of the energy conservation equation (see, e.g., McGuffie and Henderson-Sellers 2001, p. 1084). However, while equilibrium thermodynamics allow the estimation of the maximum possible mechanical work from heat it provides neither information about the actual efficiency of converting heat to work (kinetic energy) nor whether such conversion to motion actually occurs. In practice, this means that models do not define these factors from physical principles but through adjusting model parameters in order to force it to fit observations (i.e., to produce the observed wind speeds). Mostly this pertains to the determination of the turbulent diffusion parameters. An
interested reader see p. 1776 of Bryan and Rotunno (2009) for a simple example (see also here for a discussion). The principle remains the same even in the most complex models.
Thus, while there are physical laws in existing models, their outputs (including apparent circulation power) reflect an empirical process of calibration and fitting. In this sense models are not based on physical laws. This is the reason why no theoretical estimate of the power of the global atmospheric circulation system has been available until now.
The models reproduce the observations satisfactorily
As we have discussed in our paper (p. 1046) current models fail when it comes to describing many water-related phenomena. But perhaps a more important point to make here is that even where behaviours are satisfactorily reproduced it would not mean that the physical basis of the model are correct. Indeed, any phenomenon that repeats itself can be formally described or “predicted” completely without understanding its physical nature. We just need our experience to predict that in winter the days will be shorter than they were in summer. Thus, improvements in performance may be caused not by the correct physics but by an ever more refined description of the probability distributions characterizing persistent, regular behaviours. Consider for example how satellite data have made it possible to better analyze hurricane tracks allowing to judge about hurricane motion with some certainty a few days in advance, something entirely unavailable for the ancient weather forecasters. Such information is definitely valuable and useful. But it does not provide any insight concerning outcomes when the underlying system undergoes changes. For example, a climate model empirically fitted for a forest-covered continent cannot inform us about the climatic consequences of deforestation if we do not correctly understand the underlying physical mechanisms.
You should produce a better model than the existing ones
Modern numerical models of weather and climate are over half a century old. They contain huge numbers of parameterizations that summarize the work of thousands of researchers working for decades. As already mentioned, these parameterizations include the many adjustments needed to match the behaviour of the model to reality. If the physical core of the model is changed (e.g. from buoyancy- to condensation-driven), all these parameterizations will require revision. To expect a few theorists, however keen, can achieve that is neither reasonable nor realistic. We have invested our efforts to show, using suitable physical estimates, that the effect we describe is sufficient to justify a wider and deeper scrutiny. (At the same time we are also developing a number of texts to show how current models in fact contain erroneous physical relationships (see, e.g., here)).
Your effect is already present in existing models
Many commentators believe that the physics we are talking about is already included in models. There is no omission. This argument assumes that if the processes of condensation and precipitation are reproduced in models, then the models account for all the related phenomena, including pressure gradients and dynamics. This is, however, not so. Indeed this is not merely an oversight but an impossibility. The explanation is interesting and deserves recognition – so we shall use this opportunity to explain.
The circulation “motor” q unambiguously defines condensation intensity S that enters the continuity equation (see also Gorshkov et al., 2012). In a horizontally isothermal atmosphere for an arbitrary unknown condensation rate S the continuity equation has the form (see (A7) on p. 1053 in the paper)
where S and Sd are in mol m−3 s−1, Nd and γd are the molar density and the relative partial pressure of the dry air constituents. Recalling that in hydrostatic equilibrium q = –u∇p = RT(u∇N) (4) and using (3) we obtain from (6)
which the reader may recognize as the (in some quarters, notorious) Equation 34 in the paper. The main message from the above derivation is that the relative difference between S and Sd is itself of the order of γ: (S – Sd)/γd = S = q/(RT).
In current models in the absence of a theoretical stipulation on the circulation power, a reverse logic is followed. The horizontal pressure gradients are determined from the continuity equation, with the condensation rate calculated from the Clausius-Clapeyron law using temperature derived from the first law of thermodynamics with empirically fitted turbulence. However, as we have seen, to correctly reproduce condensation-induced dynamics, condensation rate requires an accuracy much greater than γ << 1. Meanwhile the imprecision of the first law of thermodynamics as applied to describe the non-equilibrium atmospheric dynamics is precisely of the same order of γ. The kinetic energy of the gas is not accounted for in equilibrium thermodynamics.
It is an interesting situation. The precision of the first law of thermodynamics is sufficient to determine condensation rate to the accuracy of γ. This accuracy is more than sufficient to allow existing models to be fitted to reproduce realistic precipitation rates. But at the same time the precision provided by the first law of thermodynamics is in principle insufficient to quantify our condensation-induced dynamics. This two-faced result is striking and has implications for models: Suppose that a modeller develops a model of atmospheric circulation that assumes that heating rate gradients are the only driver. If this model presents some unrealistically high wind velocities (due to some unanticipated effect) then this behaviour can readily be suppressed with only minor modifications. The model can thus accommodate phenomena for which it lacks any intrinsic relationships without any red-flags being raised.
We showed in our paper that following this route (i.e., reproducing condensation dynamics from condensation rate) requires a thorough theoretical analysis of the condensation rate behavior (see Section 4.2 and Appendix in the paper). As we discussed, no adequate theory for condensation rate exists in the current models. Therefore, as the fitting process cannot anticipate all situations (combinations and/or values of key variables etc.) there must be occasions when the omission of the relevant physical processes is revealed. The pertinent example here is in the analyses of alternative parameterizations of condensation rate: applying different cloud microphysical parameterizations in hurricane models produces systems that differ from each by over 40 mb in their central drop of pressure (with 55 mb being the mean figure for the pressure drop in hurricanes) (e.g., Deshpande et al. 2012). Generally, the situation is such that condensation-induced dynamics are neither present in modern models nor have their impacts been studied in an adequate manner.
Summary and outlook
The Editor’s comment on our paper ends with a call to further evaluate our proposals. We second this call. The reason we wrote this paper was to ensure it entered the main-stream and gained recognition. For us the key implication of our theory is the major importance of vegetation cover in sustaining regional climates. If condensation drives atmospheric circulation as we claim, then forests determine much of the Earth’s hydrological cycle (see here for details). Forest cover is crucial for the terrestrial biosphere and the well-being of many millions of people. If you acknowledge, as the editors of ACP have, any chance – however large or small – that our proposals are correct, then we hope you concede that there is some urgency that these ideas gain clear objective assessment from those best placed to assess them.
JC comment: This is an invited guest post, which follows up on Makarieva et al.’s previous blog post at Climate Etc. Comments will be strictly moderated for relevance and for civility.
Condensation of steam causes pressure drop (when water turns into steam, it occupies about 1600 times the volume, at standard temperature and pressure). This was the first industrial steam-powered device (developed in 1698 by Thomas Savery), It used a (partial) vacuum from condensation to raise water from below, then used steam pressure to raise it higher.
Yes, this simple expansion and contraction allows me to heat my single-pipe steam-heated home without any parasitic pump costs.
The theory is published: now we need evidence.
Much of this discussion has been about theory. Our paper and our blog were about theory too – so that makes sense. But I suspect many readers are interested in evidence. Dr Held too asked for “evidence” to pass his “high bar” – we rejected the argument as a point of principle. The question at issue then was whether we had presented a case coherent and interesting enough to answer: it is a theory. Theories come first the evidence comes later.
But that does not mean we don’t have extraordinary evidence.
We wrote a little about this in the paper (most points below can be explored by looking at the reference list there or at http://www.biotic-regulation.pl.ru/index.html), but it may be useful to highlight a few again here so you can make your own assessments. What is our evidence so far? How does out theory match reality?
For me the most powerful evidence comes from looking at how rainfall varies as we travel inland from the coast (over relatively flat terrain): Why does rainfall not decline over forest? It declines over non-forest in a relatively constant manner that is easy to understand (This seems to be a global pattern: see the figure in my previous blog here http://judithcurry.com/2011/03/30/water-vapor-mischief-part-ii/). Recycling is not an explanation – it would reduce the rate of decline but it could not prevent it. There is no alternative explanation at present.
This effect – the drawing of rain into continental interiors – requires a biologically functioning forest so we would predict that the effect will be smaller over boreal forests in deep winter (when the forests are metabolically inactive and not transpiring moisture). Observations support these predictions. There is no alternative explanation at present. See, e.g. 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, Ecol. Complex., 6, 302–307, 2009.
Our paper (discussed in this post) shows that we can estimate the power of global atmospheric circulation. This is the first ever such estimate developed from first principles and, though intended as a rough estimate, is remarkably close to the measured values. No alternative theory can currently explain this value.
Where we have good data on forest loss and rainfall change there are some observations suggesting a regional decline in rain regularity (as we would predict). See E.g. Webb TJ, et al. 2005. Forest cover-rainfall relationships in a biodiversity hotspot: The Atlantic Forest of Brazil. Ecological Applications 15: 1968–1983.
The work by Anastassia and co. (not me!) on hurricanes is also impressive: it shows that the condensation generated pressure gradients can give a physically and analytically consistent model of how such storm systems function and can be used to estimate several characteristics from first principles. E.g. Makarieva, A. M. and Gorshkov, V. G.: Condensation-induced kinematics and dynamics of cyclones, hurricanes and tornadoes, Phys. Lett. A, 373, 4201–4205, 2009.
So the score-card so far is 7:nil in favour of our theory (I rate the hurricane work as three points … but even if you don’t 5:nil is a good margin). That’s a good score line. Extraordinary? Well I acknowledge too that the search for counter-evidence is in its infancy.
So now the theory can and should be tested further. All those who think it is right, all those who think it is wrong and all those who are uncertain but recognise why it matters, can I hope agree that the ideas should be tested. That is a shared goal.
Sorry I posted this in the wrong place … see below: http://judithcurry.com/2013/01/31/condensation-driven-winds-an-update-new-version/#comment-291350
Finally the potential energy of the atmosphere is being considered. The observed reduction in the height of the atmosphere is the work available to the earth that moves energy from the upper atmosphere to the warmer surface. Temperature of the stratosphere decreases and surface temperature and sea level rise.
To get something out ahead of the horde:
This is bound to gather some criticisms here. Critics should note that the authors’ tone is not combative.
Also, maybe some of the HVAC engineers have been onto something for a while. I am not, however, referring to the first 2 responses here. ‘Nuff said.
Although the paper is interesting, it revolves around the adiabatic assumption of the process where dQ=0. The authors, however, have not demonstrated that the lower atmosphere is adiabatic invariant.
Please take note that the first three equations in the post summarize the key points. The effect is based on the difference in the vertical scale heights h and hv of saturated water vapor and whole air, respectively. This difference exists independent of whether the atmosphere is adiabatic or not. In fact, it becomes larger if the temperature lapse rate grows (i.e. when it is 6.5 K/km rather than moist adiabatic).
Please correct me if am wrong. In the first term of the Equation (3) above and for q, you considered energy of air pressure and that of gravity for an arbitrary air parcel and no sensible heat from the surroundings. Therefore, you inherently made the assumption of adiabatic process without justification. You must prove that it is an adiabatic process.
My first reaction is that the condensation of water vapor
– removes a little gas from the air
– releases relative to the that volume of gas a huge amount of latent heat
Thus it seems obvious that the main effect of condensation is in the release of heat while the reduction of gas is a very minor additional factor.
Perhaps I should have said that based on the above I agree with them who say that the effect is there but is small and adds nothing significant to the standard theory.
Perhaps. The lower the estimates of climate “sensitivity” become though, the greater it would increase in significance. What actually happens in the moist air portion of the atmosphere in any case is significant and obviously not handled very well in the current generation of climate models. Unless of course you choose to cherry pick one or two, here and there :)
‘We may believe, for example, that the motion of the unsaturated portion of the atmosphere is governed by the Navier–Stokes equations, but to use these equations properly we should have to describe each turbulent eddy—a task far beyond the capacity of the largest computer. We must therefore express the pertinent statistical properties of turbulent eddies as functions of the larger-scale motions. We do not yet know how to do this, nor have we proven that the desired functions exist’. Thirty years later, this problem remains unsolved, and may possibly be unsolvable. ‘ http://rsta.royalsocietypublishing.org/content/369/1956/4751.full
The scale issues for these processes preclude computation and the statictical functions are unknown and perhaps unknowable. Energy approaches seem reasonable.
Do you care to quantify that? Water can constitute up to 4% of surface vapor pressure.
Pekka Pirilä: Perhaps I should have said that based on the above I agree with them who say that the effect is there but is small and adds nothing significant to the standard theory.
Globally, the effect is approximately the same order of magnitude as the hypothesized CO2 doubling effect on climate. There are a lot of such small cavities in the standard theory, and collectively they may overwhelm the small CO2 effect. For that reason, your judgment of “adds nothing significant” is premature.
There’s no extra effect.
What happens is that evaporation adds water vapor to air increasing it’s volume. That moist air rises ans loses gradually that water vapor resulting in precipitation. When that’s over the volume is back to the relatively dry air volume. The only consequence of that all is that the speed of the uplift has to be just a tiny bit higher than otherwise.
There’s no significant extra influence on anything else.
There are the related well known effects: latent heat transfer and condensation leads to the moist adiabatic rather than dry.
Transporting water up and getting it back as rain takes work that’s dissipated in the process.. That’s one part of the work that the “atmospheric heat engine” must produce to keep circulation going.
These are important and well known effects. The paper does not add anything significant.
Pekka Pirilä: Perhaps I should have said that based on the above I agree with them who say that the effect is there but is small and adds nothing significant to the standard theory.
…………….. later ……………….
There’s no extra effect.
The effect is “there” but it is not “extra”?
I had some initial reactions on the paper. They were not very definitive but the basic idea was the same that i now see much more clearly.
What’s wrong with the paper is that they do not not discuss properly the mechanisms in the atmospheric context where pressure is one independent variable. It’s a control variable, not the outcome. The way they obtain pressure gradients does not make sense because of this.
Condensation is certainly there and it’s important because of the release of latent heat, but the loss of gas molecules is compensated automatically by rather small vertical adjustments in the atmosphere and does not drive anything of the kind the paper proposes.
Pekka –
That moist air rises and loses gradually that water vapor resulting in precipitation.
Will you please elaborate on that sentence for me? My comprehension is that a rising moist air parcel reduces in temperature and volume gradually, but condensation occurs rather abrupt when the air parcel temperature reaches the dew point. That would be the reason for flat-bottomed clouds, no? Thanks.
The paper is about the phase where air has reached saturation. When saturated air cools part of the water condenses and the air remains saturated at the lower temperature. (In real atmosphere the air gets a little oversaturated, but for this consideration it can be thought that oversaturation is not possible.)
Because the cooling is gradual also the condensation is gradual.
Thanks Pekka. Your explanations mostly make sense. I am still confused about flat-bottomed clouds.
Hi Pekka –
Just thought. Is it possible that in the absence of adequate cloud condensation nuclei an air parcel might become super-saturated? That might explain the flat-bottomed clouds. I don’t know. Thanks.
Blueicetosea, “That might explain the flat-bottomed clouds.”
The flat bottom is related to condensation temperature. The supersaturated condition should be fairly common since dust or some particle is generally need to start condensation.
The typical southern afternoon clouds form and build in patches even though the humidity is pretty uniform. If you consider only the saturation vapor pressure, as water condenses there would be a pressure differential so water vapor would flow toward the condensation. Since precipitation is not a given, the condensate flows out level with the condensation temperature. That is the way I understand what the authors are saying which is not all that exciting.
What could be exciting is that if the individual water saturation pressure can be consider independently of overall atmospheric pressure, they you could predict how clouds would form more reliably and what extreme energy potential is possible. An inch or two of pressure differential is pretty impressive with the right area.
Thank you, Cap’n
Where does the heat go? To the surrounding dry air and remaining water vapor, which expands, doing work? I am struggling with this.
Pekka, yes, I came to a similar conclusion when this paper was first out. The latent heating from condensation raises the pressure by an order of magnitude more than the vapor loss decreases it. In an ascending condensing air parcel, this pressure rise leads to expansion, reduction in density, and buoyancy. It has the reduced density more because it is warmer, not so much because it lost some vapor.
Thank you for your comment. Perhaps you might wish to clarify. Heat cannot be compared to volume of gas, they have different physical units, one cannot be larger than another. We might need to think of some realistic physical process where both effects of condensation are manifested and judge their significance from measurable parameters.
In our work we show that the effect of removing a little of gas produces a significant dynamic power. Similar estimates for latent heat are lacking in the meteorological theory, and the difficulty associated with obtaining such estimates was appreciated by people early in the first half of the 20th century.
Heat affects temperature. Condensation breaks the ideal gas law as a useful equation of state for the area where the condensation takes place, because knowing PV and n no longer tells you what T is.
The exception to that is if the temperature rise is small, and if that is the case then the condensation has no effect.
You may be making valid points, but I am not sure I see how they are related to the problem in question. Heat affects temperature, ok, but winds are driven by pressure gradients (as per Newton’s second law). Note — not by temporal pressure changes, but by spatial pressure changes. What kind of pressure gradients are produced by latent heat? I’ll tell you in advance that there is no answer to this question in meteorological textbooks. And the reason is physically clear.
There’s is no reason for the very small loss of gas would create significant pressure effects. The little it does goes against the much stronger effect through the influence on temperature.
The claim that vertical movements of gas could not compensate the changes from loss of gas molecules lacks all justification. The whole phenomenon is related to vertical movement of gas and changes that very little. There’s nothing that would make that difficult and require some strong winds.
The phenomena discussed in the paper are such detail issues that they are probably in some way included in meteorological models. How well the modelers succeed in modeling the details is seen in the success of weather forecasts (and lack of it when too much is expected). Climate models do not go to such details but use parametrizations based only indirectly on physics fundamentals.
So, Pekka, you’re basically repeating argument #3, which they already dispatched in the opening post. Care to respond to the counter-argument that was already laid out for you when you posted?
qbeamus,
What do you mean by the counter-argument?
I think that there is an experiment which can help to demonstrate the physics involved.
Take a large, sealed, dry air filled box. Place two pans, one hot (20c) towards a lower edge and one cold (2c) towards an opposite upper edge.
In this configuration the air will stabalise into a circulation pattern which can be measured for flow velocity and represents the energy flows from hot to cold via air thermal changes.
If two water filled pans are added on top of the two plates then additional water vapour will be added to the system, evaporating from the hot pan and condensing in the cold pan.
This will speed up the flow rates in direct proportion to the extra work being done to transfer energy from hot to cold via evaporation/condensation as well a air thermal changes.
Although condesing on a water surface will be slightly different from condensing out into air supported droplets I believe the overall effect should be similar.
How would ‘standard’ physics explain the differences other than by the mechanisms described in the paper?
Standard physics is that condensation causes pressure drop. But I like your experiment.
If true how would this new paper modify the theory of Ferenc Miskolczi?
If I’ve got it right, it wouldn’t–the two have nothing to do with each other.
I have a question. This post begins by saying:
Has the paper been altered to address the criticisms raised in earlier discussions? If so, could we get a brief explanation of the response to that criticism?
I love it. Scientists who understand thermodynamics.
Not just thermo. There’s a transport part of this, as well. This is the part of knowledge space occupied by physical chemists and chemical engineers, and where physicists struggle.
This old nuc likes it as well. Pressure changes. Temperature changes. Condensation. Flow. Energy.
Reblogged this on Tallbloke's Talkshop and commented:
Important post
‘Roughly speaking it is the power that remains after the water vapor has “lifted itself”. The value of q represents the volume-specific power of the “motor” that drives the atmospheric circulation.’
Colour me skeptical. I am assuming that this power is the enthalpy of condensation and is dissipated radiatively higher or lower in the atmosphere with the height largely determined by the momentum of the rising air mass.
Coastal winds are driven more by temperature differences between land and ocean. They happen where there is desert and forest – although both do influence the availability of moisture in the atmosphere and therefore precipitation and dew.
Hadley cells are driven by temperature differentials between the equator and higher latitudes. Walker Circuklation is driven by ocean temperature differentials in the Pacific equatorial region. Polar cyclones are caused by planetary rotation. Tropical cyclones are spun up by Coriolis forces.
I will think on it.
Chief, the main thing to me seems to be cloud formation. If the the air is saturated it is saturated, why does there need to be an aerosol factor? By setting c=cs they are looking only at a 100% relative humidity condition. If a cloud doesn’t form then because of lack of aerosols to start formation, chill that sucker down some more to form ice crystals and it should make its on substrate to form. Add aerosols and you reduce the energy required to initiate condensation i.e., form clouds.
If CO2 or anything else causes surface warming which in turn increases the ethalpy moist air, that would change the temperature required for saturation. More water means condensation starts at a higher temperature, lower altitude. Lower altitude, higher density, the clouds can absorb more energy producing an increase in “upper” level convection with a greater likelihood of mixed phase clouds. Which can result in hailstone or ice crystal recycling, basically an atmospheric heat pipe. Leading to all sorts of neat stuff potentially, like say SSW events and ozone depletion.
Trendberth BTW, missed 20Wm-2 that is associated with clouds and surface energy absorption. Of course, he is just a scientist, not an HVAC engineer :)
I have no clue if this group has figured all that out, but there are some serious flaws with the application of radiant physics on a chaotic water world.
It was indeed a curious publication process. The editor in his publication statement said:
“the editor concluded that the revised manuscript still should be published – despite the strong criticism from the esteemed reviewers…”
…
” the handling editor (and the executive committee) are not convinced that the new view presented in the controversial paper is wrong.”
So the editor invited the esteemed reviewers (and in at least one case, some persuasion required) to give their views, and then said he didn’t believe them, but gave no reasons.
Indeed. And depending on your point of view, the editor may have been wearing a white hat, or a black hat.
Nick Stokes: So the editor invited the esteemed reviewers (and in at least one case, some persuasion required) to give their views, and then said he didn’t believe them, but gave no reasons.
Thereby opening the discussion to the readership. The reviewers and their friends remain free to publish criticisms.
The reason for peer-review is to ensure quality, to make sure that the papers have been critically read before being published (by fresh sets of eyes), and to prevent the most obvious forms of cronyism.
We know that is small fields it is impossible to prevent cronyism anyway.
I see no problem with the editor doing THEIR job and making the final decision and then letting science work the way is is supposed to work. We have enough gate-keeping as is. I’m surprised that Mosher in particular is worried about the reviewers given their experiences with the BEST paper. This took 2 years, so the decision to publish anyway was not taken lightly. They did not disregard the reviewers comments. In fact, they published a special note, thus ensuring that this paper will get a LOT of scrutiny from the readership. Considering the butt-loads of crap published in all kinds of fields, not sure why this one should be excluded because it is a different way of looking at a problem. That does not usually keep papers from being published.
It’s a journal, not the Vatican.
I remember the blog discussion from the Air Vent in 2010 and the comments at the ACPD site. You may be correct that this has been a curious review process. The only thing more bizarre however was the fierce opposition to the discussion paper, and your comments still stand out. I believe that it was in part the nature this opposition that influenced the decision to publish this paper. A case of “methinks the lady doth protest too much”.
It was also this unusual opposition that drove me to conduct some simple empirical experiments back in 2010. I found that all that it would take for the Markarieva Effect to work would be for a rising moist air mass to be slightly more diabatic than a dry air mass. This of course would be the case because a rising moist air mass is full of radiative gases that can emit as IR some of the latent heat released during condensation.
I now believe that is why the paper had to be trashed. It was getting too close to the main role of radiative gases in atmospheric circulation, both horizontal and vertical. After being driven to physical experiment by the thousands of comments, wholly unsupported by empirical evidence, on the M2010 discussion paper, I have continued to experiment in convective circulation in gas columns in a gravity field.
I have found that without radiative energy loss from the atmosphere at altitude, convective circulation stagnates. When convective circulation stagnates in a gas column heated at the base, the gas column heats. The gas column can be heated and cooled a separate locations at the base, but it will still run hotter than a gas column heated at the base and cooled higher up.
This is the critical role of radiative gases in convective circulation, energy loss at altitude. Without radiative gases, strong vertical convection below the tropopause would stagnate and our atmosphere would heat. Radiative gases cool our atmosphere at all concentrations above 0.0ppm. Adding radiative gases to the atmosphere will not reduce its radiative cooling ability.
Nick, I may not have found this out if not for your opposition to the M2010 discussion paper.
Here is a diagram to illustrate that point –
http://oi48.tinypic.com/6zy1ky.jpg
The figure on the left shows normal convective circulation occurring. As you can see radiation of IR to space is critical to this continued circulation. The figure on the right shows what would happen shortly after the atmosphere lost its ability to radiate IR. Convective CIRCULATION “stalls” or stagnates, and the atmosphere heats. (this is just before the atmosphere goes isothermal and surperheats)
Konrad
I’m not in a position to judge whether or not this study has identified a major factor driving winds in our climate system, but this statement you made sounds like a big “mouthful”:
This sounds to me like a direct challenge to the hypothesis that radiative gases create a greenhouse effect in our atmosphere.
Have I understood that correctly?
(If so, I can see why this paper could be seen as a challenge to the prevailing paradigm, and has thus caused a lot of discomfort and opposition.)
Max
Manacker,
Yes, I am making a direct challenge to the failed hypothesis that radiative gases create a greenhouse effect in our atmosphere. Those proposing that CO2 emissions will heat our atmosphere do so on the basis of critically flawed equations. The “basic physics” of the “settled science” never correctly modelled the role of radiative gases in tropospheric vertical convection. They never modelled a moving atmosphere, or the role of gravity in biasing conductive flux between the surface and atmosphere. (the surface is more effective at conductively heating the atmosphere than conductively cooling the atmosphere).
It is said that extraordinary claims require extraordinary evidence. The AGW hypothesis essentially is an extraordinary claim that adding radiative gases to the atmosphere will reduce its radiative cooling ability. To date no supporting extraordinary evidence has ever been produced. What the M2010 incident shows is not only did some of the AGW promoters know their hypothesis was rubbish, but they have known for years.
Well, Douglas, they weren’t addressed, and there are many. But let me take just one. The assumption that, in the heading of Appendix A1:
“Linearity of condensation rate over the molar
density Nv of water vapor”
It was point five in my first discussion entry. It’s still here, and emphasised; the equation is S = CNv.
Now physically this is nonsense. There’s always some humidity, but it mostly isn’t raining. Condensation doesn’t occur at all until Nv nears saturation.
The justification given is the first order molecular kinetics of condensation. But this is wrong for several reasons. First it ignores the back reaction, evaporation. In unsaturated air, that is faster, so you get no condensation at all.
Secondly, no time scale is mentioned. In fact, these are nano-second scale reactions. If S = CNv really did apply, the atmosphere would dump its water on that time scale.
Of course it can’t. That’s the problem with naive kinetics. You have to look for the rate limiting process. And with condensation, it’s usually the rate of removal of latent heat. On the scale of the atmosphere, that takes many seconds to hours. Molecular kinetic issues are insignificant.
I agree 100%. The bulk of the atmosphere, N2 and O2 are the real GHGs. The radiatively active gases cool the atmosphere by radiating the atmospheric energy to space, just like roof windows in a greenhouse.
Edim,
Bingo. N2 & O2 are the true “greenhouse” gases . H2O and CO2 are the broken panes in the greenhouse. There will of course be a backlash from the defenders of the “cause” to increased discussion of this. Watch out for “baffegab”. If you see direct linear flux equations being used for modelling do not be snowed. Unless you see a complex program running such equations iteratively on discrete moving air masses you are not seeing correct modeling of the role of radiative gases in our atmosphere. You have it right. I suspect you are an engineer. If so you will know to question yourself, always check your work and anticipate that you may make errors. What you may not know is that other less ethical people exploit this to make you doubt yourself. They pose as as a “consensus”, they pretend to be many voices. This is the Alinsky technique. There are actually far fewer supporting the Knights of Consensus than you may think. They can be stopped. Science can be saved.
Nick I don’t think the editor has said he doesn’t believe the reviewers just that they shouldn’t be the last word on this subject. The authors accept that they shouldn’t be the last word, encouraging further debate and criticism, there is no reason why the reviewers can’t also take such a reasonable position.
But what is the point of having a review then? Why not just publish whatever comes in? In the interests of further debate and criticism?
But then again, what does “accepted” mean? The paper is already available for reading. Accepted should mean someone thinks it OK. Not the reviewers. And if the editor has some reason for thinking so, he isn’t saying.
But OK, I guess the paper is now “peer-reviewed”.
Nick Stokes, why not just compromise on it? They it say 4Wm-2 impact, you say 0, make it 2 and throw in a +/- 1 uncertainty. That’s climate science right? Average all the guesses :)
Nick
I agree the editor’s message is unusual and the delay was atypical — but it does look like the peer review process did what it was meant to do. It gathered feedback including significant criticisms from acknowledged experts. We (the authors) responded in detail to each and revised (i.e. improved) the paper to reflect all the inputs. Then the editor made a judgement. They were asking themselves if there was sufficient grounds to reject the paper: it appears they were unconvinced. I presume they weighed all the reviews, replies and changes — then they took the decision that is theirs to take. That all seems normal.
You are a publishing scientist right? I’m sure you acknowledge that papers that are eventually accepted often receive harsh harsh criticisms at some stage in the review process? That is normal. The authors can then defend, adapt the paper, and/or appeal. That is all quite normal. In our case this is all online so you can of course form your own opinions. But the editor is the final judge — they are not required to write a detailed assessment. That too is normal.
What is unusual is the fact that this time the editor offered any explanation at all. I don’t think I have ever seen that before.
Here is a fun blog here you might enjoy (a contrasting view): http://blogs.bmj.com/bmj/2013/01/28/richard-smith-the-editor-thinks-your-paper-is-nonsense-but-will-publish-anyway/
From the EGU.eu website
” after having passed a rapid access peer review process manuscripts submitted to EGU two-stage-journals will be published first of all in the “Discussions” part of the website of that journal being then subject to interactive public discussions initiated by alerting the corresponding scientific community. The results of the public peer-review and of the interactive public discussions are then used for the final evaluation of the manuscript by the Editor and, eventually, for its publication on the website of the actual journal.”
Nick it’s not just the reviews but the whole ‘discussion’ that the editor takes into account when deciding whether to publish. Presumably the editor believed the responses the author gave to the criticisms had some merit. And just like an author knows that their work may be deemed to have merit or not surely a reviewer also knows that their criticisms and recommendations may be deemed to be strong enough to block publication or not. A reviewer is not given absolute power in these situations. From my reading of the rules the reviewers comments (and the wider discussion) are a tools to be used by the editor, who has the final say. I’m sure in most cases editors give great weight to reviewers words and they know they are risking antagonizing them if they don’t, so I guess the decision isn’t taken lightly.Isaac Held from what I’ve seen from his blog responses seems like a fairly level headed person, I’m sure he can shrug off this ever so slight prick to his ego. BTW the editor doesn’t seem like some idiot him self (http://nenes.eas.gatech.edu/CV.pdf).
Nick I think you’re being too prescriptive,
HR,
It’s a paper of mathematics and physics. People in the Journal discussion and at blogs have pointed to shortfalls in the mathematics and physics. There is essentially no mathematical defence of the work by anyone other than Dr M. If Dr Nenes has something to say about why the mathematics is “not wrong”, I think he owes it to the reviewers that he invited, and to the journal readers, to say what it is.
Nick “shortfalls in the mathematics and physics” — we acknowledge your various comments have been helpful in pinpointing areas needing clarification but as far as I can see your points were addressed. We answered each. That helped us which we acknowledge. If you are not satisfied please be specific — I am assuming you have seen all the changes and replies (additional appendix etc). I know it takes time and energy but we do appreciate the efforts.
Many thanks
Douglas,
Sorry – long thread – my reply is in the block above.
I think Eli made the best point the other day. Who in their right mind would review again for this journal?
For the most part i think most scientists will just ignore the work, but the reviewers are the ones who have been ‘harmed’ here.
> Who in their right mind would review again for this journal?
Judy, perhaps?
“steven mosher |”but the reviewers are the ones who have been ‘harmed’ here”
I hope not. This looks like an issue that Judy is well placed to answer (as one of the harmed referees). Active publishing researchers are engaged in these kind of debates on a daily basis and most develop fairly thick skin — and in this case there was never a hint of disrespect to anyone.
Do you suggest that a journal process should never contradict a referee? Or that journals should set aside their editor’s independence to attract referees? Its an idea, but it might discourage researchers with novel ideas from submitting their ideas.
This is a good place to highlight again how grateful we were to the referees. We welcomed their feedback. We have said it before and repeat it again.
We knew in advance that Dr Held was likely to be critical (he has published work that assumes very different atmospheric mechanisms) but we also recognised that critical scrutiny would be good for the paper (held us sharpen up the logic, spot flaws etc). Each and every technical point Dr Held made was addressed in our responses. You can see our replies at the APCD site. We specifically reviewed the study by Spengler et al. (2011) that Dr Held holds as a comparison to ours and we explained the flaws in its physics (see http://www.atmos-chem-phys-discuss.net/10/C14894/2011/acpd-10-C14894-2011.pdf). We also argued why our ideas should not be given a “higher bar” than conventional ideas — even though that may seem reasonable at first glance (See http://www.atmos-chem-phys-discuss.net/10/C15085/2011/acpd-10-C15085-2011.pdf).
To expand a bit. It is unusual, but not unknown that an editor will publish against the advice of anonymous referees, or that a grant will be funded that does not score well. However, doing so when the names and reviews of the referees are public is not going to be taken well by anyone approached by Nennes to do a review. Even JC’s review did not endorse the paper, merely saying that it was interesting but flawed.
By adopting the open review process, the EGU took on a heavier obligation to the reviewers.
To give some detail concerning the peer-review process in EGU: the reviewers may decide if they publish their review anonymously or not. The reviewers are not obliged to reveal their names. Since all the correspondence concerning the paper between the Editors and the reviewers is privileged, there is no difference in this aspect with the conventional procedure.
Eli “Even JC’s review did not endorse the paper, merely saying that it was interesting but flawed.”
Can you point out where she said it was “flawed”? I was a little confuse so just checked back and didnt see that.
I thought she was encouraging. She said “The paper is interesting and provocative, and these ideas should be developed.” and made four specific suggestions where we could strengthen our case — each of which we responded to. She certainly appeared to think it was a potentially valuable contribution.
> There are four major issues that need to be fixed before the paper is accepted for publication: […]
http://www.atmos-chem-phys-discuss.net/10/C11235/2010/acpd-10-C11235-2010.pdf
Douglas, the “harm” done to reviewers can take on many forms, not merely disrespect. For example, based on this this experience folks might look at reviewers differently than they did before. For example, I now have to choose.
1, was the editor mush headed
2. Was reviwer X an ineffective critic.
in one case the editor suffers a harm in the other case the reviewer suffers.
I think Eli’s point stands and I haven’t seen any credible challenge to it.
So help me out here. Explain to me why Eli is wrong. Another point. Basically the reviewers wasted their time. Since we rely on them to do science, those of us who pay their salaries have an opinion on this.
Willard
(greetings again)
I don’t know if you have done much reviewing but I would judge that a positive review if I had written it. These four issues were all specific constructive points – stated clearly. You offer what you think is needed to improve it and you expect the authors to address these (or explain why not).
My concern here is the use of the word “flawed” which may have a rather damning flavour which I did not see in the review. I guess by your criteria every paper every subjected to a rigorous review will now be called “flawed”. I used the word differently — if I was a reviewer and called a paper “flawed” I don’t think I would offer a clear list of fixes. Maybe it is simply a question of language.
Steven Mosher — Well if my previous answer didn’t help, and you require everything to be so clear cut that researchers cannot disagree unless one or other is “mush headed” or “ineffective” I am not sure I can persuade you. The world of science I live in is very different: it thrives on the dynamic of contrasting ideas and competing theories — that is a positive force. We seek out criticism and feedback because that is what drives us forward and keeps us from dogma. I don’t look for “mush headed” or “ineffective” critics I look for the ones I admire and respect. So I don’t recognise the scenario you offer. In Ireland we argue for the fun of it too … kind of banter — but it is also a way to engage and probe people’s ideas and I would often play the devil’s advocate simply to see where the argument goes and what we can learn. But certainly it is different in other cultures … maybe what you are used to is very confrontational.
If you think it is a key point perhaps the best option is to ask Dr Held and Dr Curry. You, Eli and I are simply speculating. We don’t know.
Science must be open to new ideas but it must also take into account what has been learned earlier. When anyone proposes that there are major faults in the earlier knowledge she or he cannot expect that others pay much attention to that unless sufficiently evidence is given on the significance of the new findings and on that the new approach does not contradict earlier confirmed knowledge.
Atmospheric physics has been studied and its understanding has been developed by innumerable scientists over long period. Jumping in and claiming that a mechanism that is well known but considered a very minor factor would actually be very important and require major changes in thinking is something that requires solid arguments in a well organized form that answers clearly the obvious questions. A paper that falls severely short of that is not worth publishing.
Pekka Pirilä
Hmmm so certain? No doubts at all?
Let’s try a thought experiment please. Imagine a world where our theory is correct but few know it as they like the theory they have. You are on our team, you know the theory makes good sense, and you are keen to get the ideas accepted. Together we publish the physical reasoning and various evidence and write papers highlighting the gaps and flaws in the conventional ideas. Now … just for the sake of argument how should we then convince a sceptic that our theory is worth engaging with (not believing necessarily but accepting it as a reasonable possibility worth testing or evaluation). Be as specific as you can. How should we do that in our imaginary world?
Douglas,
Your problem is totally in that you have not presented evidence at a level that would convince others.
There’s nothing new in your basic physics. Thus everybody accepts that. What you must do to get accepted is to put the physics in full context and analyze extensively enough the behavior of the large scale atmospheric phenomena. With the present level of evidence people don’t believe you and they never will, if you don’t provide the relevant evidence. Nobody else needs to care as long as they are not convinced at all of your claims.
The tools are there in the atmospheric models. If you cannot use them that’s your problem. You cannot expect that others would bother until you have given evidence accepted by them.
Douglas,
I believe the quote justifies Eli’s claim. If you prefer, you can replace ‘flawed’ with ‘with four major issues that need to be fixed’. I don’t think Eli will mind.
Mosher your assumption here seems to be that all concerned here are fragile flowers and the whole process is ego driven. The clue for all concerned is that the process is OPEN. This is a fact not hidden from anybody. If the reviewers can’t handle their opinion being challenged in an open manner they shouldn’t be reviewing for EGU. If the reviewers fail to read the rules of the EGU review process and feel harmed by judgements that fall within those rule then they shouldn’t be reveiwing for EGU.
It’s clear this review process is unusual but the editor has not stepped outside the EGU’s own rules. It should be pointed out the editor himself doesn’t seem like some idiot from his online CV. Mosher I think you assume the participants in this process (including the reviewers) are more fragile than they actually are, strange given your own robust style.
Thanks WIllard
I’ll consider that an agreement! Reassuring to know you can adjust your claims in the face of evidence.
Most welcome. Now your turn. Here’s Eli claim again:
> Even JC’s review did not endorse the paper, merely saying that it was interesting but flawed.
This does not entail that the flaws were redhibitory, as you are now suggesting. Nor does it entail that Judy would need to make an adversarial comment to use the word “flaw”.
Your concern does not seem to be about semantics, Douglas, but about public relations.
What difference does it make who reviews this stuff, if it is only a box checking exercise? They will probably have to recruit reviewers from the same suspects who review for G&G. If G&G does in fact have reviewers.
Willard “Your concern does not seem to be about semantics, Douglas, but about public relations”
its about both.
We wrote the paper and now the blog and hope for a rich discussion with anyone interested – so certainly we value public outreach. I guess that is what you mean “public relations”.
Accuracy matters to me in any case — as a scientist that is part of my value system. A request for clarification should not be called something else (e.g. an error or flaw requiring repair). Certainly it matters more if the work of my colleagues and myself is subject to such misrepresentation (then I have a greater responsibility to offer the correction).
Is that what you want me to acknowledge? I acknowledge it.
Hope that helps. I do appreciate the mischievous spirit but plan to focus on more substantive issues now.
Dear Douglas,
Thank you for the kind words.
Let me summarize our conversation here so far:
[Mosh] Who in their right mind would review again for this journal?
[willard] Judy, perhaps?
[Douglas] Do you suggest that a journal process should never contradict a referee?
[Eli] It is unusual, but not unknown that an editor will publish against the advice of anonymous referees, […] However, doing so when the names and reviews of the referees are public is not going to be taken well by anyone approached by Nennes [from Georgia Tech, let it be noted] to do a review. Even JC’s review did not endorse the paper, merely saying that it was interesting but flawed.
[Douglas] Can you point out where she said it was “flawed”? I was a little confuse so just checked back and didnt see that.
[willard, quoting chapter and verse] “There are four major issues that need to be fixed before the paper is accepted for publication: […]”
[Mosh] So help me out here. Explain to me why Eli is wrong. Another point. Basically the reviewers wasted their time.
[Douglas, to willard] I don’t know if you have done much reviewing but I would judge that a positive review if I had written it.
[Douglas, to Mosh] Well if my previous answer didn’t help […] I am not sure I can persuade you.
[Let’s skip Pekka’s here, for convenience.]
[willard] I believe the quote justifies Eli’s claim. If you prefer, you can replace ‘flawed’ with ‘with four major issues that need to be fixed’. I don’t think Eli will mind.
[Douglas] I’ll consider that an agreement! Reassuring to know you can adjust your claims in the face of evidence.
[willard] Thanks. Your turn. [Eli’s claim] does not entail that the flaws were redhibitory, as you are now suggesting. Nor does it entail that Judy would need to make an adversarial comment to use the word “flaw”.
[Douglas] A request for clarification should not be called something else (e.g. an error or flaw requiring repair).
***
I wonder who’s looking the most mischievous there, Douglas. Really.
Nick’s point raised by Judy’s was not a mere request for clarification. As far as I can see, it has not been addressed, except by excusing yourselves with something like “Oh, science shan’t ask for perfection. It’s just a theory anyway.”
You got lucky, Douglas. Please don’t push it.
And right now, with me here and at Eli’s, you are pushing it. Too much defensive rope-a-dope. See how you’ve been rope-a-doping in our little chats.
I say chats in plural because it is the second time, now, in a few days. Here was the first one:
http://rabett.blogspot.ca/2013/01/atmospheric-chemistry-and-physics.html
Little details, perhaps. Enticed by mischievousness, I know, I know. But as auditors ought to say:
http://climateaudit.org/2005/10/29/is-gavin-schmidt-honest/
Thank you for playing,
Due diligence,
w
Yes, as I mentioned on the previous thread, Isaac Held rejected this, the authors gave a response via ACP, and Held never got a chance to respond before the editor accepted it. In other journals, a reviewer that rejected a paper would at least be able to tell the editor whether the responses to his review were adequate, but ACP doesn’t seem to have that second round procedure, which is very strange. The responses to Held did not address his main concerns at all, so I am sure he would have still rejected it, because the paper hardly changed from the one he reviewed.
Jim “The responses to Held did not address his main concerns at all, so I am sure he would have still rejected it, because the paper hardly changed from the one he reviewed.”
It would indeed be interesting to know what he might have said — and your analysis may be correct. But note that referees do not reject papers — they offer a case for doing so.
Let’s be specific — which point(s) in particular do you think should have led to a rejection?
Thanks
Douglas, his main reason for recommending rejection appears to be a valid one that the paper’s equations do not prove the case it is trying to make, and there is some incoherence in the arguments. Since the paper was about equations, it suffers from not successfully proving its point.
Jim D “it suffers from not successfully proving its point”
I believe we addressed each specific point that Dr Held raised. It is hard to take this discussion further without you being more specific. A theory is a theory — proof comes later and is based on evidence.
“So the editor invited the esteemed reviewers (and in at least one case, some persuasion required) to give their views, and then said he didn’t believe them, but gave no reasons.”
Nick, there are essentially two levels of criticism that a reviewer can supply, criticism of individual components or of the finished product; the model.
You might not like the model presented by the authors, but if all their steps are correct, then it should be published.
Here the authors have come up with a large number of components, which it appears stand up to mathematical and physical testing. The authors then stick them together and come to some sort of model.
A referee can find fault in the former, blocking publication. However, you can’t block a publication just because you think that a model is wrong.
It’s rather like accepting that all the components of an airplane will work, but if they are put together as suggested by the engineer the thing will still not fly. It might or it might not. If you have no empirical test, you have to let the authors give their view.
Judy,
If this paper is found to be in line with observations, does this mean that the General Circulation Model need to be revised?
Thank you, Nabil, for asking this question. I would like to add a further thought. If the weather forecast models are revised, might it be possible for them to be more accurate?
I agree. The weather forecast has been terrible. They projected about freezing temperatures for the weekend. We enjoyed 50 and 60 degrees instead.
Gcm or at least the literature have a difficult task reconciling the metrics with two underlying theories of opposing signs with say synoptic storms and baroclinicity due to the reduction in meridional gradient in a warming world,and the increase of water vapor,hence life birth cycles .
Gcm cannot resolve at mesoscale levels (which is the scale length of weather systems) which makes prediction both difficult and problematic as when the coarse grain resolution is enhanced we move beyond the physical laws per se.
Nabil,
avoiding revision of GCMs was the reason many thought the Knights of Consensus were fighting so hard to trash the M2010. It may be more serious than that. All it takes for the Makarieva Effect to work is for a rising moist air mass to be more diabatic than a dry air mass. This may involve radiative energy loss. Possibly IR from all that water vapour. The Knights of Consensus cannot allow any heresy involving atmospheric circulation being driven by radiative gases so they tried to trash the paper.
True, true… as important as cycles are phases–i.e., nominally–those under the surface of the Sun that mostly are hiddent to us–and, importantly what we see right under our noses here on water-Earth.
many good comments can be quoted: In current models in the absence of a theoretical stipulation on the circulation power, a reverse logic is followed. The horizontal pressure gradients are determined from the continuity equation, with the condensation rate calculated from the Clausius-Clapeyron law using temperature derived from the first law of thermodynamics with empirically fitted turbulence. However, as we have seen, to correctly reproduce condensation-induced dynamics, condensation rate requires an accuracy much greater than γ << 1. Meanwhile the imprecision of the first law of thermodynamics as applied to describe the non-equilibrium atmospheric dynamics is precisely of the same order of γ. The kinetic energy of the gas is not accounted for in equilibrium thermodynamics.
I think that this paper, and the scholarly discussions addressing it, constitute a step toward filling one of the “cavities” of the science.
Thank you for picking this up. We consider this an important point. It was not made clearly in the paper, so it is what actually belongs to the “update”.
Judith Curry
While I am unable to judge whether or not the proposed mechanism is a major factor in driving winds, it certainly sounds plausible and worth further investigation.
It’s just another one of the many uncertainties out there on what drives our climate and weather.
Is this something that could be tested experimentally?
And, as Nabil Swedan has already asked, should this be incorporated into the GCMs?
Seems like it raises more questions than answers.
Max
manacker,
When wind falls down, it loses potential energy. It takes energy to raises the wind back up and maintain the circulation. Definitely, it is the latent heat of condensation that provides this energy, there is no other. The lower atmosphere is adiabatic invariant and exchanges no sensible heat with its surroundings. Th authors of the paper failed to prove that the lower atmosphere is adiabatic invariant, and this is my main critique of the paper.
Nabil Swedan
Thanks for response and explanation.
Max
Should it be in GCMs?
1. unconfirmed by empirical study
2. probably wrong
3. if correct, inconsequential.
so, re framing, should unconfirmed, likely wrong, inconsequential physics be put into models?
Some GCM are open source, believers should knock themselves out and stop pestering people with confusingly notated papers. write some damn code
Thank you for your interest in our paper. Even if negative, it makes people think.
I think we addressed your point on writing some code in our post (“You should produce a better model than existing ones”).
There is theory and there are models. They are not equivalent. Theory comes first.
What does it mean in our case? We present a theoretical estimate of circulation power that fits the observations. Such an estimate does not exist in the meteorological theory on which the current models have been built. So where the current models are just fitted to reality (because of lacking theory), we offer a testable quantitative framework. And this is not about some detail — it is about a key parameter of atmospheric circulation: its power.
This is at least interesting, as admitted by even most critical peers (see, e.g., p. C14689) in the review of Dr. Held. At most it means a re-appraisal of our understanding of how the atmosphere works. So I see no harm if our work receives some attention.
“unconfirmed, likely wrong, inconsequential physics” was it Steven? Then why did the Knights of Consensus, Joel Shore, Nick Stokes, Eli Rabett and Jim_D all charge into the melee back in 2010? Surly if the paper was “inconsequential” that wouldn’t be required. The paper wasn’t inconsequential. Anastasia had accidentally got too close to exposing the big lie of AGW.
All it would take for the Makarieva Effect to work, would be for a rising moist air mass to be more diabatic than a dry air mass. And of course it would be as it is saturated with the most important radiative gas in our atmosphere, H2O. But if radiative gases could have some role in horizontal air circulation, what would be the consequences for vertical circulation? Further investigation would reveal that radiative gases are critical to all vertical convective circulation below the tropopause. Without this our atmosphere heats.
To keep pushing a failed hypothesis as fact years after you knew it was wrong is rancid behaviour. Trashing science in another discipline to defend the “Cause”? Simply putrescent.
I would vote for wrong rather than inconsequential because a loss of vapor without latent heat release would result in an increase in density and loss of buoyancy as dry air filled in to replace the lost vapor and equalize the pressure (dry air is denser than vapor). The key process for buoyancy generation is the latent heat.
Anastasia. you did not address my point about writing code.
1. you will address this point when you do IN FACT write some code.
That is ‘addressing’ the argument.
2. You couldnt explain the math well enough for a third party to code, see Nick stokes comments.
Merely writing words that come after mine does not constitute addressing the issue. your equations were not clear enough for third parties to understand ( so nobody could code up your stuff for you ) and you refuse to do the work yourself. You have not addressed the issue. Write some code and you will. This isnt addressed my writing more words its addressed by writing more clear math or your own code
Steven, ” your equations were not clear enough for third parties to understand ( so nobody could code up your stuff for you ) and you refuse to do the work yourself. ”
I do not think that we deserve this reprimand. In our ACPD paper (2010) we introduced a new physical concept. No code at this time, just discussing basic physical ideas and demonstrating their potential significance.
Now then we took this concept, wrote the full system of equations and solved them for two particular problems: hurricanes (where vertical velocity can be neglected) and tornadoes (where vertical velocities are essential and cannot be ignored). Unlike in the existing hurricane models, we did not use any a priori fitted parameters to match the observations. But in both cases we obtained meaningful results.
We have been doing some relevant additional research as well. We remain hopeful that there are interested researchers who will join us in developing this topic. It might be a matter of time.
So, of course, if you command “write the code” and I do not, this can be interpreted that I have not addressed your comment. But my point simply was that “writing the code” and obtaining a meaningful scientific result is not the same. The latter can sometimes go without the former.
Steven Mosher
This is a discussion about basic science. We have never demanded you or anyone to put this in a GCM (so the reaction seems strong). The arguments included above explain why we don’t want to do it either. Even if we felt able (and I for one certainly dont) that is not our interest at this point. We have aken considerable efforts to show why this is a powerful mechanism and hope that it will be evaluated. If you cannot follow our derivations and logic this discussion is an opportunity for us to clarify.
The controversy is not about the immediate local consequences of condensation or about any other issue of such basics, it’s about the claim that you present a previously disregarded mechanism for driving winds. For that you should present analysis that’s relevant and strong on this level. That may well require rather extensive use of atmospheric models. It’s your duty to do that if you wish to be taken seriously.
Pekka Pirilä
Rmember we are offering a theory to be evaluated. We are not claiming final proof. Its an important difference. The model idea is one I really dont get at all.
In any case I think we do what you are asking already. We characterise the pressure forming process in analytical form and then show that it has sufficient power to drive global atmospheric circulation. (All without models).
Non-transparent models do not and cannot offer strong evidence (read the blog above). Real world data do. Evidence is building up too (see http://www.biotic-regulation.pl.ru/index.html). e.g. There are various estimates for hurricane systems all based on simple analytically defined physics that match with observations. There are also empirical matches with forest cover rain and seasonality. I dont know if you have looked at all those.
You may be offering something to be evaluated, but you should not expect that it will be evaluated in any depth until you can give good enough reasons for others to do that.
People have all kind of ideas and theories. A very small minority of them will ever be evaluated in depth by scientists. In most cases that’s the correct fate for those theories. The very few exceptions must be introduced well enough for getting past the first hurdle. It seems that you haven’t fully succeeded in that even though you got published. So far my feeling is that this is the right fate for your ideas.
Pekka Pirilä
One last shot (late here):
Maybe it is easier to look at this in terms of specifics. We indeed address an effect which Dr Held noted has been “traditionally considered to be small”. We provide the first detailed analytical account of this effect and find that it is not small but capable of doing significant work by accelerating air masses.
In section 4.3 in our paper: “Regarding previous oversight of the effect” we consider and explain how the idea that it was small can be explained if it actually isn’t.
Our demonstration of it being big enough to care about comes from showing that the power of this effect over the world is sufficient to generate atmospheric circulation. It’s a rough estimate but it is a good one — if it was “small” it wouldn’t work (its also the first time such an analysis from first principles has been offered … your models cannot do that). There are other estimates from some of my co-authors showing similar success with certain aspects of hurricanes. So, now, why is this all so minor for you? These all sound interesting – right? If not what’s wrong?
Pekka Pirilä | February 1, 2013 at 7:52 am |
How condescending. Perhaps you meant evaluation by *other* scientists. I’d use stronger language than “condescending” but it would probably get snipped.
You choose what you allow yourself to be pestered with don’t you?
Just ignore it if you think it inconsequential even if right. It’s not like your opinion matters in any event.
Douglas.
The effect, if correct, is inconsequential. The way for you to respond to this objection is to actually do some work or write in such a way that others can do the work. You’ve done neither. So,don’t assume that you have addressed the concern when you have not. Its pretty easy. Just say
“we refuse to do the work necessary to incorporate this into a GCM and we refuse to describe what we think in such a way that others could implement it”
I’ll return, for example, to the comments made by Nick:
“But your justification was based on molecular kinetics, and you incorporated it in a continuum equation. Now you seem to be saying that, if that’s wrong, it’s still true on some larger averaged scale. But what is the justification for that? And how can an averaged result be incorporated in differential equation maths?”
Simple question; How can an averaged result be incorporated in differential equation maths?
Your possible answers are.
1. Show how
2. Say it cant be done.
Mosher everybody participating in the scietific endeavour realises that there are limits to every study. It’s common for critics to demand more and more evidence to support work. But it’s mistaken. These authors have what they have and are putting it out for consideration. The common refrain is “beyond the scope of this study” and the authors have adequately expressed that. Stop pestering them for something you know they cant provide.
Steven Mosher,
“The effect, if correct, is inconsequential.”
Not really. Water, water vapor and ice are the source of the majority of the heat capacity that regulates heat flux. The effect may be small compared to an over estimated “average” radiant impact, but 0.8% is significant compared to 1%, which is about the “negative” latent feed back at the “true” surface.
Mosher,
Leaving aside the question of who, ultimately, is correct, I have to say that, from this layman’s perspective, at least, you’re coming off as completely bent. Your insistence that the first stage of a challenge to the GCMs should be to re-write them is just insane. David’s observation that the current models were built up by thousands of people, over years, if not decades, seems to me to dispatch that suggestion, all by itself. Obviously, re-writing will require a concerted effort of numerous people, and, therefore, long before that re-writing begins, a process of devloping a concensus to do so is a prerequisite. And this disussion seems, clearly enough, the first step of a the process for developing that concensus.
This process has already cast light on degree to which the models are built on fudge-factors and false assumptions, and I’m already greatful for that enlightenment. The way they’re held up as “proof” of the concensus model seems particularly threadbare to me, now.
Steve,
You are starting to sound like you have a stick up your butt. The concept of theory coming before code writing does not appear to be as big a deal as you seem to be making it.
You don’t think much of the theory. Fine. Why attack it or Dr Makarivea on the issue of model code?
Steven Mosher: . if correct, inconsequential.
Why do you say that? If correct, it is the same order of magnitude as the hypothetical equilibrium response to a doubling of CO2 concentration.
Steven Mosher “The way for you to respond to this objection is to actually do some work or write in such a way that others can do the work”
Thanks for the suggestions.
Note that all the authors but myself are working in a foreign language. We have sought to publish in a discipline with conventions very different than our-own. Now we have taken all the steps needed to publish our ideas in your language in your discipline in your journals and we have stopped by to discuss it with you. We do that because we think you might be interested (isn’t that why you are here?). But now you want us to build your models too. Demanding us to do it all ourselves seems a little like CERN engineers demanding the particle theorists should shut up about their theories until they have made their own colliders.
I’ll speak personally: what motivates me at this point is the question whether these theories are true and what it means in terms of land-cover and rainfall reliability for millions of people around the world. GCMS don’t help (they will not clarify the truth of the physics or be accurate enough any time soon to define concerns). So I have other priorities. If someone else wants to do it I can acknowledge value in the GCM model option. I do see why climate scientists need to do simulations to test ideas — but it cannot provide the kinds of proof that I find convincing (epicycles were long considered the best way to forecast planetary motion).
So, what would I hope for with your GCMs? I would hope we find GCM people with skills and interests who wants to do it, have the time to invest and, if the equations are a problem, can ask guidance. Doesn’t sound too hard. Lots of science is collaborative these days.. Let’s hope that I’m right.
You do sound like you have tried to follow our paper which I do appreciate — if you wanted to try and model the ideas yourself I believe that help would be available.
I think Mosher wants to see some can openers and phenomenological curve fitting. Calling Dr. Pratt!
I agree with the basic premise that condensation does involve an effect on atmospheric pressure but that is simply a mirror image of the effect of evaporation lower down.
The question seems to be whether there really is something new in this that is not already implicit in the known workings of the water cycle and adiabatic ascent and descent within the main high and low pressure cells that together form our permanent climate zones.
It is clear that the water cycle enhances the adiabatic movements within an atmosphere and perhaps this is the first attempt to quantify and explain that enhancment in detail.
Certainly potential energy is at the heart of the matter and latent heat is indeed a form of potential energy so this could well be consistent with my own ideas regarding the water cycle as a regulating factor.
Whether it is a complete answer to the question as to how the atmosphere might stay at the same temperature despite variable radiative characteristics of constituent gases may be doubtful (I think it is primarily a pressure based process involving adjustments in atmospheric volume) but it may well be a useful step forward in explaining how the parts of the system fit together.
It is my understanding that radiative gases provide an additional radiative route for energy loss to space that non radiative gases fail to provide.
Thus with GHGs in an atmosphere the circulation can slow down because more of its job of maintaining top of atmosphere energy balance is done for it by those radiative gases.
The Makarieva paper makes an attempt to quantify the extent to which the water cycle and the phase changes of water could further enhance system stability by accelerating (or indeed decelerating) energy loss upward in response to more (or less) GHGs in an atmosphere.
“with GHGs in an atmosphere the circulation can slow down”
On the contrary, without radiation to space from the upper troposphere atmospheric convection would stagnate. In this great heat engine, the major heat input is at the earth’s surface and lower troposphere, the “boiler’ so to speak. The upper troposphere is correspondingly the “condenser”. There have been many “heated” discussions on various blogs as to the hypothetical structure of the atmosphere composed of no greenhouse gases, only non radiative gases.
I can’t see how convection would stagnate in a non radiative atmosphere because there would still be a declining pressure gradient with height leading to a lapse rate and convection.
Stephen, the greater the lapse the more stable the atmosphere. Write that down.
Can anyone please explain for me (i.e. in simple, non-technical terms) the significance of this paper to the estimates of climate sensitivity that come from the models?
Does it mean that, if this new physics were incorporated in the models, the models would say climate sensitivity is higher or lower than what they are currently saying?
Please explain why (higher or lower)?
Could this close the gap between the climate sensitivity estimated by models and climate sensitivity estimated from empirical evidence (as in the draft IPCC AR5, WG1, or perhaps even down to what Nic Lewis is suggesting)?
The ‘sensitivity’ is defined as the average of multi-model ensembles. Each of the members of the ensemble is a non-unique solution of a set of nonlinear equations. The ‘member’ is selected on the basis of ‘a posteriori’ solution behaviour. It looks plausible so it is in. So sensitivity is based on what seems plausible – think of a number – and not an any unique, deterministic solution to an equation.
Sound mad I know – which is why I persist in quoting people like James McWilliams and Tim Palmer.
Robert Elliston,
Thank you for your comment. What I am asking is what difference, if any, the physics presented in the paper would make to the climate sensitivity obtained from the models? I am referring to the pdfs, that are derived from the models, in Box 12.2, Figures 1 and 2 here: http://www.stopgreensuicide.com/Ch12_long-term_WG1AR5_SOD_Ch12_All_Final.pdf.
‘Atmospheric and oceanic computational simulation models often successfully depict chaotic space–time patterns, flow phenomena, dynamical balances, and equilibrium distributions that mimic nature. This success is accomplished through necessary but nonunique choices for discrete algorithms, parameterizations, and coupled contributing processes that introduce structural instability into the model. Therefore, we should expect a degree of irreducible imprecision in quantitative correspondences with nature, even with plausibly formulated models and careful calibration (tuning) to several empirical measures. Where precision is an issue (e.g., in a climate forecast), only simulation ensembles made across systematically designed model families allow an estimate of the level of relevant irreducible imprecision…
In each of these model–ensemble comparison studies, there are important but difficult questions: How well selected are the models for their plausibility? How much of the ensemble spread is reducible by further model improvements? How well can the spread can be explained by analysis of model differences? How much is irreducible imprecision in an AOS?
Simplistically, despite the opportunistic assemblage of the various AOS model ensembles, we can view the spreads in their results as upper bounds on their irreducible imprecision. Optimistically, we might think this upper bound is a substantial overestimate because AOS models are evolving and improving. Pessimistically, we can worry that the ensembles contain insufficient samples of possible plausible models, so the spreads may underestimate the true level of irreducible imprecision (cf., ref. 23). Realistically, we do not yet know how to make this assessment with confidence.’ http://www.pnas.org/content/104/21/8709.full
Any change in a model can produce divergent solutions that are not predictable beforehand – it is the nature of the nonlinear Navier-Stokes equations.
‘Lorenz was able to show that even for a simple set of nonlinear equations (1.1), the evolution of the solution could be changed by minute perturbations to the initial conditions, in other words, beyond a certain forecast lead time, there is no longer a single, deterministic solution and hence all forecasts must be treated as probabilistic. The fractionally dimensioned space occupied by the trajectories of the solutions of these nonlinear equations became known as the Lorenz attractor (figure 1), which suggests that nonlinear systems, such as the atmosphere, may exhibit regime-like structures that are, although fully deterministic, subject to abrupt and seemingly random change.’ http://rsta.royalsocietypublishing.org/content/369/1956/4751.full
We are not at the stage of having ‘systematically designed model families’ – thus the first hurdle for precision is not cleared. Solutions are literally chosen subjectively from many possible solutions and the range of possible solutions remains unknown.
I have said this many times in many different ways – but the understanding of dynamical complexity remains elusive. These models are most certainly chaotic in the sense of theoretical physics as shown by Lorenz in the early 1960’s.
This work suggests to me that the warming effect of rising CO2 is much stronger than previously thought. If warming results in enhanced winds that draw heat from the surface high into the atmosphere then the surface will need to warm more in order to achieve equilibrium.
I would urge climate scientists to test this mechanism, among others, in their GCM models so they can find out exactly how much higher climate sensitivity would be.
Modern
“How much higher climate sensitivity would be”.
(Or, “how much lower”.)
Max
Modern Science Ideas,
Climate sensitivity is a hypothesis and concept only resulting from the greenhouse gas effect hypothesis. We went by this hypothesis because we had no data back in the 1830’s. We are in 2013 and have solid data to develop theories and laws of the climate and we should not settle for a hypothesis, particularly on a subject that can greatly affect our llivelihood. I think we should start looking seriously at papers similar to this one.
Parody?
Hard to tell due to Poe’s Law:
“Poe’s law, named after its author Nathan Poe, is an Internet adage reflecting the idea that without a clear indication of the author’s intent, it is difficult or impossible to tell the difference between sincere extremism and an exaggerated parody of extremism.”
quote
I would urge climate scientists to test this mechanism, among others, in their GCM models so they can find out exactly how much higher climate sensitivity would be.
unquote
I agree. [Insert winky smiley here.]
JF
Dave Springer, thank you for Poe’s Law: i’d not heard of it.
Consider the profile of a typical humid atmosphere. As we rise in the profile, both pressure and temperature fall, eventually the water condenses and two things happen: the condensing water releases its latent heat and pressure drops because codensed water has far less volume than the gas it replaced.Dynamically we now have a new situation. We are in the micro climate of a cloud. The latent heat released will tend to reheat the gases in the cloud and this can create an unstable situation. That it does create an unstable situation is evident because clouds continue to exist. Clouds are nothing but condensed water vapour..The latent heats of evaporation and condensation are the same, so this situation can continue so long as there is a supply of humid air from below, the cloud will grow, or shrink if the supply reduces.
If you accept the above theory then the ‘new’ theory certainly applies in the lower troposphere, but breaks down at the cloud level.
“the ‘new’ theory certainly applies in the lower troposphere”
I was thinking of fog. But littlr vertical or horizontal movement occurs in thick humid fog. It is like being in cloud, so that discredits the ‘new’ theory at all levels in the troposphere.
The heating that occurs from condensation reduces the environmental lapse rate. When the environmental lapse rate is smaller than the dry adiabatic lapse rate the atomosphere is conditonally unstable and some weather event is a likely consequence in the restoration to stability.
Do you know the spectrum of the photons emitted when gaseous water undergoes the phase transition?
Do you know the absorption/light scattering properties of gaseous and solid water in the low pressure atmosphere?
Thank you, David, for ypur explanation in terms of lapse rates.
Thank you, DocNartyn, for your reply.
Yes, the phase transitions result in both enission and absorption of photons as kinetic energy, for water. For a stable cloud they would be about equal. No, I don’t know the spectrum other than it is IR..Yes, water is a so-called greenhouse gas absorbing IR, around 40% of the planet’s radiatiion. For climate implications see my website above..
In the past I have read commenters saying things like:
“Coastal winds are driven more by temperature differences between land and ocean.”
“Hadley cells are driven by temperature differentials between the equator and higher latitudes.”
So, is that so?
Here is an small exaggerated example:
I have two horizontally adjacent parcels if air, both with an identical specific gas constant, R.
Parcel one has a pressure of 10,000 Pa and a density of 1 kg/m³.
Parcel two has a pressure of 20,000 Pa and a density of 2 kg/m³.
Which way doth the wind blow between the two parcels?
( For those that may not realize the relationship being addressed, both parcels have identical temperatures by the ideal gas law of P/ρ = R·T )
Some would say no wind will blow at all, they have equal temperatures. But I tend to think they are simply viewing a one-dimensional vertical-world but applying that concept horizontally. With gravity involved, this is true vertically in a planetary atmosphere, no vertical wind would exist. The density, ρ, will track the pressure, P, without an energy density differential between the two. Seems this does not apply when viewing the same example horizontally.
Perhaps some of the scientists and engineers here may help me shed some light on this curious question, I think it is critical in this substance of this paper.
I tend to think the wind blows from parcel two to parcel one because of both the higher pressure and the higher density of parcel two.
I might also add that you can easily think of a crossed example of two parcels with different temperatures but identical pressures, then the densities are different.
Also, there is an example with two parcels with different temperatures and pressures but the densities are identical.
Which takes the precedence when looking at the horizontal wind, the pressure difference, the density difference, or always the temperature difference?
I would say pressure is always the ruling factor in all cases.
There may be a tendency for temperature to follow the pressure but also, because of the densities, this may cause this rule-of-thumb to fail. Can any physicist or meteorologist out there either agree or disagree with that? Just trying to firm my understanding in these areas.
A Google search for the phrase “exact and inexact differentials” finds many thousands of explanations, in both articles and books, of this subtle (yet crucial) thermodynamical distinction.
And so it is concerning that, beginning with equation (1)
Makarieva et al. have (inexplicably) adopted a notation that fails to make this crucial distinction.
Concern increases further when these authors ignore the vast engineering literature on the flow of steam (which is precisely the domain where the effects they are describing are most prominent).
This paper makes extravagant claims, and had a tough time in review. This could be because:
(1) the reviewers are demanding an extraordinarily high standard of evidence and clarity of exposition, before accepting transformative findings, and/or
(2) the paper is obscurely written, with poor notation, and scanty references, and/or
(3) interpreted literally, the physics is just plain wrong, but the author’s notation is so poor that they themselves do not realize it, and/or
(4) once minor notational infelicities are corrected, the authors have found a novel path to rederive well-accepted thermodynamical models.
Conclusion On the evidence, it is entirely plausible that (1,2,3,4) all are correct! And that is why (quite properly) much further work will be required of the authors, along with independent derivation of their results by other workers using different methods, before these ideas are accepted — if indeed these ideas are not old wine poured into a new (bad-notation) bottle, or alternatively, just plain wrong.
Not my feild, but I’ll go for 3, based on their other work in the biological feild, which I have had a bit of a look at – ‘just plain wrong’ sums up some of their claims there.
Wacky stuff, might be another reasonable summary.
I’d like to have a look at the biology work you mention. Linky?
http://www.biotic-regulation.pl.ru/bre-vers.htm#versus3a
or this
http://www.bioticregulation.ru/offprint/genvar11.pdf
Nevermind. Found it. Read it last year. Biotic Forest Pump.
Thanks anyway! Didn’t see your reply before posting I’d found it already. I skimmed it last year and didn’t think the effect would change the price of tea in China, so to speak, and didn’t spend any more time with it as a consequence. Not feeling any different about it now either.
Michael
The stuff you cited on “biotic regulation of climate” is interesting (I wouldn’t call it “wacky”).
The hypothesis that the biota exerts a real and significant regulating effect on our climate sounds reasonable, even if it’s not corroborated by any empirical evidence.
Where the paper gets into trouble is toward the end. Let’s go through the last paragraphs:
(Unsubstantiated. Physical evidence for this claim?)
(Purely conjectural. There is no physical evidence to support this suggestion, which even IPCC is beginning to back away from.)
(Purely conjectural. This is based, among other assumptions, on an unsubstantiated premise above on human-induced extreme weather.)
Given this, it is remarkable that the climate stability phenomena receive so little attention within the modern scientific community, theoretical biologists included. The biological theory just accepts like axioms statements that, if re-evaluated, may completely undermine its current theoretical foundations.
(Too much uncertainty regarding the existence and magnitude of “biotic regulation” of our climate and the purported negative human impact on this mechanism; no mention of possible positive effect on biota of higher CO2 concentrations and/or slightly warmer temperatures.)
So it’s a nice idea, but the conclusions are flawed IMO.
Max
They are saying that the loss of water vapor reduces the pressure and effectively sucks the air up regardless of buoyancy. This does not account for the environmental temperature and how stable the atmosphere is to convection. According to their theory, this pressure gradient operates just as strongly in any conditions. What really happens is that the latent heat release warms the ascending air which only rises conditionally on the warming being strong enough to keep the lifted air buoyant relative to the environment. Convective instability is about the best known and validated process in meteorology. Weather forecasters rely on being able to tell whether a temperature profile supports deep convection.
Jim, thank you for your persistent interest.
“Convective instability is about the best known and validated process in meteorology. ”
Had it been the case, there would have been no problems in accounting for moist dynamics in current models. But the problems are very serious precisely with reproducing the observed circulation intensity in the moist atmosphere (be that, for example, hurricanes or monsoons).
But I agree that it is certainly the best known theoretical concept. It was introduced when there were no models and no program code, and still is believed to have merit. We, on the other hand, offer another one and argue that this new one is more important.
“What really happens is that the latent heat release warms the ascending air which only rises conditionally on the warming being strong enough to keep the lifted air buoyant relative to the environment.”
This is only half of the story. In simple words, in order to lift a moist air parcel, you must draw a dry air parcel down: this is what circulation is about. Since the air descends dry adiabatically and becomes warmer than the surroundings, there is an upward force acting on it that prevents the descent. The potential energy associated with the positive buoyancy of your rising parcel must be enough to allow the warm dry air parcel to descend. When you calculate what is left to generate kinetic energy, you will find that under most conditions the two processes basically cancel each other, so their overall impact on dynamics is negligibly small. Basically, it is a well-known point for many years.
For our effect there is no such compensating process in the descending branch of the circulation.
The ideas of convective instability are founded on basic thermodynamics and don’t need models to prove them. If your theory can distinguish convective stability from convective instability, I would be very surprised, but even then, it will only be matching the old theory that works already. For air to rise, it only has to be warmer than its surroundings, which is how thermals work. Clouds are like thermals but are assisted by latent heat release. This is all very well explained already and needs no new theory because there are no gaps in the understanding. If you think there is an unexplained phenomenon, your paper has done a poor job of describing what that is exactly, because there is no point of a new theory unless there is something unexplained by current theories.
“For air to rise, it only has to be warmer than its surroundings, which is how thermals work.”
This is only true for dry air and demands the existence of an external horizontal temperature gradient. As you know, in hurricanes for example such a gradient is absent.
So, let us keep focused on moist processes. For a moist adiabatic ascent, it is not enough that the rising air parcel is warmer than the surroundings. It is also necessary that the descending air parcel is cooler. Meanwhile as the descent occurs dry adiabatically, the dry air parcel tends to be warmer too. The net result is that the degree to which “clouds are assisted by latent heat release” is on average negligible.
Jim “because there is no point of a new theory unless there is something unexplained by current theories.”
As we clarify both in the paper and in the post, current theories do not provide a quantitative estimate of circulation power. Our new theory does.
Anastassia, your concept of buoyancy is not correct. Hurricanes only work because the rising air is warmer than the environment, even if only one or two degrees. The environment does not have to descend much because there is more of it, so a small descent by a large area compensates a large ascent in a small area (updraft core).
Jim, “The environment does not have to descend much because there is more of it, so a small descent by a large area compensates a large ascent in a small area (updraft core).”
These are all ideas that are plausible at the first sight but, as we emphasized in the paper and the post, lacking a theoretical proof. When the descent occurs on a large area, however small the local work against positive buoyancy of the descending air parcel, total energy expenditure will be comparable with potential energy released in the updraft because of being integrated over a large area.
My sense is that many of the commenters cannot quite get their arms around your concepts..wacky stuff is one characterization. I kind of like the fact that you are challenging conventional thinking,. Many in the scientific community nearly 100 years ago thought Einstein’s ideas were wacky as well. Keep pushing the envelope.
Jim,
Thermals over the ocean are next to nil. There’s very little difference between SST and the surface air temperature. On average it’s less than 1C. Over dry land the difference can easily be 40C. High winds happen in the dryest of environments and are driven by thermal convection but over the ocean it’s all about water vapor which ascends because it’s lighter than air not because it is warmer. Condensation at altitude then causes a pressure drop which is by far the most significant driver of vertical motion – bouyancy (over the ocean) is like the starter motor which sets the main engine in motion. Over dry land there is no main engine only the starter motor but that starter motor can be surprisingly powerful because the temperature differential between ground surface and air in contact with the ground can become very large under a hot sun. Over the ocean you could have the sun as close as it is to Mercury and it won’t make the water much warmer than the air so long as there’s water available to evaporate.
“Since the air descends dry adiabatically and becomes warmer than the surroundings,”
Air temperature can be equal to that of surroundings at best. Otherwise you would be creating energy, and this is impossible.
Anastassia,
What is your estimate of air circulation power?
Nabil
Our estimate is 3.5 W per square m (1-2% global solar power) – see eq 41 in the paper for details
1-2% of solar power converted to wind power sounds about right. I’ve seen the figure mentioned in wind turbine discussions about how much potential power there is to harvest.
Jim D
For air to rise, it only has to be warmer than its surroundings, which is how thermals work.
incomplete understanding, as a soaring pilot I have spent 1,000s of hours observing thermals from inside them, it is not common but neither is it unusual to find the mean temperature of the air in a thermal remain at or lower than the surrounding airmass at all altitudes from source through to top of convection.
Of course lapse rate is the driver of airmass stability but some thermals obtain their buoyancy due to higher water vapour content, usually caused by strong insolation causing evaporation over saturated ground resulting in localised higher levels of WV at thermal source.
I am certain of only one thing in life, there is far too much certainty
Yes, there may be a situation where buoyancy is achieved by the water vapor perturbation alone (the well known virtual temperature effect). In either case, buoyancy is the driver of the thermal.
“We have described a new and significant source of potential energy governing atmospheric motion. Previously, the only such recognised energy source was the buoyancy associated with temperature gradients.”
OMG. Is this piece of 19th century physics new in this field? Gravity is a weak force, molecular forces are strong.
Moving water into the gas phase, 1 micron away from the surface of droplets (evaporation) needs the same amount of work required to lift it to a height of 230 km above Earth’s surface. In case of ice crystals (sublimation) it is 264 km. These values are more than an order of magnitude higher, than tropospheric thickness, therefore potential energy storage in the atmosphere is dominated by the water cycle.
Also, mass of water evaporated (and recondensed) annually is roughly the same as that of the entire atmosphere. Most of it (~90%) never reaches the surface, but re-evaporates in mid air. Atmospheric distribution of water is extremely non-uniform on all scales.
Back-of-the-envelope calculations do have their merit.
Back of envelope calculations are of huge importance. They are the first step in separating the wheat from the chaff. In engineering and computer science, in my experience, these are referred to as “sanity checks” i.e. we do some very rough calculations that tell us if there are any fundamental errors leading to physically impossible (insane) consequences. If whatever it is passes the sanity checks then we can get down to the more detailed analysis looking for more subtle flaws.
Anastassia’s comment:
“This is only half of the story. In simple words, in order to lift a moist air parcel, you must draw a dry air parcel down: this is what circulation is about”
As far as I can see, Jim D et al have not addressed this point. Until they do, in words as directly simple as the quote above, this discussion just remains as shadow boxing
The vertical exchange of a less buoyant parcel above with a more buoyant parcel below is the source of kinetic energy in thermals. It comes from potential energy. The same happens with convection when conditional instability is considered because moist air releases latent heat and effectively becomes warmer than the dry air it replaces it higher levels. The dry air has no problem descending when warm thermals rise through it. The paradox would be if the potential energy somehow couldn’t be released. It is an unstable situation, like standing a pencil on its point.
Sorry Jim, a statement of the bleeding obvious. And not different to Anastassia’s comment except in verbosity
How does this refute the paper under discussion, exactly ?
The paper at least appears to claim that it presents something essentially new, but that whole process is only slightly influenced by the change they make.
Maybe you can explain what you understand by Anastassia’s statement. I think they are denying that buoyancy is a possible source of vertical motion, but maybe I misinterpreted it.
There’s a good measure for determining the importance of their addition. It’s given by the ratio of pV to latent heat of condensation. Both are in units of energy. p is the total pressure at the point considered and V is the volume of the condensing vapor. This ratio is small in all cases in the atmosphere.
Nick Stokes | February 1, 2013 at 1:08 am |
“It was point five in my first discussion entry. It’s still here, and emphasised; the equation is S = CNv.
Now physically this is nonsense. There’s always some humidity, but it mostly isn’t raining. Condensation doesn’t occur at all until Nv nears saturation. ”
Thank you for your comment.
The problem that has to be addressed here is the problem of spatial scale. Formally, condensation rate S enters the differential equation but we know that in the real atmosphere we are always talking about some macroscopic scale. E.g. the one determined by the spatial resolution of the model.
If the considered scale harbors the circulation cell as a whole (i.e. both the descending and the ascending branch), then S and Nv characterize the mean condensation rate in the considered cell. So despite the mean Nv is unsaturated, the mean condensation rate is not zero.
Accordingly, in our estimate of the global circulation power we use global mean precipitation P which characterizes both regions where the air is ascending and precipitation is high and regions where the air descends and precipitation is low.
If we take a closer look on the circulation differentiating the areas where water vapor is saturated and where it is not, we will need to consider that the condensation-induced pressure gradient spreads outside the condensation area to drive the circulation as a whole. Eq. (4) in our post becomes generally valid in its integral form (when both parts are integrated over volume V occupied by the circulation), while its local form is somewhat modified.
Please note also that to observationally verify Eq. (3) in our post (which contains the basic physics) you do not need to know anything about condensation rate at all.
But your justification was based on molecular kinetics, and you incorporated it in a continuum equation. Now you seem to be saying that, if that’s wrong, it’s still true on some larger averaged scale. But what is the justification for that? And how can an averaged result be incorporated in differential equation maths?
Pekka Pirilä | February 1, 2013 at 2:50 am |
“There’s is no reason for the very small loss of gas would create significant pressure effects. The little it does goes against the much stronger effect through the influence on temperature. ”
This is a qualitative statement which lacks a proof. If there existed a theoretical estimate of global circulation power based on “the influence of temperature” we could compare it with ours and decide. But such an estimate does not exist leaving the field open for theoretical research.
“The claim that vertical movements of gas could not compensate the changes from loss of gas molecules lacks all justification. The whole phenomenon is related to vertical movement of gas and changes that very little. There’s nothing that would make that difficult and require some strong winds.”
The point is there is continuity equation, and when the gas goes upward, it disappears from below. Thus re-establishment of hydrostatic equilibrium in the area of condensation leads to the appearance of a pressure drop at the surface. Such that there is now a pressure difference between the condensation area and the surroundings. It drives the wind.
But then it must be noticed that water can disappear higher up only when it’s first introduced at the surface. This circulation of water is to a large extent local and occurs in one rising column. The volume of gas in that volume is increased by a couple of percent (or less depending on temperatures), but that does not influence much what happens elsewhere and that does not influence much the horizontal pressure gradients.
I looked at all the messages written while I was sleeping only after writing my above comment and another comment higher up. This reading brought up that similar points have been made by many others.
The principal conclusion is that the paper does not present any coherent view of the issues. It’s incomplete and leaves most of the essential issues untouched. Therefore it’s not immediately obvious from the paper, how badly it’s wrong. Reaching this conclusion requires that the rest of the physics is taken into account. What’s discussed in the paper is just a small detail that has a totally different and far less important role when put in the connection with the rest.
A detailed and very accurate calculation of the atmospheric flows of moist air must take into account also the effects related to the volume taken by water vapor both when water vapor is added by evaporation and when it’s removed in condensation, but these effects are very minor corrections and not a source of anything significant.
“This circulation of water is to a large extent local and occurs in one rising column.”
This is not so. There is a large-scale atmospheric moisture transport which dictates the large-scale pressure gradients. We discuss this in Section 4.5 “Evaporation and condensation” (p. 1049).
Note that if your statement were correct, rivers could not exist. Rivers are the manifestation of a long-distance moisture transport. Moisture evaporated from the ocean condenses over land.
That’s the reason that I wrote “to a large extent”. Everything you discuss is part of that “large extent”.
You seem to be missing totally the overall picture of the phenomena you are discussing.
“A detailed and very accurate calculation of the atmospheric flows of moist air must take into account also the effects related to the volume taken by water vapor both when water vapor is added by evaporation and when it’s removed in condensation”
I agree. At this point we say: “it is significant, we have a theoretical estimate and it agrees with observations.”
You say
“but these effects are very minor corrections and not a source of anything significant.”
I think that if you could provide some specific justification for this statement, it could move the discussion forward. But certainly your opinion statement is valuable per se.
I have only the intuition and experience of a physicist in judging what factors are important to include in a calculation.
Here the main reasons for the conclusion is that the case is an uplift of most air (that’s the only place where your effect is present). The whole process is formed from the heating and evaporation at the surface and from the uplift convection where the condensation occurs. The vertical motion is essentially unrestricted for that case and controlled fully by the density of the air at various altitudes and by the energy available at the bottom to heat the air and to drive evaporation.
Heating and adding water reduces the density at the bottom. That leads to the uplift. Air from some side must flow in to maintain the pressure at the bottom. The amount of water in the uplift reduces the flow from the side because it adds to the volume and because evaporation takes most of the energy available reducing the overall flow of air.
Higher up the water starts to condensate. That adds sensible heat to air and reduces the temperature drop by altitude (moves from dry adiabat to moist adiabat). The gradual condensation affects also the density increasing the density of the other gases in the remaining air.
The above changes mean that the volume of flow is reduced where partial pressure of water vapor is low and is a little larger than that where we have more water vapor.
if you wish to say anything of what’s going to happen you must describe fully all the above main phenomena. Only then can you start to conclude anything on the influence on winds. Based on what I write above the effect on winds is almost certainly small and may actually reduce them rather than add to them,
In short: Solve first the vertical flows fully. Then you can look at what happens horizontally.
Pekka Pirilä
I have only the intuition and experience of a physicist in judging what factors are important to include in a calculation.
How can you tell that, in this case, the effect is too small to be significant? Compared to the gross flows of energy through the system, radiative and non-radiative transfer, it is a small effect. It also is not, as you say, closely related to the “overall picture”. However, the hypothetical effect of doubling CO2 is also a small effect compared to the gross energy flows, and the only tie-in with the “overall picture” is the hypothetical equilibrium effect.
Unless you have a stronger and more detailed analysis than what you have presented so far, your judgment that the effect is insignificant is premature. I think you (or someone) would have to show that their effect is not “there” (a word you used previously), or that they have overestimated it by an order of magnitude. If the effect is “there”, then it is deserving of an accurate estimate. This paper is a first attempt.
After that message my thoughts of the basic error in the approach have got more specific as I have explained.
Pekka Pirilä | February 1, 2013 at 5:09 am |
I have only the intuition and experience of a physicist in judging what factors are important to include in a calculation.
Thank you for taking time to explain your view. Let me explain mine at the same qualitative level.
First, in your view what happens when moist air ascends, you have not considered the effect we are talking about. When an air parcel occasionally moves upward, its pressure changes as prescribed by the ambient conditions (e.g. of hydrostatic equilibrium). But water vapor due to condensation behaves differently and its vertical pressure change deviates significantly from that of the other gases. This creates a non-equilibrium pressure gradient that does work on the gas. We quantify this work and show it is significant.
Second, the effects of changing density and latent heat release in the process that you describe (of a one-dimensional ascent of a moist air parcel) are well-known and discussed in Section 3.4 in our paper “Comparing forces due to condensation and buoyancy”. Such considerations do not say anything however about the real world pressure gradients and the real potential energy associated with latent heat release.
This is because in the real world when the lapse rate is not dry adiabatic (and it is not dry adiabatic in an environment where condensation is at least distantly present), the descending air experiences an upward force being drier than the surroundings. This force apparently suppresses the circulation and must be suppressed by an equivalent force acting in the uplift. The net effect (the work of both forces) can be close to zero. This descending branch is, in a way, a “burden” on any circulation planning to rely on latent heat.
So despite seemingly huge amounts of latent heat, its dynamic effect can be negligible. Inferences drawn from consideration a one-dimensional vertical flow are, in this case, misleading.
I do not believe that these guys are ‘climate scientists’ at all.
Their writing style and logic is too direct and understandable to the educated layman for them to be true members of that tribe.
Latimer “I do not believe that these guys are ‘climate scientists’ at all.”
Most of us are not.
That’s precisely why your work is being attacked by concensus ‘climate scientists’ and thus why your work deserves publication and attention.
I’m with Curry and question the magnitude of the effect but not its existence. Condensing water vapor causes a pressure change which in turn produces kinetic energy. The power stroke of early steam engines were driven in this manner – a piston rises drawing in low pressure steam then the cylinder is cooled causing the steam to condense and a negative pressure then pulls the cylinder down. The cylinder is connected to a rocker arm the other side of which is attached to a water pump pulling water out of mine shaft so mining can continue. That’s pretty much the first commercial steam engine. The negative pressure can approach ambient air pressure (~14.7psi at sea level) although in operation the cylinder wall stays pretty warm which reduces considerably the pressure differential. On a very large piston a few psi adds up to a lot of power. Those first steam engine pistons were hundreds of square inches. A source of cold water, which after startup could be the evacuated mine water, was required to cool the cylinder for the power stroke.
Now then, it’s nowhere near a new idea that the same thing happens in the atmosphere only the power produced by condensation is pumping air horizontally in the atmosphere instead of moving water out of a mine shaft. In light of that I have to echo Pirila’s comment about not knowing exactly what this adds to current knowledge.
I’d also dispute the author’s position saying thermals expend all their energy in vertical motion with none left over for horizontal displacement. Strong horizontal winds exist in the very driest of environments (consider the Antarctic interior) which seems to be a simple proof that thermal updrafts do indeed generate horizontal displacement in surrounding air masses.
“I’d also dispute the author’s position saying thermals expend all their energy in vertical motion with none left over for horizontal displacement. Strong horizontal winds exist in the very driest of environments (consider the Antarctic interior) which seems to be a simple proof that thermal updrafts do indeed generate horizontal displacement in surrounding air masses.”
Thank you for your comments. Note that I did not discuss dry thermals. I discussed convection based on latent heat release. There is a difference. In dry thermals the lapse rate of the ascending and descending parcel is the same. So if the ascending parcel has a positive buoyancy (being warmer), then the associated potential energy can be available to drive the descending parcel down. The descending parcel following dry adiabat does not experience any upward force hindering its motion.
When there is latent heat the situation is different. The ascending and descending air parcels have different lapse rates. If there is a mean rate of 6.5 K/km, the parcel that ascends moist adiabatically will be warmer than the environment. But so will be the parcel that descends dry adiabatically. It will experience an upward force that can suppress the circulation.
David Springer: I’m with Curry and question the magnitude of the effect but not its existence.
In that case, shouldn’t the next step be to explore how to compute better estimates?
Congratulations to Anastassia Makarieva, Victor Gorshkov, Douglas Sheil, Antonio Nobre, Larry Li and to the editor and executive committee of Atmospheric Chemistry and Physics for successfully getting new information past the gatekeepers of knowledge by saying:
“the handling editor (and the executive committee) are not convinced that the new view presented in the controversial paper is wrong.”
What a sad day for science that so much effort was devoted to blocking its publication.
With kind regards,
Oliver K. Manuel
Former NASA Principal
Investigator for Apollo
Gatekeepers of knowledge seek to publish only “settled science.”
There is no “settled science.” The imperfections in all science are identified by open discussion.
George Orwell advised in 1946: Discussion prevents tyranny : http://omanuel.wordpress.com/about/#comment-2204
Thanks Dr Manuel – your support is sincerely appreciated.
Douglas,
that was the ‘kiss of death’ from Mr Iron Sun.
Judith Curry
It appears to me that the Makarieva et al. study has caused some excitement – and maybe ruffled some feathers along the way. Thanks for posting it here.
The best indication that it might be a serious challenge to the prevailing consensus paradigm is the strong reaction it appears to have gotten (and is getting here) by the proponents of the consensus.
Leaving aside some frivolous characterizations (“wacky stuff”, “physics is plain wrong”, “the paper is obscurely written”, etc.), or some non-specific criticism (“where’s the code?”, “the paper does not present any coherent view of the issue”, “this is all very well explained already and needs no new theory”, etc.), the most serious challenge has been that the described effects may exist, but may be too minor to be of any real significance.
As I understand it, Makarieva and co-authors have acknowledged that their findings and conclusions are novel and controversial and that, as a result, current GCMs do not take these findings into consideration.
They state that they hope that “these ideas can gain objective assessment from those best placed to assess them”.
This makes perfect sense to me and (IMO) would necessarily include gaining empirical evidence from reproducible experimentation or actual physical observations in our climate system to validate (or falsify) the proposed mechanism, as well as to quantify its effect.
Isn’t that the way that science is supposed to work?
Max
Max – Isn’t that the way that science is supposed to work?
Yes
Abusing a bag of wind is such fun.
Eli
Fun perhaps, science no
This is a blog, have you not noticed?
manaker.
The issue isnt how science operates. Scientists will ignore the paper, unless they can make sense of it better in a way that the reviewers could not.
Frankly, if Nick Stokes, Lucia, and Held could not make sense of it, then I don’t think anyone else will waste their time disproving it OR using it.
In short, it cannot be incorporated into other science until the issues are addressed in a SUBSTANTIVE way.
The real issues are.
A) the publishing process, specifically this journal.
B) cranks who latch onto papers they dont understand in the hopes that this paper will bolster their crank ideas.
C) cranks who assert that this paper that is “not wrong” somehow has arguments on accepted science.
But yes, science will operate normally and ignore bad stuff that just happens to get published.
“A) the publishing process, specifically this journal.”
Would you have been happier if they had gone with anonymous alleged reviewers in one of OMICS hundreds of journals? I bet OMICS would have created a journal just to publish this paper, and given them a discounted price. With regards to the BEST paper, you made it clear that the review process don’t mean squat, it’s getting the box checked that matters. Why haven’t you and Nick, rabette, et al just ignored this paper?
> I don’t think anyone else will waste their time disproving it OR using it.
I can imagine lots of people using it.
Not that it won’t was any time.
Not that it won’t waste any time, that is.
E.g.
> Yes, I am making a direct challenge to the failed hypothesis that radiative gases create a greenhouse effect in our atmosphere.
http://judithcurry.com/2013/01/31/condensation-driven-winds-an-update-new-version/#comment-290677
Don,
I’ll suggest that you refrain from misrepresenting my position and from distorting the facts.
“Held could not make sense of it,”
Held cant make sense of Chekroun either,or the importance of the devils staircase in winding maps as he readily admits,others such as Crucifix,understand the heuristics of M13,more readily such as the ability of Biological complex systems to adapt the environment to their liking.
“Leaving aside some frivolous characterizations (“wacky stuff”, “physics is plain wrong”, “the paper is obscurely written”, etc.),”
“physics is plain wrong” is not a frivolous characterization.
“Isn’t that the way that science is supposed to work?”
No. You’re supposed to get it right.
> The best indication that it might be a serious challenge to the prevailing consensus paradigm is the strong reaction it appears to have gotten (and is getting here) by the proponents of the consensus.
In that case, I propose we give Christopher Monckton, 3rd Viscount Monckton of Brenchley a Nobel prize of his choosing.
I think I have worked out why talk of ‘pressure’ has become confusing.
For a planet with a fixed atmospheric mass and a fixed gravitational field there can be no absolute change in total pressure.
All that can happen is that from place to place there can be pressure variations in the horizontal plane relative to the vertical plane.
Thus, considering a single parcel of air of indeterminate initial volume:
i) That parcel of air can be caused to expand relative to adjoining air parcels either by direct input of more solar energy where insolation is uneven (as it always is) or indirectly by the injection of potential energy in the form of latent heat of evaporation carried by water vapour.
Once it expands it pushes against the adjoining parcels so pressure increases in the horizontal plane but pressure in the vertical plane remains the same so the expanded and lighter air parcel moves in the direction of least resistance which is upward.
In the vertical plane viewed from the surface one then has lower pressure because the rising air is less dense and lighter. The higher the column of rising less dense air goes or the faster it rises the lower surface pressure will become in the vertical plane.
At the same time the changed pressure relationship between the vertical and horizontal planes will set up an air flow which serves to bring in new air low down to replace at the surface the air that has risen.
In extreme scenarios there can be tornados or hurricanes.
Note that from a meteorological point of view that is the formation of a low pressure cell because pressure in the vertical plane has dropped in order to relieve the increased pressure in the horizontal plane.
ii) Now let’s look at the other side of the same scenario which is the condensation process that Anastassia is considering.
When condensation occurs that parcel of air shrinks relative to adjoining parcels when it loses its latent heat of evaporation via condensation.
Having shrunk it no longer pushes against adjoining air parcels. Instead it gives way and pressure in the horizontal plane falls. Again, pressure in the vertical plane stays the same so the contracted and heavier air parcel moves in the direction of least resistance which is downward.
In the vertical plane viewed from the surface one then has higher pressure because the falling air is more dense and heavier. The higher the column of descending less dense air or the faster it descends the higher pressure will become in the vertical plane.
An air flow will be set up whereby the descending air replaces the air at the surface that has gone to replenish the nearby rising column and there we have a circulation.
So, in light of that it is potentially misleading to say that condensation causes a reduction in pressure because one assumes that such reduction in pressure has occurred in the vertical plane so as to accelerate convection.
In fact, condensation only results in a pressure reduction in the horizontal plane and the only effect of that lateral pressure change is to complete the downward half of the circulation already provoked by the initial uplift.
Note that I said that the initial uplift could be caused by any provision of more energy at the surface.
That means that the process I have just described is applicable with or without a water cycle. The only effect of a water cycle is to increase the sensitivity of the system as a negative response to surface heating.
The water cycle assists stabilisation of top of atmosphere energy balance and reduces the need for a more vigorous circulation.
The effect of CO2 and other non condensing GHGs is just the same but without the added efficiency provided by phase changes.
All GHGs make it easier and not harder for the global circulation to match energy in with energy out at top of atmosphere.
@Makarieva el al
Try this “journal” for publishing your paper. Scuttlebutt has it they’ll publish anything so long as you send them a check for the page fee.
http://www.scitechnol.com/ArchiveGIGS/articleinpressGIGS.php
Aw, C’mon Ward. Dont’cha think you are being too hard on the Beaver?
Not as long as he’s still got a pulse.
Sorry Dave no checks were sent. no fees. no fee wavier. Consider yourself put on notice.
I admit that my physics is not good enough to appreciate the details of Makarieva
et al, and I have tried to read the comments to augment my understanding; with limited success. But it seems to me that if this paper is right, then the current climate models are wrong; in a fundamental way.
Now the IPCC has set March 2013 as the deadline for new papers to be considered for the AR5; and we have not reached this date. Is this paper one that the authors of the chapter in the AR5 dealing with climate models must ignore, or admit that the output of the models might not be as robust as they had previously believed? And in which case, the certainty expressed on the SPMs needs to be downgraded.
Is this just another case where we have reason to believe that in past reports have exaggerated the certainty with which they have expressed in the SPMs is far greater than the science warrants?
Wow, talk about “redefining peer review as we know it”. Very intense thread.
What a refreshingly “Scientific” post this is.
Nick Stokes | February 1, 2013 at 3:57 am | Reply
“But your justification was based on molecular kinetics, and you incorporated it in a continuum equation. Now you seem to be saying that, if that’s wrong, it’s still true on some larger averaged scale. But what is the justification for that?”
My brief response is that the justification is the same as for the smaller scale. If S is proportional to Nv where condensation actually occurs it will unlikely become proportional to the square power of Nv (or anything else) on a larger scale.
Now let me put your question in a perspective. The physics of condensation-induced dynamics is summarized by Eqs. (1)-(3) in our post. This physics has been described in detail starting from our first paper in HESS on this topic.
In our ACP paper we investigated how the same effect can be derived (or understood) from, and agrees with, the mass conservation equation. Our physical justification for condensation rate S (34) was that (1) it should only depend on vertical velocity and (2) must have total air (rather than dry air) subtracted as the reference because of the hydrostatic adjustment of air as a whole. These plausible propositions do show (at least to us) that our approach is coherent and makes good sense. One needs to investigate the process from different sides. We did so and were satisfied with the result.
Now when the readers, including yourself, said (much to our surprise) that the grounds on which we wrote S (34) are not convincing, we undertook additional efforts to investigate its form. We additionally clarified in the Appendix that S (34) agrees with the proposition that S is proportional to Nv. This also makes good sense, because condensation is indeed a first-order reaction. If we, say, found that on a macroscopic scale it somehow becomes proportional to the second power of Nv that would be a little alarming.
Note now that S (34) follows directly from (3) and the energy conservation equation (4) in the blog. So you can either derive (4) (Eq. 37 in the paper) from S (34), or you can obtain S (34) from consideration of the potential energy associated with the ascent of condensing vapor as per (4) (cf. expression for condensation force fc on p. 1044).
We consider an interesting finding in the ACP paper (besides the Hadley estimate) the expression showing how S and Sd are related via gamma to determine the horizontal pressure gradient (Eq. 6 in the post).
“My brief response is that the justification is the same as for the smaller scale. If S is proportional to Nv where condensation actually occurs it will unlikely become proportional to the square power of Nv (or anything else) on a larger scale.”
But as I said, the justification on the smaller scale is completely wrong. Precipitation does not occur proportional to humidity. Nothing like it.
This does not support the proposition on a larger scale.
“If S is proportional to Nv where condensation actually occurs”
S (condensation rate) occurs at essentially one value of Nv, saturation. And it does not have a single value at that point. In saturated air you can have anything from mist to a downpour. It depends on what is happening to the latent heat.
Nick it would help those of us who are non-physicists to have citations to the definitive statements you make;
‘Precipitation does not occur proportional to humidity’
You see I simply have no idea what you mean by this statement. Over what timescale are you using; the second, minute, hour or day domain?
Doc,
It’s a statement in the paper. S=C*Nv. It goes into the system of continuum equations. If there are caveats about scale, they should be stated there, and be taken account of in the math reasoning.
But I don’t think it makes sense on any scale. There’s humidity in the desert on a fine warm day. Where do you see the proportional condensation rate?
There’s cpnstant reversal of burden of proof here. What’s the case for S=CNv? It’s there now in a peer-reviewed paper, I suppose. Anything else?
Nick, please correct me if I got you wrong, but do you really want to spread the message here that we put C = const in S=CNv? Why to create this confusion? Please check back in the text on p. 1054, Eq. (A11). It says that S = wkvNv, with C = wkv being constant with respect to Nv only. It does not depend on Nv, that’s all. Let’s recall that even in molecular kinetics if a reaction is first-order with respect to some substance, the proportionality coefficient is not a universal constant, but depends on the other reaction parameters like say temperature or concentration of other substances.
In our case these other parameters are the vertical velocity and the degree of deviation of water vapor pressure from hydrostatic equilibrium. In the unsaturated atmosphere it is zero. Of course this is not formal molecular kinetics, but the value of this result is that the macro-scale pattern, with S being proportional to Nv, is consistent with what we can expect from a micro-scale consideration. BTW for evaporation E that is not first-order in Nv you will never be able to derive anything of similar kind. It is perfectly plausible that micro-scale properties show up at a macro scale. It may or may not be convincing for you, but there is no error here.
So you can argue that “there are no plausible grounds to believe that S = CNv” and I will argue that there are. The point is that the relationship that is obtained from S (34) is empirically testable for anyone to decide on their own. It also has a different but totally coherent physical interpretation as given per Eq. (3) in the blog. I expand on these points below.
Anastassia, Since there is so much confusion, perhaps a simpler analogy might help.
The saturation vapor pressure of air at sea level and 25C is roughly 1 inch of hg (33millibar). If were possible for all of that to be converted to velocity pressure, that would produce a wind speed of roughly 5 kilometers per minute. I that about what y’all are predicting?
Yes, this is the scale of maximum velocities that are to be observed where the turbulent friction is minimal. This is what we predict for hurricanes.
For a large-scale circulation where friction is significant especially at the surface such velocities cannot be observed. Here what we predict is that the rate of generation of kinetic energy by the large-scale pressure gradient u∇p (i.e. local power (work per unit time) of the pressure gradient force) is determined by condensation intensity as given by formula (3) (or (5) if you integrate over z).
This can be empirically tested by either directly observing velocities and pressure gradients or using a more indirect approach and estimating the dissipative power of smaller-scale eddies. Such empirical estimates of circulation power are available in the literature.
Note also that for hurricanes (3) is also valid, so velocity in hurricanes is an additional prediction.
Anastassia Makarieva
Thanks for posting your study here.
Your hypothesis seems to me to sound plausible, despite some objections posted here.
A major question raised has to do with the magnitude of the effect.
What do you see as the next step to either validate or falsify the hypothesis that this mechanism can be a significant driver of winds?
Can you foresee reproducible experiments that could be done or actual physical measurements that could be made to validate or falsify your hypothesis?
It seems to me that this would now be the next step.
Do you agree?
Max
Thank you for your interest. I think that when people appreciate the physics of this effect, many original studies may be designed that we may not foresee right now. One specific thing I’ve been thinking about is that there are some behaviors in the existing numerical models (e.g. of hurricanes) that remain obscure right now, but which can be explained within our approach.
Regarding empirical data, the most straightforward thing is that we give a testable formula for circulation intensity based on precipitation. We have applied it to the mean circulation power and average hurricanes, but it is possible to check it against temporal variations in both regional circulation intensity as well as on individual cyclones. One will need an independent estimate of dynamic hurricane intensity, which can be derived from the observed pressure gradients and radial velocity.
Anastassia Makarieva
Thanks for your response.
You mention possible ways to get empirical data to either validate or falsify your hypothesis regarding wind intensity based on precipitation.
Are these basic data being gathered today?
I assume that if these measurements show no direct correlation between wind speed and precipitation that this would essentially falsify the hypothesis that this mechanism plays a major role in driving winds.
However, if they demonstrate an repeated direct correlation between precipitation and wind speed, will this be enough in itself to validate your hypothesis, or would additional empirical evidence be required in your opinion?
Thanks for taking the time to answer my questions and lots of luck in being able to move this to the next step.
Max
Similarly, it doesn’t take a great thermodynamicist to recognize poor notation, imprecise reasoning, scanty references, and dubious conclusions.
The harsh judgment of science For good and sufficient reasons, the article “Where do winds come from?” will fail to attract talented young scientists.
Fan
When you critique a paper, you should get specific
Generalities about “poor notation, imprecise reasoning, scanty references, and dubious conclusions” are meaningless.
State specifically what you find “dubious” and why, including your reasoning.
Otherwise your critique sounds like senseless bad-mouthing.
Max
> Makarieva et al. have (inexplicably) adopted a notation that fails to make this crucial distinction [exact and inexact differentials].
http://judithcurry.com/2013/01/31/condensation-driven-winds-an-update-new-version/#comment-290673
To accord with modern thermodynamic notation, alter Eq. (1) — and all of the subsequent expressions and conclusions that derive from Eq. (1) — to eliminate references to the (inexact) heat differential
, in favor of the (exact) entropy differential 
.
——————
scanty/outdated references
imprecise/foggy reasoning
dubious conclusions.
Reason Bad/outdated notation
——————
Hopefully this common-sense advice is helpful, Manacker!
‘Before discussing the generation of the standard thermodynamic potentials, we briefly summarize the basics of statistical mechanics. We will show how the Legendre transform enters thermodynamics through the Laplace transform of partition functions in statistical mechanics.
Equilibrium statistical mechanics is based on the hypothesis[2] that for an isolated system, every allowed microstate is equally probable. The high probability of finding a particular equilibrium macrostate is due to a predominance of the number of microstates corresponding to that macrostate.’
Quite apart from the number and extent of non-equilibrium processes in the atmosphere and that Earth’s climate is not an isolated system – the paper you reference seems more concerned with the relation of quantum states to macrostates. The concern of the current paper is evidently the macrostates of atmospheric circulation. Please -your vague references to irrelevant science and mathematics is a mere distraction. One that seems to have quite juvenile and unworthy motivations.
Fan
If you are truly interested in getting your critique of the Makarieva et al. study across, I’d suggest that you address your specific points of disagreement directly to the author(s), not to me.
Max
Inexact differentials enter thermodynamics in the production of work and heat. There is a change of state defined by end points but the path is unknown. This is opposed to ballistics where an exact path can in theory be calculated. It is clear that these processes relate to the former rather than the latter and some estimate is given of work and heat produced.
Is there a pattern emeging here of juvenile and irrelevant comment made from dubious motivation?
I would like to raise an issue. Go to Makarieva’s web site,
http://www.bioticregulation.ru/pubs/pubs2.php
Look at the other journals where their papers on this general topic have been published:
Theoretical and Applied Climatology, Journal of Experimental and Theoretical Physics, Physics Letters A, International Journal of Water, Proc Roy Soc Series A.
Papers on other topics are published in Science, PNAS, etc.
This is not the publication record of a ‘crank’, it is actually a very impressive publication record.
Why have the biotic pump papers been published in Phys Lett without any apparent problem, but are resisted by the atmospheric sciences community? The point is that the physical reasoning is sound (which the physicists seem to get). The issues that the atmospheric scientists have are raised in the post (e.g.current climate models work fine, yeah right).
We can’t fully test this idea unless we formulate a general circulation model in multi-component multi-phase form, formally treating water as a separate component in a multi-fluid flow. I believe that we need to do this anyways, I have suspected for a long time that the very high water vapor feedback is an artifact of the approximations made regarding the water phase in the current climate models. These are not an issue on weather timescales, but accumulate over climate time scales.
So in my book, this hypothesis will remain as an open issue until something like I suggest above is done. Is their paper sufficient motivation for building such a model. Probably not for anyone outside their research group. But sorting out what might be going wrong with water vapor feedback and actually getting the moist thermodynamics correct in climate models should be a high priority. Instead, climate modelling efforts are moving in the direction of throwing more ancillary elements into an earth system modeling framework.
Not every weak and/or foggy and/or wrong skeptical critique requires rethinking the foundations of a discipline. The path forward for Makarieva’s and colleagues is evident, specific, and fair:
(1) repair the starting (bad and/or wrong) mathematical notation, and
(2) extend the (scanty) references, then
(3) clarify the (foggy) reasoning, and finally
(4) draw (amended) conclusions.
That Makarieva and colleagues have not used the past to years to undertake these (entirely reasonable) repairs is grounds for scientific concern, eh?
Is this a lot of work? Of course it is! Might the conclusions change substantially as a result? Yes, that is entirely plausible.
http://judithcurry.com/2013/01/31/condensation-driven-winds-an-update-new-version/#comment-290847
Again – inexact differentials apply equally to heat and work and entropy. There being some correspondence between these entities – and that the path between states is not definable by means of a functional relationship.
Given your inexact comprehension of basic maths and physics – and all inclusive links to a few quite irrelevant references – might not this require some futher consideration on your part?
No. Please be aware that this assertion is entirely mistaken, Robert I Ellison!
• Entropy
and internal energy 
are state variables, hence their differentials 
and 
are said to be exact.
• Heat
and work 
are not state variables, hence their differentials 
and 
are said to inexact.
Note that exact differentials “
” and inexact differentials “
” are commonly denoted by distinct symbols .
A severe criticism of the Makarieva et al analysis — and perhaps a contributing factor in their article’s foggy physical reasoning? — is that their article’s notation (inexplicably!) does not respect this crucial thermodynamical distinction.
Conclusion Imprecise mathematical notation facilitates foggy reasoning and dubious conclusions.
Thank you for helping to clarify this point, Robert I Ellison!
What’s perhaps even worse is that they discuss pressure changes from internal subprocesses in a setting that’s undefined if the pressure is allowed to change for internal reasons.
‘In thermodynamics, a state function, function of state, state quantity, or state variable is a property of a system that depends only on the current state of the system, not on the way in which the system acquired that state (independent of path). A state function describes the equilibrium state of a system. For example, internal energy, enthalpy, and entropy are state quantities because they describe quantitatively an equilibrium state of a thermodynamic system, irrespective of how the system arrived in that state. In contrast, mechanical work and heat are process quantities because their values depend on the specific transition (or path) between two equilibrium states.’
‘An inexact differential or imperfect differential is a specific type of differential used in thermodynamics to express the path dependence of a particular differential. It is contrasted with the concept of the exact differential in calculus, which can be expressed as the gradient of another function and is therefore path independent. Consequently, an inexact differential cannot be expressed in terms of its antiderivative for the purpose of integral calculations i.e. its value cannot be inferred just by looking at the initial and final states of a given system.[1] It is primarily used in calculations involving heat and work because they are not state functions.’ Both quotes rom wikipedia
‘The First Law is a conservation law,- Energy can neither be created nor destroyed. It requires that we balance the energy budget when we describe a change in state. A change in energy content (dE) is accompanied by the performance of work (w), and/or the transfer of heat (q) between the thermodynamic system under consideration, and its surroundings.’ http://www.life.illinois.edu/crofts/bioph354/thermo_summary.html
A change in energy is the system involves a change in state.
‘The description of entropy as energy dispersal provides an introductory method of teaching the thermodynamic concept of entropy. In physics and physical chemistry, entropy has commonly been defined as a scalar measure of the disorder of a thermodynamic system. This newer approach sets out a variant approach to entropy, namely as a measure of energy dispersal or distribution at a specific temperature. Under this approach, changes in entropy can be quantitatively related to the distribution or the spreading out of the energy of a thermodynamic system, divided by its temperature.
The energy dispersal approach to teaching entropy was developed to facilitate teaching entropy to students beginning university chemistry and biology. This new approach also avoids ambiguous terms such as disorder and chaos, which have multiple everyday meanings.’ again wikipedia
And of course I am aware of the notation used for exact and inaxact defiiferentials. The distinction between state and process variables are irrelevant to the discussion of exact and inexact differentials. They are quite seperates things. So your fallacy FOMBS is that of a red herring.
These are quite basic concepts of maths and physics and are indeed, again, irrelevant to the discussion.
Judith, I tried previously to get opinions as to whether the IPCC can afford to ingore this report in the chapter on models in the AR5. Do you have an opinion on this issue?
yes, they can ignore it. The editor published it over the objections of reviewers. Objections that were not met. he thought the paper would generate interesting discussion. The IPCC summarizes the science.
What would they say “there is a paper that makes unsubstantiated claims”
Heck, there is a lot of grey literature that says all manner of crazy things.
Furthermore it’s questionable whether the paper would be climate science or important for climate science even if it were correct. It would affect meteorology much more directly than climate science.
You have gone off the deep end, Steven. What about papers that are rejected by a legitimate journal and then are shopped around, until they land in the pay-for-play journal of last resort. Box checked, no problem.
Reviewers don’t get to make the decision on what gets published. This paper’s box has been checked. Remember what that means?
Jim,
What price will the IPCC pay for ignoring this paper, or any other paper they choose to ignore?
Don, you write “What price will the IPCC pay for ignoring this paper, or any other paper they choose to ignore?”
Precisely. This is the point I think I am getting at. This paper is by no means unique this time around. It might merely be one of several they will have to ignore, in order to preserve the certainties which which they express their conclusions in the SPM to WG1 of the AR5.
Yes, Jim. And did you see this:
“The IPCC summarizes the science.
What would they say “there is a paper that makes unsubstantiated claims” ”
The IPCC only cite papers that have been “substantiated”, Jim. Substantiated dogma.
Pekka,
It would affect the weather, which when added together over extended periods makes what, Pekka?
If climate models would be built fully from first principles that would be relevant. In practice those parts of climate models where this paper would have some influence are in most cases not built from first principles but are presented by parametrizations made to agree with empirical data on the relevant subprocesses. A new theory does not change the empirical data.
Understanding of the physics of the atmosphere does certainly have its role also in choosing the parametrizations, but it’s difficult to see, how this paper would affect that even if it were true.
Thank you for the clarification, Pekka.
Will the IPCC publish James Annan’s comments on the IPCC procedures and the reduced estimate of global temperatures following inclusion of data from 2000 on
Steven Mosher
“Grey Literature?”
How about IPCC’s AR4 report?
“Shades of grey at work”.
Max
Don,
it’s pretty simple. The IPCC does a summary of the science that is relevant to the topic at hand. The FOD and SOD were open for review.
Anyone could be a reviewer. All a reviewer would have to do is put in a comment:
‘I think this paper is relevant to the claims you have made”
If the paper was relevant then the lead author would probably write something about it. This paper has been out there submitted for a long while. No skeptic made any comments to the authors that they overlooked it. Go figure. And it would be really hard to figure out how exactly it was relevant. As the editor said.. its not wrong and people should probably continue to discuss it. One cannot conclude anything from a ‘not wrong” claim. Now, if it correct, you might be able to show that models could be improved. But you would have to show that. basically, a ‘not wrong’ premise leads to no conclusion, at least in traditional logic.
Don:
“You have gone off the deep end, Steven. What about papers that are rejected by a legitimate journal and then are shopped around, until they land in the pay-for-play journal of last resort. Box checked, no problem. ”
Sorry but the results paper was not rejected by JGR. The paper was submitted. we withdrew it when they said they would not consider it until the methods paper published. results. methods. two different papers.
Handling water fully in a model based on fundamentals requires clearly a two-phase model as precipitation and re-evaporation are real significant processes.
The problem of this paper is that it does not handle the mathematics and its relationship with the physical system well enough. The weaknesses in that have made it possible for the authors to avoid seeing the errors in their reasoning and it makes it excessively difficult for the others to figure out exactly where the actual errors hide.
There are good reasons for the practice that the choice of free variables is specified explicitly in thermodynamic derivations and that the variables held constant are also often marked explicitly in the formulas that contain partial derivatives. There seem to be problems in the paper that have their origin in not following these practices.
It seems that a relatively simple physical model can be constructed to verify this theory. If verified experimentally, then construct a numerical model of the experiment and scale up from there.
One step at a time.
From my scientific perspective, Climate Science has been all about confirmation of a grand theory rather than solving small problems and figuring things out.
Obviously something needs to be done about GCMs when it seems every day we see more and more human forcings added on top of CO2 with absolutely zero ability to simulate the apparent low transient feedback sensitivity. Unless , of course, you crank up the aerosol knob to 11.
But hey! That’s how we *do* science.
Judith Curry
You are correct in that is not the publication record of cranks. Further, even if insignificant in magnitude, the effect (if real) would be at a minimum be an interesting excercise in theory and quantification. So what is the brouha about? Maybe it has something to do with what the paper MIGHT mean.
For example. it might mean that biotic-regulation must be taken very seriously. And it might mean that given a healthy, vigorous biosphere, the burning of massive amounts of fossil fuels is in fact relatively harmless. It might mean that significant ocean pollution or forest clearing threatens all life on the planet. It might mean a change in the status quo.
The impetus for crank science can from many quarters, meta-physics, religion, politics, money and/or power. It would be best for all of us refrain from lobbing the crank bomb and simply work on what the science DOES mean.
@Curry
“The issues that the atmospheric scientists have are raised in the post (e.g.current climate models work fine, yeah right).”
Yeah, it’s like deja vu all over again. Nobody’s interested in working on chemical evolution models either. The current ones work fine (yeah right). The “current ones” are pretty much Miller-Urey from the early 1950s where he zapped a methane ammonia atmosphere in a flask with electric arcs for few weeks and got some amino acids to fall out. The concentration of organic molecules was far too dilute to do anything interesting. That requires concentration and there’s nothing but hand-waving speculation about a mineral lattice doing it. Nothing in nature has been shown able to concentrate the amino acids enough so they can start combining into interesting polymers. Just as bad, the atmosphere Miller used is now thought to be probably not what the early earth’s atmosphere was like. An alternate model, RNA-first, is in a similar state but the can’t figure out how nature could make all four that are needed nor can they figure out how nature could concentrate them sufficiently. Without concentration of the reactants any lucky two or three unit polymers that form fall apart before they bump into more base units. And polymers hundreds of bases long configured just perfectly to have any chance of doing anything interesting. It’s a pretty pathetic narrative when you look behind the curtain. All that’s really fairly sound science is common descent and then just handwaving about how novelty is generated as descent happens. Recombination, sure. There’s a fair amount of plasticity. Nothing new needs to be invented to go from a shrew to an elephant just rearrangements of parts, sort of like cannabillizing a couple motorcylces to make an automobile. But try to get a description of how something harder happened, like how did prokaryotes turn into eukaryotes. What’s the path to go from a bacteria with circular DNA an no nucleus to the packed structures isolated inside a cell nucleus?
The answer is “It must have happened somehow because evolution is true”. Global warming works that way too. Like how can there be this pause we’re seeing when the accepted range of climate sensitivity and GCMs can’t duplicate it without subtracting CO2 induced warming. Well, it just had to happen somehow because CO2 induced warming is true. No doubt is cast on the truth of the underlying assumptions. They’re true by fiat. Hail to the climate kings.
Isn’t tht just precious? Welcome to consensus science, Curry. It gets worse.
I’m not sure it was the larger point you were trying to make, but the thing that jumps out at me from this post is the difference in reactions between Climate Science(TM) and other scientific groups. A very non-scientific reason, but it still smells like rat.
Precisely Qbeamus!
I note this phenomenon in practically every discussion on climate I see. In general skeptics seem to use discourse appropriate for scientists discussing science while true believers seem to prefer discourse appropriate for politicians discussing politics.
Judith Curry
Your point seems well taken.
I cannot judge the scientific validity, as you can, but it appears to me that this needs further investigation.
You mention the need for model simulations “formally treating water as a separate component in a multi-fluid flow”. This could possibly clear up some of the present uncertainty surrounding water vapor (and cloud) feedbacks.
Wouldn’t it also be appropriate to look for actual empirical evidence to either validate or falsify the hypothesis that precipitation drives winds, as Anastassia Makarieva has suggested?
Max
I admit unease with making any claim based on the status or authority of the authors. It is absolutely true that Anastassia has an impressive publication record. But should that be relevant here in any formal sense? Its the same issue with us being outsiders or insiders of the mainstream, or what we think of global change. While I am sure these factors — or the perceptions of them — have influence, shouldn’t we be striving for objectivity based on the value of the ideas? How does this impact young scientists and their choice of disciplines? As an outsider I think climate science needs to be a little more open and a little less defensive — less preoccupied with authority and more with ideas. I do fully understand that it is hard to do: there is so much we could be looking at that we need easy ways to filter it. But a more open attitude and a more level playing field looks desirable from where I am. Just a thought.
“Why have the biotic pump papers been published in Phys Lett without any apparent problem, but are resisted by the atmospheric sciences community?”
You were part of the resistance.. Perhaps you can answer.
Read my review, and read my previous statement, not to mention consider the fact that this is the 2nd post on this topic on my blog. I clearly do not dismiss this, whereas others are prepared to dismiss it.
> Read my review […]
As if Nick hadn’t by now. But reading Judy’s review, we certainly can tell that Judy read Nick’s review.
> I clearly do not dismiss this, whereas others are prepared to dismiss it.
Putting resistance does not entail dismissal.
Saying:
> There are four major issues that need to be fixed before the paper is accepted for publication: […]
ipso facto puts a burden on the authors. This burden offers a resistance. Compare raising four major issues with:
> I accept the paper on the condition that the author polishes their punctuation and clean up their bibliographical entries.
***
Have these issues been fixed? How so?
Bear in mind that MiniMax might be overhearing. He likes specifics.
Crickets.
Is it too much to ask if Judy’s issues have been fixed?
Morning crickets.
Nick, what are your views on this post?
http://judithcurry.com/2011/09/24/water-vapor-feedback-evaporation/
Dallas, which aspect? Evaporative processes move heat around, true.
Is that all this paper is about? Land use changes the way things move heat about. So if you look at the surface albedo and see “TREES”, you might want to think mobile heat capacity instead of radiant forcing, at least inside a moist air envelope.
It’s difficult to get hold of the reasoning as the normal order of looking at the processes is reversed.
The normal way of considering the role of adiabatic condensation in ascending air and how that’s related to pressure and temperature changes goes essentially as follows:
1. the air ascends
2. the pressure and temperature are decreasing
3. the decreasing temperature makes saturated vapor condense releasing latent heat
4. the release of latent heat makes the drop in temperature smaller but cannot reverse the sign of the change
5. the pressure is not affected by the condensation as the pressure is determined by the surrounding atmosphere (we are considering an adiabatic process and that’s slow enough for that)
The above points describe essentially similar changes as the paper discusses in chapter 2 except for one essential difference: The only source of pressure drop is the ascending movement of air not to the condensation of water.
As I started this comment it’s difficult to get hold of the whole argument of the paper, but so far I have concluded that the fault is in the handling of the pressure changes in the atmosphere. The pressure of a volume of gas in atmosphere is not determined by what happens in that volume but by the atmosphere outside that volume and in particular by the mass of air above that volume in the simplest case.
The real atmosphere is not that simple when circulation is taken into account. The paper purports to tell where the winds come from, but the analysis does not present any self-consistent picture that could describe the overall circulation and winds.
The mathematics presented does not describe what happens in ascending moist air. The formulas do not apply to the process they are supposed to describe. There’s confusion on what’s changing and what’s not. The calculations should have been made using partial derivatives and defining always carefully what are the free variables. When that’s not done at every step thermodynamic derivations are usually wrong as they seem to be in this paper.
And some seemingly want to include the paper in the next IPCC report. While awaiting empirical evidence. Which would most likely come in the form of results from climate models.
Going trough the argument I feel confident that something essential is missing. (I have sometimes erred while equally confident, but being in such a good company makes that less likely in this case). Figuring out from a paper where the gap in logic is, is often much more difficult than getting convinced that something is missing, when the paper does not present all steps of reasoning explicitly and does not describe precisely, how the equations are connected to the physical system that’s being discussed. It’s not surprising that papers that seem to be wrong leave also gaps in the reasoning.
It’s true that many good papers are also terse and difficult to understand by non-specialists. They are, however, written following practices of the particular field of science and thus accessible to other scientists of the same specialty. From the list of publications of Makariaeva we see that she has been very productive and published in a variety of fields. People with such a coverage may well end up writing articles that are not easily accessible by anyone – positively or critically.
To define the output of models as empirical evidence is to misuse the term. This is especially so where the models are based on nonlinear equations. Simple energy models such as the one presented seem much more useful in developing a theoretical understanding of these processes which remain unsolved in a realistic understanding of the limits of current sceince.
‘Finally, Lorenz’s theory of the atmosphere (and ocean) as a chaotic system raises fundamental, but unanswered questions about how much the uncertainties in climate-change projections can be reduced. In 1969, Lorenz [30] wrote: ‘Perhaps we can visualize the day when all of the relevant physical principles will be perfectly known. It may then still not be possible to express these principles as mathematical equations which can be solved by digital computers. We may believe, for example, that the motion of the unsaturated portion of the atmosphere is governed by the Navier–Stokes equations, but to use these equations properly we should have to describe each turbulent eddy—a task far beyond the capacity of the largest computer. We must therefore express the pertinent statistical properties of turbulent eddies as functions of the larger-scale motions. We do not yet know how to do this, nor have we proven that the desired functions exist’. Thirty years later, this problem remains unsolved, and may possibly be unsolvable.’ http://rsta.royalsocietypublishing.org/content/369/1956/4751.full
Pekka’s fallacy – btw – is the argument by way of demanding impossible perfection. 1st by way of unsatisfiable gut feelings – and secondly by way of complaining that the unsolvable has not been solved and therefore the concept is likely to be wrong.
JCH
Sorry, JCH, you’ve got that wrong.
“Results from climate models” are not to be confused with “empirical evidence”.
Two kettle of fish, JCH.
Models are only as good as the input that has been fed in (GIGO).
Max
I must be a little dumb, but I can’t understand
‘3. the decreasing temperature makes saturated vapor condense releasing latent heat
4. the release of latent heat makes the drop in temperature smaller but cannot reverse the sign of the change’
So based on my understanding of water I state:-
Now lets have a simple water and N2 atmosphere. Initially, all the molecules are bashing each other and pairs of hydrogen bonding water molecules are broken before they are hit by a third slowly moving water molecule.
As the temperature falls the statistical possibility for pairs of water molecules to form increases, and so there is an increased possibility for a slow speed collision between a pair and single water molecule. All of a sudden we have three hydrogen bonded water molecules. The three water molecules require a fast moving molecule hitting them to be broken up, but can capture a range of slower moving water molecules. And so it goes, the bigger the hydrogen bonded water cluster, the easier it is to form a bigger cluster. You make use of this positive feedback nucleation cascade in a cloud chamber.
Now what I do not know, and would like to know from you Pekka is this, based on your two points.
For point 3). How do growing nucleation sites emit ‘latent heat’. In what form is this energy? Is it infrared radiation? If so, what is the spectra of this radiation?
For point 4). I was under the impression that Boyles law was at work. As the water undergoes a phase transition both pressure and temperature will drop. How can you be so certain that the ‘small’ drop in temperature isn’t due to a large change in pressure?
Doc, thank you for the questions. Particularly your 2nd to last paragraph questioning point 3. I
Naively, the hydrogen-bonded water droplets might lose some or all of their energy (latent heat) via IR radiation as the energy in the molecular orbitals drops to a lower level (??)
Alternatively, it is only the water molecules already at the statistically low end of the kinetic energy distribution which condense? Thus at a molecular level no energy is transferred, but the latent heat release is a result at the thermodynamic level?
Or, the kinetic energy of the water vapor molecules is carried over into the droplet somehow, but quickly dissipated to the gas molecules via collisions occurring immediately after the condensation?
DocMartyn,
There is an attraction between water molecules. That means that the energy of a water molecule with the same kinetic energy is lower in liquid than in gas. When a molecule comes from gas and gets stuck in the liquid, energy is released. Originally that energy goes to kinetic energy of molecules in liquid but soon it’s shared also with the surrounding gas when gas molecules bounce from the liquid.
When water is condensed in constant volume (like in a closed bottle) the pressure drops and temperature is raised as potential energy is released.
When water is condensed at constant pressure the volume is decreased and temperature raised again. This is the process in atmosphere.
When the condensation of saturated water vapor is driven by decreasing temperature some energy is released again but less than is taken by the process that leads to the decrease of the temperature. In ascending air the temperature is lowered by adiabatic expansion as heat is transferred to work that the expansion does in pushing other air out of way.
not sure I understand
this i do understand. see what I wrote above “Or, the kinetic energy of the water vapor molecules is carried over into the droplet somehow, but quickly dissipated to the gas molecules via collisions occurring immediately after the condensation?”
Pekka,
Above, when I said I don’t understand, is this the same as what I asked Doc Martyn, “the energy in the molecular orbitals drops to a lower level”?
So, energy could be released by IR, but at atmospheric densities it is more likely released in droplet-to-gas collisions?
BillC,
The part that you are “not sure you understand” is just another way of saying the same thing that you understand. It’s just based on the notion that the energy of the molecule is the sum of the kinetic energy and potential energy of the interaction between molecules.
Pekka,
Thanks. Doesn’t this make heat release by IR radiation part of the process at a molecular level, or is heat release by dissipation via gas to droplet collisions much more dominant?
Billc,
The term “orbitals” refers normally to the state of electrons, not atoms of a molecule. The electronic transitions occur usually at higher energy, typically at X-ray energies, but sometimes at UV or visible light.
Molecules in gas have discrete vibrational and rotational states that are related to visible, IR and microwave radiation. Molecules in liquid don’t have such clean discrete states but they do still interact more strongly at frequencies close to such vibrational states. The states are not discrete because neighboring molecules are so close to each other that no vibrational state has time to fully form before it’s destroyed again.
BillC,
Both GH gas molecules and molecules in liquid emit and absorb IR, but an excited molecule in free space may live rather long before it emits radiation. States of CO2 that radiate 15 µm IR have a lifetime of around 0.5 seconds in free space. The time between collisions is by a factor ob billion shorter than that.
The lifetime is only average. Thus one molecule in billion does emit radiation before next collision. There are all the time billions of billions CO2 molecules in vibrationally exited state in a small volume of air. Thus there’s a lot of emission going on in spite of the fact that only one case in billion leads to that.
Respectfully I ask for some more clarification
‘There is an attraction between water molecules. That means that the energy of a water molecule with the same kinetic energy is lower in liquid than in gas. ‘
I have no idea what this means. I know that during a phase transition there is the release of energy, I wish to know what for this energy is in.
‘When a molecule comes from gas and gets stuck in the liquid, energy is released. Originally that energy goes to kinetic energy of molecules in liquid but soon it’s shared also with the surrounding gas when gas molecules bounce from the liquid.’
Are you saying we observe loss of energy from the droplet in the form of infrared radiation? If that is your contention, then I don’t understand the next part of your description.
‘When water is condensed in constant volume (like in a closed bottle) the pressure drops and temperature is raised as potential energy is released.’
How can the temperature rise? The kinetic energy in the system has been transformed by the phase transition into infrared light, photons. These photons will generally escape the volume as liquid and gaseous water have low absorptions. A large fraction of these photons will go up and off into space.
Pekka,
thanks. 9 orders of magnitude does it for me (in terms of explaining why collisions are the dominant mechanism for transfer of the latent heat)
DocMartyn,
Infrared radiation has a negligible influence on what happens to molecules in air. (I just wrote a comment describing that as something like one part in billion). IR is important on macroscopic scales, not on micro scale.
The binding energy between molecules in liquid is essentially a form of chemical binding energy. In case of water molecules it’s strong enough to make two-molecule dimers stable in free space (but very easy to break up by weakest collisions).
In phase transition from gas to liquid chemical energy is released exactly as chemical energy is released when fuel burns. Only the quantity is less.
These sorts of explanations on a molecular level I think are important to this discussion. There are many here who understand classical thermodynamics but don’t try to understand this stuff. For me, it is helpful in the climate context to clarify thermodynamic or meteorological concepts using energy conservation at a molecular scale. Of course, the answers or translation to the phenomena on a large scale usually necessitate empirically (experimentally) determined coefficients to partition energy transfers.
Pekka, have you seen this
http://www.sciencedirect.com/science/article/pii/S0012825210000176
“This paper considers the emission of infrared characteristic radiation during the first order phase transitions of water (condensation and crystallization). Experimental results are analyzed in terms of their correspondence to the theoretical models. These models are based on the assumption that the particle’s (atom, molecule, or cluster) transition from the higher energetic level (vapor or liquid) to a lower one (liquid or crystal) produces an emission of one or more photons. The energy of these photons depends on the latent energy of the phase transition and the character of bonds formed by the particle in the new phase. Based on experimental data, the author proposes a model explaining the appearance of a window of transparency for the characteristic radiation in the substances when first order phase transitions take place. The effect under investigation must play a very important role in atmospheric phenomena: it is one of the sources of Earth’s cooling; formation of hailstorm clouds in the atmosphere is accompanied by intensive characteristic infrared radiation that could be detected for process characterization and meteorological warnings. The effect can be used for atmospheric heat accumulation. Together with the energy of wind, falling water, and solar energy, fog and cloud formation could give us a forth source of ecologically pure energy. Searching for the presence of water in the atmospheres of other planets might also be possible using this technique. Furthermore, this radiation might explain the red color and infrared emission of Jupiter.”
DocMartyn,
I hadn’t seen that and it’s difficult to tell what’s its relevance really is. I tried to find other papers on the subject but found very little that’s not authored by Tatartchenko and Perel’man alone or with coauthors . One master’s thesis and a article written by Chinese authors (both written in very bad English) refer to this paper. The earliest articles by Perel’man are from the early 1970’s.
Judging by myself the value of the paper is beyond my interest. Thus lacking further significant information on that I’m just not convinced that the phenomenon is as important as Tatartchenko implies. It’s to be expected that IR can is emitted and absorbed at phase boundary when molecules move from one phase to the other but it’s not at all obvious that the effect is really significant. I would expect that a major effect would appear in various empirical data sets and that the effect would therefore be well known, if it were so significant.
One factor that seems to influence against its importance is the frequency range that’s rather high for thermal radiation. Thus it may be a relatively major source of IR at those frequencies without any significant influence on the total energy balance.
Pekka, at the moment the water cycle appears to consist of sensible heat causing the liquid to gas phase transition, water molecules rising, water condensing and releasing the latent heat as molecular collisions.
However, if even a small fraction of the latent heat is converted directly into infrared photons it changes the energy flow through the system; the Earth is cooled as IR generated in the atmosphere.
If I were a physicist I would know or be attempting to find out what from of energy transduction occurred when water gave up its latent heat during the phase transitions.
DocMartyn,
If that effect is significant it manifests itself very clearly in the spectrum of the IR radiation leaving the atmosphere. There would be a significant excess in IR radiation in rather wide peaks around 1.537 µm and 2.1 µm or in terms of wavenumbers around 6500 1/cm and 4761 1/cm. Normal thermal radiation at these wavelengths is extremely weak, solar radiation is much more important. The paper tells that the peaks exceed the background radiation by a factor of ten. The background thermal radiation is, however, at these frequencies only one millionth of the peak value or less. Thus the importance seems to be 10/1000000, or totally insignificant.
This is the conclusion at the moment and that would explain well, why this phenomenon has not raised more interest.
Anastassia said:
“in order to lift a moist air parcel, you must draw a dry air parcel down: this is what circulation is about.”
I don’t think that can be right because water vapour is lighter than air.
Thus a parcel of air into which water vapour is injected will rise without any change in ambient temperature.The reduction of density from surface upwards is what initially reduces air pressure as measured from the surface.
The circulation starts with that initial uplift. At the point of condensation the corresponding descent begins in order to complete the circulation.
Condensation might release energy and reduce pressure locally but the increase in density of the air bereft of water vapour plus the drawing in of similarly dry air from the sides will result in a recovering air pressure as measured from the surface.
You can’t reduce surface pressure when evaporation occurs then reduce it again when condensation occurs.
It would be a neat form of perpetual motion machine though.
Stephen Wilde | February 1, 2013 at 1:34 pm | Reply Anastassia said:
“in order to lift a moist air parcel, you must draw a dry air parcel down: this is what circulation is about.”
I don’t think that can be right because water vapour is lighter than air.
Ah, I thought Anastassia was including that in her circulation.
I’ve just read judith’s: “Why have the biotic pump papers been published in Phys Lett without any apparent problem, but are resisted by the atmospheric sciences community? The point is that the physical reasoning is sound (which the physicists seem to get). The issues that the atmospheric scientists have are raised in the post (e.g.current climate models work fine, yeah right).”
I think the reason “atmospheric scientists”, I’m putting this in quotation marks because I think this refers to “climate scientists” whose basic physics I dispute, are resisting this is really quite simple, it includes condensation and as I’ve gone to some effort to explain, the Water Cycle is missing from the energy budgets and I give the following as examples of their narrative:
http://www.learner.org/courses/envsci/unit/pdfs/unit2.pdf
“Two processes remove CO2 from the atmosphere: photosynthesis by land plants and marine organisms, and dissolution in the oceans.”
In other words there is no rain in the AGWGreenhouseEffect’s carbon cycle. You can hardly expect people with no rain in their world to argue rationally about this paper which postulates condensation driving winds.
Here’s another aspect: http://vixra.org/pdf/1104.0013v1.pdf
“This H2O negative-feedback effect on CO2 is ignored in models that assume that warm moist air does not rise and form
sunlight-reflecting clouds, but remains as humid air near sea level, absorbing infrared radiation from the sun, and
approximately doubling the temperature rises predicted from atmospheric CO2 increases. This false positive-feedback
(amplification) due to the assumed non-bouyancy of warm air is vital for greenhouse effect climate disaster predictions”
The climate models don’t have any mechanism for producing the clouds they argue about; they don’t have any gases bouyant in air, not even water vapour which as you say is anyway lighter than air.
I continue to find the varied input in such discussions confusing, I think it’s mainly because there is no solid agreement on the basics. So for example, some throw in that the paper is giving standard already well known to meteorology, but this is clearly, from my two examples above, not the case that this is standard or well known for “climate scientists”.
There is a naive interpretation of Anastassia’s comment that works for me:
““in order to lift a moist air parcel, you must draw a dry air parcel down: this is what circulation is about.”
in this case, if parcels are assigned by mass and not volume (perhaps not conventional) then the descending air masses must be dryer, because precipitation has fallen.
Pingback: Condensation Driven Winds « the Air Vent
Resent with typos corrected
I want to document my views on the issue raised by the Anastassia Makarieva, Victor Gorshkov, Douglas Sheil, Antonio Nobre, Larry Li
paper.
This is, I agree, a thought-provoking paper, however, I feelthere are substantive misinterpretations of the physics.
First, it is inappropriate to consider the hydrostatic relationship as anything but that which the vertical pressure distribution will equilibrate to when forcing is removed. The real relationship, as I show in detail in my modeling book,
Pielke, R.A., Sr., 2002: Mesoscale meteorological modeling. 2nd Edition, Academic Press, San Diego, CA, 676 pp.
is that the hydrostatic assumption is only accurate when the vertical acceleration is much smaller than magnitudes of the buoyancy and the vertical pressure gradient force. This is easier to achieve when the horizontal scale of a circulation is much larger than the vertical scale. Scales larger than a few kilometers are always essentially hydrostatic in their pressure distribution.
Thus, equation 2, for example, does not appropriately represent the physics.
Second, I agree – when precipitation occurs, mass is removed. This is an important issue raised by Anastassia, Victor and colleagues. This will create a density (mass) reduction in that column, which will result in small scale dynamic pressure effects (i.e. producing acoustic waves) as the atmosphere adjusts to the new spatial distribution of mass. There is no partial vacuum, however. I recommend several of our papers where Mel Nichols and others have examined this issue in considerable quantitative detail –
-Nicholls, M.E. and R.A. Pielke, 1994: Thermal compression waves. I: Total energy transfer. Quart. J. Roy. Meteor. Soc., 120, 305-332. http://pielkeclimatesci.wordpress.com/files/2009/09/r-160.pdf
Nicholls, M.E. and R.A. Pielke, 1994: Thermal compression waves. II: Mass adjustment and vertical transfer of total energy. Quart. J. Roy. Meteor. Soc., 120, 333-359 http://pielkeclimatesci.wordpress.com/files/2009/09/r-161.pdf
Schecter, D.A., M.E. Nicholls, J. Persing, A.J. Bedard Jr., and R.A. Pielke Sr., 2008: Infrasound emitted by tornado-like vortices: Basic theory and a numerical comparison to the acoustic radiation of a single-cell thunderstorm. J. Atmos. Sci., 65, 685-713 http://pielkeclimatesci.wordpress.com/files/2009/10/r-327.pdf
Removing water by precipitation will lower the pressure in that column but only as this process continues. On spatial scales of a few kilometers and larger, the system will be very close to hydrostatic balance even with this effect. Even for smaller scales, except for the most vigorous convective motions, the dynamic pressure is still only a small fraction of the hydrostatic pressure.
The drop in pressure if all of it was suddenly removed, would be significant. Consider that at sea level the weight of the atmosphere (the hydrostatic pressure) is equivalent to about 10.3 meters of water. Precipitable water (the amount in a column of atmosphere if it were all condensed out) is almost always less than 0.0635m (a near record value).
If the surface pressure were 1000mb than the reduction in pressure would be 993.8 mb. However, i) all of this water would not be removed, and ii) air would flow in as soon as the pressure started to fall. If this fall in pressure is what the authors mean, I agree this is an interesting issue.
However, its response would be essentially a hydrostatic response and not from a “partial vacuum”. It would be straightforward to test this with modeling by calculating the pressure change over a region due to the hydrostatic pressure changes due to the removal of precipitable water by precipitation. In one run, ignore this effect; in the other run include it. In RAMS runs, we account for this loss of mass in calculating the pressure fields.
Removing water by precipitation will lower the pressure in that column but only as this process continues, and it will change the pH, which could have significant effects.
Gas does not have a pH.
But, water does.
liquid water, not water vapor.
Water vaour clearly is a gas. But the condensation clearly is acidic.
I very much agree with Pielke that the effects of condensation and precipitation upon the pressure field operate on far smaller scales, both spatially and temporally, than Makarieva et al. imagine. After all, both processes are quite highly localized. Thus on the scales of Benard cells and tornadoes and perhaps up to those of cyclones, convection in particular should be affected by their mechanism. At the scales of the Walker circulation and quasi-permanent features such as Hadley cells and the Azore high, however, this mechanism must be negligible. Certainly it does not drive the geostrophic winds. At best their theory will find realistic application in modeling local weather rather than planetary climate.
‘Most circulation patterns on Earth are much wider than they are high, with the ratio height/length being in the order of 10-2 for hurricanes and down to 10-3 and below in larger regional circulations. As a consequence of mass balance, vertical velocity is smaller than horizontal velocities by a similar ratio. Accordingly, the local pressure imbalances and resulting atmospheric accelerations are much smaller in the vertical
orientation than in the horizontal plane, the result being an atmosphere in approximate hydrostatic equilibrium (Gill, 1982). Air pressure then conforms to the equation…’
So the first part of Roger’s comment is addressed in the proviso that this applies to systems where the horizontal scale exceeds the vertical. Most circulation systems that is – Including Hadley cells – and all the major rain bearing cloud morphologies.
http://minerva.union.edu/failinge/earths_convection_cells.html
Although we understand broadly patterns of circulation – http://minerva.union.edu/failinge/earths_convection_cells.html – if condensation amplifies the circulation it is an interesting addition to the theory.
Preciptitation is a little more complex given that rainfall is quite often remote from the region where condensation happens.
Dear Roger
Thank you for your comments. I would like to once again most gratefully acknowledge your previous discussions of the relevant physics with us.
Equation (2) in the post does not presume a violation of hydrostatic equilibrium. Moreover, all the derivations are made assuming that the hydrostatic equilibrium is exact. (Minor deviations observed in the real world will not change our derivation).
Rather, Eq. (2) describes the (observed) non-equilibrium distribution of water vapor. If water vapor was the only gas in the atmosphere and obeyed this distribution, this upward pressure gradient force would make the gas accelerate up to very high velocities. A similar one-dimensional flow is observed on a small (laboratory) scale in the so-called heat pipes. The vapor accelerates from the hot end of the pipe, where there is evaporating liquid, to the cold end of the pipe where the vapor condenses. Velocities resulting from condensation-induced pressure gradient are huge, which determines the high heat transfer coefficients of these devices.
In the atmosphere the flow is three-dimensional, the water vapor is not alone and there is hydrostatic equilibrium. But still water vapor undergoes condensation and features a non-equilibrium distribution. Our proposition is that the potential energy released per unit time from vapor condensation (which is most visible in the 1-D case) is also present in the 3-D atmosphere. It is described by q (3). Because of hydrostatic equilibrium, vertical motions cannot be significant. Thus, all this potential energy produced goes into horizontal pressure gradients and horizontal winds, as per conservation energy equation (4).
So we agree with you fully that the hydrostatic readjustment is a key process. Namely this process leads to the formation of horizontal pressure gradients that we estimate.
Regarding the investigation of this effect in current models our view is that there are theoretical caveats, some of which we described in our blog. Additionally, because of condensation there is always a pressure readjustment in the column that occurs independently of the droplet fallout. (E.g., recently Spengler et al. (2011) addressed this problem in a 1-D numerical simulation.) So simply removing or adding precipitation (i.e. modifying the ∂p/∂t term as per the mass sink studies) does not, in our view, capture the relevant physical difference between a condensing and a non-condensing atmosphere.
Regardless of the rights and wrongs of the actual paper, I find it to be an extremely valuable paper. I believe the editors chose to publish it for the very reason that it puts new ideas about what drives climate out there – right or wrong, it has people thinking about things they otherwise would not and that is a good thing.
While the effects described may be small as some have suggested, it may also be the case that these small variations are at least in part responsible for “natural variation” – an area that seems well overdue for investigation.
We are rightly suspicious of monocausal explanations. And, there is not enough computing power on Earth to resolve a model embodying all of the relationships even if we knew all the variables. The real problem is, Western academia refuses to abandon a monocausal explanation for global warming and the reason for that refusal lies outside of science.
Enormous uncertainties persist with respect to the role of clouds in climate change. Moreover, models that strive to incorporate everything, from aerosols to vegetation and volcanoes to ocean currents, may look convincing, but the error range associated with each additional factor results in near-total uncertainty. Yet, there is a greater concern. Throughout the history of science, monocausal explanations that overemphasize the dominance of one factor in immensely complex processes (in this case, the human-induced emissions of greenhouse gases) have been inevitably replaced by more powerful theories.Philp Stott
It’s funny to read all the messages here that express belief that this paper has got so much criticism because it would put climate science in doubt, and be a problem for IPCC.
I cannot see, how that would be true even if the paper would be right. The paper is not about climate or climate change but about dynamics of the atmosphere on short time scales. It might affect models used for weather forecasting and dynamics that occurs at a level that’s not described in most climate models at all.
Pekka –
Your points seem generally true to me. However, I believe that if the paper were to provide support for significant biotic regulation of temperature then that would also diminish CO2 sensitivity. So, that does not pose a problem per se, depending upon IPCC objectves.
blueice2hotsea | February 1, 2013 at 5:33 pm | Reply
Pekka –
Your points seem generally true to me. However, I believe that if the paper were to provide support for significant biotic regulation of temperature then that would also diminish CO2 sensitivity.
Try this:
http://tallbloke.wordpress.com/2013/02/01/new-study-lends-support-to-makarieva-et-al-biotic-pump-theory/
Thanks tallbloke.
It is true that these miroprocesses of evaporation, condensation and cloud nucleation are not included in models as deterministic equations but are included as paremetisations. Getting the parameters right is of course one test of the plausibility of the model.
The other of James McWilliams tests of plausibility is ‘a postiori solution behaviour’.
Additional couplings introduce futher structural instability into the models – but presumably the subjective expectations about the range of plausible solutions would remain.
Peika said, ” It might affect models used for weather forecasting and dynamics that occurs at a level that’s not described in most climate models at all.”
well, if it turns out to be useful for weather forecasting and dynamics at that level, perhaps it would prove useful in reducing the rather large uncertainty the GCM have with clouds and aerosols?
This spreading effect they mention is kinda like that “ground plane” issue I have asked about. How with increased temperature the average altitude of the start of condensation would lower and that the lower cloud base would stimulate upper level convection and advection. That is one of the issues with a radiant model based on a moving “surface” I pointed out, Frame of Reference and all that.
Now if the models do accurately model the cloud dynamics, reduction in average cloud height, increased rate of upper level convection and the impact of advection on the radiant “resonance” I think it is called, this paper is a complete waste of time :)
I agree. The warmists are for some reason concerned and protest too much.
I don’t think that “the warmists” are really concerned.
The issue is that scientists don’t like grandiose claims that imply that they have been stupid. When they see such claims they require good arguments to support the claims, and keep on protesting when no good arguments are given but the claims have not been withdrawn and even presented again trough getting published.
Of course most scientists don’t like the grandiose claims that imply they have been stupid and that’s the point – good science can be and is rejected only because it implies that the orthodoxy has been stupid. And bad science is protected and cheered if it’s consensus, no matter how stupid.
But, do you think that it’s wrong to tell that erroneous grandiose claims are erroneous?
No of course, but what’s erroneous (or not) is decided by observation and experiment, not by consensus and opinions.
A scientific paper that cannot support its claims is erroneous.
It’s erroneous even, if the claims find support later.
Edim, you write “I agree. The warmists are for some reason concerned and protest too much.”
Is this reminiscent of the furor over the original paper by Livingston and Penn? This was rejected for publication, and the authors were persuaded to publish it on line. It has now been amplified, and published in a proper journal. It was about sunspots, and only peripherally about climate, but the rumor was that the implications for a potential Maunder type minimum was the main reason why the paper was rejected in the first place.
Pekka says>
“A scientific paper that cannot support its claims is erroneous.”
Agree again. That makes all the scientific papers that claim AGW is true erroneous – they cannot support the claim. However, nature remains the final authority.
Edim,
Normal scientific papers are much more specific than that.
Their conclusion may add evidence for or against strong AGW, but they don’t take more general stands on AGW.
It’s obvious that this paper does not take any stand on AGW, not for or against or purport to support any view on AGW.
Steven Mosher | February 1, 2013 at 12:03 pm |
“Simple question; How can an averaged result be incorporated in differential equation maths?
Your possible answers are.
1. Show how
2. Say it cant be done.”
Steven, I thought it might help if you clarified your idea of “SUBSTANTIVE”. What do you mean? What will satisfy you? For example, with this question you sound like you (and Nick) have something significant in mind. What is it?
All the equations are written in the differential form. They are there to be solved. The fact that they are applied to final volumes in the real world is no news. What “averaged result” do you want to incorporate and where?
Of course it is possible to take any argument irrespective of its relevance and appear as if you found a crucial error. But it is not very productive.
Simple. answer nicks question and show the math. you had two choices.
Pick one.
Steven Mosher | February 1, 2013 at 11:46 am | Reply
“In short, it cannot be incorporated into other science until the issues are addressed in a SUBSTANTIVE way.”
I think that it might clarify the situation also with the question why our paper was accepted if we undertake the following exercise. Let us take a paper which Nick Stokes and, by inference, Steven Mosher are comfortable with (find it SUBSTANTIVE) and see the difference with ours. We have one such example, it is the hurricane model of Bryan and Rotunno (2009). It is mentioned in the post and was mentioned in previous discussions at the Air Vent and, if I am not mistaken in my recollections, Nick praised it.
In this model the standard set of equations plus some additional numerical schemes are combined to find the dependence of some variable of interest (in this case, hurricane velocity V) on other variables and parameters, of which one is the turbulence length scale hl. We can for simplicity formulate this as V =f_1(hl), with function f_1 determined by the system of equations that are solved in the model. So, V=f_1(hl) is what the model produces. If we knew hl (e.g. if we take it from observations), we could predict V.
But the key problem is that hl is not an observable parameter. Nobody knows how to measure it experimentally and independent of V. So what is done in this model, this parameter is instead chosen such that the value of V the model produces matches the observations. But as we have chosen hl such that model-derived V matches the observations, we can no longer test our function f_1 that relates hl to V — is it at all reasonable? Thus, we cannot judge about whether the model physics is right or wrong. The model is not falsifiable. This is what we meant when we wrote that models are not physically based but calibrated.
Now let us turn to our work to see where and how it differs. We apply some physical arguments to get a relationship that links the dynamic circulation power q (3) or Q (5) to observable atmospheric parameters. We can write it, for example, like Q = f_2(P), where P is precipitation. We show in the ACP paper that function f_2 can be obtained in two ways: by making a couple of entirely plausible assumptions about condensation rate S or directly by considering the non-equilibrium pressure gradient of water vapor as per Eqs. (1)-(3) in our blog. Nick has been actively challenging our first derivation, he does not find some of our arguments plausible. That’s fine and constructive, and all of his specific points are well taken. Now then, on the basis of these criticisms, Nick perpetuates the idea (that looks SUBSTANTIVE to Steven) that our results are wrong and have been manipulated or “made up”. I am not certain about this last quote, but I recall having seen it somewhere, so if this is not true, I apologize in advance.
However, any interested reader can see at once that there is extremely little space for manipulation in our approach. There are only four key variables: vertical velocity w, water vapor pressure pv, and the two scale heights hv and h. They are all perfectly observable and measurable and combine elegantly to give an estimate of circulation power. Thus, even if someone, like Nick, is not convinced by how we derived our function f_2, the validity or invalidity of our propositions can be easily checked by empirical evidence — the ultimate criterion of truth in science.
This is a possibility which is lacking in our counter-example of the hurricane model (which, in contrast to ours, Nick and Steven find SUBSTANTIVE). There is no way of determining of whether function f_1 (which summarizes the model physics) is valid or not.
Such an empirical check in our approach is facilitated by the fact that both w (vertical velocity) and pv (water vapor pressure), and hence precipitation P, vary within wide limits in the atmosphere. So if our function f_2 were wrong, by changing the key variables greatly the prediction of circulation intensity that it yields must go weird.
But what do we observe instead? Instead we observe that f_2 yields meaningful predictions over a very broad range of precipitation values. No existing model demonstrates such a skill. If you think a moment, it’s really impressive. Is such a result worthy of publication? Certainly, yes.
This is what I speculate might have played a role in our paper getting published. The paper itself can be poorly written and have too many equations, the authors can be confused about some less essential points, they may have made a terrible error of writing heat dQ instead of entropy dS in an otherwise adiabatic equation, but having spent two years on disentangling this mess, the Editors might have became certain that the result is there. And it is important.
There is also the third (main) problem: these now-acknowledged flaws unaccountably have persisted for two years, despite the sincere input of multiple reviewers.
Many climate-change scientific collaborations routinely post multiple, steadily-improving versions of their work to the arxiv server for everyone to see … this practice greatly improves the quality scientific discourse … perhaps you and your coauthors should do this too?
In the long run, that openness might bring good results, Anastassia Makarieva! Because it is evident (to everyone) that you have tackled an interesting problem. What is less evident (to everyone) is whether your mathematical methods are valid, and/or your inferences are logical, and/or your conclusions are correct.
Don’t let another two years go by! Post a longer, clearer, better-validated analysis to the arxiv!
It is clear that comparison of you to Kyuna – a cutesy, bubble-pop, Korean pop princess is well warranted.
It is less clear that you have any command of, maths, physics or natural philosophy. Greater horror still is the repetition of your eggregious bad faith.
Please no one has shown any fault in the maths or physics. Roger suggested it was different in very systems – but that was specifically excluded to concentrate on major systems. Pekka is going wit his gut. Steven is demanding that it be numerically modelled before it becomes real for him. What did we do before computers? And you are really quite silly.
Please allow me to differ, Robert I Ellison! It is evident (to me anyway):
• Anastassia Makarieva’s post was sincere in acknowledging that the article’s Eq. (1) is thermodynamically flawed.
• Does repairing that flaw (and other flaws) alter the article’s conclusions? That remains to be seen (obviously) … at present, no-one knows!
• Posting manuscripts to the free-as-in-freedom arxiv server is acknowledged by the world’s most eminent scientists and mathematicians to be good collegial practice!
Aren’t all three of these points plain math/science common-sense, Robert I Ellison?
“Please no one has shown any fault in the maths or physics. “
Out of curiosity, do you think that precipitation rate S is proportional to humidity N_v? That this has been demonstrated?
Judith, and Fan,
Fan’s deliberate mischaracterization of Dr. Makarieva’s comment seems to me to have gone beyond the bounds of courtesy. His repeated posts also seem to have gone beyond the rules against spamming. I’m embarrased, on hehalf of the forum denizens, and I apologize on his behalf to Dr. Makarieva.
Not at all Kyuna – they are merely examples of your eggregious bad faith following silly comments on inexact differentials and Legendre transforms in thermodynamics. Someone might take you seriously – but someone who combines cluelessness with a bubble-pop princess persona does it for me.
Natural notation in thermodynamics is a useful mathematical language to learn, Robert I Elison! Old thermodynamical ideas are thereby clarified, and many new thermodynamical ideas can be expressed in no other language!
Cheif, All the nit-pickery about notation is desined to fill and confuse the thread so that relevant coments such as the one from Peter Berenyi http://judithcurry.com/2013/01/31/condensation-driven-winds-an-update-new-version/#comment-290686 get lost in the waffle from the likes of Kyuna-Fan
Tallbloke, is not precisely the opposite the case? Because of …
So perhaps the wisest course of Makarieva and her colleagues, is to recruit a strong mathematician to: (1) repair the article’s bad/(wrong?) notation and then (2) repair the article’s imprecise/foggy/(wrong?) reasoning.
Obviously objective (2) cannot achieved before objective (1). Because there’s no getting around it: correct thermodynamical computations requires all of the elements of mathematical maturity.
Too many elements of (thermo-) mathematical maturity are missing from the Makarieva et al. manuscript, and missing too from critical reviews of it … and missing especially from skepticism of climate-change physical science in general!
That is why climate-change skeptics especially should study and practice assiduously the elements of (thermo-) mathematical maturity. Doesn’t that plain advice amount to pure scientific common sense, Tallbloke?
If dear – you used the wrong term in nit picking Tallbloke. This assumes that there is anything at all relevant – instead of the wildly improbable – indeed quite claims on inexact differentials and Legendre transforms in thermodynamics. To add to that we now have Arthur C Clarke and Ruppeiner geometry via wikipedia. It has all the hallmarks of an internet search for terms that include thermodynamics and links to things that on inspection turn out wholely irrelevant to the issue a hand.
It leaves the impression of comments motivated less by truth and science and more by some psychological disorder. That it is all delivered in a cutesy, bubble-pop princess persona is bizarre.
Anastassia, be advised that Steven Mosher is a drive-by shooter who never comes back to face criticism of his illogical assertions. He is a waste of your time.
Anastassia and co-authors, it comes down to whether you believe undilute ascending air parcels follow the moist adiabat which requires latent heat release in its derivation from the thermodynamical equations. If this is not your field of research, you may not have realized that the moist adiabat is central to understanding atmospheric convection and is now routinely used by weather services successfully to estimate updraft buoyancy, or convective severity, and to predict cloud-top heights. The concept of CAPE (convective available potential energy) is a well established consequence of the moist adiabatic lapse rate, and CAPEs over several thousand J/kg correspond to strong updrafts and severe convective situations when they are seen in atmospheric soundings, especially in conjunction with vertical shear that helps organize the convection into long-lived cells.
Jim, we are aware of CAPE. In our paper a whole Section 3.4 is all about comparing our effect with CAPE.
“CAPEs over several thousand J/kg correspond to strong updrafts and severe convective situations when they are seen in atmospheric soundings, especially in conjunction with vertical shear that helps organize the convection into long-lived cells.”
Yes, but as we discuss in the paper on p. 1045, very strong moist updrafts form in low CAPE environments as well. This suggests that perhaps CAPE is not the thing that actually decides.
The point I made several times in this thread is that CAPE calculated for the ascending parcel is not a measure of potential energy available for circulation because of the energy losses in the descending branch. (When the parcels that descend dry adiabatically become warmer than the environment.)
‘In meteorology, convective available potential energy (CAPE),[1] sometimes, simply, available potential energy (APE), is the amount of energy a parcel of air would have if lifted a certain distance vertically through the atmosphere. CAPE is effectively the positive buoyancy of an air parcel and is an indicator of atmospheric instability, which makes it very valuable in predicting severe weather. It is a form of fluid instability found in thermally stratified atmospheres in which a colder fluid overlies a warmer one.
While I was there – wikepedia had a nice little animation.
http://en.wikipedia.org/wiki/File:Convective_instability_animation_12Z_21Z_Jan08.gif
A skew-T plot showing a morning sounding with a large hydrolapse followed by an afternoon sounding showing the cooling (red curve moving to the left) which occurred in the mid-levels resulting in an unstable atmosphere as surface parcels have now become negatively buoyant. The red line is temperature, the green line is the dew point, and the yellow line is the air parcel lifted.
Btw – Anastassia thank you – and what a beautiful name you have.
There was a bit of discussion about disciplines. A broad knowledge is essential to understanding just about anthing in climate. Narrow disciplines are just so 20th Century.
Unfortunately the discussion in your sections 3.3 and 3.4 is mistaken because when you create the dry and moist columns you apply a diabatic cooling process (like refrigeration) which is an entropy reduction. This would be OK, except different coolings are applied to each column. As we know, cooling at constant volume (constant height of the columns) leads to a pressure reduction and so the dry column which becomes cooler should have lost more pressure than the moist one in this diabatic process. Taking this into account, you would have seen that the pressure effect you are looking at is smaller than that due to your cooling process that you neglected. The net effect is consistent with the idea that latent heating has a bigger positive pressure effect than vapor reduction’s negative effect.
In ‘The Australian’ newspaper today,Feb 2,, 2013, almost a half
page coverage ( p22) of ‘Where the Winds Come From’ under
the heading, ‘Branching out on climate’…. a theory based on the importance of forests iis threatening the established ideas. The
article includes comments by Douglas Shiel. professor of forest
ecology. Would this theory have received media publicity 3or 4
years ago, I wonder?
blogged here:
http://tallbloke.wordpress.com/2013/02/02/makarieva-et-al-make-the-headlines-with-where-do-winds-come-from-paper/
Please see Tallbloke’s blog for my comments (not here where it is a distraction)
i) Pre condensation a given volume contains:
a) Water vapour and b) A quantity of air.
ii) After condensation that same given volume contains:
a) The same amount of water vapour in condensed form and b) That initial quantity of air and c) More air and water vapour imported from the surroundings.
Isn’t ii) going to be heavier than i) for a net increase in surface pressure ??
What has happened is that a nominal pressure reduction locally at height caused by contraction has imported more molar material into the original volume for an increase in pressure at the surface below.
Isn’t tthat a serious problem for Anastassia ?
Dear Dr. Makarieva,
how do you explain several prominent features of the global wind field?
1. winds are strongest above mid latitude oceans
2. there is a sharp drop in wind energy at continent boundaries
3. winds are low above continental areas except over deserts
4. winds are weakest over the tropics where precipitation rate is highest
5. the windiest continent is Antarctica, also the driest
see Annual 50 m Wind Speed (July 1983 – June 1993)
Thank you for your question. Let me give some context to my replies below. Our theory constrains the rate at which the kinetic energy of winds is generated. In hydrostatic equilibrium this rate q defines the work per unit time of the horizontal pressure gradient, see Eq. (4) in the post. It is given by u∇p, i.e. it is proportional to the horizontal velocity component u_p that is parallel to horizontal pressure gradient.
Thus, Eq. (4) does not constrain the velocity component that is perpendicular to the pressure gradient (e.g., the geostrophic wind). Nor does it separately constrain u_p and the pressure gradient, only their product. In order to find all the velocity components one equation is obviously insufficient. So, while q in (4) is proportional to precipitation P, this equation cannot be interpreted as a kind of rule of thumb “where precipitation is high, wind is strong”. Eq. (4) provides a constraint on the generation of dynamic power, but we certainly need additional information on dynamics as per Euler and N-S equations to solve the problem in full.
After this preambule, I do not think that my replies below will be anywhere surprising.
1. Strongest winds in midlatitudes. Unlike the Hadley cell, the midlatitude circulation cell goes against the gradient of solar power (with the ascent occurring where there is less solar power and hence less potential evaporation). This “forced” cell is heavily impacted by the dynamic processes in the much larger tropical area, with geostrophic winds enhanced by a larger Coriolis force at a higher latitude.
2. there is a sharp drop in wind energy at continent boundaries
3. winds are low above continental areas except over deserts
Surface roughness is higher on land than over the ocean, and is minimal on land in deserts. Note that condensation-driven winds, like any circulation, have an ascending branch (where it rains) and a descending branch (where it is dry). So the existence of winds in deserts per se is not an argument against condensation-driven dynamics.
5. the windiest continent is Antarctica, also the driest.
Antarctica is outstanding not only in being the driest. Unlike in Hadley cell where the ascent and hence precipitation occur over a longer radius (because of low latitude) than the descent, in the polar cells situation is the opposite. So there is an opportunity for the dynamic power that is generated in the midlatitude rain belt of the ascending air to be concentrated over a relatively small area in the area of descent over the poles. Concentration of wind energy produces large local winds over the poles.
An excellent and provocative paper – I do like ideas that make me think!
I couldn’t see where in the equations the effect of suspended (or teminally falling) water droplets on surface pressure was handled. These are the forces exerted on the air by the water droplets suspended in it. I may have missed something, or in the discussion here.
The surface pressure on a patch of ground is the weight of all the material above it, minus its integrated density times acceleration. When water vapour condenses, the water droplets are in freefall, and accelerate downwards. This explains the drop in pressure. Only the air has to be held up by the ground.
But after a few moments the air resistance counters the acceleration, and the droplets reach terminal velocity. The droplets in clouds are supported by the air, which is in turn supported by the ground via air pressure. This implies that suspended droplets contribute to surface air pressure. Only when droplets fall freely, independently of the air, is the pressure reduced.
Thus, I would expect a transitory pressure drop as water vapour condenses and starts to fall, the pressure returns to normal as it reaches terminal velocity, then as the droplet evaporates again the increase in pressure from the expanding water vapour balances the loss of the downward drag forces. What the net effect of all this would require more detailed calculation and more cloud microphysics than I have time for, but my suspicion is that the effect would be less than anticipated based on freefalling droplets.
What have I missed?
NIV, same questions that I have, I think this can only be addressed by trying to incorporate these processes in a model and evaluating what happens (on short and long time scales) both with and without these processes.
See this paper by Bannon
http://journals.ametsoc.org/doi/pdf/10.1175/1520-0469(2002)059%3C1967%3ATFFMOM%3E2.0.CO%3B2
Note, when I pointed this paper out to Makarieva, she thought the way Bannon handled the falling drops was absolutely wrong. I haven’t looked into it enough to judge, but these are issues that modelers have been ignoring and we can’t tell yet whether all this can be ignored in large-scale weather/climate models
Judy, thank you so much for the opportunity of this discussion at Climate Etc.
Indeed, we find that the derivations presented by Bannon (2005) are incorrect. We have made a detailed analysis of this here. Along the way it turned out that there are related but different inconsistencies in the multi-phase flow treatments outside the atmospheric science domain. E.g. if you read Section 5 in that text, this alternative derivation is incorrect for a reason different from that of Bannon (and this derivation contradicts Bannon’s), but consistent with what is written in multi-phase flow textbooks. There is also controversy between Bannon (2005) and Ooyama (2001). From what we have learnt so far, it looks like it will take some effort to set the record straight now.
Yes, I think digging into these issues and sorting them out will help integrate your ideas into the more traditional frameworks
“I haven’t looked into it enough to judge, but these are issues that modelers have been ignoring and we can’t tell yet whether all this can be ignored in large-scale weather/climate models”
Our experience has been that there is no sufficient interest in the underlying physics. For example, there are two conflicting derivations of the equation of motion in the presence of droplets, the one of Bannon (2002) (sorry 2005 above was an error) and Ooyama (2001). They directly contradict each other, have different physical meaning. A decade has passed without any analysis having appeared in the meteorological literature that would have addressed the contradiction.
I agree that no reactive motion occurs when droplets start to fall. Initially, when the speed of the droplet is very small and increasing with the acceleration of nearly g it doesn’t have any effect on pressure, later the drag leads to an effect that corresponds to the mass of the droplet when the limiting speed has been reached.
I’m not absolutely sure but I think that the error in Bannon’s paper can be expressed equivalently in a different way as follows:
The two phases that occupy a specific volume at a specific time are not bound together well enough to allow for writing the conservation law of the momentum in that way for a volume that contains both phases. On the contrary each droplet falls independently. They cannot be considered as a body of material as a volume of gas can. The outside pressure on the volume of the gas is not pressure on the droplets. Droplets exit continuously the body at the bottom and enter at the top. They may also be created or destroyed and they may grow or contract within the volume. These processes require handling in a way different from Bannon’s equations.
Agreed, in addition much cloud converts to rain and as the rain hits the ground, the pressure would drop due to reduced mass in the column. The paper neglects to mention any of this process which is likely larger than the one they postulate.
Thank you for your comment. You are right, in the paper there is nothing about droplets. The effect that you are talking about can be theoretically estimated and shown to be insignificant. I can only give you my word on that right now because we still plan to publish it.
But there is another effect associated with the droplets that exists even the droplets experience freefall and do not interact with the air. If we imagine circulation as an elevator — where one part is ascending and another one is descending, we note that the descending mass balances the ascending mass such that the total work of the elevator motor against gravity is zero.
The same with circulating gas: the descending branch serves as a weight for the ascending branch.
Now imagine that we have condensation, which means that some mass ascends as vapor, converts to liquid and falls down in the region of ascent. Our “gas elevator” balance is now broken. There is less gas mass to descend than it has ascended. Thus, the circulation “gas motor” has to do some additional work to push the descending air downwards. This additional work depends on the height from which the droplets are falling in the ascending branch. Incidentally, this work coincides with the amount of frictional dissipation related to precipitation.
This is what we have already investigated and you can read the details here, although the text is not final. In brief under observed conditions the effect does not reduce the circulation power as estimated in (3) by more than 30% (upper limit), but this figure grows with temperature suggesting a suppression of circulation at higher temperatures.
Thanks Anastassia, I will await publication with interest.
Assuming I’m understanding it correctly, your objection to Bannon isn’t quite what I was talking about. I agree that when the droplets are first formed in freefall, there is no reactive force on the air to balance the weight of the droplets that is causing them to accelerate downwards.
What I had meant, (and what I had assumed Bannon meant although I haven’t parsed the paper in enough detail to tell), was the situation when the droplets have reached terminal velocity – the upward drag force on the droplet has to be balanced by a downward reactive force on the air equal to the weight of the droplets. This adds to surface pressure. I don’t understand how it can be negligible.
It’s Newton’s second law. If the only external forces on the atmospheric (air +droplets) column are the upward surface pressure and it’s weight, then the only way the surface pressure can change is if the atmospheric column accelerates. I agree that initially freefalling droplets do so, that there is no reactive force on the air when they do, and this must lead to a pressure drop. But it only lasts as long as the droplets accelerate downwards. When the droplets are at terminal velocity there’s no longer any acceleration. For the air around the droplets to maintain the pressure drop, it would have to be (net) accelerating downwards instead. Is this the case?
Nullius in Verba, indeed what is discussed in our critique of Bannon (2002) is the reactive motion term, while you are talking about has the form of ρlg term.
Yes, what you are saying about one droplet is correct, but when you translate this to a large scale there are complications.
E.g. you say “The surface pressure on a patch of ground is the weight of all the material above it,”
If there is one droplet above me, what is the area on which its weight is projected on the Earth? I.e. how do you define the size of the patch of ground above which the droplet is? It looks like it is arbitrary, isn’t it.
“Thus, I would expect a transitory pressure drop as water vapour condenses and starts to fall, the pressure returns to normal as it reaches terminal velocity, then as the droplet evaporates again the increase in pressure from the expanding water vapour balances the loss of the downward drag forces. ”
Spengler et al. 2011 (we discuss their work in our ACPD comment here) made an attempt to investigate this for 1-D case. What they found — the pressure at the surface drops immediately upon condensation and never returns to the initial value.
The droplets being accelerated by gravity influence the pressure less by the amount that corresponds to the acceleration by Newton’s law. The reduced pressure influences the rest of the atmosphere and the reduction in pressure is removed trough that. We end up with the same pressure at the surface as without this effect.
What I write above is vague as it does not define, what is the general state of the atmosphere and the surface. Therefore it’s not really possible to make statements like the one I make above.
This is perhaps the most fundamental problem with your paper. You compare columns A and B similarly without well defined environment where they exist. It.s not possible to make statements about horizontal pressure gradients when the overall setting has not been defined.
As I explain in a lengthy comment below, very much that you describe is restatements of common knowledge. Deciding what is different requires well defined description of the overall settings. The described settings must agree on essential points with the real atmosphere.
Now all the strong conclusions are in the air (sic) without clear connection to what happens in the real atmosphere.
“If there is one droplet above me, what is the area on which its weight is projected on the Earth? I.e. how do you define the size of the patch of ground above which the droplet is? It looks like it is arbitrary, isn’t it.”
It would depend on the details of the flow of air around the falling droplet, but there would be a definite answer. The drag-deceleration of the droplet will send out a pressure wave into the air, which when it contacted the surface would constitute the raised surface pressure – the circle of contact would define the pressure footprint.
There are various physical analogies that might be helpful to explore ideas. There is an old scientific riddle about a truck driver carrying a load of birds across a weak bridge. The bridge is strong enough to carry the truck, but not the truck with its load. How does he get across?
The answer usually given is that he bangs on the side to alarm all the birds into flight. With them flying freely, the load is lightened and the truck can cross safely.
But this only applies if the cages are open. If the truck is airtight, the truck will weigh the same with the birds flying or settled, as the downdraft is equal to the bird’s weight.
Another example would be a parachutist in a large, sealed container of air. If his weight was not exerted on the base, one could transport the parachutist and container upward, and then recover the same amount of energy by allowing the container only to descend, giving perpetual motion.
I had a look at the linked document but couldn’t see Spengler’s analysis. Can you be a bit more specific? Thanks.
NIV, and the more obvious example is that we are not flattened every time a plane flies overhead. The key to the hydrostatic assumption is that it only applies when horizontal scales are much larger than vertical scales. A wide enough rain or cloud area would have a hydrostatic loading effect on the surface pressure. Unless the discussion is restricted to wide columns, we can’t consider the hydrostatic pressure to be an accurate approximation.
That particular answer to Held where Spengler is referred to presents basically the standard derivation that leads to moist adiabat. There’s nothing new in that and making the interpretation that they prove the impossibility of condensation in combination to latent heat dominance presents total misunderstanding of the standard theory.
As I have already written twice, everybody agrees that condensation is connected to decreasing temperature. The latent heat dominance does not contradict that. It means only that the rate of cooling is so much less that we end up with moist adiabatic lapse rate that’s typically about half of the dry adiabatic lapse rate.
Whether the effect of falling droplets is significant depends on the time it takes them to fall in comparison to the time it has taken for the air to ascend. If the ratio is very small the influence of the droplets on pressure can be neglected. The same value can be expressed also as ratio of water in droplets to that as vapor.
In other ways precipitation is certainly important. One interesting question is to what extent they fall in the rising column and to what extent the end up outside that in drier air where evaporation occurs rapidly and changes the properties of adjoining air.
Precipitation efficiency, which is the ratio of surface precipitation to column condensation, is an important parameter in determining hurricane intensity, and it is the surface pressure effect that is at the center of this. There have been papers on this issue. This efficiency is a function of microphysical processes affecting the rain production rate.
I have to admit that I hadn’t gone trough the whole article before writing my earlier comments (and I’m not totally trough even now but far enough for what follows). Going trough the paper does not explain to me why it has been written as it has, and going trough the paper raises questions on why the authors write, what they write. They seem to be implying that their statements would provide some essential new insight where I don’t see anything new.
The mathematical derivations start in Chapter 2.1. There they present a derivation that must be essentially the same as that used in textbooks to derive the moist adiabatic lapse rate (I haven’t worked out the details). They don’t go trough all the steps to the lapse rate, but what they do is included in that derivation.
They end the chapter “This proves that water vapor condensation in any adiabatic process is necessarily accompanied by reduced air pressure.” That’s essentially equivalent to the well known fact that adiabatic condensation occurs always in ascending convection where the parcel of air moves to lower pressure and cools. Thus the observation is really well known.
Next they put the sentence “Adiabatic condensation cannot occur at constant volume” as the title of the next chapter. That chapter contains only an obvious corollary of the previous one. Again very well known.
Next chapter on non-adiabatic condensation tells also only very well known facts.
In Chapter 3 they derive the barymetric formula for dry and moist profiles (nothing exciting here). The main differences are due to the very different lapse rate that makes the density of the moist profile fall faster than that of the dry profile. (With increasing altitude the pressure drops and that makes the density go down, but the temperature drops also and that has the opposite effect making the decrease in density with altitude the weaker the larger the lapse rate is.)
Adding the influence of water vapor concentration to the density profile makes the difference between the densities of the two cases a little larger, because water vapor lowers the density as H2O molecules are lighter than average for dry air. The main deviation in the density profiles is still due to the difference in the lapse rate. The lapse rate effect is discussed in every textbook on the subject, the small additional effect due to changing water vapor concentration not. To get the order of magnitude lets assume that the vapor pressure at the surface is 30 mbar. That reduces the density by 1.2% in comparison with dry air. At the altitude of 5 km the saturation pressure could be 7 mbar and its influence on density 0.3%. Thus we have an effect of 0.9% in the density.
At the altitude of 5 km the pressure is a little more than 500 mb with a little influence from the temperature profile and much less from the moisture profile. The temperature would be almost 50 C less than at the surface based on the dry adiabatic lapse rate and roughly 25 C less based on moist adiabat. Thus we find about 10% difference in the density as the first order effect.
The density of the moist atmosphere is lower at the surface and the difference grows with altitude as the moist atmosphere is warmer. Above the altitude of about 0.5 km the difference in lapse rate is the main reason for the density difference. As the moist column has a lesser density, it’s pressure drops more slowly with altitude.
The differences in moist and dry columns are essential foe many weather phenomenons. Everybody is likely to agree on that. My understanding is that the only point that the paper tries to make is that the influence of water vapor content on the density of air has not been taken into account in all calculations. Further they imply that its influence would be a major one. I cannot see, how that could be the case taking into account also that this effect weakens with increasing altitude.
How they reach such conclusions from the changes described above is another question. My view is that they don’t give sufficient emphasis on discussing the physics but perform calculations that are not really applicable for the physical case. Such calculation may get spurious strong results by applying some unphysical constraints, which exclude natural processes that occur in vertical direction and are therefore forced to introduce excessively strong effects horizontally. It’s possible that they use altitude as vertical coordinate in a situation where geopotential height would be more appropriate. I have not tried to figure out, whether this is the reason.
When you get to such small impacts,just about any approximation can lead to spurious results. Consider geopotential height, fine for weather, but for climate 50 meters of sea level change is nearly a half degree of error trying to estimate a 1 degree impact. So if you use geopotential height, what baseline do you select?
Then as snow line and glaciers retreat, the moist air envelope would expand changing the virtual temperature and the CAPE in the higher latitudes, SSW events?
The models are using yard sticks when they need micrometers.
Most of the controversial parts of the paper are in the manipulations following Eq 34. A major mathematical fallacy is responsible for these.
Eqs 32 and 33 are conventional conservation of mass equations. Eq 34 looks like a consequence, but has been formed by a first principles argument involving reference volumes. This is just another way of doing cons mass, and should yield the same result. There is a slight error. In the correction -Nv/N (&partial;N/&partial;z) she uses ordinary air (N) as the non-condensable reference. But it isn’t, quite. She should have used Nd, dry air, and then the equations would be entirely consistent. But they are close.
The major math fallacy is that she proceeds with the algebra as if the different mass conservation equations are independently true. This introduces a new equation to the effect that the discrepancy is exactly zero. All sorts of things can follow from that.
The analogy I use is, suppose you have a derivation that yields x=1, and another that yields x=0.99. For many purposes, that is not a problem. But if you insist that both are true, then you can subtract (0.01=0), multiply by 100 (1=0), multiply by the national debt – well, you get the idea.
Nick,
My last paragraph and your last paragraph present the same argument.
I continue to claim that the errors originate in not defining the overall settings. There are equations but their relationship with the full problem are not well defined. Therefore additional assumptions are needed and these assumptions but these assumptions are not formulated explicitly.
Another point was the reply to Held’s comment the Makarieva linked to in one of her most recent comments. That was a pure strawman argument based apparently on misinterpreting standard thinking.
Words “.. and these assumptions ..” should be taken off.
I try to be a little more specific of the issue where I see the problem.
In order to discuss horizontal pressure gradients a plausible overall setting is needed. Looking at two columns (moist and dry) tells only that they are so different that they cannot coexist in the same atmosphere without much more that must be taken into account.
We have here the comments of Roger Pielke Sr along the same lines telling that a variety of mechanisms is always triggered by a combination of such columns. Horizontal winds related to the columns cannot be restricted to them alone. Taking the removal of water vapor from the air into account does not create any fundamentally different situation, it only changes a little the quantitative details. By far most of the structure results from the very different lapse rates which are determined mainly by “the dominance of latent heat release” as Held put it.
The paper may present something that changes a little the quantitative details, it does not present any new overall mechanism to tell “Where the winds come from”.
Nick, thank you for taking time to once again summarise your points.
I think that it might be productive for you to decide whether you want to persuade your readers in a mathematical or a physical fallacy present in our paper. It is easy to see that you cannot have both, so this takes away from the strength of your critique. This seems to be a persistent uncertainty. E.g. while you started your comments to our paper as “Mathematical doubts”, you ended them with (my emphasis) “I still can’t see the basis for Eq 34, and in particular what extra physics makes it independent of Eqs 32 and 33.”
For example, you now say that the fallacy is mathematical. However, anyone able to perform basic algebraic operations can see at once that all the equations in our paper are mathematically consistent. Condensation rate S in the continuity equations (32)-(33) is undefined. It is an unknown function. So in order to gain anything from the continuity equation, one needs to specify S. This is done in Equation (34). If you are concerned about mathematics, you can formally take Eq. (34) as a postulate. From this postulate and the continuity equation you then unambiguously obtain Eq. (37), which is the main result. I am sure that you will not be defending the statement that Eq. (34) can be mathematically derived from the continuity equation.
So, I suggest that you should instead insist that our paper contains a physical fallacy. It is more logically consistent. You should say that the assumptions that we involve to justify Eq. (34) are not convincing or anything obvious to you. To this I will respond that to us those physical assumptions do seem plausible. Moreover, I will say that there is a totally independent set of physical considerations that give the same result (Eq. 37) without involving the continuity equation at all (like Eqs. (1)-(4) in the post, see also the unnumbered expression for f_C on p. 1044). I will also say that the obtained result agrees with observations. At this point I think we will need to leave the readers of the paper to think for themselves not being able to add anything significant.
I should perhaps add that your case would be stronger if you not only be saying that our assumptions behind Eq. (34) are not convincing or following from nowhere, but rather showed that they violate some well-established physical evidence or equation or, even better, gave an alternative expression for condensation rate S.
Nick, let me consider your point about this mysterious small magnitude in somewhat greater detail.
It might be a vivid picture but to make it relevant you might need to give your readers some specific details: how does it actually pertain to our work, if at all? What is this small magnitude that you are talking about?
In the meantime, let me present our take on this small magnitude issue. It is in fact very interesting even if unrelated to our main conclusions.
In a horizontally isothermal atmosphere with ∂Nv/∂x = 0, the continuity equations (32)-(33), where S still remains an unknown function and no Eq. (34) is in sight, can be re-written as follows (see Eq. (6) in the post and (A7) on p. 1053 in the paper):
-u∂N/∂x = (S – S_d)/γ_d.
Here S, I emphasize, is an unknown function and S_d is expressed as the vertical derivative of the dry air relative partial pressure γ_d ≡ p_d/p, and N is molar density of air. There are two points here: first, that γ_d is very small and second, that S_d captures the main terms in the condensation rate, so S and S_d are very close whatever S is. (The ultimate reason for that is that the vertical pressure gradients are much larger than horizontal pressure gradients.)
Owing to the ideal gas law p = NRT, ∂N/∂x is proportional to the horizontal pressure gradient. So what we can learn from the above equation is that the horizontal pressure gradient is a function of two close magnitudes divided by a very small magnitude. So if we derive S from some independent considerations, we will need to ensure a very high precision of those considerations, to be able to have an accurate estimate of the pressure gradient. As we argue in the post, the first law of thermodynamics that is present in the standard set of equations solved in the current models, cannot provide the needed precision because of inherent physical limitations. This is the reason why, in our view, the condensation-induced dynamics has not been systematically studied or previously described from a theoretical viewpoint.
Note that all the above reasoning does not involve any assumptions about condensation rate S. It is generally valid.
“I think that it might be productive for you to decide whether you want to persuade your readers in a mathematical or a physical fallacy present in our paper. It is easy to see that you cannot have both, so this takes away from the strength of your critique.”
Not at all. The physics fallacy is believing the new equation (34) is based on physics other than mass conservation. You have often claimed that, but never said what that other physics is.
But the math fallacy is then using it as an independent equation. You already have sufficient equations in Eq 32 and 33. Eq 33 determines S. You then add Eq 34 as another equation in the same variables as 32 and 33. The system is overdetermined.
Proceeding from there just leads to nonsense results. For example, I noted that if you combine Eq 34 with 36, you get S=u∂N/∂z. That is, precipitation rate is equal to a wind velocity component times an air density gradient. This doesn’t require the presence of water at all. Rain out of dry air!
In terms of mysteries, we now have another one. You now say that
u∂N/∂z=-(S-S_d)/γ_d. That is a different expression again from that yielded by 34 and 36. That is the problem when you force solution of an overdetermined system. You can get anything.
Nick, why not to perform elementary algebra first? You have now announced another mystery, but try to put Eq. (34) into the equation above and you will get (37). Or you get (37) from (34) and (36). They are all consistent.
So your claim
“That is a different expression again from that yielded by 34 and 36.”
is incorrect.
I’ll reply to the rest of your comments a little later. Thank you again for rasing these issues here.
(I missed the start of this very interesting (in hindsight) thread by three days so it’s taking me a while to get caught up with the earlier comments.)
@PP: “This proves that water vapor condensation in any adiabatic process is necessarily accompanied by reduced air pressure.” That’s essentially equivalent to the well known fact that adiabatic condensation occurs always in ascending convection where the parcel of air moves to lower pressure and cools. Thus the observation is really well known.
1. I would accept your rephrasing if stated as “adiabatically rising moist air eventually results in condensation.”
Obviously there can be no condensation with dry air. Moreover condensation is not a result of decreasing pressure (quite the opposite in fact) but of decreasing temperature. Relative humidity decreases with isothermally decreasing pressure, but not as fast as it increases with decreasing temperature when both are the result of increasing altitude. It’s therefore the rising that’s important.
2. I would not however accept that this rephrasing is equivalent to AM’s phrasing “water vapor condensation in any adiabatic process is necessarily accompanied by reduced air pressure” because the latter carries with it the implication that if significant condensation occurs at a fixed altitude, with no other effect such as failling rain, then there is a reduction of air pressure. This can’t be the case for the obvious reason that pressure is determined solely by the mass of air and water above it, which is unchanged by condensation. Instead condensation can only occur as a result of decreasing temperature and/or increasing pressure, with temperature trumping pressure when both are decreasing due to altitude.
When the condensate starts falling is another matter, but that’s separate from condensation itself, which is all that the analysis of sections 2 and 3 of the paper treat.
3. Nothing has been said about the rate at which condensation occurs, however caused. Even if it had been the case that condensation reduced pressure, this change in pressure could in principle be so slow that its contribution to wind would be negligible compared to the other influences.
I don’t see any problem with the rest of this long comment of Pekka’s, but it was an early comment anyway and superseded by his later comments after reading more of the paper.
“Instead condensation can only occur as a result of decreasing temperature and/or increasing pressure, with temperature trumping pressure when both are decreasing due to altitude.”
This is wrong. Air containing water vapor, cooled to (and under) its dew point, will not undergo homogenous condensation (supersaturation). Only condensation nuclei (ions, hygroscopic condensation nuclei…) will induce condensation.
I don’t know if you’re a native speaker of English, Edim, but “only” means the opposite of what you seem to think it means. “A only if B” means that B is a prerequisite for A, not a cause of A.
No, I am not a native speaker, but I think you know what I mean. No condensation nuclei, no homogenous condensation. Even deep under the condensation point.
http://puhep1.princeton.edu/~mcdonald/JEMcDonald/mcdonald_ajp_30_870_62.pdf
I understood exactly what you meant. My point was that you seemed not to understand what I meant when you wrote “This is wrong.”
Hi Anastassia – Thank you for your reply. The fundamental issue that I disagree with you on is perhaps captured by your statement that
“all the derivations are made assuming that the hydrostatic equilibrium is exact.
This is indeed the issue. Accelerations do occur even in a hydrostatic system. So do dynamic pressures. This is what causes the rapid adjustment to the pressure field when density changes, such as when condenstation occurs.
Roger
I have tried to formulate essentially the same argument in my way. Yours is probably much easier to understand.
I just want to congratulate our hostess for putting such a challenging paper on this blog. Having read some of the commenters posts for a couple of years now on hundreds of topics, it is evident that this paper has really stretched them all. Science has to be the winner from all this.
I think most commenters need to take a much better look at this problem before unloading, including some of the original reviewers. The main problem, I think is that Anastassia and her colleagues have done a lot of thinking about this already and have a hard time now reconnecting with the rest of us, although they are doing a lot of effort. I think it would help a lot if some people who are now very critical do an effort as well, because most of the criticism is, I am sorry to say, superficial. You really need to try to do some of your own sample problems besides the ones given in the paper. So far, i have not really found anything that goes against conventional physics although I would readily admit I am not an expert in this area. And about these comments about “providing code”: what else could it be to it then to do exact bookkeeping of water in the gas and liquid phases in weather prediction models and use that to calculate or correct the hydrostatic component of pressure?
Precisely the same could be said of the folks at Principia Scientific International (PSI), eh?
That is why PSI folks are well-advised to verify their ideas by writing codes. Because turing ideas-and-equations into simulation code is a good way to come to grips with imprecise mathematical notation that is accompanied by foggy reasoning!
‘Prediction (the Zeroth Law of Thermodynamics) It is a remarkable mathematical theorem, that although the mass density and the energy density both will be found (in general) to be strongly -dependent, the temperature so (numerically!) computed will be found to be independent of , which is physically to say, the temperature of a fluid column in a gravitational gradient is constant throughout the column.’
Write the code Kyuna – I am assured that such goobledegook will follow the age old princple of GIGO to the nth degree.
The way the comments are presented reflects the grandiose claims of the title of the paper. Writing such a title brings attention, and largely critical attention when the paper cannot provide solid support for the title.
It is called a theory ,or an alternative explanation of the problematic area in hydrodynamics such as hydrostatics which deals with the Euler and NS solutions (with all their accompanying problems) and where solutions to the hydrostatics are indeed very rare,and great care is needed,
The arguments that the existing theory is complete or correct is naive at best.
The existing theory is as good as their done correctly up to a minor approximation which has been done understanding well its nature.
The existing unproved theory(s) whic (Due to approximations ) admit infinite solutions, and appealing to closure arguments is indeed troublesome at best
The problem is this. folks have asked questions repeatedly starting at the air vent, then at lucias, and finally as reviewers. Folks who have worked their whole lives in modeling just these sorts of processes. Some folks well recognized in the field. Getting aswers to these questions would allow testing to proceed. without those answers nothing can be done. yet the authors refuse to clarify. They are wasting the time of smart people.
And in the midst of this they trash the work of authors who do supply working equations and they feed the doubts of the uninformed.
That is not a good thing.
Steven Mosher
Any basis in evidence? None. Anyone taking a little time to check will see that this is false.
From what I see many unanswered queries are aimed at you.
I think, Steven, that you are exaggerating. Wasting the time of smart people? As Nick Stokes mentioned here the other day, “There is essentially no mathematical defence of the work by anyone other than Dr M.” But equally there is essentially no mathematical attack on our work by anyone other than Nick Stokes. He is actually doing all the job for those smart people who want to be critical. But it looks like (and I hope that) Nick does not mind, so all is well.
Besides, many people enjoy intense discussions like this one, so I think that many smart people do follow — even if for fun. It is like watching boxing, you know. But unlike in boxing, where everyone understands what the two men are actually doing, here there is an additional bonus for the smart: the more you understand the specifics of what is going on, the more you enjoy.
I agree Dr. M. This is a fascinating discussion and quite typical for bold new ideas. Mosher is just using stupid meta-arguments, which he is fond of. I look forward to further research which is all that is needed. Hang tough in the face of empty criticism.
Steve, perhaps your comment would have more credibility if you had answered some of the remarks made in reply to your previous comments in this thread.
This is the first I’ve heard of any of this, even the post let alone the paper it claims to defend against its critics. The following is therefore to be taken with several grains of salt.
From a quick read of the post I gathered that the paper starts from sound microphysics and passes to what looks to me like unsound macrophysics.
The microphysics is that when a water vapor molecule condenses on a water droplet, its translational energy, aka its latent heat, is converted to sensible or thermal energy, which the droplet then distributes democratically to all its constituent molecules.
This much is fine.
The authors infer from this the macrophysical claim that condensation drives temperature upwards.
Arrgh. This is rubbish. (Feel free to interpret “this” as referring to the previous paragraph or this one. Choosing the latter will save you having to read the rest of this comment.)
Transitions between the solid, liquid, and vapor phases of typical substances, water being no exception, are completely reversible. Le Chatelier’s Principle (as applied generally to physical as well as chemical processes) therefore holds. For any circumstances (in particular temperature) there is an equilibrium between phases.
Equilibrium is standardly analyzed in terms of constituent proportions (in this case the two phases of water as respectively molecules and droplets sparsely populating air) and temperature. In equilibrium at a fixed temperature water molecules pass between their vapor and liquid phases in equal numbers. With decreasing temperature the equilibrium shifts to favor the liquid phase (with the caveat that the process can be very slow due to the empirically observed tendency of vapor molecules to bounce off droplets when intuition would suggest they would stick at first hit).
A complicating factor with Le Chatelier’s Principle is the notion of an exothermic reaction. For example 2H2 + O2 converts to 2H2O exothermically, a positive feedback that hastens the reaction so as to drive it rapidly away from equilibrium (witness the Hindenberg).
Likewise the assimilation of a free-range water vapor molecule into a water droplet is also exothermic. This is because the droplet converts the translational energy of the captured vapor molecule (close to Mach 1) into thermal energy, much as a meteor (way more than Mach 1) burns up in the atmosphere, or your disc brakes get hot bringing your car (definitely not Mach 1) to a halt.
But there’s a difference from the Hindbenberg: the feedback is negative. This is because increasing temperature promotes evaporation of the droplet, which quickly restores equilibrium.
When the feedback is negative like this, the temperature is not merely the dominant driver, it is the only driver.
This CE post proposes to respond to what the authors claim to be the three principal criticisms of the paper, namely that the extant models are sufficiently (i) comprehensive, (ii) effective, and (iii) comprehensive as to gain nothing new from this paper.
My criticism is none of the above. It is simply that condensation does not drive temperature in the manner envisaged in the paper at all. Rather the temperature of the atmosphere governs the rate at which condensation occurs. The bottom of a cloud is the altitude at which the temperature is low enough for the equilibrium to shift to the point where droplets become visible.
Those saying that the impact of condensation on temperature is small enough to be negligible could more accurately say that there is no impact at all. Condensation simply is not a driver.
From the perspective of Le Chatelier’s Principle, clouds happen at the whim of temperature (which in turn is governed by altitude), not conversely.
The theory is published: now we need evidence …
Much of this discussion has been about theory. Our paper and our blog were about theory too – so that makes sense. But I suspect many readers are interested in evidence. Dr Held too asked for “evidence” to pass his “high bar” – we rejected the argument as a point of principle. The question at issue then was whether we had presented a case coherent and interesting enough to answer: it is a theory. Theories come first the evidence comes later.
But that does not mean we don’t have extraordinary evidence.
We wrote a little about this in the paper (most points below can be explored by looking at the reference list there or at http://www.biotic-regulation.pl.ru/index.html), but it may be useful to highlight a few again here so you can make your own assessments. What is our evidence so far? How does out theory match reality?
For me the most powerful evidence comes from looking at how rainfall varies as we travel inland from the coast (over relatively flat terrain): Why does rainfall not decline over forest? It declines over non-forest in a relatively constant manner that is easy to understand (This seems to be a global pattern: see the figure in my previous blog here http://judithcurry.com/2011/03/30/water-vapor-mischief-part-ii/). Recycling is not an explanation – it would reduce the rate of decline but it could not prevent it. There is no alternative explanation at present.
This effect – the drawing of rain into continental interiors – requires a biologically functioning forest so we would predict that the effect will be smaller over boreal forests in deep winter (when the forests are metabolically inactive and not transpiring moisture). Observations support these predictions. There is no alternative explanation at present. See, e.g. 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, Ecol. Complex., 6, 302–307, 2009.
Our paper (discussed in this post) shows that we can estimate the power of global atmospheric circulation. This is the first ever such estimate developed from first principles and, though intended as a rough estimate, is remarkably close to the measured values. No alternative theory can currently explain this value.
Where we have good data on forest loss and rainfall change there are some observations suggesting a regional decline in rain regularity (as we would predict). See E.g. Webb TJ, et al. 2005. Forest cover-rainfall relationships in a biodiversity hotspot: The Atlantic Forest of Brazil. Ecological Applications 15: 1968–1983.
The work by Anastassia and co. (not me!) on hurricanes is also impressive: it shows that the condensation generated pressure gradients can give a physically and analytically consistent model of how such storm systems function and can be used to estimate several characteristics from first principles. E.g. Makarieva, A. M. and Gorshkov, V. G.: Condensation-induced kinematics and dynamics of cyclones, hurricanes and tornadoes, Phys. Lett. A, 373, 4201–4205, 2009.
So the score-card so far is 7:nil in favour of our theory (I rate the hurricane work as three points … but even if you don’t 5:nil is a good margin). That’s a good score line. Extraordinary? Well I acknowledge too that the search for counter-evidence is in its infancy.
So now the theory can and should be tested further. All those who think it is right, all those who think it is wrong and all those who are uncertain but recognise why it matters, can I hope agree that the ideas should be tested. That is a common goal.
@DiA: So the score-card so far is 7:nil in favour of our theory (I rate the hurricane work as three points … but even if you don’t 5:nil is a good margin)
Regardless of the margin, let 7:nil = x and 5:nil = y. Then 7 = x*nil = nil = y*nil = 5. Subtract 3 from both sides to give 4 = 2 and then divide both sides by 2 to give 2 = 1. Since you and Tony Abbott are two, you and Tony Abbott are one. No surprise there.
More seriously your theory takes various correlations and claims causality for them while ignoring that correlation is not causality.
Vaughan.. their theory has a nice beat but you cant dance to it.
Fun thanks
Indeed correlation is not causation. The process here is about falisification.
Any theory that can explain a phenomenon lacking previous explanation or can predict an observation (or correlation) that other theories cannot certainly looks better than one that fails to so.
Indeed some science philosophers think the whole point of the science process is about finding theories and then finding data (results or observations) that could potentially falsify the theory. The point is here that the data collected support rather than contradict the theory and the score-card is in our favour. If we get into more detail that may change. There may also be alternative explanations for some of these observations … so I accept there is a need to go much further if we can,
Indeed Douglas, where causation is plausible correlation is evidence of causation. So you in fact have some evidence. You have an evidence based conjecture and that is how science progresses. What is interesting is that people want to reject this conjecture for no reason except that it is such. Your answers are very good. Some of your critics seem to claim that conjectures should not even be published! They are very wrong.
> Indeed some science philosophers think the whole point of the science process is about finding theories and then finding data (results or observations) that could potentially falsify the theory.
Citation needed.
They got their theory published, doc. While you will be back at the AGU next year with the same little ole quasi-poster with the big red letters and the big red arrow. Will you take willie with you, to hold your poster?
Mosher has an in with COMICS. I bet he could get them to start a Deep Earthquake Science for Dummies journal for you. (Now let’s see what willie the yapping, sniffing stalker does, if he is not too busy sniffing and yapping at doc Douglas.)
> Dr Held too asked for “evidence” to pass his “high bar” – we rejected the argument as a point of principle.
Indeed, by completely misinterpreting Dr. Held’s point and by hinting at the fact that he was suffering from confirmation bias. I’ve told Douglas three times now and he dodged three times:
http://rabett.blogspot.ca/2013/01/atmospheric-chemistry-and-physics.html
Douglas’ promise to keep on substantive issues held for a day.
Willard
here you go
http://en.wikipedia.org/wiki/Falsifiability
http://en.wikipedia.org/wiki/Karl_Raimund_Popper
Thanks.
Now, we need a quote showing that “some science philosophers think the whole point of the science process is about finding theories and then finding data (results or observations) that could potentially falsify the theory.” You might have a tough time, since Popper’s claim was prescriptive, not descriptive.
The other day, it was Feynman and Einstein. Now, it’s Popper. You do seem to have all your talking points prepared for your entrance, Douglas.
***
Speaking of which, tallbloke’s, via the GWFP, both reproduced this interview with The Australian:
http://tallbloke.wordpress.com/2013/02/02/makarieva-et-al-make-the-headlines-with-where-do-winds-come-from-paper/
My favorite line was this one:
> Accepting our theory would basically mean the climate models are wrong. It wouldn’t mean that theories about carbon dioxide and greenhouse gasses are wrong.
Care to expand on that one, Douglas?
Unless you don’t consider that the social network (The Australian, the GWFP, Judy’s) a substantive issue.
Nice photo!
Hi Willard
Thanks for the appreciation.
Was d’Australis once d’Africa? Or was I facing too far southeast?
=========
Douglas,
A pleasure, as always.
I also acknowledge that you ignored my question, which was not rhetorical.
Please rest assured that I do recall these little things.
Another interesting quote;
“Sheil says the key finding is that atmospheric pressure changes from moisture condensation are orders of magnitude greater than previously recognised”
A finding??
In what sense was this a ‘finding’?
I feel that the phrase ‘show me the data’ may somehow be relevant.
HI Willard
Thanks for the continued interest
Sorry … I sincerely thought that the “no” answer would be self evident. (That article is not what this thread is about. Too many tangents and possible distractions (over 500 comments already) I can appreciate the flippant and theatrical ones without responding to each). Right now the most constructive action appears to be on other threads (Anastassia and Nick) — let’s enjoy it.
Michael
Good question but wrong place to ask it. Let’s stick with this blog here.
Nevermind, I know the answer – it’s a ‘finding’ in the sense of an assertion.
Michael
See the blog at Tallblokes.
Douglas,
You should pay some attention to the fact that a fatal explicit error has been found from the paper in addition to the other problems it has.
That would be minutes of my life I’d never get back.
OTT – how did a guy in ecology get caught up in this mess on atmospheric physics/chemistry??
Minutes lost? How would you tell the difference from the rest of your blogospheric life?
MIchael – “How …”
Its in the first blog link in the blog text above (climate etc.): I wanted to know if these ideas were true or not. I received diverse responses … but I also realised that they were not getting a fair hearing (scientifically reasoned). Well we’ve made progress.
Yep, saw that.
Read your first foray into this as well (Sheil & Murdiyaso, 2009).
The question remains, how did an ecologist wind up as a coauthor of a paper on atmospheric physics/chemistry?
> That article is not what this thread is about. Too many tangents and possible distractions.
Sure, let’s stick to more urgent matters like Popper and falsifiability and keep direct consequences from the theory to another time.
Michael “The question remains”
Sorry I don’t grasp your point. You know my motivation and you know I offered to provide help and have done so. What else do you need to know? I have a natural sciences background so dont find the math as scary as some though most certainly I lack the quick precision insight of Anastassia and Victor. I think for this theory there is a an advantage in coming from outside the discipline and not being weighed down by the implicit assumptions etc. What is your interest in this? Or rather why do you think the question is relevent here?
Willard
I fail to grasp your point (though yes I do appear to be allowing myself to be distracted despite myself … also a proposal deadline so nothing persona if I am tersel).
You do or you dont think a discussion of our evidence and how me might be able to falsify or find support for our theory would be useful? Clearly I would. I hear quibbles and side-tracks but not susbtance. Entertaining certainly, but unsatisfying. If you could get the cryptic and snide stuff under contriol we might even manage a constructive and substantive conversation. Not sure that’s what you want but I’ll keep an open mind.
Douglas,
Little more than curiosity.
There’s just such a jump from your usual publication topics (very interesting) to this.
And yes, plenty of people are capable of the maths, but as has been pointed out a few times on the thread, a deeper understanding of the topic (atmospheric physics) is helpful in avoiding descent into mathematically correct physical nonsense, or excited claims for newness of what is already well known by specialists in the field.
Thank you for your concerns, Douglas.
When you’ll respond to my criticism at Eli’s, we’ll see what I can do for you.
Michael “a deeper understanding of the topic (atmospheric physics) is helpful”
I agree … which is why we sought to do this (open paper, blogs etc). Despite all the suspicions — apparently many — we want to know if this idea is true. That’s it. That’s why we are here. We hope to engage persuade or be persuaded.
Willard “When you’ll respond to my criticism at Eli’s”
Done (yet again)
Nice try. Next time, try to focus on relevance:
http://rabett.blogspot.ca/2013/01/atmospheric-chemistry-and-physics.html?showComment=1360109422431#c436089895924529612
Douglas,
Since you tried to answer at Eli’s here it is:
> I fail to grasp your point […]
My point is that the consequences of the results in your paper, e.g. the correctness of actual atmospheric models, might be more fruitful that discussing the principle of falsifiability. To be able to falsify a claim is welcome, but it’s not an absolute.
There, I’ve said “absolute”. I’m sure you agree.
***
I could tell you about the need to naturalize epistemology and that holism wins in the end, but I’d rather not. I also surmise that that you’re not the substance guy. So if I ever want to discuss substance, I’ll contact Anastassia or Antonio.
Sorry everyone:
Pasted the wrong citation for the seasonal evidence of point two in my list (the effect is predicted to require a transpiring forest). I meant this one:
Makarieva, A. M., Gorshkov, V. G., and Li, B.-L.: Revisiting forest impact on atmospheric water vapor transport and precipitation, Theor. Appl. Climatol., 111, 79–96, doi:10.1007/s00704-012-0643-9, 2013.
its worth a look if you can access it.
[The one I had pasted belongs with point one (the relation with annual rainfall)]
Apologies for the confusion
The theory is published: now we need evidence …
While that’s easy for you to say it does not agree with the journal’s statement: “The handling editor (and the executive committee) concluded to allow final publication of the manuscript in ACP in order to facilitate further development of the presented arguments, which may lead to disproof or validation by the scientific community.” [Emphasis mine]
The theory is published: now we need disproof or validation.
Sorry if that was news to you, I’m just the messenger in this case. (Full disclosure: I believe this article will go the way of cold fusion, the OPERA neutrino faster-than-light anomaly, and the retroviral theory of chronic fatigue, but faster since it’s more obviously wrong.)
Vaughan Pratt “it does not agree with the journal’s statement”
Indeed. It wasn’t meant to — I do not speak for the journal. I was calling people’s attention to the idea of evidence as an alternative means to advance this theory. Having more people aware of the evidence we have (whatever they think of it) can encourage scrutiny and further work from those interested.
Does it need repeating that I concede the possibilty of a logical or analytical error (I’ve asked for that type of scrutiny many times from all kinds of people so thats on the record … indeed that is why we are here now)?
As I keep repeating too the skeptics (those willing to do the hard work, not the Moshers and Willards etc.) are actually our biggest resource here. You may be correct (a quick disproof). Let’s see. If so I shall be a little dissapointed but will be glad for the final resolution.That’s how it works right?
After so many years in public review I was less convinced that a knock-out blow is likely. If not how shall we progress? There may be readers here who find the idea of evidence interesting. That’s my hope.
Does it need repeating that I concede the possibilty of a logical or analytical error (I’ve asked for that type of scrutiny many times from all kinds of people so thats on the record … indeed that is why we are here now)?
Your paper has an interesting history. As far as close scrutiny goes, four out of five referees recommended rejection in 2009, on grounds that make complete sense to at least some of us here. Yet “here we are now” as you say, rehashing all the same objections. Perhaps the true destiny of your paper is as the Flying Dutchman of atmospheric physics.
(Incidentally I knew a Beau Sheil in Sydney in the 1960’s, we were residents of International House, and two decades later lived near each other at Stanford. Any connection?)
Vaughan Pratt “four out of five referees recommended rejection in 2009”
Thanks for the interest
I think it was one out of two (i.e. invited referees). You can check if you like. Its not so unusual with peer review — though I think noone would have objected if they had then sought a third reviewer.
A point that has not come up yet is how Dr Held was identified and invited as a reviewer. We actually helped do that ourselves in the full knowledge that he would be critical — that likely also affected the journal’s assessment. (If you get a submitted paper critical of a given theory you would, if you are an editor, send it to one opponent [someone directly criticised] and one less involved and see how each judges the work and then weigh the justifications offered accordingly). In any case the referees points were addressed. Despite the various claims repeated the paper changed quite a lot. The appendix for example was added.
“all the same objections”
No I think we have progressed. The scrutiny may ultimately leave both sides slightly unsatisfied but I do see clarification concerning what the core issues are (that is the way the system works formally). We have gone beyond saying Eq 34 is simply “wrong” to assessing what it implies and how it might be tested.
(Beau? I will check … I do have relatives in the country but have not heard of him, its not a very common surname so I would guess some connection. I’m not Australian myself)
@DiA: (quoting me) “four out of five referees recommended rejection in 2009″ Thanks for the interest. I think it was one out of two (i.e. invited referees).
Sorry, my mistake, the “rejection in 2009” was of the hurricanes paper Rosenfeld was critiquing, which only had Makarieva, Gorshkov (VG) and Li as authors — you and Nobre joined the “condensation reduces pressure” team when “hurricanes” were reduced to “winds”. I see now it’s up to six with the addition of Peter Bunyard in further reducing “winds” to “air passage” (last on the list below).
Most of the following papers invoke condensation as a mechanism, especially as one that reduces pressure. They also include refutations of arguments based on one or another violation of some thermodynamic. Given the generally negative reviews of the more recent of these, sorting out which of these mechanisms and refutations the referees are generally comfortable with seems like a nontrivial task.
Gorshkov VG, Makarieva AM, Gorshkov VV (2004) Revising the
fundamentals of ecological knowledge: the biota–environment
interaction. Ecol Complexity 1:17–36
Makarieva A.M., Gorshkov V.G., Li B.-L. (2006) Conservation of water cycle on land via restoration of natural closed-canopy forests:
Implications for regional landscape planning. Ecol. Res. 21: 897-906.
Makarieva A.M., Gorshkov V.G. (2007) Biotic pump of atmospheric
moisture as driver of the hydrological cycle on land. Hydrol. Earth Syst. Sci. 11: 1013-1033.
Gorshkov V.G., Makarieva A.M. (2008) The osmotic condensational force of water vapor in the terrestrial atmosphere, Preprint 2763, Petersburg Nuclear Physics Institute, Gatchina, 43 pp.
Makarieva A.M., Gorshkov V.G., B-L. Li (2008) On the validity of representing hurricanes as Carnot heat engine, discussed in ACPD 2008
Anastassia M. Makarieva, Victor G. Gorshkov, Douglas Sheil, Antonio D. Nobre, Peter Bunyard, Bai-Lian Li, Why does air passage over forest yield more rain? Alternative interpretations of , arXiv:1301.3083 [physics.ao-ph]
I think Beau Sheil is from the US.
Veep:”Perhaps the true destiny of your paper is as the Flying Dutchman of atmospheric physics.”
Or perhaps it will become a perennial poster at the annual AGU meeting. A recurring unphysical bad penny.
And this is not the firdt time this idea, or its variations, have been rejected on the basis of it’s errors of physics.
http://www.agrometeorology.org/topics/needs-for-agrometeorological-solutions-to-farming-problems/a-forest-as-biotic-pump201d-hypotesis-discredited-due-to-errors-in-basic-atmospheric-physics
Michael
Well we know its controversial.
See this earlier link for the key Meesters et al. texts and the reply: http://judithcurry.com/2013/01/31/condensation-driven-winds-an-update-new-version/#comment-291430
Meesters and colleagues did a valuable service in addressing the ideas in a formal manner. I wish they would do more. (The article in your link is in large part derived from Daniel and my article but re-jigged to highlight the formal Meester’s et al. review. There was then a response and a reply I think).
I think they felt it beyond help.
Their summary was- it was wrong before and its wrong now.
The present summary might be – it was wrong first, then it was wrong again, and it’s still wrong.
VP Perhaps the true destiny of your paper is as the Flying Dutchman of atmospheric physics..
DM Or perhaps it will become a perennial poster at the annual AGU meeting. A recurring unphysical bad penny.
You mean like the perennial models of Pratt,who keeps moving from statistical artifact (such as the AMO) to statistical artifact (blowing his harmonics) when wind instruments are not possible in Flatland eg (Abbot 1899,)
‘A point which has been increasingly emphasized in M&G’s successive expositions of the biotic pump theory, is the “pressure drop” which occurs on condensation. Since condensation implies disappearance of water molecules from the vapor phase, there remain indeed less molecules which exert pressure. But on the other hand, condensation heats the air parcel and hence causes faster molecular motion and a rise in pressure, which is neglected in the calculations of M&G. Actually, condensation causes not a drop but a rise in local pressure (compared to parcels at the same height but without condensation). This is accompanied with expansion and thinning, contributing to the well-known buoyancy of convective clouds. This mechanism of heating and expansion is a fact which has been very well observed, and it strongly contrasts with the one proposed by M&G in which condensation is regarded as the cause of an ongoing implosion.’
It is not clear how much of the heat that appears in water droplets is lost radiatively or kinetically in heating the surrounding air parcel. Does this heat rise and expand in these types of clouds. It appears so looking at cloud time lapse photogtaphy – but again it is not clear why in an unconfined volume that a temperature increase should cause an increase in pressure and not an incease in volume – indeed as stated in the excerpt quoted.
Contrary to the claims of the blogosphere – it seems again that no fundamental problem has been identified. This is an ‘error’ identified by Pratt and Rabbett below – but not one that holds any condensation.
What happens in condensation is not mysterious or unknown to the least. How pressure and volume react locally is not an open question.
The larger scale conditions vary and consequently the resulting weather phenomena vary, but that’s a problem on the next level. The idea of pressure drop as described in the present or related papers is simply totally wrong. That’s not proven by studying these papers beyond the observations that their argument lacks all valid basis and is based on explicitly erroneous reasoning. That their conclusion is wrong is shown by the standard calculations that take all physical phenomena routinely into account and don’t miss anything in contradiction of the claims of the authors.
That a few people make baseless claims is of zero value, when they cannot justify these claims and when these claims lead to some strong results that are totally erroneous by contradicting very well j´known facts.
It’s always good that people try to find new ideas, kudos on that to the authors, but it’s not good to be stubborn when errors in reasoning have been proven. Yes, they have been proven. That there are people on this blog who cannot follow the proofs does not weaken the proofs.
Nothing can be proven in physics to people who accept only what they understand but who don’t understand any physics.
@DM: Or perhaps it will become a perennial poster at the annual AGU meeting. A recurring unphysical bad penny.
This Monfort character (if that’s his True Name) seems to have taken a distinct disliking to me. This raises the fascinating question of how many beers I’d have to buy him to make him any friendlier.
If his ideology forbids befriending the mathematically competent I’d guess the answer would be enough for him to pass out on the floor, for which his wife would then blame me. So goes modern environmental science.
There was once a theory that rain followed the plow … only it didn’t.
http://en.wikipedia.org/wiki/Rain_follows_the_plow
My theory is that unicorns cause wind. My theory came first. Can I get paid to look for unicorns
Steven Mosher
Fun to know more about that. I guess that this fits your models? I knew there must be something. (Guess it also explains the competitive defensiveness). Now try and publish it.
Steven Mosher
Depends on what you feed them.
Max
Douglas, it is published. Everyone agreed that the theory was not wrong.
Now we are looking for unicorns to prove it. It’s how science works
Steven Mosher
If you are so certain it is wrong please take a little time to get specific as to why (as Nick has). We vould all value that. Your criticisms so far lack any technical substance. They are based on fasle claims or repetition of other people’s comments that were answered already. (I admit I like your unicorns — humour is always welcome).
Mosh’s unicorn example is a technical comment, if you consider epistemology as a technical field, which you should. It underlines the need for a mechanism.
Not only your theory needs coherence, but it needs to posit causal relationships that could make sense in the overall theories covering yours. If we have the choice between maintaining the overall picture we have of the climate and trying to decipher a paper where there is the vague promise to overthrow AGW, the choice won’t be tough to make. Just as it ignores iron suns, it’s the sun stupid, and the like.
In principle, some might try to falsify your claim. But void of specific mechanism, the threat for now.
And all that assuming that the formalities are correct, when you have people telling you they’re barely coherent. So even on the formal front, nothing is won yet, at least as far as scientific acceptance goes.
Perhaps you could ask Dr. Monkton to join his tour? Your presentation would make a nice first part.
Superbowl ate my sentence:
> But void of specific mechanism, the threat for now is quite theorical, unless of course we extend scientific practice to what the GWFP does.
Willard, ” It underlines the need for a mechanism.”
Knowing “the” mechanism is nice. Sometimes though you have to better describe the phenomena before you can determine “the” or combination of mechanisms required to explain all the details causing the phenomena. Their “new” idea is easy to evaluate, does it improve the predictability of cloud formation and precipitation?
I mean, having absolutely no clue how clouds respond to just about anything didn’t hold back Hansen :)
Now wait, she creates;
Microclimate, what you get.
If you come and plant.
==========
As you may recall from Vaughan’s death thread, Cap’n, I don’t mind descriptive work. I only mind PR machines.
But I admit that “it’s the forest, stupid” has a ring to it.
Willard, without PR you are lost in the noise. It is far than “it’s the forest stupid” though, more like its the water stupid, we live on a water world :)
Did you know that for every degree of surface warming the average cloud base height would decrease? I believe that is a negative feedback.
Forests though, which were closed to wiped out near any growing city state back in the day though, would have had some significant “local” impact. But if you look at the dark albedo of those forest from space, you would assume that they have a warming impact. I imagine that if it weren’t for the hydrology cycle they might, but then they would be dead wouldn’t they?
Oh, Coke is targeting kids again on the Superbowl!
When they’ll be unicorns in Superbowl ads, Cap’n, I’ll start to believe in them.
Perhaps Mosh should try to publish in Annals of Cryptozoology. If he does, he could try to get interviews where he’d state:
> If my theory’s correct, this could overthrow all we knew about fantasy worlds, starting with The Hobbit up to the Harry Potter series.
Perhaps Mosh’s black hat marketing skills could help him go where Douglas would never think to venture.
Willard, I think Mosher doing better with his monkey flying out of butts index :)
Actually, this paper has a great mesh with UNtopia.
Your call, Cap’N.
From where I stand, I certainly would like to see the GWPF proclaim that we need more forests.
The proposal fails at the first hurdle – the inability to frame a credible and testable hypothesis.
The unicorn proposal…
Ah. So you want to start making money with your hobby.
Are you lookin’ at me?
I am not the originator if you notice – just rejecting funding. Try to keep up big dave. Or are just stalking me?
There’s a problem with responding in a subthread:
One could be responding to the comment on the top level of it.
One could be responding to the comment just above it.
One could be responding to another comment above it.
One could be responding to just a bit of information in all of the above cases.
Quotes can help clarify all these cases.
Hey, BC, did you know that rain follows the dance? People don’t enjoy dancing in the mud, they prefer to wait until the ground has thoroughly dried out. However there’s a higher probability of rain following a long dry spell than a short one. So after a while people started to notice that rain follows the dance, giving rise to rain dances as a way to encourage rain.
What’s more it worked, as long as you refrained from dancing in the mud.
Tell it to Glastonbury – which reminded me of the Glastonbury Romance – a wonderful modern retelling of the Grail myth by John Cowper-Powys’s. There are seven stations in the Grail vision. The first is a shining fish – and a question asked. The question asked was ‘is it a tench?’ The seventh Grail vision portends floods and cataclysms. I have a bit of a history of vividly reimagining books. I swear I saw the shining fish – and laughed and asked – ‘is it a tench?’
I have evidence that unicorns cause piss and wind – http://s1114.photobucket.com/albums/k538/Chief_Hydrologist/?action=view¤t=unicorn-1.png
VP
Wrong.
Statistically speaking, rain follows rain more often than rain follows dry spells.
Think about it.
Max
@manacker: Statistically speaking, rain follows rain more often than rain follows dry spells. Think about it.
Uh oh. I hope this doesn’t mean my rain check for a dance in Melbourne is now going to bounce, Max. Bouncing rain checks are even worse than bouncing hailstones. :(
Sands of Araby
Do the petroleum dance.
Whirl up a dervish.
============
Interesting
In varirous places I have worked there is a recognition that forests attract rain. We can test that too.
Forests also attract unicorns.
(Have you ever seen one in an open field?)
Max
More points on the unicorn score card!
Max “Have you ever seen one in an open field?”
Afraid I havn’t written the code — Steven seems to know though
Can I hav the next dance. VP? Listen! They’re playin’ our song )
Hey, don’t think I haven’t seen The Crying Game, BC. >:- Slips are showing.
SM so how is yer unicorn theory falsi-fi-able?
I expect he has seen them
Seeing them would be confirmation, douglass. Her question was how to falsify. Falsifiability is a principle that separates metaphysics from science. Claims that are not falsifiable in principle are metaphysics. Unicorns are not metaphysical beings. So, it is falsifiable in principle. Find unicorns and show that they dont cause wind.. If you cant find a unicorn that doesnt mean they are unfindable in principle. If you find one and show that it doesnt cause wind, then you have falsified the theory. So, it is possible in principle to falsify the theory. That seperates it from metaphysics.
For background on the principle of falsifiability start with the logical positivists. That will contextualize it for you, as opposed to the version you might hear at the local pub or from beth
And I had thought you were joking – my mistake
See my reply to Willard above
To save you to find misplaced citations, Mosh, here there are again:
http://en.wikipedia.org/wiki/Falsifiability
http://en.wikipedia.org/wiki/Karl_Raimund_Popper
Do you recall Popper describing scientists at work?
Me neither.
Mosh,
I forgot to ask.
Do you think that Popper would say that to be on record on **The Australian** positing the conditional death of the greenhouse theory as a scientific activity?
What about being shot in front of a giant tree?
Willard.
Yes, don’t ask such embarassing questions about Popper. why study science scientifically? I did read that piece in the australian ( I think) I found that somebody was selling a finding that was “not wrong” for a lot more than it was worth. On a side note, when chatting with ravetz, we did come to a point of agreement.. the death of philosophy was announced prematurely. so, this whole debate is not a total loss.
BTW Im still meaning to get to waltons (?? was that his name ) dissertation.
Sorry guys I’ve lost track of whether you are asking anything here. I suspect not. Just mischief? Ok carry on.
No Douglas we are not asking anything. Question marks would be a clue there. We probably wouldnt ask you anything given your propensity for not answering direction questions. Instead I am pointing out a fallacy in your thinking by constructing an example to show the hole in your thinking. Sometimes instructors ask questions. Sometimes they make points. Pretending that we can only ask questions and then taking note that we dont ask one as an exit strategy is bad faith.
Steven Mosher “No … we are not asking anything ”
Thanks for confirming.
Why ask questions when easy assumptions and indirect accusations are more fun right? Its Ok, I don’t mind. Some are here for fun and why not?
I have answered all questions (as far as I can see). You in contrast have ignored most — and have a reputation for it (I’m new here but see e.g. http://judithcurry.com/2013/01/31/condensation-driven-winds-an-update-new-version/#comment-291069
Quote tallbloke | February 2, 2013 at 7:32 am | Rep:
“be advised that Steven Mosher is a drive-by shooter who never comes back to face criticism of his illogical assertions. He is a waste of your time”
Well I have disproved the last bit: “never comes back to face criticism of his illogical assertions” — should have been “almost never”– evidence and falsification have their values.
Anyway, have fun and take care.
simple beth. Show that unicorns do not cause wind. Sheesh. We have a theory. Unicorns cause wind. After copious review everyone agrees that theory was not wrong. Now we are looking for evidence. For my part I am following on on the fairy line of evidence. Fairies are friends of unicorns. Faires live in the forest. Therefore, unicorns will be found in forests close to fairies. You’ve never seen one in the open have you? And of course it ties in with the rain as well. Because when it rains you see rainbows. rainbows, fairies and unicorns. It all fits. And my parents wondered what I would do with philosophy. Find unicorns!
Steven –
If I’m not mistaken, it isn’t that everyone agrees that the theory was not wrong – it is that no one as yet has proven that the theory is wrong.
On that basis – I am trying to get published my paper speculating that a drop in pirates causes global warming. No one, as yet, has proven that theory wrong.
Joshua. yes thank you for that correction and your attention to details.
Shall I add you as co author or simply acknowledge you in the acknowledgments
Well at least we agree the ideas should be tested to be science. Though you sound a little imprecise on the details. You can chase your unicorns and we can seek constructive engagement with researchers who are open to our ideas (there are inceasingly many). Feel free to keep us posted how you go.
What Mosher and others tend too overlook,is that the fluid equations do not have a fundamental nature ,they are phenomenological equations a l imitating constraint on the prevailing theory of which very little can also be said such as Emanuel’s excursion in Flatland.
A mention in the acknowledgement will suffice, thank you.
This could be an interesting debate – about whether the paper “should” have been published.
At the two extreme ends of the spectrum, we have to equally unsatisfying positions (IMO): (1) that a study that goes against convention has a higher standard and, (2) that a study is valuable because no one has proven it wrong.
Now I think that willard has made a strong case criticizing the way that this paper’s authors take Held’s comment out of context – but even still, the question of whether or not a counter-convention study should face a higher bar is an interesting one, IMO.
The problem with this discussion, as with so many discussions in the climate debate jr. high school cafeteria food fight, is that many involved have twisted the arguments away from a matter of philosophy to a matter of furthering partisan agendas.
In the end, because there is more interest in Jell-o flinging than the philosophical debate, I am left with this: There is no answer as to whether the study “should” have been published. The editors decided to publish it. Everyone will survive. And if they hadn’t so decided, despite hand-wringing and claims or inferences of victimization, the authors would have survived, and been free to do what so many authors of papers do when their studies are rejected for publication: look to publish in another journal – perhaps one with a lower impact factor – or make substantial revisions and try to publish elsewhere, or move on to something else.
What I find amusing here is that some people want to hold hostage, the philosophical debate about tenets of good science, to score points in the climate debate war.
Same ol’, same ol’.
Joshua,
Yes, if you have followed the main author as long as I have you would know that she will never engage willard on his direct question. And then she will say she has answered him or will chnge the topic. Now, the really striking thing is that I know you know nothing about fluids. yet, in watching her avoid willard you do get a sense of how the reviwers of her math felt. BTW, I’m beginning to like your notion of accountability. Although, I think most folks think admitting they are wrong is enough.
( referncing the lynas comments you made at keiths )
Joshua “some people want to hold hostage, the philosophical debate about tenets of good science, to score points in the climate debate war”
It appears you may be right. Do you believe that outsiders publishing in an open review, open access climate journal, and now asking for scrutiny are doing something crazy? Is it possible to win them over with an idea that challanges the status quo?
‘Outsider’ sounds like a rhetorical device, not science.
Douglass –
I think that Michael’s point stands. The notion of “outsider” assumes a sense of victimhood. I get bored with all this victimhood (and focus on rhetoric). It seems to me that some combatants on both sides are more interested in vindicating their sense of victimhood than almost anything else.
That said – let’s move beyond your “rhetorical device.”
No – I don’t think that it is even remotely crazy for anyone to publish in an open review, open access climate journal – and then ask for scrutiny.
in fact, I think it is a fascinating – and at least for me very much undecided – question as to whether such action furthers “science.” I can see merits in the arguments from both sides – and I suspect that those firmly convinced one way or the other are mostly suffering from binary mentality disease.
I have some reason to question motivated reasoning here – as I read somewhere in the websites associated with your group some rhetorical reference to climates scientists wanting to impose restrictions on the global economy (paraphrasing, and open to correction) – which I wonder if you’re willing to address. Please note, questioning the existence of motivated reasoning is not the same as questioning motivation And certainly it is logical to question the potential for motivated reasoning from those who are support “conventional wisdom” – (which in contrast to the potential of your own biases – you seem willing to offer as the operational rationale for the reaction among those who disagree with your findings). And certainly we have issues related to “novelty bias” and confirmation bias etc., etc. as you and willard discussed elsewhere – unfortunately, IMO, much to briefly.
But neither the potential nor the inevitable reality of motivated reasoning renders the philosophic debate irrelevant. The philosophical debate remains, IMO, irrespective of those factors (which are inevitable) and actually irrespective of the mathematical or theoretical validness of the technical aspects of the paper.
The basic problem, as I see it, is that people (in general) are too busy finger-pointing to take note of the ubiquity of biasing influences, and too busy claiming victimhood (and bad faith) to do what is necessary to control for their own biases.
Like I said. Same ol’ same ol.
But what is a bit unfortunate here is that the notion of open review and open access offer potential as vehicles to help to control for the impat of motivated reasoning. Perhaps this Jello-flinging is basically just growing pains – but I suspect not. If I were to consult my Magic 8-Ball, I am guessing it would tell me “Outlook not good.”
Douglas –
Sorry for the extra “s.” Feel free to call me Joshu if you’d like!
Joshua
Thanks – I am genuinely interested in all this. I have worked a fair bit in different cultures and do see myself as an advocate here with only a little time to pasue and look around. I dont see us as victims (well there are moments when I do) but the outsider issue is fairly fundamental to my reading. But as you say we all have our biases and delusions.
Thanks again. I’m going to mull this over.
Douglas –
Let me know if you want to talk after mulling it over. I see you making operational assumptions about your interlocutors (e.g., that which goes along with the identification of being an “outsider” as a description of your relationship to them), and acknowledging the ubiquity (in the abstract) of biases and delusions as an indirect way of not avoiding your own influences — but the rubber meets the road when you dig into the nitty gritty. Your thoughts seem more developed and specific in reference to others. I’d suggest that such a balance is the inverse of what is required to make progress.
FYI – I would be interested in knowing more about your reference to different cultures.
Steven –
You might find it interesting that over at CaS, if you click past the jump in the Lynas post, you will see that Keith updated and crossed out another passage of the Lynas interview.
From Lynas:
At least he apologized this time, and I’d say that this explanation ranks significantly higher on the accountability scale than his explanation for his previous error. In addition to the apology, this time he explains why he made the error. Still, I don’t think that he was sufficiently accountable – the “heat of the moment” component looks like a lame excuse to me. There was no real “heat of the moment” as his very own explanation makes plain. And even if there were a “heat of the moment,” he should simply leave that out of his explanation. Does it somehow “explain” his behavior to say that he allowed his emotions to trump his professionalism?
Still no editorial comment from Keith, however, despite that the topic of the post was Lynas’ credibility.
Joshua – mulling …
Thanks – The “nitty gritty” is perhaps where we need to go then … though I am not yet sure what that is. I am grappling with this (leaving all the hard work here to Anastassia and Victor today). Interested in thoughts. My responses to yours:
This process is interesting in itself as a (?) way to nudge science along in a transparent manner. Lessons? Would we advise others to do this? Why?
“Outsiders” – I think the first person to use this label was Judy (she used it again in the review). While it carries some baggage (as any such term does) she used it for a valid reason. We clearly were not “insiders” and we didn’t know all the conventions. All groupings, however arbitrary, seem to develop a sense of ‘us vs. them’ — this has numerous implications for communication. We are not members of the audience we’re trying to reach here (the main climate science community).
My point about different cultures is not a deep one. Its about recgnising the variaton in norms and rules and how they govern interactions. When I first went to Bangladesh (very young) I grew a beard as I knew that no one would take a beardless man seriously. If I work in remote communities (e.g. in Papua) I am always careful to get properly introduced to the leaders, be respectful, repeat consistently why I am there, win trust (listen, share humour, food etc). Its easy to make mistakes: this can be hilarious or dangerous. Anyway … a blog community has its own culture. Plenty of scope for fun and disaster and for anthropologists.
“indirect way of not avoiding your own influences” — well self delusion is likely. (By chance I do know a little of the large background literature on bias and self-delusion… should make me more alert … at least I acknowledge that self delusion occurs in everyone). Can’t I just be here because I want to know if this theory is correct or not, and this is where it has led? (my claim). Where should we go to find these subconcious “influences”? Not simple.
Just thoughts.
I’ll come back to your motivation issue in a moment (you are direct without making accusations — that’s welcome).
Joshua
“I have some reason to question motivated reasoning here – as I read somewhere in the websites associated with your group some rhetorical reference to climates scientists wanting to impose restrictions on the global economy (paraphrasing, and open to correction) – which I wonder if you’re willing to address.”
Great – thanks. Good question.
Well we are not an ideological group. We simply don’t talk about that.
I can speak for myself: Certainly I am pro-forest and pro-people (I have worked with tropical conservation a long time) but I am also critical of a dogmatic approach to anything. I have often argued in favour of greater engagement with industry etc. or against imposing parks etc. when that seems counter-productive according to circumstances or data.
It would seem an implausible conspiracy: take years to come up with a convincing theory with supporting evidence and then spend a lot of time and effort getting it into a difficult journal and stopping by every now and then to invite the critics to help identify the flaws. Then say it is a distinct issue to greenhouse gas arguments. I admit a few assumptions there … but it’s not a story line I would write for a conspiracy novel. Let me know if you see a plausible narrative that fits.
Say we find gaps or errors in the current theories of how we understand how the climate works. What should we do about that? We can try and share them right? What other options are there? Good science needs critical thinking and debate all the time (for and against the status quo). It doesn’t need dogma.
Douglas –
Douglas – Part II
I believe you don’t talk about that – but how do I reconcile that against what I read about climate scientists seeking to impose policies? Sounds ideological to me.
Are you dogmatic about being anti-dogmatic? I see dogmatism on both sides of the fence in all these issues. I also see flexibility on both sides. I am skeptical of anyone who leans towards identifying dogmatism with ideology – as opposed to seeing that based on what we know about the intrinsic qualities in how we reason, we all need to control for a tendency towards dogmatism.
I don’t believe in conspiracy theories. I don’t suspect a conspiracy here. Knowing that motivated reasoning is a reality does not assume a conspiracy. The plausible narrative is that you and your colleagues are serious about your work – and that like anyone, you are inclined towards confirmation bias, towards tribalism, towards as sense of victimization.
Agree.
IMO – you’re better off leaving that last sentence off. It suggests that you are saying that someone here is arguing in favor of dogma. It becomes a self-fulfilling prophecy, IMO
Joshua
many many thanks for that. I appreciate the feedback and concede the wisdom in your points. (I am definitely dogmatically anti-dogmatic on occassion — that one is going to haunt me.)
Thanks again
(1) The paper is highly controversial, proposing a fundamentally new view that seems to be in contradiction to common textbook knowledge. (2) The majority of reviewers and experts in the field seem to disagree, whereas some colleagues provide support, and the handling editor (and the executive committee) are not convinced that the new view presented in the controversial paper is wrong. (3) The handling editor (and the executive committee) concluded to allow final publication of the manuscript in ACP, in order to facilitate further development of the presented arguments, which may lead to disproof or validation by the scientific community.’
This is from the editors comment at the end of the paper – and cites some support – we can include Tomas in that – and a bunch of people who are withholding judgement. I am in the second group solely because my mind works far too slowly too reach judgement too soon. I build a visualisation of the processes – dip my toes randomly into the math – and eventually form a whole and see if my picture matches the math. It takes a long time. But the objections to the processes I have seen seem wrong, banal, uninformed or blatantly ludicrous.
The comments descend to the trivialities of unicorns and pirates as an epistemological analogy from a plethora of epistemological wankers. Pirates and unicorns – whatever symbology is involved – have about as much place in science as God – one way or the other.
One myth that needs revealing is that this is in the spirit of objective enquiry – that there is an equivalence in confirmation bias between two diametrically opposed tribes. Whatever stories you tell yourselves superficially in the objective idiom of science. Perhaps there is – and it is far from obvious that one side or the other has any absolute claim to scientific truth. Indeed – any claim to absolute scientific truth seems a priori a social construct rather then a scientific one. On one side there is a groupthink dynamic – moral certainty, collective rationalisation, an illusion of unanimity, steorytyped views of outsiders, gatekeepers. The other side seems defined by opposition as conservatives always are.
Neither side is likely to be scientifically correct – at this stage in the evolution of climate science there is only speculation and the ‘jiggle-jiggle-jiggle or the wiggle of the path.’
Joshua what would suffice as accountability.
‘I said something without checking it in the heat of the moment?
Do you want a report on his internal state when he said it?
How do you check that?
His admission to me and his account of why he made it ( heat of the moment) seem about as deep as one can credibly go.
“I did it in the heat of the moment and you know I felt under pressure and under attack and I have a bit of an ego problem that comes from having a bad childhood..” I mean how deep do you want him to go. I suspect you will doubt whatever he says until his report of why he did it agrees with your speculation of why he did it. Maybe he was having a bad day. I can tell you I have said wrong things soley because I was in a rush, got a phone call, and hit the submit button. I suppose a real explanation would not suit you. Finally, you are asking for something that you cannot check.
So, why do you ask for accountability?
mosher –
An admission of responding “in the heat of the moment” seems like a weak rationalization to me. It seems offered as an excuse rather than simply an explanation.
These are relative issues. Certainly, if you hauled off and shot someone who did nothing deserving, because they ticked you off – saying “Well, I responded in the heat of the moment” is not considered being accountable.
Obviously, Lynas’ error is nothing like shooting someone, so the comparison is not a direct one, but I think that while Lynas showed more accountability than he did with the prior error, there is still room for improvement.
He could simply have said something on the order of: I made an accusation without having taken the time to properly investigate the facts. There is no excuse. It was unprofessional. I apologize. I re-double my pledge to not make the same sort of mistake in the future, as I realize that any time I do that, it lowers my credibility. Since my credibility was the very subject of this post, I understand and acknowledge the importance of my errors.
This reminds me of how neocons responded to the argument that it is relevant to look at how our actions w/r/t why we get attacked by terrorists: “Those libz think we should offer Osama bin Laden a blankie and a therapist.”
So saying he was having a bad day shows accountability for a professional journalist to make an accusation that he hadn’t done due diligence in investigating – in an interview where his credibility was the subject? Nope. I can’t go with that.
Sure, he could say something on the order of what I suggested above, and be lying. It is possible. But it is implausible. What could possibly be the motivation for him to lie in such a way? What would be be covering up by lying in that fashion?
The article was about his credibility.
Empiri-call-y?
simulation … much better
what’s the code?
Manaker “whats the code”
I think you are meant to write it yourself if you want to talk about unicorns
see here:
http://en.wikipedia.org/wiki/Two_Dogmas_of_Empiricism
you can find a copy on the web.
The more I look at the details of the paper the more totally it seems to be wrong.
Now I have looked at the chapter 3.3 pressure profiles in moist versus dry air columns.
In that chapter two columns are defined, moist and dry. The problems are in the handling of the moist column that seems to be physically total nonsense.
They start defining a static isothermal column with saturated moisture at the bottom. The static nature is taken literally and results in independent barymetric profiles for dry air and vapor as indicated by the sentences:
While that’s theoretically correct and consistent it’s difficult to see any relevance choosing such a column as starting point as nothing like that can ever occur in real Earth atmosphere at latitudes less than 100 km.
Then they add the moist lapse rate to the column and calculate the vapor profile in the modified column. They conclude properly that the column with moist lapse rate cannot sustain as much water vapor as the isothermal static column. As there’s less vapor in the column with lapse rate the pressure is reduced at the bottom.
The above makes sense for a column in a vertical pipe with walls that keep the amount of dry air fixed in the column. In a atmospheric context more dry air will automatically enter the column when water is removed. The formula (27) that calculates the change in the pressure in column A does not have anything to do with the real atmosphere. Nothing derived from that has any relevance on anything that happens in the atmosphere as far as can see.
It may be that little in the remaining paper beyond equations (28) and (29) depends on the equation (27), but why to include such nonsense in the paper at all?
The Figure 1 is based on the nonsense calculation and so is the discussion related to it in 3.4. Thus it forms part of the further argumentation that is therefore nonsense as well.
Pekka Pirilä
Great that you want to work through this.
Here’s some homework for you if you really want to get to grips with the two different views.
I think you’ll like this: Meesters, et al. Comment on “Biotic pump of atmospheric moisture as driver of the hydrological cycle on land” by A. M. Makarieva and V. G. Gorshkov, Hydrol. Earth Syst. Sci., 11, 1013–1033, 2007, Hydrol. Earth Syst. Sci., 13, 1299–1305, doi:10.5194/hess-13-1299-2009, 2009.
But then see also the replies here Makarieva, A. M. and Gorshkov, V. G.: Reply to A. G. C. A. Meesters et al.’s comment on “Biotic pump of atmospheric moisture as driver of the hydrological cycle on land”, Hydrol. Earth Syst. Sci., 13, 1307–1311, doi:10.5194/hess-13-1307-2009, 2009.
Hope that helps
Douglas,
I stick now with this paper. As long as I find essential weaknesses (and outright faults that are not explained/admitted) in its argumentation I’m less tempted in looking at the biotic pump.
Trying to go further in this paper seems also impossible. In my view discussion of horizontal pressure gradients in the spirit of chapter 4 can be done properly only by starting from the description of a circulating system that’s realistic enough for allowing a comprehensive analysis as an isolated system. It’s essential that the subsystem being considered does not interact with other parts of the atmosphere in a way that may affect essentially the conclusions, but which remains unknown quantitatively.
Already in an early comment in this thread I speculated that it may be impossible to do those analyses without the use of a full model of circulation. (Not necessarily a GCM of the whole atmosphere, but a CM anyway.)
Horizontal gradients cannot be discussed reliably looking only at two columns and postulating how they interact at various altitudes, a much more comprehensive approach is needed. When a system is understood in detail, it’s often possible to present simplifications known to be true based on the more detailed analysis, we can read descriptions of such in any good textbook. Trying to guess the simplified equations without the help of a more comprehensive picture is, however, likely to fail. At minimum it’s impossible to convince others of the validity of the approach.
The problems that I have with those parts of the paper that I believe to be able to judge directly make me more than doubtful on the relevance of those parts that lack the required full context.
The paper discusses pressure changes from removal of vapor as if there would be a change from a state with much more vapor to one with less. The situation is, however, discussed as stationary. Thus the amount of vapor in air does not change at all. it remains constant at every altitude. When air ascends water condensates but new water enters at the surface.
There’s no overall pressure difference from the condensation, there are different vertical profiles for the dry and the moist column. That leads to horizontal pressure differentials. That’s true, whether the change in air stoichiometry is taken into account or not in the calculation of the moist column. Taking the stoichiometry into account has a small influence that’s often left out. Otherwise the standard analysis is fine.
Pekka, yes, see my remark above (search for “refrigeration”). They make a thermodynamic error in setting up the two columns where they neglect the pressure reduction that should go with their cooling. Everything that follows that set-up is a comparison of two independent columns because of this error. There is no reason a pressure gradient between thermodynamically independent columns can tell you anything useful.
@PP (in response to Sect. 3.3, 1st par.): While that’s theoretically correct and consistent it’s difficult to see any relevance choosing such a column as starting point as nothing like that can ever occur in real Earth atmosphere at latitudes less than 100 km.
My reaction exactly. Why not just start with the environmental lapse rate ELR of two columns set to the appropriate MALR and the DALR respectively? (Possible answer: so as to define a specific level of moisture at each altitude that will become supersaturated when the ELR is raised as in the 2nd par. But if so, isn’t that a weird way of getting this effect: why not start with the MALR and just add extra moisture? At the very least some explanation of this seemingly strange choice would surely be in order.)
Sect. 3.3, 2nd par: Now the columns cannot be static: the adiabatic lapse rates are maintained by the adiabatically ascending air.
I don’t see this: what’s preventing them from being static? In the 1st par. ELR = 0 which is way less than the MALR hence the column is absolutely stable. But that’s true for any ELR less or equal to the MALR, whence raising the ELR to the MALR (which in the case of the dry column equals the DALR) as done in the 2nd par. cannot disturb stability (though it can cause condensation in the first column if it makes the air supersaturated). What would cause air to ascend in this setup? Not diffusion since all that does is to very slowly reduce the ELR, increasing stability. I don’t understand this at all.
Sect. 3.3, 3rd par: The change in pressure \del p_s in column A due to water vapor condensation is equal to the difference between the initial weight of water vapor pv(Ts) and the weight of saturated water vapor:
How could that difference be other than zero? Pressure at any altitude is exactly the weight of the molecules above unit area, independent of their phase (whether vapor or liquid), so phase changes can’t cause pressure changes. Furthermore the volume V = NkT/P should stay more or less constant since k and P are constant while NT (product of number N of molecules with temperature T) should also remain roughly constant because although N has decreased very slightly, this is offset by the corresponding slight increase in T. (This is how I would address Eli’s concern that condensation violates the ideal gas law.) Even if NT isn’t exactly constant I don’t see how it could change enough to cause any significant draft, either vertical or horizontal. (Some of this echos Pekka’s concerns.)
Sect. 4.1, 1st par. We have shown that condensation of water vapor produces a drop of air pressure in the lower atmosphere up to an altitude of a few kilometers, Fig. 1c, in a moist saturated hydrostatically adjusted column.
I would like to see Section 3 argued more rigorously before I could accept this “We have shown” claim. It seems to me that it’s much easier to show the opposite conclusion, namely that condensation doesn’t cause any change in pressure, and at most an insignificant change in volume if any.
@PP: As long as I find essential weaknesses (and outright faults that are not explained/admitted) in its argumentation I’m less tempted in looking at the biotic pump.
There are too many errors, as well as too many easily disproved claims, up to Section 3 for there to be any point in pursuing this further.
“Pressure at any altitude is exactly the weight of the molecules above unit area, independent of their phase (whether vapor or liquid), so phase changes can’t cause pressure changes.”
Only if there’s no net vertical acceleration. If the condensed liquid is in freefall, it doesn’t contribute to the surface pressure.
Condensation can’t cause a pressure change?
http://www.youtube.com/watch?v=skhSfFz28g0
If you have a pressure differential of 1″ w.c. it can produce a velocity pressure of 4005 fpm. 150 fpm only needs a vp of 0.01 ” w,c. Small pressures over large areas equal stuff moving.
NiV,
I agree on that, but presenting the issue with an accuracy where that matters we should consider also acceleration at different altitudes in the motion of the ascending air. It’s accelerated at every altitude as its pressure goes down and its density follows at a lesser rate. The condensation of water does also reduce a little this acceleration.
All these details are at a level that can be safely left out for most purposes.
capt.d., a pressure change can cause condensation and that happens naturally all the time. It may happen that condensation causes a pressure change but that requires some cooling by other means, e.g. condensation on cold glass, which we don’t see so much in the free atmosphere, so I am skeptical of this process.
“All these details are at a level that can be safely left out for most purposes.”
That’s the question.
capt.d., I will correct what I said slightly. Fog formation is a process where condensation occurs by radiative cooling, so there are processes that may do this, but these are radiation-produced clouds, not the subject of the paper in any way.
JimD, “capt.d., a pressure change can cause condensation and that happens naturally all the time. It may happen that condensation causes a pressure change but that requires some cooling by other means, e.g. condensation on cold glass, which we don’t see so much in the free atmosphere, so I am skeptical of this process.”
You really shouldn’t be skeptical of the process, just skeptical of the magnitude of the impact of the process. On a summer afternoon a “shower” can dump three inches of rain per hour. Once condensation starts in a super saturated cloud you have all the dynamics involved, but it takes water vapor to get the ball rolling. That is the biota effect, green space, local wet land restoration, reduction in non-permeable surface the whole smire.
Their size can be easily estimated with some accuracy, and I’m sure that has been done very many times by people studying atmospheric physics. I have some feeling of the answers but not at a level that I would be ready to make public even here.
The influence of condensation on the volume and density of moist air is also understood by every scientist working on these issues. Most may have given little thought on them but it’s not credible that there were not also many who have had a bit closer look on them.
capt.d., the rain is formed by condensation in ascending air. The pressure drops significantly just from its vertical motion, and maybe the vapor loss adds a tiny bit to that.
Pekka, “The influence of condensation on the volume and density of moist air is also understood by every scientist working on these issues. Most may have given little thought on them but it’s not credible that there were not also many who have had a bit closer look on them.”
That would depend on who you ask. Weather models get into more detail and have adjustments to virtual temperature etc. to make things work. Climate models assume constant condensation and neutral sensible heat exchange. You might note that sensible and latent were the two biggies that Stephens “corrected” in his energy budget plus the “window” flux which interacts with cloud moisture.
JimD, “capt.d., the rain is formed by condensation in ascending air. The pressure drops significantly just from its vertical motion, and maybe the vapor loss adds a tiny bit to that.”
that’s why the hurricane intensity models are so accurate eh :)
Concerning climate science we must take into account the level of detail they have. What can one do with the size of grid cells used in GCM’s? Certainly not much of what has been discussed in this thread in a explicit way. Whether the parametrizations include them implicitly cannot really be told.
@NiV: Only if there’s no net vertical acceleration. If the condensed liquid is in freefall, it doesn’t contribute to the surface pressure.
Indeed. That’s a lot clearer than what’s in the paper.
Not that precipitation spends much time accelerating, let alone at g (free fall). The more relevant situation is precipitation falling at terminal velocity, since that causes a downdraft countered by an updraft surrounding it. Such vertical winds can be extremely strong.
@PP: All these details are at a level that can be safely left out for most purposes.
@NiV: That’s the question.
Easily answered. The pressure lost due to accelerating rain is recovered when the rain is decelerated by hitting the ground. If m kg/sec of anything traveling at velocity v collides with an area A with no rebound the resulting pressure is mv/A. (Double that for perfectly elastic rebound.) For rainfall of one inch/hour (fairly strong), m/A = 2.54*10/3600 = 0.007 kg/sec/m2. Rain falls at a terminal velocity of 5 m/s more or less depending on drop size whence the pressure is m/A * v = 0.007*5 = 0.035 Pa or 350 parts per billion (ppb) of atmospheric pressure.
But the rain itself is the wrong thing to look at. At a pressure where air has a density of 1 kg/m3 (a few hundred feet above sea level, at sea level it’s around 1.2 kg/m3), air of velocity v striking a flat surface develops a pressure of v^2 Pa. This is because m/A in the above no-rebound formula mv/A is itself v kg/sec/m2. This is way more than the above m/A of 0.007 kg/sec/m2 for rainfall of one inch per hour, so a downdraft has orders of magnitude more influence on pressure than the acceleration and deceleration of falling rain.
But the terminal velocity of rain is measured only with respect to the air. If in falling the rain drags the air down with it, the rain might be falling 5 m/s faster than the air but the air itself can be driven downwards at speeds up to 30 m/s, see e.g. page 2 of http://kiwi.atmos.colostate.edu/rr/groupPIX/daniel/thesis.pdf . A downdraft of that speed colliding with a flat surface (e.g. the wing of a plane) will therefore develop a pressure of 30^2 = 900 Pa or about 1% of atmospheric pressure. This makes the above 350 ppb for rain hitting the ground after the downdraft has dissipated higher up look pretty puny!
For those who prefer experiment to theory there’s a straightforward way to actually measure the pressure of rain striking the ground. Put out a scale in the rain with a fine mesh held over it to break the fall of the rain without however preventing the rain from landing on the scale (by allowing it to leak through the mesh). When equilibrium is reached zero the scale. Now remove the mesh and note the weight in grams resulting from the rain striking the scale directly. Divide by 100 to convert to newtons. This is the force needed to break the fall of the rain. Divide by the area of the scale surface to give pressure.
For a platform of area 0.1 m2 (about a square foot) expect less than a gram of force, so either use a very sensitive scale or increase the catchment area to more like a square meter. Rain develops very little pressure on the ground and is therefore very tricky to measure.
@captdallas: If you have a pressure differential of 1″ w.c. it can produce a velocity pressure of 4005 fpm. 150 fpm only needs a vp of 0.01 ” w,c. Small pressures over large areas equal stuff moving.
cd, I agree exactly with your first sentence when the coefficient of drag (if that makes sense in an HVAC context) is exactly 1. In metric that would be 250 Pa at a velocity pressure of 20 m/s (assuming air at STP with density 1.25 kg/m3). (250 is exactly 1.25 * 1/2 v^2 where v = 20; the formula v^2 I was using is for a flat plate, which has a coefficient of drag of 2.)
However when you reduce the velocity pressure by a factor of 4005/150 = 27, shouldn’t the corresponding pressure be reduced by a factor of 27^2 or around 700, which would call for a vp of 1/700″ = 0.0014″ rather than 0.01″? If your 0.01″ figure is correct then I’m surprised that conventional physics gives exactly your result in the first case (assuming Cd = 1) but nowhere near it in the second.
I can also see where falling rain (which of course results from condensation) can easily cause air to move at the former speed, 20 m/s being reasonable for a strong downdraft.
What I’m completely failing to see is any connection between the associated pressure differentials that such velocities can induce when colliding with e.g. stationary air or the wing of a plane, and the mechanism of condensation itself.
Regarding the collapsing can video, how is this adiabatic? If you had no water at all but simply heated the air in the can (in the manner of a hot air balloon but with the air very much hotter) and then plunged the inverted can into the cold water, wouldn’t the decreasing pressure in the can as the water cools the air inside cause the can to collapse anyway?
As I understand the article it is claiming that adiabatic condensation decreases pressure in the atmosphere, which is the bit I’m having trouble with. The experiment in the video is nowhere near adiabatic. I’m fine with atmospheric pressure changes resulting from non-adiabatic processes, or from wind etc., but not from condensation alone.
@VP: which would call for a vp of 1/700″ = 0.0014″ rather than 0.01″?
Oops, I was blindly copying your “vp of 0.01” when you clearly meant “pressure differential of 0.01″. Likewise I should have written pressure differential of 1/700”.
capt.d., to go back to your first question. Condensation in ascending air does cause a pressure change which is net positive. This is only briefly before the air expands due to latent heating to equalize the pressure (Pielke commented on these fast adjustment processes due to sound waves). The heating has several times more effect on the pressure and final density than the loss of vapor. This is another flaw in the thinking that condensation leads to a pressure reduction, because the latent heating effect is completely missed. With zero latent heating, yes, the pressure would reduce, the density would go up as air fills in the deficit, and clouds would sink. Not very realistic.
JimD, it is like most non-linear dynamic systems, there is a “sweet spot” or bifurcation point where different variables have maximum impact. If you consider a hurricane, it can grow too large and the squall bands choke the inflow of moist air. When a storm system moves across land, the rate of moist air inflow decreases with the available moisture. The paper misses the critical dimension by assuming there is no limit to X, when the limit of X is the point.
They get into ratios of height to width later when they should mention that the limit of X on the impact is critical. Where is the “sweet spot”?
As I said before this is another case of what surface?
http://redneckphysics.blogspot.com/2013/02/which-surface-is-surface-strikes-again.html
The idea seems intuitively correct for circulations such as Hadley Cells or for the rain pumps known as rain forest. That evaporation and condensation is a dynamic process – which is pretty much as described in the paper – goes without saying and makes no difference at all.
Robert,
Makes all the difference and solves a meteorological paradox. Take the Amazon, to cite your rainforest and an example that served as insight to the development of the Biotic Pump Theory. If you look the last ten years of surface temperature both on the hotter Atlantic (http://earthobservatory.nasa.gov/GlobalMaps/view.php?d1=MYD28M), where the trades gather steam (literally), and on cooler land (http://earthobservatory.nasa.gov/GlobalMaps/view.php?d1=MOD11C1_M_LSTDA), where forest canopy is transpiring at great rates, you wonder how is that possible that winds blow from a region with warmer surface into another region with colder surface – for thousands of km inland, year round? You could think of winds accelerated by Earth spinning etc. But the tropical Atlantic trades gather on the northern hemisphere, blow South, cross unceremoniously the Equator, enter South America and travel down to latitude 30 S in the austral summer.
Our theory solves the paradox as it clearly puts and physically explains the low pressure over the Amazon (higher evaporation and condensation due to the forest) that “pulls” the tropical Atlantic trades over land, down deep into the southern hemisphere. In such a scenario, surface temperature per se is not an issue. If it works intuitively for you, and if it works physically for the Amazon as our theory demonstrates, how could it not work elsewhere?
Thank you for the explanation, Antonio.
You should become the spokeperson.
The only question is how it was missed before? Very likely because consensus science is too dogmatic and not really interested in science, but in protecting and maintaining the paradigm.
@AN: But the tropical Atlantic trades gather on the northern hemisphere, blow South, cross unceremoniously the Equator, enter South America and travel down to latitude 30 S in the austral summer.
Source, please. (Neither of the two links in your comment seem to imply any continuity in wind flow from the Northern Hemisphere to 30 S.)
Vaughan,
See climatological wind fields over the ocean around South America, (Climatology of Global Ocean Winds). The seasonal fluctuation of the Inter Tropical Convergence Zone (determined by Earth axis tilt) plays a role in determining how much air comes in, at different seasons, from the northern hemisphere into the Amazon. But the proof that air from northern hemisphere does cross the Equator is the long-known landfall of Saharan dust on the Amazon (e.g. African dust keeps Amazon blooming). The sucking-in of trade winds at the mouth [roughly between longitude 45o and 54o] of the Amazon’s south-western-bound “aerial-river” is evident, year round. Tropical Atlantic SST is always warmer than the amazonian LST.
@AN: See climatological wind fields over the ocean around South America, (Climatology of Global Ocean Winds).
Ah, thank you. Looking along the equator I see the arrows pointing inwards (hence south) towards the Amazon, just off-shore. According to the paper’s theory this direction is the result of condensation-induced pressure reduction over the tropical Amazonian rain-forest, right?
How do you show that the direction is not simply that of off-shore sea breezes? If you look along the equator to the east the arrows orient themselves upwards to point to the coast of Africa, even more strongly. Why can’t these be accounted for by weak sea-breezes into the Amazon (which having more forest is cooler than Africa) and strong sea-breezes into Africa, which being drier is hotter than the Amazon?
Also the map only shows the tropics. Where is the evidence that wind blows from the Northern Hemisphere to 30 S in the Amazon?
But the proof that air from northern hemisphere does cross the Equator is the long-known landfall of Saharan dust on the Amazon (e.g. African dust keeps Amazon blooming).
But the whole Atlantic is blowing east to west year, and the portion from the Sahara seems to be blowing more into Central America than South America. I can see that a little bit of Saharan dust would be helping the Amazon but then one would expect even more to be dropped on Central America (which might not notice however since its need for nutrients is not as critical as the Amazon’s).
The sucking-in of trade winds at the mouth [roughly between longitude 45o and 54o] of the Amazon’s south-western-bound “aerial-river” is evident, year round.
Again, how do you distinguish “sucking in” from a generally westward wind blowing across the whole Atlantic (whether pushed or pulled), slightly shifted south in the manner of ordinary sea breezes, namely by rising warmer air over the land than the air over the Atlantic.
Vaughan,
Sorry not responding promptly to your valuable questions. I’ve just fathered a baby-girl few days ago, domestic life is quite busy, I’ve had not much time to follow this thread. I guess I missed adding here the climatological wind fields for South America, which should quench your justified thirst for data. While I search and find a suitable link to illustrate my statements, believe me, ocean trades that turn in towards the Amazon continue inland, in a SW direction, for more than three thousand km til they reach the Andes barrier and split, deflecting NW (minor branch) and SW (larger branch that keep going till Argentina during austral summer). If you look, these low altitude flows counter the usual direction of the Hadley circulation at those levels over adjacent oceans, specially during the austral summer, but not only. Sorry again for missing the links, which I hope I will be able to post soon.
Correction: “ocean trades that turn in towards the Amazon continue inland, in a SW direction, for more than three thousand km til they reach the Andes barrier and split, deflecting NW (minor branch) and SE (larger branch that keep going till Argentina during austral summer).”
=> South East is the outgoing flow direction from the Amazon into meridional South America.
Antonio Nobre,
Two questions:
– What’s the evidence that the best present atmospheric models (e.g. the models of the European Centre for Medium-Range Weather Forecasts) fail in describing the Amazonian area?
– What’s the interdependence of Biotic-pump theories and the paper discussed in this thread?
Thank you Antonio. Intuition is little vague – it is more visual reasoning that I was referring to. A proto language and proto math of consciousness. Am I being mystical? After visualisation comes rigouress math or language precision.
“The words or the language, as they are written or spoken, do not seem to play any role in my mechanism of thought. The psychical entities which seem to serve as elements in thought are certain signs and more or less clear images which can be ‘voluntarily’ reproduced and combined. …. This combinatory play seems to be the essential feature in productive thought before there is any connection with logical construction in words or other kinds of signs which can be communicated to others”. Albert Einstein in a letter to Jacques Hadamard.
Now I suppose I will be accused of comparing myself to Einstein. Last week it was Walt Whitman. The Climate War has unanticipated battlefields. Good luck.
> The words or the language, as they are written or spoken, do not seem to play any role in my mechanism of thought
Einstein was very likely wrong:
http://www.ted.com/talks/steven_pinker_on_language_and_thought.html
Einstein was likely wrong? A very big call indeed.
It is called visual thinking – benzene, DNA and relativity are a few successes that come to mind. http://www.sv.vt.edu/classes/ESM4714/methods/vizthink.html
Feynman continues: “What I am really trying to do is bring birth to clarity, which is really a half-assedly thought-out-pictorial semi-vision thing. I would see the jiggle-jiggle-jiggle or the wiggle of the path. Even now when I talk about the influence functional, I see the coupling and I take this turn – like as if there was a big bag of stuff – and try to collect it in away and to push it. It’s all visual. It’s hard to explain.”
Schweber: “In some ways you see the answer – ?”
Feynman: “The character of the answer, absolutely. An inspired method of picturing, I guess. Ordinarily I try to get the pictures clearer, but in the end the mathematics can take over and be more efficient in communicating the idea of the picture. In certain particular problems, that I have done, it was necessary to continue the development of the picture as the method before the mathematics could be really done.”
I have experienced relativity as a flower opening in my mind over decades.
How do you think visual thinking works, Chief?
See item #6 of the second link:
http://plato.stanford.edu/entries/mental-representation/
http://plato.stanford.edu/entries/perception-contents/
The experience Einstein is talking about seems to correspond to what Mihaly Csikszentmihalyi calls “flow”:
http://www.youtube.com/watch?v=fXIeFJCqsPs
McWilliams might not help you here, but you certainly can copy-paste your favorite bit anyway.
The Stanford links are not even close.
Try wikipedia –
Visual thinking, also called visual/spatial learning, picture thinking, or right brained learning, is the phenomenon of thinking through visual processing.[1] Visual thinking uses the part of the brain that is emotional and creative, to organize information in an intuitive and simultaneous way.[citation needed]
Visual thinking is one of a number of forms of non-verbal thought, such as kinesthetic, musical and mathematical thinking.[citation needed]
Visual thinking may have a co-morbidity with dyslexia and autism.[citation needed]
Although the Feynman quote is more informative about the actual processes. But I note that ‘flow’ in music is one of the forms of non-verbal thought.
James McWilliams – eminent climate scientist that he is – is entirely in the realm of dense (in terms of meaning) and precise language. It took me perhaps a month to gain confidence that I understood even a fraction. Knowledge is the offspring of curiosity and diligent study. I suggest you try it some time.
Einstein confused subjective time for objective time, his relativity is nonsense.
Relativity – seriously thinking that the chef rushing around a kitchen preparing a meal is going to take longer to get to tomorrow than the customer sitting still at the table waiting for her dinner…
Chief,
Visual thinking is still cognition with semantical properties, not unlike playing chess, composing music, proving theorems in category theory or even kung fu fighting (think of the animal forms).
The brain acquired its perceptual system way before the verbal one, and it is quite clear that language has a perceptual source, as the Pinker presentation underlined. (Pinker also emphasized the interactional source, which satisfy ethological functions like dominance.) That makes me surmise that cognition is mostly situated, and that the traditional computational model of the mind is wrong. You can look for situated cognition to know what I mean by that.
***
What Einstein says rings true to me, and I should have said so. What I took objection to was the bit where he seems to say that what he experiences is beyond linguistic manifestations. According to he contemporary theories of mind I prefer, his experience could form a continuum with verbal languages.
In another life, I did some research on diagrammatic reasoning. I am a chess player, and a visual guy. I have lots of experience with autists.
My own experience is that most people tend to grasp faster concepts when expressed via multiple modalities. Think of encyclopedic dictionaries: you have images of concepts or objects, and the word expressing them below it. Logic would be easier to understand if logicians understood that formulae become non-sense quite fast:
http://www.youtube.com/watch?v=JiTz2i4VHFw
Some people can do without the words. Some other can do without the images: Pierre Duhem, for instance, was against the use of models in science. We should not infer from that that the images stop to carry meanings.
I was merely describing how my thought processes work for these sorts of physical systems. Although I am not claiming any great correspondence – both Einstein and Feynman resonate as descriptions of the process.
But the distinction is between visual thinking and communication – vision and language. Different parts of the brain.
And Myrh – there is an imperceptible time dilation in everyday activities. It is measureable. That’s all that counts. Although Einstein did say that some things that can’t be counted – count as well.
Nullius in Verba | February 2, 2013 at 1:37 pm |
NiV (is it fine if I call you this way?), thank you for your comments. I am very interested in this topic so if you have any other thoughts they’d be most appreciated. From what you said, consider that the falling droplet creates a surplus of pressure beneath itself. This local pressure anomaly sends waves in all directions, not only down to the surface but also in the horizontal directions. So where these waves actually contact the surface (what is the radius of that circle) is a big question. Any ideas?
Regarding Spengler et al. (2011) here is the link: Spengler, T., Egger, J., and Garner, S. T.: How does rain affect surface pressure in a one dimensional framework? J. Atm. Sci., 68, 347–360, 2011. They present graphs showing the time response of surface pressure following condensation. It can be seen that immediately upon condensation aloft (i.e. with a time delay set by the height and the speed of sound) the surface pressure drops and never returns to its value (while the droplets are falling down).
(Note that their latent heat idea is not correct, in our view, as we clarified in our comment. Basically what they studied is adiabatic condensation at constant volume which we show in our paper is not possible. They release latent heat as sensible. But they also have a set-up without heating.)
In all cases of interest the number of droplets is very large and they form and grow continuously. Therefore it seems clear that all the effects of pressure waves average out effectively and what’s left is the increase in pressure that results from drag forces on droplets at altitudes above the point of observation. The calculation can also be done averaging in horizontally.
Anything more complex seems totally unnecessary.
Anastassia,
I’m happy to be called NiV. It’s a lot shorter!
Yes, the pressure wave is approximately spherical, expanding at the speed of sound. The radius of the circle will depend on how high the droplet was when it started falling and how long ago that was.
I had a quick look at the link, but unfortunately it’s paywalled, so I can’t read it. The phrasing of the abstract is ambiguous.
They say:
“It is shown that the rain formation leads to a change of the surface pressure after a short period of acoustic wave activity. There is, however, no hydrostatic surface effect once the particles reach terminal velocity. It is not until the rain reaches the ground that the surface pressure decreases consistently with the mass removed by the phase change.”
It’s not clear whether “no hydrostatic surface effect” means no further change (after the pressure reduces), or no reduced pressure (i.e. it is as it was before the drop started to fall).
In a 1D setting, I would expect the pressure to drop sharply at the instant of condensation, then rise slowly back to where it was as the droplet approaches terminal velocity, and then at terminal velocity it would remain unchanged at its initial value (i.e. no pressure drop) until it hit the ground. Then with the drag loading taken off, the air would relax and the pressure drop to the lower value corresponding to the weight of the air with the droplet removed.
This fits the abstract’s phrasing, but I can see how the first and second sentences quoted above could be interpreted as a permanent drop. But what about the third sentence? Doesn’t that imply the pressure doesn’t drop until the rain reaches the ground?
The 1D setting doesn’t tell the whole story, though. In 3D, the droplet falling at terminal velocity is continuing to radiate pressure changes, although these do not change the total pressure on the surface below. They change the distribution of forces in the air so as to continue to support the droplet at its moving location. The pattern of pressure changes on the ground will be more complicated, although it will still have to integrate up to the weight of the raindrop.
You can get an idea of the shape of this pressure field by considering two expanding spheres starting at the place/time the droplet condensed, and the droplet reached terminal velocity. They both expand at the speed of sound Outside the outer sphere, you get the initial background state. Inside the inner sphere you get the usual axisymmetric Stokes flow, of fluid flowing round a sphere, shifting downward at terminal velocity. Pressure change proportional to Cos(theta)/r^2 where theta is the angle from the downward vertical.) And in between you get a transition between the two that depends on the details of the raindrop’s acceleration to terminal velocity. When the pressure wave first hits the ground it reflects, and you have to add in the reflected wave as if from a mirror image of the raindrop rising below the ground.
I probably ought to add that the above discussion makes various simplifying assumptions, like that the Reynolds number is small, which requires small droplets. Check the usual mathematical conditions on Stokes flow to be safe.
NiV, by “There is, however, no hydrostatic surface effect once the particles reach terminal velocity.” it is meant (at least I so understand) that pressure at the surface does not equal the initial column weight. (So it is in a way in disagreement with your statement.) In their Fig. 8 surface pressure when rainfall is present after an initial ~1sec jerky behaviour, is reduced to a value approximately equal to that of the final state (when droplets have fallen down). I.e. surface pressure drops immediately and never returns to the initial value despite it takes quite some time for the droplets to fall out (at 5 m/sec from a few km height).
“The pattern of pressure changes on the ground will be more complicated, although it will still have to integrate up to the weight of the raindrop.”
Concerning the waves from the droplet, given that the droplet moves slowly compared to the speed of sound, the weight of the droplet (i.e. its dynamic presence) should be spread over a fairly large area.
When the droplets fall at constant velocity they contribute by their whole weight to the mass of the atmosphere and pressure at lower altitudes. Only, when they are accelerating is the contribution less. Anything else contradicts the Newtons’s law.
@PP: Only, when they are accelerating is the contribution less. Anything else contradicts the Newtons’s law.
Exactly so. This is the basis for the reasoning in my comment above.
‘winds are driven by pressure gradients’ (comment by Anastassia Makarieva 1 Feb, 12:29am) or ‘kinetic energy generated by horizontal pressure gradients’ (main post) – I’m not so sure of that. Are not the winds ultimately driven by temperature gradients (horizontal and vertical) and the rotation of the earth? It is the movement of the air (the winds) that produces horizontal pressure gradients.
For example, the movement of the air produces low pressure systems. Not the other way round, despite the common habit in for example weather forecasts of claiming that tightly packed isobars are the cause of the strong winds with which they are associated.
But water in all its phases, and of course the transitions amongst them, is clearly very important in the climate system, and so as an interested bystander I am very glad to see this paper as evidence, I hope, of lively research in this area. The amount of attention that has been paid to carbon dioxide in recent decades (especially in politics!) seems to me out of all proportion to its importance in the system compared to, for example and perhaps especially, water.
John Shade | February 3, 2013 at 7:51 am | ‘winds are driven by pressure gradients’ (comment by Anastassia Makarieva 1 Feb, 12:29am) or ‘kinetic energy generated by horizontal pressure gradients’ (main post) – I’m not so sure of that. Are not the winds ultimately driven by temperature gradients (horizontal and vertical) and the rotation of the earth? It is the movement of the air (the winds) that produces horizontal pressure gradients.
For example, the movement of the air produces low pressure systems. Not the other way round, despite the common habit in for example weather forecasts of claiming that tightly packed isobars are the cause of the strong winds with which they are associated.
Thank you for pointing that out. I’ve now had a look at a few more explanations and it appears that winds created by temperature differentials are marginalised and one of its effects has been made the driver, pressure differences. What I had previously taken as being a poor explanation seems to be the now the usual telling.
This and Stephen Wilde’s comment here leads me to suspect that this is an outcome of the disappearance of real gases from the consciousness of climate scientists. http://judithcurry.com/2013/01/31/condensation-driven-winds-an-update-new-version/#comment-290821
These explanations may mention hot air rises cold air sinks, but why it does so isn’t forthoming – http://home.comcast.net/~rhaberlin/pwpptnts.htm
http://www.education.noaa.gov/Weather_and_Atmosphere/Weather_Systems_and_Patterns.html
These don’t even mention change in density, which the MET does on this page, http://www.metoffice.gov.uk/learning/wind/what-causes-wind
Though quite what it means by the following is anybody’s guess: “The greater the mass of air above us, the higher the pressure we feel, and vice-versa. The importance of this is that air at the surface will want to move from high to low pressure to equalise the difference, which is what we know as wind.”
What puzzled me, Anastassia, is Section 3.3 of your paper, with the two columns. It says:
“Water vapor in column A is saturated at the surface (i.e., at z = 0) but non-saturated above it (at z > 0). The saturated partial pressure of water vapor at the surface pv(Ts) (4) is determined by surface temperature and, as it is in hydrostatic equilibrium, equals the weight of water vapor in the static column.”
As far as I know, this last sentence cannot be true: it would mean that the weight of water vapor in the static column would be fixed by surface temperature, and vice versa. I was taught that the partial pressure of water vapor is just (by a good approximation) the (local) molar fraction times the total pressure.
Then (24), (26) for a saturated column are ok as far as I can judge. But the problem with the interpretation of pv(Ts) above comes back in (27): pv(Ts) is not the original weight of water vapor in the static column. In (27), it should be replaced by the integral of g times the initial water vapor density, which was not specified except that it was unsaturated except at the surface.
This is where I got stuck. I suppose it is an important issue for the suggested mechanism, i.e. the explanation of figure 1(c).
Thank you for your question. What you were taught is correct for a usual atmospheric column with a non-zero lapse rate. In such a column molar fraction of water vapor decreases with height. In our thought experiment both columns are initially vertically isothermal (see p. 1042). I.e. temperature does not decrease with height. Then water vapor follows the hydrostatic distribution and its total amount in the column is determined by surface temperature.
Sorry, please ignore comment below, that was stupid.
” In our thought experiment both columns are initially vertically isothermal (see p. 1042). I.e. temperature does not decrease with height. Then water vapor follows the hydrostatic distribution and its total amount in the column is determined by surface temperature.”
It won’t retain that distribution for long. An atmosphere, by definition, is unstable whenever the environmental lapse rate is below the saturated lapse rate. Zero lapse rate is far below saturated (wet) lapse rate. There will be a buttload of vertical motion in that atmosphere until it becomes at least neutral which, again by definition, is an environmental lapse rate somewhere between wet and dry lapse rates.
David,
Atmosphere is unstable, when the lapse rate is more than the adiabatic lapse rate, not when it’s less.
Cees de Valk,
A totally isolated column of gas under gravitation in equilibrium is isothermal and each type of molecules has its own exponential density profile. The coefficient of the exponential is proportional to the molecular weight. Thus the molar fraction of water vapor would increase with altitude.
In my view it does not, however, make any sense to consider at all an isothermal column to start with, and even less sense to consider a column with such molecular density profiles.
Pekka, I do not understand your use of the word “sense” here. If the thought experiment can be performed then it makes sense. Physics is full of such experiments.
It does not make sense to calculate the difference between the equilibrium that can never be even approached and a realistic stationary state. They find a large difference between these two profiles and imply that such large values would be significant for the physics of the atmosphere. That’s, however, totally false. The equilibrium state is totally irrelevant for real physics and calculating deviations from that is pure nonsense.
It makes just as much sense as the zero feedback equillibrium sensitivity to CO2 doubling, which some people think is very important.
David,
You are on very weak ground. You just throw claims that you don’t seem to understand at all.
In such a column, water vapor would be oversaturated if it were saturated at the surface, or not?
Ah, no, of course not, it is supposed to be isothermal
Cees, “Ah, no, of course not, it is supposed to be isothermal”
That isothermal column is I believe a thought experiment. Since there is a lapse rate, the actual column of air could not be isothermal. An isothermal layer in that column can form and expand upward and outward in a ratio limited by available water vapor. The horizontal expansion would have to be greater than the vertical expansion if the layer were stable.
It’s not only a thought experiment. They calculate deviations from that and draw strong conclusions from those deviations. That’s, what’s so crazy.
Pekka, “It’s not only a thought experiment. They calculate deviations from that and draw strong conclusions from those deviations. That’s, what’s so crazy.”
Hey, I am fluent in crazy. Using the thought experiment just provides a limit of the potential energy. The condensation “lens?” can only expand so far vertically, so the “spread” in the horizontal would be necessarily much larger with the total limited by the ideal potential energy.
How they worded that, if that is what they were doing, sucks, but I can see using it the way they did.
Pekka Pirilä | February 3, 2013 at 8:57 am | Reply
“A totally isolated column of gas under gravitation in equilibrium is isothermal”
That’s. Internal energy of a mole of gas at any give altitude is identical in a non-convecting atmosphere. Temperature is not identical measured with a thermometer as a thermometer measures kinetic energy while internal energy is the sum of kinetic and gravitational potential energy.
You should know better. A isothermal atmosphere in equilibrium is a physical impossibility. A contradiction in terms.
captdallas2 0.8 +/- 0.2 | February 3, 2013 at 10:20 am |
“That isothermal column is I believe a thought experiment.”
Perhaps in the same manner that a square circle is a thought experiment.
An isothermal atmosphere at equilibrium is a contradiction in terms. The fully relaxed state is maximum entropy and that is acheived when internal energy, not temperature, is equal everywhere.
David, I was thinking more in line with the explanation given below. It was added in response to a reviewer comment.
http://judithcurry.com/2013/01/31/condensation-driven-winds-an-update-new-version/#comment-291515
It is a refreshing paper though.
I suggest Pekka do a little more due diligence into atmospheric physics including different forms of atmospheric energy and first principles governing distribution of same.
Physics of the Atmosphere and Climate
By Murry L. Salby
http://books.google.com/books?id=CeMdwj7J48QC&pg=PA474&lpg=PA473&ots=eojLKzf4jG&dq=atmosphere+column+internal+energy
Go forward and backward from the page above. It begins with the formula to integrate upward from the surface for column internal energy. In an equilibrium column internal energy of a mole of gas is the same everywhere. The volume which contains the mole is what varies and the temperature varies by the change in volume according to (among other ways of obtaining it) the ideal gas law.
Until you wrap your head around the concept that thermometers don’t measure internal energy you will be a source of confusion to both yourself and others.
David,
You are wrong. Gravitational energy does not enter in the temperature in a convective atmosphere.
The average gravitational energy is temperature dependent in an isothermal atmosphere of thermal equilibrium. In other words the average altitude of the molecules depends on the temperature of an isothermal atmosphere. That’s how the equipartition theorem applies to gravitational energy.
@DW: It makes just as much sense as the zero feedback equillibrium sensitivity to CO2 doubling, which some people think is very important.
Touché.
At the 1930 Solvay meeting Einstein proposed a thought experiment purporting to defeat Heisenberg uncertainty by opening the shutter of a box arbitrarily briefly to allow a photon to escape at an accurately known time then measuring the resulting decrease m in weight of the whole box to arbitrary accuracy so as to estimate the photon’s energy E accurately as mc^2.
For a photon of visible light m would be about 0.000005 times the mass of an electron, a minor experimental detail that apparently didn’t bother Einstein.
Actually the tropopause is by definition vertically isothermal (if only over a kilometer or so) so it’s not purely theoretical. And it should be possible to measure zero feedback equilibrium climate sensitivity on a small scale if not on a planetary one.
I think Einstein’s box thought experiment presents the biggest challenge.
Pratt, an isothermal column is extremely unstable with a buttload of vertical motion.
Pirila, I pointed you to the appropriate chapter at Murray Salby’s book on atmospheric physics. I can lead a horse to water but cannot make him drink. Gravitational potential energy is a component of internal energy. Internal energy is used in many formula that deal with the atmosphere. This is explained in the atmospherics physics chapter I pointed out to you. If you think it not true then I suggest you go argue with Salby not me.
David Springer,
I don’t have Salby’s book but I know that part of physics. You have misunderstood something.
@David X (where X on this blog is never “Mermin”, sadly): An isothermal atmosphere at equilibrium is a contradiction in terms. The fully relaxed state is maximum entropy and that is acheived when internal energy, not temperature, is equal everywhere.
Can you (emphasis on you) prove that the presence of any field (gravitational, electrostatic, or whatever you like) can create a situation where internal energy is equal everywhere even though the temperature is not?
If you can’t do it yourself then you’ve probably just misunderstood something you read somewhere.
Given a completely closed vertical column of air in a gravitational field, closed in the sense that no energy enters or leaves it, the equilibrium state towards which it will drift (very slowly) is an isothermal one. The drifting is slow because diffusion is the only mechanism by which temperature can change.
This would not be the case if the condition ELR < ALR resulted in unstable air as you seem to believe. However you have that condition backwards: ELR < ALR is absolutely stable. This includes the isothermal case ELR = 0. There can therefore be no convection, leaving diffusion as the only remaining transport mechanism.
Pekka and Vaughan, I have to side with David Springer here. A well-mixed atmosphere in gravity is in a state known as isentropic where the potential temperature is uniform. This is the dry adiabatic lapse rate. If you take an isothermal gas in gravity and thoroughly mix it, it turns isentropic. On the other hand, an isothermal atmosphere is a stable state and won’t mix spontaneously unless you stir it by some means, but that is true of all stratified states, not just isothermal, but temperature inversions that are even more stable. The dry adiabatic lapse rate is the maximum entropy state which is achieved by mixing until it is homogeneous in the potential temperature. With no gravity this would be isothermal too.
Jim D,
There’s no question about the fact that an atmosphere with vertical convection has a lapse rate that’s adiabatic lapse rate, moist adiabatic lapse rate, or something called environmental lapse rate depending the particular conditions. No disagreement on that.
That kind of atmosphere is, however, not isentropic or in thermodynamic equilibrium. It’s typically in a stationary state, but never a thermodynamic equilibrium. In thermodynamic equilibrium the atmosphere is isothermal and the vertical profiles of each type of molecule have their own exponential dependencies with altitude. An atmosphere in thermodynamic equilibrium is stratified and totally free of vertical convection and mixing. The only vertical motion is diffusion at molecular level.
Whether any atmosphere anywhere is in thermodynamic equilibrium is questionable (answer is almost certainly that no atmosphere is in thermodynamic equilibrium), but that does not change the properties of a thermodynamic equilibrium.
Pekka, thermodynamic equilibrium is not a well defined concept to me. Radiative equilibrium does drive towards an isothermal state, but mixing goes towards an isentropic state, which is a recognized term indicating constant potential temperature (also dry adiabatic) because the log of potential temperature is basically the entropy in thermodynamic terms.
Jim,
There are few concepts that occur here that are as well defined as thermodynamic equilibrium. If You don’t know what it is, you should find out.
Thermodynamic equilibrium in gravity is different from that in no gravity. They lead to different pressure profiles and hence temperature profiles. As a term, it is too vague to define the profile by itself.
Jim,
It’s equally well defined with gravity. it’s isothermal also with gravity. The densities drop exponentially for every species of molecule at a rate given by molecular weight. The average gravitational potential energy of every molecular species is given by equipartition theorem as kT/2 per molecule.
That’s already a good starting point.
Pekka, no the potential energy starts to matter. Velocities of molecules tend to be lower at higher altitudes (cp*T+g*z is roughly preserved because Vdp=~-g*dz). As air ascends it encounters lower pressure, expands and cools. A truly mixed state is adiabatic. Once air gets into that state it cannot settle out into another profile (or unmix) because that maximizes entropy.
Jim,
They are not slower at higher altitudes. Gravity affects the density, but it does in such a way that the velocity distribution is the same at all altitudes. Molecules that are slow have a higher probability of falling than rising, while faster may also rise higher. Going through all that in detail proves that everything fits perfectly together.
I have written a short note on how that works out
http://pirila.fi/energy/kuvat/barometric_derivation.pdf
That’s not a new theory invented by me, but a well known fact, but I haven’t seen all that gone trough in this way elsewhere.
Pekka, I don’t think you are disputing that in an adiabatic convective profile, the temperature at higher altitudes is colder, so the higher molecules are slower, and are continually moving up and down without losing or gaining diabatic energy but with their temperature changing, so I think this argument is about whether the convective profile is an equilibrium profile or not. I say it is, except for conditions where the equilibrium is controlled by radiation, which is a diabatic process.
Jim,
Convective atmosphere is a totally different case. In the Earth atmosphere the situation is closer to what I have been talking about at altitudes of well more than 100 km, i.e. outside the turbopause. Up to the turbopause convective mixing dominates, above turbopause we have diffusion and an atmosphere that has many properties of thermodynamic equilibrium. In particular the density profiles are there different for each molecular species.
Coming back to the troposphere. The temperature drops with altitude, because in convection molecules do not move individually and are not subject to the automatic selection that favors faster molecules in upwards diffusion. Most importantly a rising parcel of air does work in pushing other air out of the expanding volume.
Adiabatic expansion causes cooling also without any rising movement. That’s used in some heat pumps although heat pumps are usually based on phase transition, because that leads to better efficiency.
Jim, a very simple way to show you’re right is to consider the case of a single air molecule bouncing perfectly elastically up and down on the surface, with enough energy to bounce say 20 km high. Clearly it will be moving faster in the lower half than the upper, and therefore will have more energy in the lower half.
It follows that a gazillion molecules doing this in a column, perfectly vertically without ever colliding with each other, would produce a column of air which is hotter in the lower half than the upper. QED
Unfortunately this simple proof has one little flaw: population.
Each molecule traverses the same distance in the upper half as in the lower, but more slowly. Hence it spends more time in the upper half than the lower. So even though each one is contributing more thermal energy to the lower half than the upper while it’s there, there are more molecules in the upper half at any given instant which tends to offset that effect.
If a plate is placed over the column so that molecules bounce off it before reaching their peak, it doesn’t change this overall story. What it does show however is that pressure is behaving “normally” because the force exerted on the surface by the molecules bouncing off it is much greater than that on the plate above. It also shows that lower pressure need not entail fewer molecules per unit volume.
While this isn’t intended as a replacement for Pekka’s rigorous proof (for one thing the absence of collisions makes this far from an ideal gas so PV = nRT doesn’t hold at all), it might help to undermine at least a little the refutation of Pekka’s proof that you may have had in mind.
The case that Vaughan presents is described in some more detail and with formulas in a paper by Berberan-Santos, Bodunov, and Pogliani.
Here the chapter II.B.1 Kinetic derivations, Noninteracting molecules is exactly the same case.
Vaughan and Pekka, I am not going to be convinced by non-interacting molecules because that lacks the pressure effect of collisions. Anyway defining thermodynamic equilibrium states is beside the point to me, so I won’t pursue this further here.
I think we agree that this paper and the descriptions of processes here by the authors are designed to leave us in doubt that latent heat really does lead to increased buoyancy and that buoyancy really does lead to ascent. If they wrote a paper focusing on these two points, it would be very entertaining to see how it does in the review process, because this is as much a departure from reality as their contrarian condensational pressure reduction idea.
Jim,
No need to discuss the thermodynamic equilibrium further. Related to your message I do, however, note that intermolecular collisions don’t change the pressure of an ideal gas.
Collisions are needed to maintain local thermal equilibrium, but the pressure is independent on their frequency. The local thermal equilibrium is in many ways essential, while the global thermal equilibrium is not. Collisions are certainly involved in all pressure effects but even mutually noninteracting molecules would exert the same pressure on surfaces, if they could somehow be brought to the state with the Maxwell-Boltzmann distribution of velocities.
The interactions between the molecules have a small influence on pressure when it’s necessary to apply real gas equations of state like the van der Waals equation That’s not necessary in atmospheric considerations.
“It follows that a gazillion molecules doing this in a column, perfectly vertically without ever colliding with each other, would produce a column of air which is hotter in the lower half than the upper. QED
Unfortunately this simple proof has one little flaw: population.
Each molecule traverses the same distance in the upper half as in the lower, but more slowly. Hence it spends more time in the upper half than the lower. So even though each one is contributing more thermal energy to the lower half than the upper while it’s there, there are more molecules in the upper half at any given instant which tends to offset that effect.”
But molecules of gas don’t travel distance. A molecule of gas is constantly being hit by zillions of other molecules and are constantly changing direction. An average single molecule is not going anywhere- not environment where there lots of gas molecules.
So instead of average molecule going bottom of atmosphere up to top in a minute, it’s staying basically in same location for hours or indefinitely.
Or the energy of the kinetic gas is forward, backward, upward, downward,
left, and right and this changing every nanosecond.
It’s only in lower gas density environment where a gas molecule is able to travel any significant distance, because you could say “the many” determine a single molecules direction- and when in environment there fewer molecules, the velocity of single average molecules is able to travel any amount of distance.
So the lower and density atmosphere is like a huge traffic jam- once molecule enters “it’s never going to get anywhere”.
gbaikie,
Vaughan noted that the idea was not to prove anything but to make it little more understandable that the equilibrium is isothermal. The same is true for the paper that I linked in this thread. The authors tell there as well that the purpose is to help understanding, not prove.
The note that I have written and also linked further up in this thread has a different goal. It presents a proof that kinetic theory of ideal gases under gravity has an isothermal equilibrium. Being a proof it’s certainly more difficult to understand. It does not assume that the molecules don’t interact but takes molecular collisions into account.
“gbaikie,
Vaughan noted that the idea was not to prove anything but to make it little more understandable that the equilibrium is isothermal. The same is true for the paper that I linked in this thread. The authors tell there as well that the purpose is to help understanding, not prove.”
I don’t think atmosphere is a equilibrium isothermal. At least per this definition of:
“The state of an atmosphere at rest, uninfluenced by any external agency, in which the conduction of heat from one part to another has produced, after a sufficient length of time, a uniform temperature throughout its entire mass. Also known as conductive equilibrium. ”
Now atmosphere at rest is strange idea considering the molecules bouncing about at 1000 mph.
But I am assuming what meant is there isn’t excessive heating or cooling and it’s not windy and there not all kinds of stuff going on [also generally known as weather- so “normal” and calm weather].
I would agree it’s same heat, if allowing for PE as equaling heat/energy.
So, the equilibrium one discussing is a balance of heat, if this balance of heat included potential energy of molecules at higher point in the gravity [what Vaughan described].
The problem I had with Vaughan is he said the molecules which converted the PE into kinetic energy [by falling- and kinetic energy in regards to gas *is* heat] would spend less time in a lower part of atmosphere because they traveling a further distance in shorter period of time.
Which is correct IF you bounce the molecules without inference from other gas molecules with different vectors. And if this was the case then the gas molecule at the lower of the atmosphere would have higher kinetic energy BUT because there was very few of them in a given volume [most time they would at a higher elevation] so far less mass is involved.and the gas would actually have less heat [though btw, the surface would have to be very warm otherwise the high velocity gas molecules would transfer their energy to the surface {they would not bounce perfectly, it would more like a bug spat against a windshield}].
If invert what Vaughan is talking about, it’s describing the stratosphere and higher. Where there less molecules and so less inference to distance traveled by a molecule. Or instead inverting, just consider the top of troposphere as the “surface”- it’s a spongey surface but is hard enough to faster moving molecules- and faster molecules are “selected” by the process.
Though both apply.
But in the troposphere the gas molecules are stuck in a mad traffic jam. And one could say the lapse rate is one type of measurement of this traffic jam. With the water molecules as quite significant in terms of bogging down the traffic.
Pekka Pirilä : The equilibrium state is totally irrelevant for real physics …
That’s a refreshing view.
I have written that the equilibrium state is merely an insufficiently accurate approximation to real physics.
@Matthew R Marler: I have written that the equilibrium state is merely an insufficiently accurate approximation to real physics.
Real physic on the other hand considers equilibrium to be when you’re taking as much chlordiazepoxide as the next guy. Don’t write that off.
Lest we forget dihydrogen monoxid:
http://www.dhmo.org
Matt,
It’s probably understood by everyone, but to be sure:
My statement on the irrelevance of the thermodynamic equilibrium referred to the full atmosphere, which is always far from the thermodynamic equilibrium. Thermodynamic equilibrium is a very useful and important concept for many other systems including local thermodynamic equilibrium of small volumes within the atmosphere.
I think this is another case of the which “surface” syndrome.
The condensation temperature should be the “surface” at saturation. Below that layer would be supersaturated air and above would be near saturated air.
https://lh5.googleusercontent.com/-i0UApz1xKQc/UQ5rdbDFvYI/AAAAAAAAHD4/bslBj27ttN0/s729/moisture%2520ground%2520plane%2520surface%2520evaporation.png
That would create a small pressure differential creating a circulation that would expand the “surface”. What is actually happening with the energy would require more work and it is a bit of a chicken or egg situation as to what starts the process, but that small differential could build if the conditions are right.
Just my take.
It’s just another case where formally correct mathematics is used to derive wrong physical conclusions.
Pekka, “It’s just another case where formally correct mathematics is used to derive wrong physical conclusions.”
Likely, but that may be partially due to the formally correct models. Saturation pressure is dependent only on temperature, based on the models, leading to these nasty little situations like super saturation and having to assume things because the models don’t model them very well.
http://cires.colorado.edu/~voemel/vp.html
Being off a few percent isn’t a big deal most of the time, sometimes it is though.
The detail that makes the calculation so meaningless is the comparison with the isothermal full thermodynamic equilibrium with the molecule specific density profiles. That’s so extremely remote of anything really occurring in the atmosphere that it produces totally spurious large differences.
pekka, “The detail that makes the calculation so meaningless is the comparison with the isothermal full thermodynamic equilibrium with the molecule specific density profiles.
Isothermal in the x direction and equilibrium in the z direction.
That case is isothermal in vertical direction.
pekka, “The detail that makes the calculation so meaningless is the comparison with the isothermal full thermodynamic equilibrium with the molecule specific density profiles.
Isothermal in the x direction and equilibrium in the z direction in a thin layer at the temperature of condensation which is set by surface temperature since “surface” temperature is the basis for estimating saturation vapor pressure.
It is circular because the references for saturation pressure vary with the freezing and boiling point of water that would vary with sea level and composition of the gas if I read the equations correctly.
Their state of comparison for their moist column is fully isothermal. The water content is chosen to be at saturation at the surface. The saturation partial pressure depends only on temperature, not on anything else. With increasing altitude the partial pressure of water goes down but the share goes up, because the partial pressure of dry air goes down faster.
That’s all true for the thermodynamic equilibrium. That’s, however, irrelevant because the real atmosphere is always very far from thermodynamic equilibrium both in it’s temperature profile and in maintaining essentially constant stoichiometry for all noncondensible constituents. Water vapor is different in that and its share varies widely.
As I wrote the calculation of the isothermal column is mathematically correct. What’s wrong is to use it in comparisons in the way they do. Any comparison can, of course, be made, but this kind of comparisons do not tell anything worthwhile. They do, however, claim that the comparison tells things that it most certainly does not tell. Their physics is totally wrong.
Pekka, “As I wrote the calculation of the isothermal column is mathematically correct. What’s wrong is to use it in comparisons in the way they do.”
I agree completely with that. The actually “effect” they are implying is real and the delta P based on saturation vapor pressure provides sufficient potential energy to allow a “new” approach. The devil is in the details.
Deus ex Makarichina.
===============
Nick Stokes | February 3, 2013 at 2:29 am |
“I think that it might be productive for you to decide whether you want to persuade your readers in a mathematical or a physical fallacy present in our paper. It is easy to see that you cannot have both, so this takes away from the strength of your critique.”
The last statement is a good and strong mathematical claim. Had you been able to prove it, the paper would have been rejected. This is my view. But the problem is that you did try a few times, but had not been successful — despite your ubiquitous claims to the contrary.
For example, you say today that (33) determines S and (34) also determines S. Let us take a look if this is indeed so.
(32) is an equation on ∂N_d/∂x and ∂N_d/∂z (N_d is dry air molar density).
(33) is an equation on ∂N_v/∂x, ∂N_v/∂z and S (N_v is vapor molar density and S is condensation rate)
(34) is an equation on ∂N_d/∂z and ∂N_v/∂z and S.
We additionally have ∂N_v/∂x = 0 (horizontally isothermal atmosphere).
So, we have five formal variables: ∂N_v/∂x, ∂N_v/∂z, ∂N_d/∂x and ∂N_d/∂z and S. And we have four equations. This means that if we are lucky and know S, we can possibly express any other variable as a function of S.
Now you say:
You must have made a typo here, the correct expression is S=u∂N/∂x = u∂N_d/∂x.
But please appreciate that mathematically this is no nonsense. It is exactly what we have been hoping for: to express one of the variables in terms of S. And we did it. This is a valid mathematical result.
To support your claim of a mathematical fallacy, you would have needed to demonstrate that the system (32)-(34) produces a mathematical contradiction. E.g. y = x AND at the same time y = x/2, you know. This is what overdetermination is about. For example, in your comment above you attempted precisely that: you claimed that our equation (A7) formally contradicts (34). But that was an erroneous statement of yours.
So, I would still propose that you should focus on a physical fallacy, it is a much more foggy business. This is what you are actually doing:
Please appreciate that mathematics does not know what rain is. It is indifferent. Your claim is purely physical, and the alleged “nonsense” that you are proposing is physical as well. I’ll address your physical proposition separately. It is interesting!
Anastassia Makarieva, this statement makes grammatical sense, but it does not make mathematical sense! The reason is that (e.g.) ∂N_v/∂x and ∂N_v/∂z must satisfy an integrability condition (because otherwise N_v(x,z) does not exist globally, even though ∂N_v/∂x and ∂N_v/∂z are locally well-defined locally) and thus ∂N_v/∂x and ∂N_v/∂z cannot consistently be regarded as independent variables.
It is concerning that within both the long-active (but not-yet-accepted) “PSI” community and the nascent (but not-yet-accepted) “condensation-driven winds” community, foggy verbal arguments commonly are prefaced with imprecise (and/or just plain wrong!) mathematical statements.
Recommendation The “condensation-driven winds” folks (and the PSI fols too) need to bring mathematicians on-board to (1) repair their mathematical formalisms first, and (2) then clarify their prose.
Verdict By Donald Knuth’s criterion, both PSI-theory and condensation-theory presently qualify as “Art”, but not as “Science”. So after repairing their mathematics, both PSI-theorists and condensation-theorists should verify the repaired mathematics by providing some sample code!
Donald Knuth says: “Science is what we understand well enough to explain to a computer. Art is everything else we do. […] Science advances whenever an Art becomes a Science. And the state of the Art advances too, because people always leap into new territory once they have understood more about the old.”
Almost prophetic. A lot of experimental and practical genetics is now done in laboraties the size of an integrated circuit chip where reaction sites are microscopic wells numbering in the thousands replacing test tubes, electrical currents shuffle stuff around the initial and resultant reactants, lasers read the results, and a computer controls it all.
Moore’s Law evidently applies to synthetic biology. Write that down, Sidles, you’re going to need to know it in the future.
And by the way, Knuth is wrong. When you can explain it to a computer it becomes engineering not science. We are just at the beginning of explaining biology to a computer but biology has been a science for centuries. It was perhaps still art when Da Vinci was dissecting bodies and making drawings of human anatomy but I’d argue it was science then too.
Each new generation of Boeing airliner is based upon fewer wind-tunnel models than the generation before.
Similarly in synthetic biology, each new cycle of innovation relies less upon on experiments and more upon computation than the generation before.
And in climate science, the same accelerating reliance upon computation is a plain fact-of-life, eh?
Many wet-bench biologists are distressed to hear 21st century graduate students proclaim “Experiments are for robots!” More broadly, it commonly happens 20th century scientists and engineers — who are not themselves overly possessed of mathematical maturity — are discomfited that 21st century scientific careers require mathematical maturity absolutely.
As Kurt Vonnegut says: “So it goes”!
Results obtained from the computer still play second fiddle to results obtained from reality.
Write that down, John Sidles.
Now, now, DS; you misunderestimate the power of pictograms.
=============================
Anastassia,
When it comes to proving mathematical contradiction, I’ll settle for nutty results. And I cited one.
But here is how your error plays out. Your Eq 34 is:
S=w(∂N_v/∂z – N_v/N ∂N/∂z)
From the conservation of mass equations 32 and 33, one can derive, in vector form:
S = v.( ∇ Nv – Nv/Nd ∇ Nd)
In your argument for Eq 34, you ignored horizontal velocity and horizontal derivatives. I think that needs to be justified, but doing it yields, from 32 and 33:
S=w(∂N_v/∂z – N_v/N_d ∂N_d/∂z)
The only difference from 34 is the appearance of N_d instead of N. This reflects your error of using N (moist air) as the non-condensable reference gas rather than N_d.
So Eq 34 does not represent new physics. It represents 32 and 33, with the added term N_v/N ∂N/∂z – N_v/N_d ∂N_d/∂z. This is purely due to error. But the effect of treating 34 as an extra equation is to add the equation
N_v/N ∂N/∂z = N_v/N_d ∂N_d/∂z
to your reasoning chain.
Well, almost. The horizontal components that were ignored all come in to this difference equation too. But you have added a new equation that makes no physical sense whatever.
Nick
The equation (34) is, indeed, strange. As you point, the first part of that can be derived from (32) and (33) except that instead of N we have N_d and that the derivatives with respect to x must be set to zero.
This get really bizarre when these equations are used to derive a formula for partial derivative of p with respect to x. Dropping x -dependence out in (34) it was effectively stated that nothing relevant depends on x. How can anyone imagine that an equation based on such an assumption is used in deriving the dependence of one related variable on x. How can they assume that the partial derivative dN_v/dx=0 when the partial derivative dp/dx is not zero as the latter would clearly imply also that T varies horizontally.
It’s also true that claiming that S is given by both (33) and (34) is not justified and is hardly justifiable and that all the equations (32), (33) and (34) are used to derive the next equations.
Previously I thought that they do the mathematics correctly and err only on physics, but evidently they mess totally the mathematics as well.
Pekka,
“Dropping x -dependence out in (34) it was effectively stated that nothing relevant depends on x.”
Yes. I think what has happened is that the argument for Eq 34 was made in a vertical updraft situation where it may be that those horizontal terms were negligible relative to the vertical terms retained. But the effect of the overdetermination is that these terme are asserted to be equal to the error-induced term as an equation, which then gets applied everywhere. I now realize that even without the incorrect use of N as the noncondensable reference, the discrepancy in treatment of the horizontal parts alone would have nessed everything up that follows.
Nick,
Some of my earlier comments have essentially stated that a much more comprehensive analysis is needed before any horizontal derivatives can be estimated at all, because something is needed to set the horizontal scale and other essential relationships between vertical columns.
Anastassia, I think Nick Stokes was right about this:
“This reflects your error of using N (moist air) as the non-condensable reference gas rather than N_d.”
If you replace in your (33) N_v by N_v/N_d, u by N_d*u and v by Nd*v, you get: ∇(N_v/N_d) . (u,v)= S/N_d, which has characteristics
dx/dt= u,
dz/dt= w,
d(N_v/N_d)/dt= S/N_d,
with N_v/N_d= γ/(1-γ)
Now I tend to agree with you about the nature of the “source” term: it should not be seen as an independent physical source term, but simply expresses what should happen according to the relationships in Section 2.1. That means that along the characteristic,
d(N_v/N_d)/dt= w ∂(N_v/N_d)/∂z,
so
S= w N_d ∂(N_v/N_d)/∂z = w N (1-γ) ∂(γ/(1-γ))/∂z
and so you also have your compensation for overall volume expansion right there. I think that is it, nothing new or unphysical. So you are right, your equations in 4.1 are pretty close anyway, but the one above is rigorously derived.
Nick Stokes | February 3, 2013 at 3:58 pm | Reply
Nick, “nutty results” and “mathematical contradiction” is not the same. “Nutty results” do not make a ground for accusing people of all possible things. Let’s be specific as Steven Mosher wants us to be. You cited two things (if I have not overlooked anything.) Here as I clarified your claim was based on a mathematical error. Here you spoke about rain from dry air, which has nothing to do with mathematics. (I do remember this point, it will not be ignored, just not everything at once).
So, I am glad that you have now concentrated on physics. And you are making progress, you no longer assert that (33) defines S or the system is overdetermined. Your points now is: “But you have added a new equation that makes no physical sense whatever.” Let’s discuss this.
Anastassia,
“Nick, “nutty results” and “mathematical contradiction” is not the same”
Well, if that semantics is important to you, let’s deal with the nutty results. I said that because of the lack of independence of 34 from 32 and 33, you had introduced an equation which asserted that the discrepancies were exactly zero, without justification. I see this is now made explicit – that equation is A7. (Sd-S) is the measure of the difference made by choosing N as the reference gas density rather than N_d. And the left side, u ∂N/∂x, is one of the terms you omitted in forming 34, but was present in 32/33. The others seem to be omitted by your maintaining the requirement that ∂N_v/∂x=0.
So if you had S=S_d, you would have u ∂N/∂x =0. But you also has this equal to S. So S=0? It never rains but it pours from dry air?
For more weirdness, try putting real values of C, the kinetic constant, into A10.
After a little more thought, it still seems to me that the sign in eq. (27) is wrong:
h_v(z’)<= h_v(0) (see fig. 1(a)),
therefore exact rho_v(z) <= approximate rho_v(z) (as it is an increasing function of h_v(z') in (26)),
and therefore, the right-hand side of (27) should be a lower bound to the change in surface pressure rather than an upper bound as stated.
That makes the approximation of p_A(z) in (30) a lower bound, so the pressure difference p_A(z)-p_B(z) may as well be 0 at the surface. It does not need to be negative.
Or am overlooking something?
What you are saying concerning the heights is correct. But as one can see from Fig. 1b, delta ps is defined to be a positive value. I.e., its absolute magnitude is shown and calculated in Eq. (27): delta ps ≡ -(p_A(0) – p_B(0)).
That is right, just wanted to write that, so that actually helps :). So your point of the approximation is that the adjustment process from isothermal to adiabatic profile leads to a drop in surface pressure in the saturated column, and not in the dry column, if you keep the surface temperatures the same.
I guess this initial condition is rather remote from the final state; what does a change in column weight between these states then mean, since in the beginning, there was far more water vapor present than would be realistic? Is it relevant for what happens in the real atmosphere?
As people seem to be discussing Section 3.3 “Pressure profiles in moist versus dry air columns”, I would like to provide some hopefully relevant context. In this Section a thought experiment is considered to illustrate the basic ideas and scales at work. A vertically isothermal column with vapor saturated at the surface (A) or dry air (B) is cooled such that it acquires either a moist (A) or a dry (B) lapse rate. The main purpose of this excercise is to illustrate that, because of condensation, surface pressure in column A will be lower (Fig. 1c) than in column B, while in the upper atmosphere the reverse will be true (because of the difference in the dry and moist lapse rate).
One might think — why would they need it? In fact, with this excercise we were responding to our previous influential critics (like, e.g., Dr. Rosenfeld) who denied the very opportunity that there can be a pressure fall associated with condensation. Now everybody seems to know that, but at those times things were different, and the issue was not clear to many.
That this is an objective vision of the situation is supported by the fact that approximately during the same time that our paper was submitted to ACPD (but after it was posted at arxiv.org), Spengler et al. (2011) undertook an effort that was in some way similar.
Spengler et al. (2011) did the following: they took a normal column of air with an observable quantity of the water vapor and then imagined, in a thought experiment, that all vapor between 2 and 4 km in the atmosphere suddenly condenses, the latent heat is released in the sensible form and warms the atmosphere. (So people believing that our thought experiment is far from reality should see that in an appropriate context). They investigated how the hydrostatic equilibrium sets in and found that indeed, the surface pressure drops, while the upper atmospheric pressure rises (because of warming). This result of Spengler et al. (2011) basically repeats what is shown in Fig. 1c.
I think that the work of Spengler et al. (2011) was interesting. However, Spengler et al. (2011) from their 1-D thought experiment drew far reaching conclusions about the unimportance of the condensation-related pressure drop. This cannot be done in principle for a 1-D case. Unlike Spengler et al. (2011), we used our thought experiment in Section 3.3 solely to illustrate the involved physical concepts and scales. The real-world horizontal pressure gradients that are produced by condensation are considered in Section 4.
I note the work of Spengler et al. (2011) here because it was mentioned by Dr. Held in his review as an evidence against our propositions. Besides being for 1-D case, the process considered by Spengler et al. (2011) represents adiabatic condensation at constant volume that we showed in Section 2.1 is not physically possible.
The real single process of interest to the issue that you study seems to be that of ascending air, where
– evaporation brings moisture into the air and the temperature is set to a specific value. The density is affected by both temperature and moisture level
– the air ascends. At some point (possibly immediately) the relative moisture reaches 100% and condensation begins.
– the condensation is a continuous process that goes on as long as the air ascends
– in condensation latent heat is released and water is removed from gas phase to liquid droplets (or at some point ice)
– under proper conditions the process is nearly adiabatic. If this is the case the resulting temperature profile follows the moist adiabatic.
– under conditions that lead to adiabatic transition the pressure is also very close to hydrostatic pressure at every moment and every altitude
– the loss of water from the air does not affect the pressure, it reduces the volume of the remaining air
– the speed of the convective flow is reduced due to the reduced volume of the gas.
That’s my description of the moist column.
When I wrote “does not affect the pressure” I meant the local process. The moist adiabat calculated taking the chances in stoichiometry into account differs a little from the approximate values obtained when it’s not taken into account.
The pressure levels depend also on the amount of falling precipitation. To calculate that effect the size of the droplets must be determined as the result depends on the terminal speed of the droplets.
Thanks for that explanation Anastassia
captdallas2 0.8 +/- 0.2 | February 3, 2013 at 9:40 am |
“Being off a few percent isn’t a big deal most of the time, sometimes it is though.”
I used to write code for electronic slot machines. I accidently made the odds 49:51 in favor of the player instead of in favor of the house.
I explained to my boss it was only a 2% mistake so what’s the big deal?
yep, that is what is so fun about this debate. Having one team throw out ridiculous confidence intervals while the real world laughs.
Think mass, gravity and energy supplied.
Nothing else matters.
If anything else tries to disturb the temperature (or more accurately energy content) derived from those 3 characteristics alone then all one sees is a change in circulation adjusting the flow of energy throughput to keep top of atmosphere radiative balance stable.
The stabilisation process involves the switching of energy to and fro between KE and gravitational PE and that is what determines temperature since only KE registers on sensors as temperature.
The switching to and fro between KE and PE is achieved by expansion and contraction.
All else is chaff.
With the caveat that I’m not used to thinking about the water cycle in this detail, so don’t have an easy familiarity with the terms, I’m rather enjoying the discussion – I particularly like the Anastassia/Pekka loggerheads here:
Pekka’s – “the loss of water from the air does not affect the pressure, it reduces the volume of the remaining air”
versus Anastassia’s – “Spengler et al. (2011) did the following: they took a normal column of air with an observable quantity of the water vapor and then imagined, in a thought experiment, that all vapor between 2 and 4 km in the atmosphere suddenly condenses, the latent heat is released in the sensible form and warms the atmosphere. (So people believing that our thought experiment is far from reality should see that in an appropriate context). They investigated how the hydrostatic equilibrium sets in and found that indeed, the surface pressure drops, while the upper atmospheric pressure rises (because of warming). This result of Spengler et al. (2011) basically repeats what is shown in Fig. 1c.”
So, should be grateful if any replies, from anyone, are couched in least technical language possible..
If the loss of water from the air (condensation) reduces the volume, then isn’t it also increasing the density? If the greater the density the heavier the air wouldn’t that mean pressure increased?
Which is why I can’t get my head around condensation creating an area of low pressure at the surface.
Is condensation the point of change between low and high pressure?
It seems to me that the low pressure at the surface created by the expansion of rising gases would begin to alter at whatever height there was condensation, so while it may appear to still be low pressure at the surface this is about to change as condensation gets into its stride and the colder air around the condensation will also be getting heavier. (The air heated by condensation will continue to rise, but there can’t be ‘vacuum’ around the condensation, the colder air at that level will come in to replace any lighter hotter air rising, and colder is heavier so will sink which increases the pressure at the surface.)
There’s some on “Stability of moist air – moist convection”, here:
http://eesc.columbia.edu/courses/ees/climate/lectures/atm_phys.html
p.s. ” the colder air at that level will come in to replace any lighter hotter air rising, and colder is heavier so will sink”, which is wind. Wind is a volume of air on the move.
SM Feb 3 @4.07pm on falsifiable statements.
Yeah, Steven, me and the blokes down at the pub would have
a problem with Quine’s arguments about synonyms, distinctions
leading as they do ter ter skepticism about meaning.
Regardless of Quine’s perception of transferring meaning of
synonyms, me and the blokes down at the pub take a statement
ter be ‘true’ *if it has no internal logical inconsistencies and * if it corresponds to the reality, ie the statement: ‘all birds have feathers,’
( or ‘all swans are white,) is true, if, and only if, it corresponds ter
reality. It should therefore be empirically falsifiable.
As we see it, SM, (me and t b d a t p ) yr theory concerning unicorns
and the wind can only be true if and only if the pesky little critters exist . Finnding them … “#!^*#WTF !” would not be confirmation of yr theory
but only give yer a basis fer proceeding,
Hope yer git yr grant and verify yr theory, SM.
Beth and the blokes down at the pub.)
Agreed
Beth, you need to go back to carnap. and that does not mean a rest in the in your volvo. Truth as correspondence, is an entirely different matter but never mind I can make do… You surely dont mean that for something to be true it has to be empirically falsifiable.
is the statement “for something to be true it must be empirically falsifiable’ true? that is, is it empirically falsifiable?
Opps. so you just made a statement that contains no logical inconsistencies and which cannot be falsified by empirical evidence. Opps. So, if it is true, then it is false. Opps. thats a paradox fer ya. Ya might consider that a bunch of folks waay smarter have been strugglin with the problems of truth and meaning and if the answer was simple, well, it would be simple.
That’s all there is to know, and all ye need to know.
=======
It has to be falsifiable to be science. Not everything that it is true is science.
Anastassia told Nick on a subthread we should leave to them:
http://judithcurry.com/2013/01/31/condensation-driven-winds-an-update-new-version/#comment-291748
Notwithstanding the parsomatics involved, Anasstasia’s claim seems to presume that uttering a mathematical contradiction might provide a sufficient ground to accuse people of all possible things.
But does it?
Let’s take Deolalikar’s purported negative proof of the P = NP conjecture.
Here’s how Terence Tao summarizes the whole shebang:
http://rjlipton.wordpress.com/2010/08/11/deolalikar-responds-to-issues-about-his-p≠np-proof/
In other, more mundane words, Tao is trying to preserve what’s good in that venture. If he’s proved wrong, no harm’s done. (He’s Tao, so he’s a bit above it all.) If he’s proved right, no harm done either. Perhaps some embarrassment either way, but no great deal. People are not that interested in others’ failures as much as they fear themselves.
***
There are many questions one may ask about the implications of somebody else’s work. All could be worthwhile. Attitudes must be kept in check, from all sides.
Here’s Lipton conclusion:
Climateballers should perhaps rejoice that such exciting events became their daily bread. I honestly can’t recall a single week in the last three years that did not include an incident of the magnitude of Deolalikar’s proof. At least, when seen from the Climateballers’ top-of-lungs’ comments.
***
Does every single comment deserve to be played as it was the last two minutes of the Superbowl?
Sometimes, I wonder.
Sometimes.
willard, I have been following your comments with interest, and I have been even planning to answer some, but I must confess I do not understand your attitude at all. Perhaps this is because of a different culture, e.g. “the Superbowl” that is so often referred to here does not mean anything important to me.
For example, in this comment, are you pleased or displeased with my response to Nick? I cannot say. Nick has invested a lot of time in criticizing our work and many people draw on his opinion. His reputation will be affected if, after him having proclaimed so many times that our paper is full of mathematical and physical fallacies, people take a closer look and it turns out that it contains new, solid and falsifiable propositions. So Nick can be tempted if unconsciously to exaggerate his claims to prevent people from taking a closer look. (Nobody is free of biases, me neither.) So what I am doing — I am just trying to be as specific as possible not to allow Nick to create confusion in people’s mind. (Hopefully other people including Nick play the same role with respect to myself.) There is no mathematical contradiction in our work and Nick’s claims on this point are unsubstantiated, that’s all I wanted to say.
Also, if we speak of citations, recently I knew the following one. It is from
John Milton, the blind, English poet:
To me the main message of this citation (which I do feel is correct) is that when you find something really new, you may not be able at once to find the correct form for this new thing. Should we be writing this paper today, we would have written it differently (mostly like in the post). Many things in our paper I would have today omitted, others I would have re-written — time is going, we are working and the picture becomes clearer together with the arguments. So I am also using this discussion to clarify our points to future readers. It is good to have all arguments discussed in one place — you can later make a list of links to specific comments where important things are discussed.
Anastassia,
My own explanation for the uneasiness I am causing is this:
> If you want to realize the truth, don’t be for or against. The struggle between good and evil is the primal disease of the mind.
http://neverendingaudit.tumblr.com/post/8472483540
This might be romanticized, I admit. Other explanations would be too long. I need to go.
I’ll try to respond tomorrow. Until then, please leave Nick’s subconscious out of it. The subconscious might be as real as unicorns anyway.
Anasstasia,
Here is my response to your comment.
1. I am both pleased or displeased with your response to Nick. I am pleased it is happening, and displeased it is so difficult to follow. The way the discussion gets fragmented in the thread does not help.
What I feel about the discussion is irrelevant anyway. What matters is that you both clearly exasperated one another. So what you could do, before you go, is to make a final comment, in which you could summarize every point you think should be addressed.
***
2. Most of your comment amounts to what is called, among climateballers and elsewhere, “playing the ref”. You are basically exposing your recriminations against Nick for me to blame him. You are using a strategy that, in transactional analysis, is called “Look How Hard I’ve Tried”.
I’m not an arbitre between you and Nick, Anasstasia. And quite frankly, I could not care less if Nick is wrong or right. Nick can take care of himself.
If you care about what I care, what I would care is that the summary in #1 gets done. I hope you do care about that.
***
3. I don’t think that Nick’s reputation will be affected if your paper “contains new, solid and falsifiable propositions.” Nick is Nick. Nick says what he sees. He can be wrong. If he is, you have to show it to him. He won’t concede anything. But if you do show him, he’ll (perhaps grudgingly) admit it.
Nick’s impersonating what scepticism should be: “believe it, but check it,” as Garry Kasparov always say. When I think about hockey (one of the reason I got interested in climate), I compare Nick’s style to Mario Lemieux, a player not unlike Valeri Kamensky if you can ask around. A one man army. Scores goals all by himself. Likes to play physical, to the expense of his health.
What happens to Nick concerns no one but Nick.
***
4. I have my doubts regarding the cultural difference. Superbowl is quite a big thing in itself. Besides, what I’m talking about is captured by all languages, as Steven Pinker says over there:
> Well, I said I’d talk about two windows on human nature — the cognitive machinery with which we conceptualize the world, and now I’m going to say a few words about the relationship types that govern human social interaction, again, as reflected in language. And I’ll start out with a puzzle, the puzzle of indirect speech acts. Now, I’m sure most of you have seen the movie “Fargo.” And you might remember the scene in which the kidnapper is pulled over by a police officer, is asked to show his driver’s license and holds his wallet out with a 50-dollar bill extending at a slight angle out of the wallet. And he says, “I was just thinking that maybe we could take care of it here in Fargo,” which everyone, including the audience, interprets as a veiled bribe. This kind of indirect speech is rampant in language. For example, in polite requests, if someone says, “If you could pass the guacamole, that would be awesome,” we know exactly what he means, even though that’s a rather bizarre concept being expressed.
http://www.ted.com/talks/steven_pinker_on_language_and_thought.html
Something like this is what happened between me and Douglas at Eli’s first, and then here.
I’ll make this forecast: sooner or later, climateballers we’ll hear from Douglas. He might even become a climateballer himself. I could very well see him blog about your theory, about ecology, big trees, forests, climate models, Einstein, Feynman, Popper, the IPCC. His writing style makes me envision a Lomborgian Honest Broker. An editorial line more enviro-friendly than a rational optimist, but not too far from it. Some concerns. Lots of uncertainty. Models, bad. Anasstasia says. Bad models.
Just a conjecture. We’ll see how it will get falsified.
***
5. Notwithstanding my previous criticisms, I do wish you the best of luck with your theory. Even if this idea you’re cultivating for so many years turns out wrong, at least you’ll have been impassionated by one idea. So many scientists can’t even claim to have had one single original idea. In empirical science, truth is way overrated.
***
6. I’m quite confident that this experience is quite stressful to you. So many criticisms, at so many levels. You just had a taste of Orestes’ predicament:
http://en.wikipedia.org/wiki/Orestes
Making a “list of links to specific comments where important things are discussed” would certainly be a good idea. At the very least, if will provide help you organize the information in a way that can be fruitful and redeeming. You will be able to focus on what’s constructive, and what’s less so. You could even list what kinds of comments tend to “push your buttons”, so to speak, to make sure you are better prepared next time.
7. Yesterday, I tought that the P=NP debate provided a good way to convey what I’m telling you right now. It also conveys the idea that mathematics is not only about formal derivations. It takes time before mathematical proofs become gap-less. Even in mathematics, one can be wrong without being that stupid as to miss a derivation step. It happens.
This will be my last comment on this topic,
Best of luck,
w
Sorry to have mispelled your name again, Anastassia.
It is astonishingly obvious that when it rains it storms – that evaporation and condensation is correlated with winds.
If I understand the argument, and perhaps I do not, the paper’s controversial proposal is that evaporation and condensation causes winds and causes the transport of heat to higher altitudes where it gets radiated away.
Which, looking out my window, seems obviously true as an empirical fact. I don’t get strong winds without strong condensation, and when I get strong condensation, things cool down.
Perhaps I am completely missing the point.
I don’t think you are. The consensus says either the winds are not condensation-driven or it’s already implicitely in the models.
So, let us discuss the physics behind Eq. (34) and the reference term, as per this thread of comments above.
This is how the general continuity Eqs. (32) and (33) can be combined:
S = v.( ∇ Nv – Nv/Nd ∇ Nd)
I emphasize that here S is an unknown source term. We do not know, if it is evaporation, condensation, or for example some chemical reaction that changes the amount of water vapor in the atmosphere.
Now we want to find S assuming that S represents — quite specifically! — condensation of water vapor that occurs because of cooling. We consider a steady-state case, so cooling may only happen in a saturated air parcel if it moves to a colder place. From this we propose that S must be proportional to the velocity component that is parallel to temperature gradient. Since our atmosphere is horizontally isothermal, the velocity component that is parallel to temperature gradient is vertical velocity w. Please pause at this moment to appreciate that this statement does not in any way follow from the continuity equation. It is an independent physical statement. At this point we may not even know what the continuity equation might look like at all.
Now, as our process of condensation depends on air motion (NB! it is important, it is an independent physical statement), it must manifest itself in spatial changes of the respective constituent (water vapor). In other words, it should depend on ∂N_v/∂z. (It cannot depend ∂N_v/∂x because we assume a horizontally isothermal atmosphere with ∂N_v/∂x=0). Once again, note at this point that if we were considering, say, a non-steady state with a standstill atmosphere cooling by radiation, then this proposition would not be vaild. We would then need to propose that condensation rate should be proportional to local temperature change rate or something — no grounds to expect it to depend on ∂N_v/∂z. So, I emphasize, at this point we have not anywhere relied on the continuity equation in our argument.
Now we know, again from independent physics, that there is another process that influences water vapor concentration besides condensation: it is gravity. So some part of ∂N_v/∂z is because of the gravitational expansion. We apparently need some reference magnitude to subtract from ∂N_v/∂z in order not to overestimate condensation rate. What can this reference be? (Again, no continuity equation in sight).
We propose to subtract the corresponding share γ= N_v/N of total air density change ∂N/∂z. This is motivated by the argument that we want to identify the effect of gravitational expansion that the air, including water vapor itself(!), undergoes. This gravitational expansion is determined by the hydrostatic equilibrium equation. (You remember, we assume our atmosphere to be hydrostatic). The air as a whole conforms to this equation, while the dry air component does not. This is a very essential point: dry air does not conform to hydrostatic equilibrium, so its molar density changes will not inform us about the effect of gravity.
From all the above considerations, where in no place we have made a recourse to the continuity equation, we write our Eq. (34): S=w(∂N_v/∂z – γ∂N/∂z).
At this point we can recall that there is a continuity equation and check how our expression for condensation rate relates to it. It turns out that (1) Eq. (34) is mathematically independent and (2) that combining Eq. (34) with the continuity equations (32)-(33) produces a meaningful result (37) that conforms to observations and is directly testable from empirical data.
But there is one additional thing we could do. We could check if using dry air as a reference would produce anything meaningful. We a priori believe that it cannot, because, I repeat, dry air does not conform to hydrostatic equilibrium and its distribution will not inform us about the effect of gravity that we want to subtract from condensation-induced density changes. And indeed, it appears that using S_d ≡ N_v/N_d ∂N_d/∂z as a reference produces a non-sensical physical result (although mathematically it’s ok). In this case we would have, instead of (34), that S = ∂N_v/∂z – N_v/N_d ∂N_d/∂z ≡ S_d.
But it is easy to see that Eqs. (32)-(33) (at ∂N_v/∂x =0)can be re-written as
u∂N/∂x = (S – S_d)/&gamma_d (here S is an unknown function)
So if we put S = S_d (having dry air as reference) we obtain that
u∂N/∂x = 0. In other words, winds cannot blow parallel to air pressure gradient. This result is clearly invalid.
So, this is our physical argumentation for S (34). I hope people can see that there ARE physically independent arguments. I.e. (34) was formulated without any reference to the continuity equation, only considering (1) the vertical direction of temperature gradient, (2) condensation caused by motion, and (3) gravitational expansion in hydrostatic equilibrium. But, additionally, there is an absolutely independent way of deriving (34). When the physics is right, you can come to the same conclusion from different sides. I will dwell on this independent derivation separately.
Anastassia,
You say that 34 is based on independent physics. But if so, it must have numbers associated with that physics. If the formula depends on gravity, where is g? If thermal, where are the specific heats? Or temperatures?
You say that “S must be proportional to the velocity component that is parallel to temperature gradient.” but there is nothing quantitative about what the temperature gradient is. Yet the proportionality is precise:
S = w∂N_v/∂z
All you’ve done is said that the velocity can be assumed vertical. You may say that it’s because of the layered temperature field, but that is not part of the basis of the equation. It cauld be 1D flow for any of several reasons – same equation. What it does say is that a parcel of constant volume can only change its density over a distance if it sheds mass at rate S. That is what determines the constant of proportionality – not the temperature gradient or the amount of graviry.
In math terms, if you have an increment Δz and constant w, then the nett wv flux per unit area is wΔN_v, and since water must be conserved, the volume precipitation must be S Δz. That is what gives you your first equation.
Then you correctly observe that w might not be constant. The air might be converging or diverging. So then,
S Δz = ΔwN_v = wΔN_v + N_vΔw
An extra term. How to find it? As you do, by tracking a reference gas that flows with the wv but does not change its mass.
This cannot be the air N (Nd+Nv). That is losing water. It must be the dry component N_d. The coresponding expression for it is
0 = wΔN_d + N_dΔw.
Eliminating Δw,
S Δz = wΔN_v- (N_v/N_d) wΔN_d.
and in the limit,
S = w(∂N_v/∂z – (N_v/N_d) N_d ∂N_d/∂z)
Now this is what your argument amounts to, or should. Temperature gradients, hydrostatic pressure, gravity make no quantitative contribution.
Nick, all your argument is based on the premise that
If this premise is incorrect, all your subsequent argument is moot. How do you justify it? Is this a universal scientific law?
Anastassia,
It’s you who must find the justification.
Nick tells that you have not introduced any other physical formulas than the continuity equations, twice correctly and once erroneously.
Anastassia,
“How do you justify it?”
Easily. Physics is quantitative. You always need to know, how much?
If 34 depends on gravity, then what gravity. If g does not appear, then is it in Earth? Mars? The Moon? The formula is the same in all cases. Therefore it does not depend on gravity.
Nick:
Nick, I am utterly surprised. What about parametric dependencies? They do not exist? So if there is gravity, it must explicitly show up as g? And if a formula does not contain g, it may not depend on gravity?
As I said above, total air rather than dry air is used because we assume a priori that total air conforms to hydrostatic equilibrium. Therefore, changes in density N of total air are governed by hydrostatic equilibrium condition dp/dz = -ρ g = -NM g. Using the hydrostatic equilibrium and the ideal gas law you can easily express the reference term γ ∂N/∂z via g and temperature. If you have difficulties, I’ll show you the answer.
Is your statement that
“The formula is the same in all cases. Therefore it does not depend on gravity.”
your only argument against our physical arguments for Eq. (34) or you have others?
Anastassia,
Other physical laws do exist, but you have not used them. You have taken the continuity equation once more, this time in an approximate form. Then you have required that the exact law and the approximate law are simultaneously in force and exactly. From that spurious requirement you get totally erroneous results.
“As I said above, total air rather than dry air is used because we assume a priori that total air conforms to hydrostatic equilibrium.”
I can see no relevance in that. You don’t need to use either pressure or gravity to get the missing term, which is velocity gradient. But the point is that total air isn’t non-condensable. In fact, it gains or loses exactly as much mass as the vapor component, ie S.
I think that Nick says it clearly enough, but I rephrase.
The equation (34) is not an equation that tells what the source term is. It’s just the continuity equation, only a wrong continuity equation. There’s absolutely nothing else in that equation from any physical principles.
Nick, you are concerned about Eq. (A7).
Nick Stokes | February 4, 2013 at 12:10 am |
Nick, I cannot see why you continue to interpret C as a kinetic constant, which we never did, and I clarified how C is defined above.
Anastassia,
“Nick, I cannot see why you continue to interpret C as a kinetic constant, which we never did,”
Well, you said in part A1:
“The linearity assumption is justified by the particular physical nature and stoichiometry of condensation, with gas turning to liquid: condensation is a first-order reaction over saturated molar density Nv of the condensing gas. This can be experimentally tested by considering condensation of water with different isotopic composition (e.g., Fluckiger and Rossi, 2003).”
And indeed, F&R is a place where you’ll find a discussion of molecular kinetics.
On C, you said:
“In chemical kinetics C depends on temperature and the molecular properties of the reagent as follows from the law of mass action. Since the saturated concentration N_v of condensable gas depends on temperature as dictated by the Clausius-Clapeyron law, we can ask what the proportionality coefficient C physically means in this case. Different substances have different partial pressures of saturated vapor at any given temperature – this is controlled by the vaporization constant L and the molecular properties of the substance. Note too that for any given substance (like water) the saturated concentration depends on various additional parameters including the curvature of the the liquid surface and availability of condensation nuclei.”
No mention there of ” the vertical velocity and the degree of deviation of water vapor pressure from hydrostatic equilibrium”.
But the main thing is, if your justification for S=CN_v isn’t first order molecular kinetics, then what is it? It can’t be everyday observation.
Hey, VP @ 12.22pm,
There’s a sign o the ballroom wall,
‘Sorry we don’t take checks.’
Rain-dancer, ( I also dance with wolves.)
Before addressing Nick’s points about S = 0 and rain from dry air, which are very worth addressing, I will now show how S (34) can be derived in a different way than wha we described in our paper as above.
This complementary physical picture is presented in the blog. We start from the proposition that the potential energy associated with the non-equilibrium vertical pressure gradient of water vapor is what drives the circulation. As I explained above, in a 1-D case when there is only saturated water vapor and a temperature gradient, this potential energy is very visible and drives significant air motions. We propose that in 3-D case this potential energy — if condensation is present — does not mysteriously disappear, but continues to drive the gas motion. Nobody has ever looked at this process before, this is indeed a new and testable proposition.
This potential energy released per unit time proportionally to vertical velocity determines the atmospheric circulation power q (the motor) as per Eq. (3) in the post, where the opposing effect of gravity is accounted for. Next, we assume as before that the atmosphere is in hydrostatic equilibrium. This means that kinetic energy is generated by horizontal pressure gradients only (with vertical pressure gradients balanced by gravity). Equating q to u∇p as per Eq. (4) immediately yields Eq. (37) in the paper (which is our main result). As one can see, no continuity equation has been involved.
Note that these energy considerations do not involve any mysterious small factor that is causing confusions, but unambiguously determine the circulation power as q = wp_v (1/hv – 1/h).
Now then combining it with (32)-(33) for a horizontally isothermal atmosphere we naturally obtain S (34). So S (34), originally formulated from different considerations (although again involving hydrostatic equilibrium), is consistent with independent energy considerations. This overall consistency is also described in the paper, e.g. the condensation-related force f_c as formulated on p. 1044 is equal to f_c = S RT/w.
(Note that if the atmosphere is not horizontally isothermal, S (34) will not follow from (37) (or Eq. (4) in the post). But in this case q (3) will also be diferent, because the non-equilibrium water vapor pressure gradient will not be vertical. But this is a separate story beyond the horizontally isothermal case considered in the paper.)
Anastassia,
You are deriving a formula for horizontal pressure derivative assuming that:
– the temperature does not depend on x
That’s not true, when the horizontal pressure derivative is not zero.
Your derivation is based on an assumption that contradicts is result.
Your assumption is unfounded and contradictory. Could do do anything worse on that point.
The few concrete points I have brought up are not just irrelevant nitpicking. They seem to be at the heart of your paper. There’s almost nothing left when they are taken into account.
You have accepted, in a way, one of the criticism claiming that it’s not really relevant as the whole chapter is just additional comments. Strong claims have, however, been presented in your paper based on that erroneous part.
This is probably more essential for the rest, because here you take advantage from the fact that from one contradiction it’s possible to draw anything. Forcing a non-zero to zero has such a power.
Steven Mosher | February 4, 2013 at 1:47 am |
Steven, willard never asked me any questions. Nick did, even you did even if refused to clarify what you meant, and I am doing my best to answer. But to do you a pleasure if you will formulate what the direct question of willard is, I will respond (of course, if I know the answer and if the question is pertinent to science).
The question is the main one in that sentence, Anastassia:
> Is it too much to ask if Judy’s issues have been fixed?
http://judithcurry.com/2013/01/31/condensation-driven-winds-an-update-new-version/#comment-290959
If you read my comments as you say you did, you should know that no one answered it yet. I have a feeling you know why I’m asking this: it is related to Eli’s comment a bit earlier, to which you commented yourself and against which Douglas has nothing more than diverting away with an equivocation on the word “flaw”.
For the sake of openness, I can tell you why I’m asking this question: I want to know to which extent editors courteously trampled on the reviewers’ public comments and gave you a free pass.
***
If the editors decided to disregard the reviewers’ comment and publish the paper (we can help Douglas and invoke Feyerabend’s “anything goes” here) my question remains to be answered by the authors and the reviewers.
I’m not asking you to respond for Judy, who might be too busy to respond to a question that might be lacking in substance, if we’re to believe Douglas’ judgement on substantiveness. Considering his glosses on falsifiability, others might consider that this question deserves to be answered.
***
As you can see, I can be quite forthright. Since I’m not sure my interlocutors want everyone to notice how the limits of their justified disingenuousness is being underlined, I try to remain cryptical. But when they ask for it, I will do the best I can to say what I mean in the clearest way I can.
Sorry to include some jabs at Douglas, but quite frankly, he would deserve worse than that. Since you’ve read Steve’s, I’m sure you know what I mean. His uninspired excuses are doing you a disservice. I don’t know why he would think that his tricks would escape philosophically-minded people, a cast who is basically paid to read mischievous arguments.
I’ll cast my lot with those with lots of caste to cast.
=========
Willie, me boy
Sounds like you have prepared a legal “plaidoyer” for the party that would be aggrieved by the publication of a paper suggesting condensation driven winds (whoever that might be).
Fuggidaboudit, Willie, let’s get the thing tested rather than quibbling about it.
Max
willard, thank you for your question. Our reply to Judy’s review is public, so you can read and decide for yourself.
The review has four points. Point one is addressed in our reply to Nick Stokes posted 26 January 2011. An appendix was included into the revised text with additional considerations on Eq. 34. We did not change the notations, because the physics of our approach, that is related to the ideal gas law, is best illustrated based on molar densities rather than mass densities. We generally believe that the standard notations that introduce a gas-specific constant to the ideal gas law are misleading. They are obscuring the fundamental nature of the universal gas constant.
Point two suggested an alternative between “This needs to be demonstrated either in the context of a more comprehensive scale analysis that includes the Navier Stokes equations” and “numerical model simulations using mesoscale or weather or climate models.”, the purpose of which to clarify the degree to which the effect “matters”. While the paper has been under review, two of us made two detailed accounts of condensation-induced dynamics on two drastically different spatial scales — hurricanes and tornadoes. Note that in these two papers we not only derived wind velocities, but also theoretically determined turbulent friction coefficient — something in principle unachievable in conventional simulations (where this coefficient is parameterized). The results of these papers were included into the revised text and discussed in a broader context of the other findings.
Additionally, we indicated that there is a different way of responding to the reviewer’s concern of estimating how the effect matters. A new section and a new figure (Fig. 2) was introduced comparing an effect that is known to matter (CAPE) and our effect. This is section 3.4 in the revised text.
Point three — we introduced a more extensive discussion of the ideas surrounding Hadley cell as indicated in our reply to JC.
Point four — in our response we presented a literature-based discussion defending our statement in the paper concerning the surface-specific nature of evaporation. Since the reviewer indicated that it anyway unlikely impacts our conclusions, this discussion was not included into the revised text.
Please if you have any specific questions, I’ll be happy to respond as time permits. I’ve basically done my program here — Nick has just got the idea how to (in)validate our propositions empirically, so the others will be able to do so as well. That will inspire critical and constructive thinking about our work. That has been our major goal.
Regarding Steven McIntyre’s blog, indeed it is where I learnt the blog culture (it was the first blog in my life where I commented) and I am grateful to people there from whom I learnt a great deal.
Please note that all the statements of Douglas concerning our paper represent a co-ordinated position of all the authors, including myself. If there are any particular things that especially concern you, we can discuss them and clarify misunderstandings.
On a personal note, some of your more philosophical statements are really interesting and it would be interesting to discuss them in greater detail. But as I said earlier today I do not feel your language well enough to chat in a more relaxed way. E.g. Max thinks you are preparing a legal case against us. But anyway thanks for your interest.
Thank you for your response, Anastassia.
Anastassia –
Feeling willard’s language is a challenge for all of us at times. He can be….cryptic at times.
I recommend that you take what Max says with a grain of salt.
I’m truly sorry. That’s because my comments are made like this:
It’s OK willard – but only on those occasions that you fulfill both basic requirements: that you are responding in the heat of the moment, and….
max,
Spend some time reading more of willard. read some of the source material he refers to. One minute he might appear marginal and cryptic, but after a while you can see what he sees. Resist the urge to say it makes no sense. For a long while I felt that way. Then I looked at it like a puzzle. That helped. Then I thought.. Tobis likes him. Tobis is no dummy ( although we disagree ) maybe I should push harder on Mosh to understand willard. In the end, that extra effort is worth it. And no that does not mean I agree with everything he says. But rather, that what he has to say is worth the effort to think first, read more completely and ask clarifying questions before going off half cocked. Put another way,
I always check the comments links to see if he has commented. Always a good bet. Not a sure thing, but a good bet.
Steven Mosher is right Max. Willard’s posts are always worth checking out, He is IMO a genuine sceptic who holds both sides of the debate to account.
Thanks.
We’re all OK.
Even these two Cryptics:
http://rabett.blogspot.ca/2013/02/on-priors-bayesians-and-frequentists.html
Authors disagree about the authenticity of the dialog.
Anastassia, I posted something yesterday under Nick’s comment about Sec 4.1 showing that your claim that condensation and a horizontal pressure gradient sustain each other can be proven exactly from the thermodynamics in Section 2.1: (also in your mail)
If you replace in your (33) N_v by N_v/N_d, u by N_d*u and v by Nd*v, you get: ∇(N_v/N_d) . (u,v)= S/N_d, which has characteristics
dx/dt= u,
dz/dt= w,
d(N_v/N_d)/dt= S/N_d,
with N_v/N_d= γ/(1-γ)
Now I tend to agree with you about the nature of the “source” term: it should not be seen as an independent physical source term, but simply expresses what should happen according to the relationships in Section 2.1. That means that along the characteristic,
d(N_v/N_d)/dt= w ∂(N_v/N_d)/∂z,
so
S= w N_d ∂(N_v/N_d)/∂z = w N (1-γ) ∂(γ/(1-γ))/∂z
and so you also have your compensation for overall volume expansion right there.
I think that is it, nothing new or unphysical. So you are right, your equations in 4.1 are pretty close anyway, but the one above is rigorously derived.That saves a long Appendix :).
Your claim proven as a theorem now means basically (with some radiative cooling at the top for the return flow) that it does not need confirmation by numerical modelling, but rather, application, to see how it plays out in real atmospheric problems.
If I need to explain this further please let me know
Cees, please see my comment above with the justification of the S term.
The justification has already been proven wrong.
Cees,
The issue is not whether they are close (34 and 32/33), but whether they are independent. because that is how they are treated, 35 and 36 are just linear combinations of them.
So what can we conclude from the above.
The paper studies certain previously extensively studied processes of the atmosphere taking little advantage from the earlier literature. They start from the correct fundamental equations of physics but without specifying initially the overall physical setup. The first part of the derivations are done correctly but a few strange sentences are included, and one of them is used as a strawman argument against criticism of Held.
Next they make an artificial comparison that’s totally irrelevant, but used erroneously to justify the importance of the effects they study.
Then they write three equations, the third of which should be derivable from the previous two, but that one is written erroneously and is therefore in contradiction with the previous two. This contradiction from the error of the third equation is used to derive results that are essential for the continuation.
(There’s one thing that I’m personally happy. Before going trough the paper in any care I made intuitive conclusions on what kind of errors the paper must have. Those inclusive conclusions are now proven to be correct. It’s nice to get confirmation that my intuition still works.)
This excerpt from my earlier comment tells perhaps best about my reasoning and role of intuition in that:
Based on more traditional ways of looking at the same physical situation I was virtually certain that their results were seriously wrong. That made me to speculate on the possible nature of the error in the paper. The most likely explanation was a erroneous assumption on the role of “horizontal” and use of that to get a spurious constraint equation.
That’s exactly what the set of equations (32), (33), and (34) does. There’s an extra constraint related to handling horizontal differences versus vertical ones that’s created by using simultaneously all three equations. That’s just of the nature I speculated about in the excerpt.
Nick Stokes | February 4, 2013 at 12:10 am |
“Nick, “nutty results” and “mathematical contradiction” is not the same”
No, Nick, the difference between mathematics and physics is not semantic. Mathematical coherence is necessary (but insufficient) for a theory to be valid. If there is a mathematical contradiction, i.e., say your system of equations yields x = 0 AND x = 1 at the same time, there is nowhere to go from here. As I already said above, publishing such a paper would have been of course a bad editorial mistake.
If, on the other hand, the system of equations is mathematically coherent, and unambiguously yields x = 0, then you can make the next step and try to verify your theoretical result by empirical evidence, i.e. to see if x is indeed zero.
I just want to emphasize once again that the system (32)-(34) is mathematically coherent. Your claims about the system being overdetermined or there being any mathematical contradiction are unsubstantiated.
The equation (34) is nothing else than an erroneous continuity equation.
If you disagree tell what is the alternative justification that you have. The one that you have given so far confirms my and Nick’s interpretation.
No, 34 is just a version of 32/33 with some terms (horizontal components) omitted, with the argument that they can be neglected against the main terms. There’s a mistake, but i’s a distraction to dwell on that. I’m not sure if dropping those terms is justified, but this is commonly done in continuum analysis (when justified). What is never done is put the original equation together with its simplified derivative into the same set of equations.
The dropped terms were dropped because they were small relative to the main terms, not because they are known to be absolutely zero. There is no justification for such a belief, but keeping both equations asserts it, because the equation and its simplifier can be subtracted. And in your case, that becomes explicit with equation A7. And the results following Eq 36 reflect that.
Try to follow my comment Nick (Feb 4 2:28 above) and you see that something almost equal to Anastassia’s result follows exactly as a consequence of mass balance and the assumption of saturation (which fixes the vertical gradient of N_v/N_d). So her result is correct to a very good approximation and is easily made exact. You just can’t avoid this result.
And as you see, I used your suggestion for looking at N_v/N_d :)
Her S is just a term which can be used to match the mass balance to the constraint imposed by saturation, which fixes the vertical gradient of N_v/N_d. That is all there is to it.
Cees,
Yes, your formula is exactly the same as the ,a href=”http://judithcurry.com/2013/01/31/condensation-driven-winds-an-update-new-version/#comment-291818″>one I derived. It is a consequence of conservation of mass, and could have been derived directly from 32 and 33 by dropping horizontal components.
The issue is, having simplified 32/33 to 34, you can’t use both. In fluids, if viscosity is unimportant, you can solve with the Euler equations rather than Navier-Stokes, dropping the viscous stress term. But you can’t use both.
one I derived. – broken lik fixed
Cees,
The derivation, in better notation, direct from 32/33 was given in my first comment in the ACPD discussion. It’s Eq 5.
SM 1.14pm Carnap and cars? Don’t drive a Volvo meself, prefer a
souped up Ford. Re my comment, SM, I didn’t say that for for
‘something (a process,situation?) to be true it has to be empirically falsifiable,’ What i said was that for a ‘statement to be true it has to correspond to reality and this means being subject to a process of
verification /falsification. Hmm, verification?
As there’s a serious discussion going on here I’ll have a go at
responding re logical positivism in me simple fashion anon on
‘Open Thread.’ Sm. Bc
Anastassia,
your last reply refers to a lengthy physical explanation of what the source term should be. That is all fine, but if you follow my derivation, you see that your explanation only gives an approximate expression, whereas the one I gave is very simple (just the geometry of the hyperbolic equation), and gives an exact result and not an approximation; the change in time of N_v/N_d along the characteristic CAN BE NOTHING ELSE then w times the vertical gradient in N_v/N_d: the latter is already fixed because of saturation, so it is there already and you just read it off along the characteristic. This argument supersedes all other considerations you have come up with. So it is really very simple.
and I am not saying this to suggest that there is anything trivial about your results: on the contrary, I think it is profound and original and may have a big impact on a lot of issues in weather and climate; having looked at it in a little more depth only makes it better. So thank you for posting and discussing it here.
On this point I disagree totally.
The paper makes an explicit error in the derivation of formulas (36) and (37) when it requires that the exact continuity equation and the approximate continuity equation are simultaneously valid. That requirement leads here to results that are fully spurious.
There’s nothing of interest left, when the erroneous derivation is removed.
Heh, nothing of interest. Only a little physics. pfah.
=========
The paper does not tell anything that’s both new and valid about physics.
We shall observe.
=======
Well, no,I was wrong here.
I just want to go over some basics as I see them.
H20 gas condenses on some kind of surface- H2O gas doesn’t just condense with other H2O gas. Though H2O gas condenses on liquid
H2O.
All gases of gas mixture can be regarded to have a partial pressure
or vapor pressure. So what control evaporation of water is existing
vapor pressure of the air in contact with the liquid water and the temperature of the air. Wiki:
“The vapor pressure that a single component in a mixture contributes to the total pressure in the system is called partial pressure. For example, 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.”
So if pressure is constant it, the partial pressure of H2O could increase, but also temperature in constant, a increase in pressure also corresponds
increase in partial pressure [or decrease in pressure, decreases H20 partial pressure.
Or if had atmosphere of pure H2O gas, 1 atm of pressure requires temperature of 100 C- or if the gas was cooler 100 C, H20 gas does condense with other H20 gas molecules. And lower pressure say 1/2 atm pure gas H20 gas would condense at lower temperature.
H2O at above -150 C will evaporate into a vacuum or in atmosphere which has zero partial pressure of H20 gas.
The water phase diagram:
http://en.wikipedia.org/wiki/Phase_diagram
At somewhere around 23 mbar if you had atmosphere of pure H20 gas and
it’s at 20 C, then H2O gas could condense with other H20 gas. But in gas mixture it doesn’t- it need a some liquid or solid to condense to.
Or at least that the way I understand it.
Though these gas molecules “try” to stick together, but the energy of all the other types of gases disrupt this from happening. Or H2O gas is not ideal gas because instead bouncing around freely it’s sticking to other H2O molecules and being forced apart by other molecules of gas.
But if somehow enough H20 gas molecules become a liquid droplet of water there some critical amount of molecules- as wild guess, say 1000 molecules formed as liquid water- in which H20 gas can more than just briefly condense onto such a droplet of water.
So if water molecule formed with say 5 molecules of H2O liquid droplet, within some period measured in nanoseconds, and if this other gas molecule condensed making 6 molecules, than they may fly apart within nanoseconds or less than a second. But if one molecule of gas join with droplet of 1000 or more molecules of water then it might not evaporate for many seconds- a water molecule and many water molecules may stay as water molecule for a long period time.
Or if the there enough water molecule in droplet the water acts similar to a drop water from dripping facet or pool of water- which a portion of it is evaporating and condensing- but with H20 gas molecules in a gas mixture most of gas molecules are transiting from gas to liquid state within a time frame of something like less than a second.
So start with water surface. 1 mm above the surface there is H20 gas.
Some Water molecules from liquid are becoming a gas and lasting in this gas phase lasting for less an 1 second before returning to the liquid, some portion of this H20 gas gets beyond 1 mm, they may go to 1 cm, or 1 meter or 1 km [etc] high without returning to the liquid water surface, and have zillions of transitions from gas to liquid state, before become a rain drop and falling into the ocean.
And so water droplets starting with 2 molecules of H20 to billions of molecules of H2O.
A Rain drop 3 mm diameter: “4.716278×10^20 molecules”:
http://www.had2know.com/academics/how-many-molecules-drop-water.html
Now, a rain drop remain buoyant until it reaches a certain size, and cloud with updrafts can have larger raindrops kept buoyant and rain rain may increase it’s size falling.
Now a raindrops without updrafts which comprised of “4.716278×10^20 molecules” is going to fall, but smaller droplets are going float without
updrafts. Fog is can be not fall much in quite still air and it’s small droplets of water- 1/1000th to 10th mm diameter range.
And 1/1000th mm is about 1.0 x 10^10 molecules.
So roughly if less than 1 trillion molecules it will be buoyant in air.
And somewhere less the 1 million molecules they will not be very visible or noticeable.
And these smaller droplet aren’t going last very long- they will have very fast evaporation rate in dry enough air.
So lots of them near ocean surface. If live near an ocean you get salt corroding everything- so have these water droplet- not formed from condensation [that would be pure water and not a problem] but wind and waves mechanically making these droplets, plus high humidity of general environment preventing them from evaporating quickly.
Now read stuff about this salt from ocean is suppose to have something to do with cloud formation, but my experience is you have to be fairl close to ocean to have this corrosion issue and and can’t imagine much salt getting above 1000 meter in elevation.
And trying over how water vapor condenses, but this getting long, so
anyone got any answers, that would be good.
Gbaikie,
You are right in that water does not condense in pure air immediately the saturation point is exceeded. Some supersaturation is needed before droplets start to be formed. When the relative humidity exceeds by more than 10% the saturation level the condensation proceeds rapidly under ideal conditions. In less ideal conditions that occurs earlier.
In ascending convection of the real atmosphere the supersaturation may be as high as a few percent, but rarely more. That’s largely due to the presence of aerosol impurities.
An old paper about on homogenous nucleation:
http://puhep1.princeton.edu/~mcdonald/JEMcDonald/mcdonald_ajp_30_870_62.pdf
Nick, I see, so we agree what it should be. Nevertheless, I the whole point of Anastassia was to demonstrate that in a 2D motion, saturation is proportional to a horizontal pressure gradient times a factor relating horizontal and vertical scales of velocity. I think this point is indisputable. And her approximation is rather accurate anyway. So we are really talking about finer details. It should not be the case that these prevent publication of an overall very good and original paper with potentially a high impact.
“And her approximation is rather accurate anyway.”
Let me go back to my earlier analogy. Start with an equation that says
x=0.9999. So far so good. Then make some simplifications. You get x=1. That is rather accurate anyway.
So put them in your equation mix. We can subtract: 0=.0001. Well, OK. Then multiply by 10000 – you can always do that. So 1=0. Getting a bit nihilist…
That’s what happens when you solve an exact expression together with a rather accurate approximation.
Or even worse if you manage to get the difference between the exact value and the approximate value in the denominator
1/(1-0.9999)=10000
That’s what occurs in the paper for some quantities.
No the point is based on a spurious constraint and not valid at all.
Cees, thank you very much for your support. But I am sorry to disappoint you — if you use your equation instead of Eq. (34) and solve it together with (32)-(33) and ∂N_v/∂x = 0, you will get a physically meaningless result:
u∂N/∂x = 0.
Mathematically it is fine. But it means that over an isothermal surface the winds never blow along the pressure gradients. This is certainly not true, as in hurricanes for example radial convergence is responsible for all precipitation. This shows that your equation and Nick’s, and the logic behind your derivation, is physically incorrect — irrespective of whether Eq. (34) is correct or not.
Your mistake (as well as Nick’s) is that you disregard the importance of hydrostatic equilibrium in determining the change of total air density and believe that Eq. (34) is somehow related to (32)-(33). It is not. It does look similar, but there is a profound difference.
Note also that here there is an independent derivation of (34).
” if you use your equation instead of Eq. (34) and solve it together with (32)-(33) and ∂N_v/∂x = 0, you will get a physically meaningless result:”
Indeed you do. It isn’t independent. And adding in a simple error doesn’t make it more meaningful.
Anastassia,
You continue to attempt making something from nothing. You make unjustified assumptions and derive spurious results from them.
Separating vertical and horizontal cannot be done like that. You must present full equations without unjustified assumptions that any particular variable does not depend on x to get correct equations. Your results are spurious consequences of failure to do that.
You are also totally wrong in claiming “Nobody has ever looked at this process before, this is indeed a new and testable proposition.” as that has been done more correctly in every standard derivation of moist lapse rate. There are approximations in those, but not errors like in yours.
Nick Stokes | February 4, 2013 at 7:30 am | Reply
Nick, what you are talking about is a mathematical contradiction. As I repeatedly said in this thread, the system (32)-(34) is coherent. You cannot derive anything like 1=0 from it. So this claim that you are perpetuating is incorrect. I said it several times, but you were changing topic — see here, here and here — instead of conceding this obvious point.
If you are able to produce anything like 1=0 from (32)-(34), please do. I am sure that people will appreciate it. If you cannot, why to repeat this unsubstantiated claim?
Anastassia,
Cees and I have independently derived the correct form from Eqs 32/33. Pekka agrees. Isaac Held derived the same form (Eq 1). It’s elementary. Yet you insist that we’re all wrong and your very similar form is the only one that is right, because it uses different physics. Yet you cannot point to any quantitative contribution that that physics makes. Your explanations are arm-waving.
Nick, you have changed the topic once again and have not replied to the mathematical contradiction issue.
I not merely insist that you are wrong, I give a very clear reason — your derivation is wrong, because it produces a physically meaningless result,
namely -u∂N/∂x = 0. So, whatever you think of my derivation, yours is incorrect. (Dr. Held referred to your comment on this issue, by the way.)
Ok, you say that my explanations are arm-waving. But you did not show any error in these explanations. You just said that it is wrong because it does not contain gravity. And added that you see “no relevance” in our derivation. That sums up your critique so far. But to “see no relevance” and show an error are different things.
Your own derivation of S is incorrect, as shown by the result its produces, u∂N/∂x=0. You see, it is not just “no relevance” or accusation of “arm-waving” from my side, it is simply a statement that contradicts the evidence. Our Equation (34) produces a result that makes a lot of sense;
-u∂N/∂x = S.
It is also empirically testable — and has been tested and gave meaningful results. Besides, it is derived independently from two independent set of physical considerations here and here.
If you continue to ignore this evidence, there is little difference from arm-waving.
Anastassia,
There are, indeed, fully obvious explanations for reaching the equation (34) as an approximation. I cannot imagine any connection with an equation of that form to anything else than the continuity equation. You have not given any plausible derivation for that equation from physics beyond the continuity.
That’s one point, but there’s also the second point that you must justify fully every claim that some partial derivative over x is zero. You have not given proper justification for the assumption that N_v does not depend on x. That would be the case if nothing else would depend on x either, but then pressure would also be independent of x. Now you assume that the partial pressure of dry air component does depend on x but that of vapor does not. That could be explained if the temperature would not depend on x, but having a different overall pressure means that you lose all justification for that assumption as lower pressure means more adiabatic expansion and that means a lower temperature.
You derive results that prove your assumptions wrong. Your derivation is contradictory and worth nothing.
Anastassia,
If you want a contradiction, here it is. In the notation of A7, our form is S=S_d. Your A7 is S-S_d=γ_d w ∂N_v/∂x. Now in deriving 34, you set horizontal derivatives to zero, including N_v. So S=S_d. We’re right! That’s a contradiction, isn’t it?
You say we’re wrong, despite not being able to say where (three independent derivations). You say so because it leads to a ridiculous result. I wearily point out that that is exactly what Pekka and I have been saying. Using 34 with 32/33 willlead to ridiculous results. That’s the point of my x=0.9999 analogy. x=0.9999 isn’t wrong. x=1 isn’t wrong. But together they lead to a ridiculous result..
You just add in an obvious error to get more complicated ridiculous results.
Anastassia,
Nick’s derivation does not produce such a result in general settings. The derivation is also just valid mathematics. If it does produce such a result in your case that only proves that your case is wrong.
Anastassia, perhaps you should look more closely at the meaning of isothermal in x. If X is very large, winds would go to 0, there should be a range of X that allows maximum pressure differential.
I’m afraid I did not read the whole 4.1 yet … Well at least N_v/N_d must vary with x in some way for (34) to be right. I need to look at this section again and read all what has been written here before I can make sense of it.
quote
[] looking at how rainfall varies as we travel inland from the coast (over relatively flat terrain): Why does rainfall not decline over forest? It declines over non-forest in a relatively constant manner that is easy to understand (This seems to be a global pattern: see the figure in my previous blog here http://judithcurry.com/2011/03/30/water-vapor-mischief-part-ii/). Recycling is not an explanation – it would reduce the rate of decline but it could not prevent it. There is no alternative explanation at present.
unquote
Dr Curry,
I believe this is within your area of expertise. A (large?) proportion of CCNs are derived from the oceans — sulphate from DMS and salt from spray/bubbles. Are there measurements to quantify CCN numbers wrt distance from the coast?
Over forests there may well be extra biologically-produced aerosols (B. Nozière had a paper recently about the Amazon). If inland clouds are deficient in aerosols, my naive expectation would be that precipitation rates reduce, but near forests which are adding extra CCNs then it would not.
JF
‘The sun is very sultry and we must avoid its ultry-violet rays.’
H/t N. Coward perambulating through Plum’s Orchard strewing sulphate CNNs.
========
An’ I wanna gracias to Anna v for insights here.
================
Nick Stokes | February 4, 2013 at 8:39 am |
Nick, it is your second mathematical error made in this thread.
(A7) is not S-S_d=γ_d w ∂N_v/∂x.
(A7) is S-S_d=γ_d w ∂N/∂x.
So your contradiction does not exist.
Yes it does. You put all horizontal derivatives to zero.
Actually in the paper it is S-S_d=γ_d w ∂N_d/∂x.
Nick, why not to be clear? You’ve made an error, concede it. If you can derive a formal contradiction from (32)-(34), please do. You’ve tried twice alreadyy.
This is incorrect.
I do not even try to find errors in your derivations. A priori I am absolutely open to accept your derivation and discuss it. But when you produce your result, -u∂N/∂x = 0, it is nonsense. Everybody, including yourself, agrees that it is nonsense. So your derivation is incorrrect.
You, instead, are trying to question my derivation without understanding its physical content. It does not seem to you relevant or convincing, fine. Like me saying nothing about your derivation, you are not able to say anything wrong about mine, just unsatisfied with it. Fine, I leave it to you. But unlike your nonsensical result, the result I produce has a lot of sense and quantitatively conforms to observations:
-u∂N/∂x = S.
The error is in your paper, not in Nicks’s derivation.
-u∂N/∂x = S is not shown to be valid, when ∂N_v/∂x = 0. That’s a spurious result that seem to result from the other errors of the paper.
Well, let me wearily put it another way. Is u ∂N/∂x=0 a ridiculous result? At one level yes. You are waving it as a contradiction that nullifies our correct and orthodox algebra.
But your similar result S = u ∂N/∂x? That brings rain out of dry air? That is nuts, but you insist it is not a mathematical contradiction. So you’re fine. But how is ours different again?
But what is happening here is the following. As Pekka and I have said over and over, if you have an equation 32/33, and strip out a few terms (horizontal derivatives) to get an approximation 34, at least our correct version, that’s OK. But if you use both equations together, then you can subtract them and get that the difference, the terms you stripped out, are exactly zero.
u ∂N/∂x was one of those terms. You are insisting that the other derivatives really are zero. So u ∂N/∂x =0 is exactly the product of combining the equation that had it with the approx version that doesn’t.
If you’ve assumed all horizontal derivatives are zero, you can’t complain when the algebra throws that back at you.
OK, your dry rain result was S = u ∂N_d/∂x
The equation -u∂N/∂x = S is nonsensical also in the way that we know by certainty that in the middle of a wide column x-derivatives are zero. Thus there would be no precipitation in the middle of a wide rising column of moist air.
Demarcation in science is falsification of hypotheses through tests.
As David Springer is wont ter say, ‘Write that down.’
Logical Positivism is about a criterion of meaningfulness for
statements, can you ‘verify’ yr statement?’ Trouble is you can’t.
Hume said ‘No way, no matter how many times yr try, plus one fer
effort but ‘fail.’ Nassim Taleb wrote a book about black swans, no
unicorns though. Logical positivism jest led nowhere.
> Logical positivism jest led nowhere.
Neither did falsificationism.
Perhaps you’d trust an Aussie on that one:
> Sorry, Pop, but Popper’s not popular in philosophy.
http://neverendingaudit.tumblr.com/post/3252880609
Interesting to see that philosophy has become a popularity contest – I admit to not being aware of these developments
(I know I know not substative …)
Popper is worth reading, Douglas, but like it happened with Hegel for Russell, it’s a good idea to snap out of it.
To take another Aussie’s:
http://www.utilitarian.net/singer/by/19740502.htm
Willard
Thanks. Genuinely interesting. (We did Popper as undergraduates though most of what I have read is second hand). The value of trying to falsify predictions remains a valuable principle of scientific progress in my books — especially if can be done without bias and prejudice (quite a lot that we could discuss). We could talk about paradigms etc. too, but really I was trying (am trying) to turn from theory to empirical here. It was an invitation.
As Willard notes beth Falsification led nowhere also.
Let me see if I can explain through a series of examples.
you know the law of gravity. If not, jump up. we will wait, you’ll come down.
So, lets suppose that you decided to test that old law of gravity. And you dropped an apple and it did not fall. What can you conclude?
You have three choices really.
1. Conclude the law of gravity is false ( falsified )
2. Question, your data.
3. Revise your theory.
The very fact of disagreement between hypothesis and observation doesnt tell you which is wrong. Historically, in fact, you can find many examples of scientists ignoring ‘falsification’. And they ended up being correct.How? and why? Well fundamentally because you cannot test a theory in isolation.
The decision about which choice to take (1,2 or 3) is fundamentally a pragmatic decision. 1. How important or central is the theory to our understanding, that is how much other science rests on it. 2) how easy is it to simply repeat the test, and 3, what kind of auxilary hypothesis need to be added to explain data that doesnt exactly fit.
The notion that there is a simple way to test theories in isolation from all other beliefs doesnt really hold.
Lets take an example from climategate, which I detail in the book.
Briffa found tree rings that diverged from following temperature.
He had three choices.
1. Call the data bogus ( hide the decline )
2. Challenge the very science of tree rings.
3. find an auxilary hypothesis to explain the freaky data.
Briiffa Did #1. Skeptics claimed number 2.. and recently two seperate groups have explained the divergence, saving the data and the larger theory. Science does not work by falsification. observation that disconfirms a theory tells you nothing in and of itself EXCEPT that you have a problem:
1. wrong theory
2. bad data.
3. incomplete theory.
Which one you choose is not pre ordained. There is no rule book, no “method” that tells you what to do in advance. the decision is situational. the decision is pragmatic. The fact of disagreement between theory and observation tells you nothing, it merely presents you with a choice.
your next reading would be in foundationalism in epistemology.
Also have a look at the Moorean Shift.
Translation – You can’t falsify CAGW. We have spent years being condescending to, and impugning the integrity of, anyone who disagreed with us. Our egos could not take the hit of being so wrong on what we have made the most politically important scientific issue of our day. Therefore there is nothing that could convince us we were wrong…I mean falsify CAGW.
Simples.
Please no one has shown any fault in the maths or physics. Roger suggested it was different in very systems – but that was specifically excluded to concentrate on major systems. Pekka is going wit his gut. Steven is demanding that it be numerically modelled before it becomes real for him. What did we do before computers?
I quite agree with the Chief’s statement above.
2 years ago I did a quite extensive review of the maths in the Anastassia’s preprint.
Independently N.Stokes did a similar analysis and we came to a similar conclusion.
After having studied the published paper, I see that the fundamentals didn’t change so the analysis from 2 years ago still holds. As I have seen several rather wild and unsubstantiated statement about the mathematics, I would remind the conclusions.
Everything happens in the section 3.; For commodity of blog posting I will use “d” instead of the partial differentiation symbol.
1) The system is NOT closed.
– The equations 32 and 33 are just continuity.
– dNv/dx = 0 follows from the isothermal hypothesis
– N=Nv+Nd is a definition
– Equation 34 is phenomenological and at the core of the whole paper
That makes 5 equations and 6 unknown functions (u,w,Nv,Nd,N,S)
The system is not closed and admits an infinity of solutions.
It is obvious that the system must be indetermined. If we used a source term in Navier Stokes as unknown, the N-S system would also become undetermined and in practice when there is a source/sink in N-S it must be prescribed.
2) Navier Stokes closes the system
2 years ago I suggested à Anastassia that she adds N-S to close and solve numerically the system. Unfortunately this has not been done.
The reason why N-S closes is that we add 2 equations for the 2D case considered here and one new unknown function, the pressure.
So we finish with 7 equations and 7 unknowns what gives a closed system.
3) The system is consistent.
Both the closed system with N-S and the indeterminate system in the paper are consistent. Beside that, all 5 (or 7 with N-S included) equations are independent. More specifically the equation 34 can’t be deduced from other equations.
From the mathematical point of view, there is no problem with the paper. Closing the system by introducing N-S and solving numerically for the 2D case would demonstrate existence and properties of the solution (and of S). This would also close any discussion about the mathematical side of the paper which is clearly correct.
From the physical point of view the only possible discussions are :
– equation 34. If it is considered as a reasonable approximation of the condensation dynamics then all conclusions follow. Equation 34 can also be validated (or refuted) empirically. To my knowledge this has not been done yet.
– the isothermal hypothesis linked to dNv/dx=0. This is lies at the core of R.Pielke’s comment. Clearly as long as we consider a 2D model what is the case here, this condition can be imposed and is justified for the idealized atmosphere. The real atmosphere doesn’t obey (strictly) this constraint and it is not clear how far the idealized solution approximates the real behaviour. However it is safe to say that most of the “critics” of the paper here have not a clue about the answer to this question either.
For the casual readers not very deep in fluid dynamics, I advice a rule of thumb. If a poster doesn’t deal with these 2 points, then he either didn’t understand the paper or he is only mudying waters by being intentionally off topic. You can and should skip such posters.
The first point, that Eq. 34 can be validated or refuted empirically was true two years ago, too. So, let’s get on with it.
=============
Yep, that is about the size of it.
Tomas,
In principle the equation (34) might happen to be true for some unknown reason, but assuming that something does not depend on x and then deriving a result that’s not consistent with that assumption ca never be valid. Thus the derivation of equations (36) and (37) is guaranteed to be invalid.
It’s also not reasonable to write a paper that’s supposed to derive results from theory and do that in a way dependent on accidental validity of a formula that has not been studied.
“More specifically the equation 34 can’t be deduced from other equations.”
I, Prof Held and others have shown by simple algebra that an equation very like 34 can be deduced from 32 and 33. You say that AM’s 34 should be regarded as empirical. I suspect that demonstrating hers as distinct from the cons mass consequence would be difficult.
The dispute isn’t much over the correctness of 34, but whether it is independent from 32 and 33. If it isn’t, you can’t use both. And if it can’t be distinguished from our dependent solution, that is also unpromising.
“This would also close any discussion about the mathematical side of the paper which is clearly correct.”
What do you think of the proof of the claim that S=CNv? ie precipitation proportional to humidity?
“Equation 34 can also be validated (or refuted) empirically. ”
Given the messiness of the atmosphere and our inability to model from strict physics principles, it can only be validated or refuted empirically.
Two issues: Do we observe that condensation produces winds and ground level cooling.
Does equation thirty four describe condensation producing winds and ground level cooling
From that equation we can derive the result that no rain is possible in the middle of moist uplift and many other obviously wrong results.
We have made observations that do not agree with such predictions.
I will herewith correct the typos Tomas –
Please no one has shown any fault in the maths or physics. Roger suggested it was different in very small systems – but that was specifically excluded to concentrate on major systems. Pekka is going with his gut. Steven is demanding that it be numerically modelled before it becomes real for him. What did we do before computers?
I realise that I mistyped what could change the sense of the comment concerning the isothermal hypothesis for some. Of course the right sentence is : “This is at the core of R.Pielke’s comment”
Dear All,
As many of you got to know having read our paper and this thread, the main result of the system (32)-(34) and horizontal isothermy, ∂N_v/∂x = 0, is:
-u∂N\∂x = S.
Please see the algebraic describtion of the system here. Here S is condensation rate, u is horizontal velocity, and N is air molar densityy.
In the vector form and going to pressure p = NRT this main result is
–u.∇p = SRT.
Please note that in its differential form it is an extraordinarily strong statement. Being so strong, it can be easily used for empirical validation of the underlying physical ideas. It also allows one to judge about the relevant spatial scale of condensation-induced dynamics. What does it mean? It means that if we know condensation rate S from independent considerations (e.g. observations), we know u.∇p and vice versa.
We have shown that integrating this equation globally, using the observed mean global precipitation, produces the observed value of the global circulation power. So in its integral form it does match the real world. The same is true for hurricanes and tornadoes.
But what about local scale? For example, we know that S = 0 where the air descends (e.g., in anticyclones or at 30 lats). This makes a very strong prediction: where S = 0 (no convection), the winds must be geostrophic or the pressure gradient must be zero.
(Of course in the real world zero is a relative thing, so it should be compared to what happens where S is not zero).
So, for example, S = 0 predicts negligible pressure gradients and wind velocities in anticyclones compared to cyclones. It predicts geostrophic winds in the descending branch of the Hadley cell etc.
But an even more interesting case is where we know a priori that u.∇p = 0. This condition can be unambiguously derived from circulation geometry. For example, both radial velocity and pressure gradient are zero in the center of the hurricane. Thus, our equation predicts that in such a point S = 0. It thus predicts the hurricane eye, where the air must therefore descend.
Likewise, where the two Hadley cells meet along the equator, both u and ∇p = 0 from obvious geometrical considerations. Thus our equation predicts S = 0 in a long narrow zone (something similar to an eye) and the existence of a streamline discontinuity (as per hurricane eye) in a narrow zone near the equator. This may help explain the observed narrow equatorial westerlies.
I forgot to mention that the above comment addresses Nick’s concerns about “rain from dry air” or ∂N/∂x=0, or S = 0, which I promised to address above.
“This makes a very strong prediction: where S = 0 (no convection), the winds must be geostrophic or the pressure gradient must be zero.”
This is indeed a strong prediction. S=0 means zero precipitation, not zero convection. There are large parts of the world where it is almost always not raining. So it says that there the winds must be almost always geostrophic or the pressure gradient zero. That should indeed be easy to test.
Just a further comment on the nonsense of this result. Despite some belief here, this paper is not based om meteorological observation. It is a theoretical paper based on the properties of gases. And there is nothing special about water. It would apply to any condensable gas.
The formula in its original form was u ∂N_d/∂x=0. N_d is dry air, so no moisture need be present on the RHS. It’s equivalent because it is assumed that ∂N_v/∂x = 0, and N=N_d+N_v.
But again, this describes the behaviour of gases generally. If no water need be present, then what is S the condensation of? It could equally be alcohol, sulphuric acid etc.
Anastassia,
You have still not presented any derivation of (34) from something known.
You have still not explained, how you justify the assumption that atmosphere is isothermal horizontally, but allows for horizontal pressure gradients.
A more natural situation is that the temperature drops with the pressure as it does always in adiabatic expansion and that both the pressure and the partial pressure of vapor vary horizontally when pressure varies horizontally. You cannot just assume that this is not the case. You must give solid justification for this unlikely situation before you can make such assumptions (which are in reality certainly wrong.)
“Thus our equation predicts S = 0 in a long narrow zone (something similar to an eye) and the existence of a streamline discontinuity (as per hurricane eye) in a narrow zone near the equator. This may help explain the observed narrow equatorial westerlies.”
What on Earth are you talking about? In the equatorial trough of low pressure one finds either doldrums or converging EASTERLY trade winds.
Hmmmm. Does altitude matter?
========
Season matters. The easterlies they mention are more common in winter which tend to shift the Saharan dust flow south of the equator. In summer the dust tends toward the north, hopefully, reducing hurricane impacts a 25N 81W :)
Kinder fun goo-goo-googling the Bodele Depression.
============
It’s a good thing there isn’t a large population just east of that Depression susceptible to the siren song of wind farms. The rain forest in South America would have shriveled short of the Brazilian Air Force doing Air Control & more over the farms.
===============
Right, Cap’n, that dust fertilizes North America too. So, USAF for air control, and rain forest natives for the dirty work on the farms.
============
@John S: What on Earth are you talking about? In the equatorial trough of low pressure one finds either doldrums or converging EASTERLY trade winds.
Easy mistake for theoreticians to make. I would have made it myself in responding to Nobre earlier had I not taken the precaution of writing “westward” instead of “westerly” so that there would be no confusion (hopefully). Similarly care is exercised in the (long) first sentence of Wikipedia’s article on tropical waves.
As best as can be surmised from her words, Makarieva’s mistake is not simply a matter of nomenclature. She is talking, it seems, not about transient tropical waves, but about the seasonally persistent SW monsoon, as if it were driven by the trough paralleling the equator.
John S. | February 4, 2013 at 8:23 pm | Reply
AM” This may help explain the observed narrow equatorial westerlies.”
John S. | February 6, 2013 at 10:10 pm | Reply
Thank you for your comments (also the one below the comment of Roger Pielke). I am talking about equatorial westerlies.
@AM: Thank you for your comments (also the one below the comment of Roger Pielke). I am talking about equatorial westerlies.
How do equatorial westerlies support the theory that Sahara nutrients are being blown into the Amazon? I would buy that India is benefiting from them, but the Amazon? My brain is melting down.
My brain would recover if it turned out that Russian sailors mean something different by “westerlies” than those who sail in the Atlantic.
The dated AMS glossary definition of “equatorial westerlies” that you reference is hardly unequivocal, indiscriminately mixing synoptic -scale and climatic-scale features found near the equatorial trough—which is as persistently a barotropic zone as can be found on Earth. While the mechanism that occasionally develops synoptic-scale westerlies may perhaps involve contributions from intense condensation events such as cyclones, the recurved trade winds that form a seasonally persistent SW monsoon in many NH regions are the product of intense continental heating during the boreal summer. (This heating can be concealed In tropical rain forest station records by the fact that near-ground temperatures are typically several degrees cooler than those at the top of the canopy.)
Certainly in West Africa the persistent Saharan low cannot be attributed to condensation. The SW monsoon winds along the Guinea littoral are almost perennially at maximum development in August; deep convection and preciptation, however, peaks in the spring and fall transition months, with a local minimum in August (locally called the “little dry season”). This persistently mismatched timing pattern argues strongly against condensation being an important climatic factor in wind development there. Similarly, ordinary thermal considerations adequately explain near-equatorial winds in the Pacicfic Although I’m little familiar with conditions off the Indian Ocean, I would be surprised if anything radically different occurs there.
John S. | February 7, 2013 at 8:26 pm |
Thank you for your detailed response. It is precisely this type of considerations that will help identify the importance of our mechanism in wind generation. As I noted in response to Berényi Péter, the interpretation of condensation-induced dynamics as a rule of thumb ““where precipitation is high, wind is strong” is not 100% precise. Instead, condensation rate S determines the product S = –u∇p in a condensation-induced circulation. But since any circulation has a downdraft part where S =0, for that part our equation predicts that u∇p = 0. This means that either winds are negligible, and/or winds blow along the isobars, and/or the pressure gradient is absent. Therefore, observing circulation over those driest parts of the Earth where S = 0 we can decide whether this prediction is true or not.
Considering the region of Sahara and studying wind and pressure patterns there, we can come to one of the two conclusions: either u∇p = 0, which would yield support to the statement that ciculation in Sahara is driven by condensation elsewhere. (E.g. nobody disputes that the descending motion at 30N and 30S is driven by the ascending motion at the equator, whatever the physical reason for the latter is). If, on the other hand, u∇p ≠ 0 (not zero), then it will mean that Saharan circulation is not condensation-driven, but, instead, is driven by the temperature gradient between land and sea as commonly believed. That is possible.
In this latter case we could use this circulation as an example of a temperature-driven circulation to judge what kind of temperature gradients are needed to generate the observed winds. We will notice (we can see e.g. Chauvin et al. 2009.) that these really huge temperature gradients between Sahara and the ocean generate very mild winds and very little inland moisture transport. The temperature gradients that characterize global circulation (e.g. Hadley cells) are just non-existent in comparison — what kind of winds could they produce? Seen from this perspective, it comes as no surprise that global models fitted to reproduce global circulation cannot reproduce monsoons (land-ocean interactions) and vice versa. So my point here is about the need of a quantitative theory-based analysis rather than the usual qualitative considerations. In other words, it is needed to add to the analysis of the type performed, e.g., by Chauvin et al. 2009. another dimension — local precipitation, to study wind, pressure AND condensation patterns together. And I agree that making a clear distinction between different time scales can be very important in such analyses. I am not in the very least claiming any whatsoever complete knowledge of the relevant evidence — I am learning all my way, so thank you again for your insights.
John S. | February 7, 2013 at 8:26 pm |
In our 2013 paper on forest moisture transport (the manuscript can be found here) we investigated the seasonality of precipitation across the tropical Atlantic ocean in a rather detailed way (e.g. see Fig. 2). The pattern is indeed very complex. There is, for example, a seasonal longitudinal migration of the ocean-to-land precipitation ratio minimum (Fig. 6). I agree that a comprehensive analysis of such patterns and evidence will help decide that.
Let me ask your opinion on the following pattern. At 25S in Australia, if we look across the continent, there is a dry and a wet season, with a very pronounced difference in precipitation. This is shown in Fig. 6c of the above paper. The Australian land surface warms to huge temperatures, very similar to what happens in Sahara. Why does not this induce updrafts in the continential interior that could bring there some moisture? Indeed, independent of whether the season is dry or wet, the inner continental precipitation is always significantly lower than over the ocean to the East.
If you now look at the boreal zone in the NH (Fig. 6b) summer precipitation remains constant and exceeds the oceanic precipitation over seven thousand kilometers. How can we explain this difference from a conventional perspective?
Note that my question is not about why Australia is dry and why the boreal zone at 60N is relatively wet in summer. It is about why precipitation over land is much lower than over the ocean in one case, while the reverse is true in the other.
Thank you again.
Vaughan Pratt, “How do equatorial westerlies support the theory that Sahara nutrients are being blown into the Amazon? I would buy that India is benefiting from them, but the Amazon? My brain is melting down.”
Not yet. Since the equatorial westerlies/easterlies are indications of the state of the Quasi-Biennial Oscillation, which is linked to Sudden Stratospheric warming, your head starts getting cooked once you get into the gravity waves that transport all that energy to the polar heat sinks :)
Since the Amazon basin is producing the lion’s share of deep convection, which tends to drive the QBO, it will suck in Saharan dust until things shift.
This is R Gates turf, perhaps he will stop in.
In treating your equation for condensation rate S as if it were the ultimate arbitrer of whether circulation in a region is condensation-driven, you neglect a crucial consideration: the possible absence of condensable water in the atmosphere. One need only look at Mars to find Hadley cells and huge dust storms on a planet without condensation.
Here on Earth, the same physics applies. Compared to the winds of tropical thunderstorms that are responsible for most of the local condensation and precipitation along the Guinea littoral, the monsoon winds are indeed usually gentle. At the temperate latitudes of boreal forests, the rain that falls during the passage of frontal systems may have condensed a week ago a thousand miles away. And western intensification of warm-water currents produces entirely different precipitation levels compared to opposite coasts washed by equatorward-bound cool currents. The entire climatic problem cannot be reduced to simple equation with highly restrictive presumptions.
Sorry for the brief reply, but week-end plans await.
Not to get into a hypothetical “how many angels can dance on the head of a pin” debate here, BUT:
It seems to me that climate science is full of major uncertainties and that the Makarieva et al. paper just highlights one more possible uncertainty on how our climate works, which still needs to be tested (validated or falsified) empirically.
To quibble about the accuracy of the assumptions and equations at this point seems senseless to me. Suffice it to say that this is not a “crackpot hypothesis”, so that it is worth publishing and testing empirically.
We’ve got far greater “uncertainties” in far more critical climate assumptions.
The model-predicted value for (2xCO2) equilibrium climate sensitivity (ECS) is arguably the most important single basis for IPCC’s CAGW premise, yet it is on extremely shaky grounds, as recent studies are showing.
This value is estimated, by IPCC AR4 as likely to be in the range 2 to 4.5°C with a mean estimate of 3.2°C, and is very unlikely to be less than 1.5°C.
In AR4 WGI Chapter 8 (p.633) IPCC tells us how the models have arrived at the predicted value.
Including water vapor feedback, lapse rate feedback and surface albedo feedback, but excluding cloud feedback, the IPCC models predict a value of 1.9°C ± 0.15°C.
This includes a negative lapse rate feedback and a strongly positive feedback from water vapor based on the assumption that water vapor will increase with warming to essentially maintain constant relative humidity.
Using the data cited by IPCC, the relative temperature impacts are roughly:
+1.0°C No feedback
+1.8°C Water Vapor
-1.3°C Lapse Rate
+0.4°C Surface Albedo
+1.9°C Sub-total
Including cloud feedback the predicted value is stated to be 3.2°C ± 0.7°C.
This is based on a fairly strong positive feedback from clouds, which adds 1.3°C to the ECS estimate.
Actual physical observations have shown that net cloud feedback is very likely to be negative rather than strongly positive and that the net water vapor plus lapse rate feedback is around half of the amount estimated with assumed constant relative humidity.
Adjusting the midpoint of the range of the IPCC estimate but maintaining the spread between the upper and lower end of the range, we would have:
Spread = 2.0°C to 4.5°C = 2.5°C or ± 1.25°C or ±40%
Midpoint adjustment for clouds = 3.2°C – 1.3°C – 0.4°C = 1.5°C
Plus adjustment for lower net WV+ LR feedback = 1.5°C – 0.5*0.5°C = 1.25°C
So, with these adjustments we would have 1.25°C ± 0.4*1.25°C ~ 1.3°C ± 0.5°C
(This very rough calculation checks with recent estimates based on actual physical observations, rather than simply model simulations.)
The upper end of the range would indicate that the maximum warming from human GHG emissions could approach the “magic” 2°C allowable limit, while the lower end means that we will have no appreciable GH warming at all.
Lots of “uncertainty” there, right?
And we are quibbling about the accuracy if the equations used here when the whole CAGW house of cards is based on assumptions that are so dicey that they are essentially meaningless?
Let’s stop the quibbling and get on with the testing (first in models, as our hostess has suggested, and then empirically).
And, while we’re at it let’s define ECS more accurately based on empirical data (as recent studies have done), rather than simply model predictions.
Max
Max,
It does not highlight anything else than that a totally wrong paper can somehow get published, not without special comments by the editor and contrary to the views of the reviewers, but even so it got published.
Pekka
You fret about “a totally wrong paper [with marginal impact on the ongoing overall climate debate] getting published”.
But you keep quiet about the publication of IPCC’s supposedly “gold standard” AR4 (WGI, WGII and WGIII) climate report, which is loaded with “totally wrong” assumptions, estimates and projections.
A dichotomy I find hard to understand.
Max
Max,
I have argued strongly also against a paper of Hansen and his coauthors. I’m ready to argue against any article that I am convinced to be wrong. Not surprisingly such obviously wrong papers come more often from people who are not specialists in the field they write about.
@PP: I’m ready to argue against any article that I am convinced to be wrong. Not surprisingly such obviously wrong papers come more often from people who are not specialists in the field they write about.
Does the principle “arguing with idiots makes you an idiot” generalize to “arguing with non-specialists makes you a non-specialist?” Maybe not, you may just be a retired specialist like Pekka with tons of time on your hands.
I prefer the principle “arguing with anyone for more than hour without getting anywhere makes you an idiot.” Much less judgmental: after all you may be the one at fault. ;)
Having gotten exactly nowhere arguing with certain denizens here who shall remain nameless, I’ve cut back greatly with them. I find it much more profitable to argue with the likes of Pekka, Jim D, and captdallas because the ROI is way greater: less need to argue in the first place, and when the need does arise consensus on who’s right and who’s wrong is reached must faster.
Now we are engaged in a great condensation argument, testing whether the theory that it reduces pressure adiabatically can long endure. We are met on Climate Etc. and have come to dedicate a portion of that blog as a final resting place for those arguments if any that were deemed unsound by mutual consent, that science might live. It is altogether fitting and proper that we should do this.
Several people have suggested that an experiment be done to see whether adiabatic condensation reduces or increases pressure. Although I’d previously argued against this on the ground that those proposing that it reduced pressure were theoreticians and should therefore be judged on the merits of their theoretical arguments, I’m starting to come round to the other viewpoint.
Until an experiment has been designed and conducted to resolve this vexatious question, this torrent of papers inferring endless consequences of the premise that adiabatic condensation reduces pressure will forever haunt the skies of atmospheric physics like the legendary Flying Dutchman.
Vaughan,
As we know for complex systems it’s a general truth that no model is right but some models are useful. Therefore the relevant question about this paper is not whether it’s strictly correct but whether it is correct enough to have value trough that. Therefore it’s not a problem that they use equations that are not exactly right, but we must figure out whether the deviations from exactly right are such that the results remain useful.
This fact has made me emphasize the extent by which their results contradict well known physics on essential points. On this basis we can learn that the paper presents derivations that can very easily be extended to produce results that contradict very severely well known physics. The clearest case is in my view the result
S=u∂N/∂x
that is just a few steps from what’s written explicitly written and that’s a formula presented by Anastassia Makarieva as well. S is the condensation, i.e. it’s the source of rain. Thus they derive the result that am uplift cannot have condensation and cannot produce rain without a horizontal density gradient. That’s physically total nonsense, there’s no more need for empirical proof that their theory leads to contradiction with facts known to be true both theoretically and empirically. The required empirical tests have already been made and they confirm that the model is wrong on an essential point.
To me it’s also significant that I was from the beginning convinced that their results were throughout contrary to what I considered empirically known truth. I soon figured out what’s the most likely nature of the error that may have led to such totally wrong results, and then I found that they have, indeed, made an explicit error of exactly that nature. Thus I “knew” first that their work leads to totally false physics, and then later the detailed way the paper reaches those results. When the results alerted the knowledge that the paper must be wrong, it’s clear that it’s not just an inconsequential minor error or approximation.
All the necessary experiments to prove that the paper is totally wrong have already been done, no new experiments need be formulated.
A paper is totally wrong when it’s main results are empirically seriously wrong and when those results are consequences of an explicit error in the theory.
@VP: Several people have suggested that an experiment be done to see whether adiabatic condensation reduces or increases pressure. Although I’d previously argued against this on the ground that those proposing that it reduced pressure were theoreticians and should therefore be judged on the merits of their theoretical arguments, I’m starting to come round to the other viewpoint.
@PP: A paper is totally wrong when it’s main results are empirically seriously wrong and when those results are consequences of an explicit error in the theory.
On that, many of those discussing the paper at ACPD would appear to be in complete agreement with you. The experiment would be only for the benefit of those against whom you are arguing here, since evidently theoretical arguments don’t work for them and they are demanding experiments.
As to whether experimental demonstration would be any better, on further reflection I seriously doubt it. Those incapable of recognizing elementary theoretical errors, call them “slow” for want of a friendlier term, cannot be persuaded by any experimental demonstration because there will always be theoretical reasons why the demonstration is flawed. The slow ones will be unable to distinguish sound theoretical objections from unsound ones, leaving them no better off than before the demonstration.
My own reason for arguing any point on Climate Etc. is not to persuade the skeptics, who have persuaded me that they are beyond persuasion, but merely to put my own views on record in a setting where there are sufficiently strongly held opposing views as to show that my view is at least not vacuously true but requires more thought in order to accept it. In cases where my view is wrong, the one needing more thought is me, which I’ve been finding a great way to learn a subject.
Anon, Max, the baying Jamesian from Japan begins to deafen & dumbinform the angelique gabrielle.
================
Sources tell me that MiniMax will soon dispute James’ criticisms of Lewis.
In 3, 2, 1, …
Preview, please. Ever the twain shall meet.
=========
More like the train meeting a bunch of water vapor feedback. In a cloudbank.
=======
kim
Is that the gravy train you’re talking about?
Looks like it’s not only headed into water vapor feedback problems, but the wheels look like they are starting to come off, and the bandwagon appears to be heading for the ditch as well.
Are we staring a train wreck in the face?
Max
Pachauri Jones, you’d better, watch your speed.
============
max
‘And, while we’re at it let’s define ECS more accurately based on empirical data (as recent studies have done), rather than simply model predictions.”
Models have not been the source of estimates. Go review all the evidence over the years. paleo and observational dominate.
You didn’t get the memo, moshe; Gavin is giving up nightmares.
===================
Coming to the separate question on what’s new in what the paper is supposed to calculate.
Compare the derivation to that of Caballero’s lecture notes chapters 3.14 and 3.15. It seems that there’s nothing new in the physics taken into account in comparison to these rather short lecture notes. (Only some “new” conclusions that result from errors of the paper).
Pekka, wouldn’t the utility of the paper depend on what is more accurately measurable, water vapor or pressure? The magnitude of the saturation pressure difference is consistent with impact.
What’s the physical problem that’s being considered? It’s what happens in the uplift of moist air. That’s the process that’s described in the standard derivation of moist adiabat. In the form Caballero presents it the condensation is fully taken into account. The only term left out is that which concerns the liquid after it has been formed. That term is left out also in the present paper. Thus the two derivations describe exactly the same physics. There’s nothing more in the present paper.
So far we have discussed one column. That’s of course not everything. Take any book of atmospheric science (I have Wallace and Hobbs: Atmospheric Science, An Introductory Survey). There you can find a lot of discussion on how to continue from understanding behavior in single columns. Any such text discusses the kind of issues that must be considered to estimate horizontal pressure gradients or other interesting further results. The paper discussed here tries to get some results of that nature directly making some simple assumptions. The problem is that those assumption are wrong and so are the results obtained from those assumptions.
In what’s correct in the paper there’s nothing that has not been considered widely, when they deviate, they deviate because they make unsupportable and wrong assumptions. (Or in one case in that they calculate totally irrelevant results,)
Well, the one dimensional example does absolutly no justice to the complexity. Their assumptions require real limits, which I don’t think they are denying.
The actual driver of their “effect” falls apart as the horizontal increases past a point which would cause their “condensation layer” to break apart. It looks to me that using Pws or Nws, would help determine those limits and knowing those limits would be valuable.
As Tomas pointed out, they have to close the system somehow. Personally, using N-S and trying to actually “solve” would be a major PITA. Using data and “fitting” would be a much simpler way of determining the boundaries.
Granted, the “effect” is small, but with the way things are going, we will be looking at 3rd order effects before long.
Hmm …and I’m gettin’ nowhere either. Better take a nap in the car.
But not while it’s rolling…
(That’d be worse than getting your hand caught in a steel door.)
Thx fer reminder, Max, I’ve turned off the ignition, hand brake’s on :)
Anastassia Makarieva
I’ve got to hand it to you.
You are bravely and determinedly defending the paper you have co-authored.
To me (as an innocent bystander) it looks like you are holding your own.
No one here has been able to directly refute the logic or approach in your paper, although a few have tried.
I’d say the thing to do now is to test your hypothesis – first with climate models, as Judith has suggested, and (more importantly) with empirical data that either falsify or validate it.
Questions to you:
– How could your hypothesis be falsified (Popper)?
– What empirical evidence could you could you see that would corroborate your hypothesis (Feynman)?
– What do you see as the next steps?
Pardon me for being so direct, but I think it is time to test your hypothesis rather than simply try to find mistakes in the logic or equations used.
Max
Max,
You accept empty rhetoric to solid arguments. Therefore you haven’t noticed explicit proofs of fatal errors in the mathematical derivation.
Derivation that cannot justify it’s assumption and that leads to internal contradiction is just wrong. That some don’t accept that and that you cannot see that does not change the facts. No-one has been able to fill the gaps in the derivation. And a derivation that has such gaps is not a derivation.
Pekka
Assuming you are correct, and the “condensation driven winds” hypothesis of Makarieva et al. is incorrect, that will be shown by empirical testing.
If it is not falsified by empirical testing (Popper) we are back to square one.
If it is corroborated by empirical testing (Feynman) we are getting closer to having a “corroborated hypothesis”.
Why do I say all this is important?
Because exactly the same goes for the IPCC CAGW hypothesis, which is backed by the GH theory and some model predictions, but which also has not yet been tested empirically, yet which is being used as the justification for completely revamping the industrially developed world to wean it from GHG-causing fossil fuels..
Again, one must ask the questions:
– What would it take to falsify the IPCC CAGW hypothesis (2xCO2 ECS of 2°C to 4.5°C)?
– How could this hypothesis be corroborated by empirical evidence?
Isn’t that the way the scientific method is supposed to work?
And with the world economy hanging on the IPCC assumptions, wouldn’t it be wise to first clear up the uncertainty in the hypothesis upon which they are based before undertaking actions whose effectiveness and unintended consequences we cannot even estimate today?
Isn’t that the “no regrets” approach? What’s wrong with that logic, Pekka?
Max
Some people will so regret losing their fears, that they can’t.
=========
Well, Max, there’s one part that’s easy to refute empirically.
The claim that the precipitation rate is proportional to humidity.
Look about on a sunny day.
Max,
The phenomena are part of textbook atmospheric physics and important in meteorology. Models have been developed and tested all the time to get better weather forecasts and to understand weather phenomena better and better. All the physics that the present paper tries to take into account is included in that work. In that sense the physics has been tested and will be tested.
On the other hand, who would spend any time in testing specifically claims of a paper that’s obviously wrong?
Pekka and Nick
I am not defending the hypothesis itself.
I’m just saying, assuming it is not simply a “crackpot theory” (which I do not believe has been shown here), that it should be tested (models and empirically).
If either of you want to refer to it as a “crackpot theory”, go ahead and do so.
But, from all I have read here (especially the comments by our hostess), I do not believe that it has been demonstrated that this is a “crackpot theory”.
So it should be tested.
That’s all.
Max
Max,
It’s not crackpot theory in the sense that the paper is based on correct physics fundamentals. It’s wrong through making false assumptions and almost certainly the error that they have picked one equation in approximate form while they are also using the precise form of the same equation. This combination is particularly bad as the small difference between the two forms is taken as a real physical phenomenon.
In the same connection they mix formulas that are valid when there are no horizontal deviations with formulas that allow for horizontal deviations. Again they have a set of equations that leads to crazy results as they don’t notice what kind of error is done by that. These two “small” errors turn out to create strong artifacts that have no basis in real physics and come solely from their erroneous mathematics.
kim
Yes. The fear of losing one’s irrational fears.
I think psychologists call it “compulsive hobgoblin loss anxiety syndrome” or “phobophobia”.
Max
.
I could add a more positive comment on the paper.
They do discuss many important issues and try to make estimates on those factors. Thus I don’t see anything wrong with the goals of the work.
But good intents and discussing relevant physics does not make the results correct when there are severe technical errors that distort all the outcome severely.
That the field cannot be familiar to them comes out from the fact that they are not alerted by the implausibility of their results. They accept and write out results that create the immediate reaction of the specialists: We have done all that analysis and we know those results are wrong – and they, indeed, are wrong.
Specialists may have their own biases and may widely neglect some significant issues, but in general they have the correct feeling of what’s wrong and what’s right in their field. Outsiders may bring in new ideas, but much more often they repeat unknowingly well known ideas and quite often they make severe errors that they don’t recognize as specialists would.
What would be your criteria to determine what counts as a crackpot theory, MiniMax?
willard, Pekka might give you a clue if he answers Max @ 11:58 AM.
==========
Here’s how I read correctness, MattStat:
> The challenge faced by Computer Science is very similar to that faced by genetics. Our first and entirely non-trivial task is to understand what a computer program does. As for other engineering artefacts, the externally visible aspects of program behaviour can be codified as a formal engineering specification, expressed in the relevant technical terminology. An explanation of how a program works can be formally expressed in terms of types, assertions and other redundant annotations. They serve as internal specifications, attached at all the major and minor program interfaces. The correctness of the explanation can in principle be checked by a program analysis tool known as a program verifier. It uses automated logical and mathematical proof techniques to check consistency between a program and its internal and external specifications. A program verifier can play the same role in Software Engineering research as the automatic tools that are now essential or even obligatory in other branches of Engineering, to check the soundness and safety of engineering designs, long before they start construction. An adequately specified and annotated program, which has passed the scrutiny of an automatic program verifier, is said to be a verified program. The verifier offers highly credible evidence that the program will work in accordance with its specification.
http://comjnl.oxfordjournals.org/content/50/3/254.full
Vaughan can correct me if I’m wrong.
Pekka Pirilä
I could add a more positive comment on the paper.
They do discuss many important issues and try to make estimates on those factors. Thus I don’t see anything wrong with the goals of the work.
But good intents and discussing relevant physics does not make the results correct when there are severe technical errors that distort all the outcome severely.
In this discussion, some approximations have been asserted to be false because they disagree with other approximations. I would like to see published a point-by-point rebuttal to the paper with the claims of the paper and the counter-claims by the critics laid out side-by-side with the supporting references (author, title, equation number) and alternative calculations (where Makarieva et al have been called in error.)
In most of the climate science that I have read to date the misfit between theoretical calculations and measured results has seldom been addressed. It is asserted without empirical justification that the calculated results are known to be good enough. Here Makarieva et al try to refine the knowledge in one limited sphere of the climate science, and the principle objection to their work has been that it is different from what others already did. Some really detailed and carefully measured empirical results in the areas where the results are different would be extremely helpful.
Matt,
The problem is not that they use a worse approximation than someone else. The problem is that they require that both the approximate formula and the exact formula be true at the same time. From that requirement they derive totally wrong results like that no rain is possible at a place where we have the strongest rain.
manaker
‘If it is not falsified by empirical testing (Popper) we are back to square one”
unicorns cause winds. You cant show the logical problem. you cant show the math problem. It’s falsifiable. but we dont have empirical evidence.
Therefore?
My theory is on par with theirs with regards to the standards you use.
Pekka Pirila: The problem is that they require that both the approximate formula and the exact formula be true at the same time. From that requirement they derive totally wrong results like that no rain is possible at a place where we have the strongest rain.
If that is what you wrote, then I have missed it. I shall try again to reread all of your posts in order.
Willard(@nevaudit): Here’s how I read correctness, MattStat:
I don’t see how that applies here. Nearly all of the equations in physics are approximations, and those used in climate science that I have seen to date are all based on some simplifications (equilibrium, for example; or ignoring the fact that 22% if incoming TOA radiation is absorbed in the upper atmosphere.) From those approximations are derived equations that are also at best approximations to the relationships that are of interest. Always the accuracy of the result must be assessed before it is relied upon for the achievement of a practical goal.
@Matt: Some really detailed and carefully measured empirical results in the areas where the results are different would be extremely helpful.
It’s a theory paper, Matt. Theory papers have their own criteria: is the reasoning sound, and can the opposite conclusion be drawn with a shorter argument? (By default proofs of respectively P and not-P are resolved in favor of whichever has the shorter proof pending further refinement of the reasoning.)
At a faculty meeting not so long ago we were discussing one theory researcher’s algorithm that had made a huge breakthrough by being the first improvement in decades on the previous best. I naively asked if someone had programmed it and was informed that my question made no sense as it was a nonconstructive method. Silly me. Theoretical computer science has made great strides since I moved on to other areas (hopefully just a correlation).
Vaughan Pratt: It’s a theory paper, Matt. Theory papers have their own criteria: is the reasoning sound, and can the opposite conclusion be drawn with a shorter argument?
It would still be good to have, in the future, empirical measurements good enough to test the theory’s accuracy.
MattStat,
I’m not sure what I have in mind works in our context either. In formal specification, correctness tells you your program will work as advertized:
> In theoretical computer science, correctness of an algorithm is asserted when it is said that the algorithm is correct with respect to a specification. Functional correctness refers to the input-output behaviour of the algorithm (i.e., for each input it produces the correct output).
A distinction is made between total correctness, which additionally requires that the algorithm terminates, and partial correctness, which simply requires that if an answer is returned it will be correct. Since there is no general solution to the halting problem, a total correctness assertion may lie much deeper. A termination proof is a type of mathematical proof that plays a critical role in formal verification because total correctness of an algorithm depends on termination.
http://en.wikipedia.org/wiki/Correctness_(computer_science)
I believe it applies here, because the criticisms, from what I can undestand, underline semantical properties of the algebraic terms. This explains why, for instance, Isaac Held says thinks like:
> There is no other physics in this expression. In particular, there is no thermodynamics, which would come in at the point that one determines which regions are condensing and which are not, or when estimating the mixing ratio gradient by assuming that the atmosphere is on a moist adiabat in the region of upward motion.
http://www.atmos-chem-phys-discuss.net/10/C14687/2011/acpd-10-C14687-2011.pdf
Held seems to looks at the equations and ask what they denote in the scientific field they wish to be applied. After checking the terms, Held seems to say: “it does not compute”.
I’m quite willing to consider that this is not exactly what is done here. This is just my way of understanding the use of “correct” in our discussion.
Willard(@nevaudit): Held seems to looks at the equations and ask what they denote in the scientific field they wish to be applied. After checking the terms, Held seems to say: “it does not compute”.
Thank you for the link to Held’s review. In it he refers to the fact that some of the issues are unclear, and refers to his “hunch” and “guess” that Makarieva et al make incorrect assertions. Hunches and guesses have to be tested as well. After reading Held’s review, in which he recommended against publication, I would have recommended for publication, had I been the editor. He also referred to Makarieva et al making explicit equations for some effects previously considered (he does not say why) to be negligibly small. And, though he decries appeals to authority, he writes that Makarieva et al have written equations contrary to text books, without telling what the evidence is that the text books are more accurate. He also exaggerated the main claim of Makarieva et al, who conclude that the quantitative effect of what they have presented is small relative to total energy flux.
I do not conclude that Held asserted “does not compute”; I conclude that he disagreed with the authors. Especially when he addressed effects that previously had been considered negligibly small.
Thank you for your comment.
I hope you realize that the expression “does not compute” helped me interpret the quote and not the rejection itself, and that this example was supposed to illustrate the concept of correctness, MattStat.
There is at least one other meaning of the word “correct” I could illustrate by analyzing what you’ve just done.
@manacker: No one here has been able to directly refute the logic or approach in your paper, although a few have tried.
Max believes he is a better judge of reasoning about physics than the likes of Pekka or Nick.
Thanks, Max, you made my day. ;)
The logic is easily refuted and Pekka, Nick, Isaac Held and others have already done so. You’ve demonstrated a great many times that you are no judge of sound reasoning in physics whatsoever.
This article makes many elementary errors, some minor, some more serious, some fundamentally wrong. Ideally one would like to start with those errors that are most egregious and most easily seen to be errors, but identifying them takes more effort than merely picking errors at random.
Vaughan
Our hostess has stated that this is no crackpot theory.
I cannot judge the correctness of the equations, etc. (as I have previously stated), but, since it is not a crackpot theory, it deserves to be tested.
Has the co-author, Makarieva, conceded that Pekka and Nick have valid criticisms, which invalidate the paper?
I have not seen evidence of that here. Have you?
The proof of the pudding, VP, in physics (as well as anywhere else) is…(you know the rest).
It’s like Feynman said (I paraphrase) it doesn’t matter who says it’s right or wrong, it’s the empirical evidence that counts.
And that is my point.
Period.
Max
PS Just like the IPCC CAGW hypothesis (or your “millikelvin” extrapolation to 2100) also deserves to be tested empirically, even though many individuals might find one or the other part of it extremely doubtful. Got it?
Theoretical analyses can be right or wrong. No experiment is needed to decide that.
Pekka Pirilä
Theoretical analyses can be right or wrong. No experiment is needed to decide that.
That is puzzling. Perhaps you mean that derivations of conclusions from hypotheses may be correct or incorrect. Empirical results are necessary to determine whether the conclusions are correct..
Matt,
I didn’t mean that being right or wrong is the only qualities that they may have. They may also be relevant or irrelevant, they may be about real physics or something else.
You may continue the list.
Here the relevant point is that if the theoretical analysis is wrong, that can be known without any empirical evidence. This paper is wrong and this paper is wrong so badly that it’s results are not of further interest.
It puzzled me too. An experiment or (shudder) an observation may improve a theoretical analysis, right or wrong.
===================
Their hypothetical two columns are irrelevant to anything that happens in the atmosphere. This is the most succinct way of putting it, and has been said by many here already in various ways. They might do the thermodynamics of these two columns right (?), but it doesn’t apply to anything. It is a little like the difference between pure mathematics and applied mathematics.
@manacker: Our hostess has stated that this is no crackpot theory.
Who so far has claimed this is a “crackpot theory?” Certainly not me. Many including Judith have expressed concerns about the paper’s reasoning, but it’s not like the authors are claiming faster-than-light communication, perpetual motion, an antigravity shield, or dinosaurs inside our hollow Earth. Most of the reasoning is sound, which makes it harder to identify the unsound bits.
I have not seen evidence of that here. Have you?
Yes. This is a purely theoretical paper claiming (a) a problem with the extant theory and (b) a theoretical solution to that claimed problem. If the reasoning is wrong then all that is needed is to point out why it’s wrong, either by identifying a specific incorrect step in the reasoning or better yet by showing that the claimed result contradicts well-established facts. If you read through all the ACPD discussions since 2009 you’ll see that both of these have been done in spades. I don’t see why ACP now thinks there is still room for doubt.
If the only permitted way to refute an incorrectly argued theory paper is to carry out an experiment, as you seem to be suggesting, instead of simply pointing out the errors, it would make science a hundred times as expensive as it already is.
@Jim D: Their hypothetical two columns are irrelevant to anything that happens in the atmosphere.
True, but Section 3.1 wasn’t claiming it was. The paper is unclear as to the point of that section, but my guess would be that it was to give one straightforward way of specifying an initial loading of the columns with water vapor. This loading has the property that the vapor does not condense in the isothermal state but does condense (seemingly at all strictly positive altitudes) with the MALR thermal profile introduced in Section 3.2. I don’t have any problem with that, other than that surely there are simpler ways of specifying a supersaturated preloading. I don’t see what’s special about this particular one.
Vaughan Pratt
“Let’s not test this hypothesis because it would be too costly”.
Bad logic.
We are spending billions on climate research. Testing this hypothesis is not going to bust the piggy bank.
I almost get the sneaky suspicion that there is a fear that this hypothesis might pass the test and that’s why there is so much resistance to simply testing it.
It’s out there. It’s been published. Lots of folks are taking shots at the theory or equations. Fine.
Now let’s go the next step and see if it passes the test. No big deal. And we’ll all be smarter afterward either way.
Your “millikelvin” poster deserves the same.
That’s the “scientific method”, as I learned it.
But we have beaten this dog to death and it’s time to move on.
Max
I have to say – this is one of the more fascinating arguments that makes the rounds in the “skept-o-sphere.”
These “billions of dollars” amounts to a drop in the bucket, if we want to consider something like the GDP. How much of a tax increase for the top .01 percent of our public would cover these “billions of dollars?” Even if we think the danger of climate change to be unlikely – what kind of risk assessment would say that even a small chance of dangerous climate change would not merit such an expenditure?
It is like when Judith focuses on the lack of money for stabilizing and modernizing our capacity for longer-term weather prediction — by pointing to the money spent on studying climate change – as if that were the problem.
Neglected from mention is the short-term mentality imbedded in anti-government rhetoric. Indeed, many of those who most strongly advocate against the kinds of government spending that would enhance our long-term weather prediction capacity fall into political associations that are strongly associated with climate “skepticism.”
MiniMax answers to the emphasized bit:
> If the only permitted way to refute an incorrectly argued theory paper is to carry out an experiment, as you seem to be suggesting, instead of simply pointing out the errors, it would make science a hundred times as expensive as it already is.
as if it was not following what has been said before the emphasized bit.
@manacker (putting words in my mouth): “Let’s not test this hypothesis because it would be too costly”
That’s not what I said, Max. I said that if the only permitted way to refute an incorrectly argued theory paper is to carry out an experiment it would make science a hundred times as expensive. I’m glad to see you’re fine with upping the science budget by a factor of a hundred.
Given that pencils are two for a dollar at Amazon while planes to measure atmospheric circulation are considerably more than that, even just to rent, my figure of a hundred may have been on the conservative side. No one rents pencils.
VP implies that Max should not discuss the science as his credentials are lacking. One thing that bothers me is that the scientists mentioned (including yourself) may be suffering from the opposite of confirmation bias – shall I call it “defending the cause”?
Max is free to present his views on science, but I don’t think that makes it right to present a supposed quote of what another has written changing the wording so much that the message is seriously distorted.
Max should also avoid other explicit statements on the discussion as facts when they may mean only that he does not know what has been presented or that he has not understood it, while it should be perfectly understandable with his background.
I will go with Pekka , on this one. He seems even less impressed than he was with VP’s foolishness.
Hmmmm, applications, an empiric(shrug) sort of thing, might be profitable.
===================
> VP implies that Max should not discuss the science as his credentials are lacking […]
Citation needed.
@PD (putting words in my mind if not my mouth): VP implies that Max should not discuss the science as his credentials are lacking.
Peter, I am unaware of ever having said such a thing. If you feel I implied it somewhere, please quote the offending sentence.
Max himself states from time to time that his credentials are lacking, but I’m not aware of others who’ve responded to that by saying or even implying “well, in that case why are you commenting here?” Max has the same right as everyone else here to discuss the science, actually more in my opinion because he does so politely.
@MM: That is puzzling. Perhaps you mean that derivations of conclusions from hypotheses may be correct or incorrect. Empirical results are necessary to determine whether the conclusions are correct.
In mathematics, Matt, no one gets any credit for an unsound proof of a true theorem.
How do statisticians judge that sort of thing? And how would you as a statistician recommend that climate scientists judge it?
Willard and Vaughan asked me for a citation.
“mplied” is not the same as “said”
Start of quote”Vaughan Pratt | February 4, 2013 at 1:27 pm | Reply
@manacker: No one here has been able to directly refute the logic or approach in your paper, although a few have tried.
Max believes he is a better judge of reasoning about physics than the likes of Pekka or Nick.
Thanks, Max, you made my day. ;)
The logic is easily refuted and Pekka, Nick, Isaac Held and others have already done so. You’ve demonstrated a great many times that you are no judge of sound reasoning in physics whatsoever” end of quote
My point was a trivial one but it was brought on by VP’s name dropping and general tone towards Max’s posts. IMO Max is one of the few commenters on this blog who is genuinely seeking to find common ground, and while we all know that he is rationally sceptical of AGW alarmism he deserves more respect that was accorded to him by VP on this occasion.
I was surprised and disappointed that VP had written this because I had always believed that he was a pretty cool commenter who generally engages well with people who disagree with his views, as can be seen in the recent thread on his poster on millikelvins.
@PD (quoting my comment to Max): “You’ve demonstrated a great many times that you are no judge of sound reasoning in physics whatsoever.”
Well, perhaps that was excessively blunt (I’m from the 1960’s Australian school of bluntness), even though Max has agreed from time to time that he doesn’t follow some of these more technical issues.
Let me rephrase it then. It was in response to Max’s claim “No one here has been able to directly refute the logic or approach in your paper, although a few have tried.” How about this?
“I much admire Max’s keen ability to find flaws in those refutations of the paper’s logic for which no one else has been able to find any flaw. I just wish he’d share his findings with us.”
@PD (responding to Pekka): intense resistance to change of the orthodox science seems to be a feature of many of the paradigm shifts that have occurred in recent, and, of course, not so recent history.
This is certainly true, witness Angstrom’s resistance to Arrhenius’s theory of CO2 forcing, Einstein’s resistance to Heisenberg’s uncertainty principle, Linus Pauling’s resistance to Shechtman’s quasicrystal theory, and other like examples. French aeronauts were unable to accept the idea that two bicycle mechanics could have developed a working heavier-than-air flying machine until five years after its first flight! As Max Planck said, “science advances one funeral at a time.”
But people also resisted Blondlot’s n-rays, Pons and Fleischmann’s cold fusion, the OPERA faster-than-light neutrino anomaly (which by September 2011 had been judged statistically significant at the 6-sigma level, the same level used to justify the more recent observation of the Higgs Boson), and plenty of other theories that did not withstand close examination.
I conclude from this that “intense resistance” to a new theory is an irrelevant metric. Perhaps a few funerals are in order, but another possibility is that that those responsible for a given new theory are incapable of debugging their own theories.
When all the support for a new theory is coming from those who haven’t put a lot of work into trying to understand that theory in any technical depth, to the point that they can’t even begin to answer questions about the theory from those that do have the requisite technical depth, the odds of the theory being a serious contender for a radically new theory of anything are extremely slim.
The ultimate litmus test however is whether the paper is accurate. In that respect the paper under discussion fails miserably, being littered with errors. Pekka has pointed out a few, but there are way more. After you’ve found a few, the return on the investment of finding them all starts to become vanishingly small.
I’m content with merely refuting the central thesis of the last half-dozen papers coauthored by Makarieva, Gorshkov, et al, namely that adiabatic condensation reduces pressure, in the simplest possible terms so that even Max might have a chance of following it: see my comment on this here. Looking forward to engaging with anyone volunteering to refute my refutation. :)
Thanks for your response VP. I’m not a troll and almost never worry about nits. My comment didn’t make a big deal of anything really and the main point was that even scientists may have biases which could affect their judgment as well as people like Max and myself, who may not be up to speed with the science and making unfounded assertions.
I guess the takeaway about new advances in science is the fact that very rarely do bad theories get past the the peer review process and this is a good thing. This peer review process, however, hasn’t been implemented properly in the case of AGW as postulated by the IPCC and this is one of the main reasons why there is continued resistance from sceptics.
@PD: This peer review process, however, hasn’t been implemented properly in the case of AGW
Now that we have publication of “Where do winds come from” as a benchmark, which AGW papers would you cite as an equivalent illustration of improper implementation of peer review?
One way of comparing them would be by rate of errors per page. That could be informative in itself.
Pre condensation a given amount of volume contains water vapour and air.
After condensation the same volume contains the original air plus condensed water plus more air drawn in from the surroundings.
That must giver a higher surface pressure mustn’t it ?
Evaporation reduces pressure but condensation increases it.
One cannot have reducing pressure from evaporation then reducing pressure again when condensation occurs.
Doesn’t make sense.
For it to make sense the surrounding air would somehow need to be prevented from flowing in after condensation occurs but that clearly doesn’t happen.
One could say that the pressure of the surrounding air would be reduced but the surrounding air is essentially the rest of the atmosphere so any pressure reduction would be imperceptible.
Evaporation injects potential energy into an air parcel in the form of latent heat thereby reducing density and weight to below the global average.
Condensation then restores the weight of the air parcel back to the global average by removing the latent heat.
I can’t see any common sense basis of support for the proposition that condensation reduces pressure except as a very local short lived phenomenon whilst surrounding air moves back in.
Condensation adds heat, and as I mentioned in a comment somewhere above, that causes increased pressure more than the loss of vapor decreases it, so it becomes a loss of mass in the column because of latent heating.
JimD, that assumes that the heat released by condensation is absorbed in the volume of the dry gas. So are we going to neglect that atmospheric window energy when convenient?
The radiative heat transfer is far too weak to affect these considerations.
Pekka, “The radiative heat transfer is far too weak to affect these considerations.” Possibly, but there was a rather large 18 to 20 Wm-2 discrepancy in the energy budget, so I would not be too confident in that.
A far as the volume/pressure, in a 30 C saturated volume of air at sea level the specific volume is ~0.9 kg/m^3. Removing the moisture reduces the specific volume to ~ 0.865 kg.m^3 or by about 4%. In order to increase that volume of dry air to 0.9 kg/m^3 would require heating by 14 C or about 1400 meters of adiabatic compression. That would be a fair down draft if the volume was’t heated by the release of the latent heat, kind of like a micro burst.
The latent heating going on in a cloud is at least two orders of magnitude higher than the slow radiative processes.
Here we are considering weather phenomena, not climate. Radiative imbalances play extremely little role in the behavior of a moist uplift.
capt.d., using your numbers, you appear to have removed 35 g/m3 of water vapor (an extreme case for sure). How much latent heat does that release? It turns out that each g/kg condensed can warm air by 2.5 C. Therefore your 35 g/kg will warm the air by 87.5 C, easily enough to return the volume beyond where it started.
JimD, “capt.d., using your numbers, you appear to have removed 35 g/m3 of water vapor (an extreme case for sure). How much latent heat does that release? It turns out that each g/kg condensed can warm air by 2.5 C. Therefore your 35 g/kg will warm the air by 87.5 C, easily enough to return the volume beyond where it started.”
30 C is not that extreme in the tropics. 35 C (95F) surface temperature at 80% relative is pretty normal around here in the summer. Having the surface temperature drop to 75 to 78 during a rather exciting downpour with a light show and the occasional floor show (lots of ‘spouts) is pretty normal too.
Condensation removes latent heat from a parcel of air. The condensed out particles then hold the energy in liquid form and since the liquid can hold far more energy than the air at the same temperature there is no change in temperature.
However there is a contraction of volume but that immediately pulls in more air from the surroundings so the original volume then contains more mass than before which must give a rise in surface pressure rather than a fall.
I can’t see a way out of that for the authors of the article.
No, condensation warms the air. This is how clouds gain buoyancy. Maybe you mean the heat is initially concentrated in the droplets that form, but it doesn’t stay there for any measurable time before it is conducted to the surrounding air. There is a net gain of energy due to the water bonds that are formed.
Thanks Jim. I was unhappy with my comment and so have been looking it up and have confirmed that the energy released by condensation warms the air.
Condensation has a net warming effect on the then dryer air whereby the energy has been removed from the air as latent heat and converted to sensible heat.
Air containing water as vapour is always lighter than dry air.
The air without water in vapour form must therefore become more dense and must still fall unless the extra sensible heat warms it to such an extent that it becomes as light as the air containing water vapour previously was.
In fact the sensible heat or a portion of it gets radiated to space so the heated air parcel never becomes as light as it was when it contained water vapour so it becomes denser and heavier and must fall.
Additionally there has been a contraction of the water vapour to liquid which pulls in more dry cold denser air from the surroundings to replace the water vapour that has been removed.
The net result must be more molar mass in the original volume and a net rise in surface pressure.
To avoid that outcome one would have to suppress all radiative loss so that the air parcel after condensation is as light as the air parcel before condensation.
That doesn’t happen due to the net upward radiative flux.
It is not possible for the sensible energy released by condensation to cause enough heating to prevent the air from contracting cooling and descending so the net result must be an increase not a decrease in surface pressure.
This paper actually proposes that the sensible heat released by condensation warms the dryer air to a point where it becomes even lighter than air containing water in vapour form.
That is impossible and that, I think, is the fundamental flaw.
Stephen, I am going to disagree with your characterization of the paper. My big complaint about it was that it did not consider the pressure increase due to heating, only the loss due to vapor, which you suggest correctly causes contraction and negative buoyancy. How did they get it so wrong? It looks like they say the pressure drop doesn’t cause contraction, but instead a large-scale low-pressure center that results in low-level convergence and ascent.
Jim.
I think they are saying that condensation and rainfall results in contraction high up that causes a pressure drop over the whole forested area thus inducing a flow of moist air from oceans to above the forests which then generates condensation and rainfall to renew the loop.
For the reasons I put forward the contraction cannot possibly cause a pressure drop. It must instead induce a pressure rise.
Condensation is indeed associated with low pressure but the cause is the earlier evaporation rather than the condensation. Condensation is the opposite of evaporation in all respects.
As regards your point that a release of heat to sensible form should increase pressure rather than reducing it you should bear in mind that such an increase in sensible heat would reduce density and increase buoyancy which gives low pressure.
The reality however is that the release of sensible energy cannot give enough fresh buoyancy as to overcome the loss of buoyancy when the water vapour condensed out.
That is fatal to their paper in my opinion.
Thus condensation must result in an increase in surface pressure and if the rising column of fresh vapour rich air is continued then the descending column is displaced to an adjacent location as we see all the time in the real atmosphere.
It looks like Jim D is coming to the same conclusion about the paper’s basic premise that Dan Rosenfeld did when he refereed the paper more than three years ago, see Appendix 5, pages S12436-7, of the October 2009 Interactive Comment. Rosenfeld’s objection was simply that condensation increased pressure, contrary to the authors’ belief that it decreases pressure.
Rosenfeld’s analysis answers the question I raised here as to which direction NT (product of number of molecules and temperature) moved during condensation. His answer: sharply upwards, proving his claim.
Since the first sentence of Section 2.2 of the paper claims that Section 2.1 has somehow immunized the paper against this objection, perhaps the objection needs to be restated in a manner that makes it even more transparent than it already is, and therefore that much harder to reject. I’ll number the steps so they’re easy to refer to.
1. The following analysis is of a parcel of supersaturated air at an altitude of 620 hPa (about 4 km) and 280 K (7 C).
2. As can be verified from any online steam table or dewpoint calculator, 7 C is the dewpoint for a water vapor partial pressure of 10 hPa. (Being even more old-fashioned than our resident HVAC expert captdallas I looked it up in Table C-2 of my copy of Potter and Somerton, Thermodynamics for Engineers, Schaum’s Outlines.) This is the only value of the Clausius-Clapeyron relation we need here, pay no attention to all those complicated formulas.
3. Initially the parcel is supersaturated with a relative humidity of 200%.
4. This corresponds to an initial water vapor partial pressure of 2 * 10 = 20 hPa.
5. The partial pressure of the (dry) air component is therefore 620 − 20 = 600 hPa. (These initial conditions should be very close to a point in the first column in Section 3.2 of the paper at an altitude of around 4 km.)
6. For each mole of air we have initially 20/600 = 1/30 moles of water vapor. From here on I’ll assume the parcel contains 31/30 moles, namely 1 mole of air and 1/30 moles of w.v.
7. The mass of w.v. is 18/30 = 0.6 g, needed for step 10. (And the mass of air is 28.97 g but that’s not needed in the following.)
8. At 200% RH, half the w.v., namely 1/60 moles or 0.3 g, will condense out to bring the RH down to 100%, leaving 1/60 moles of water still in vapor form.
9. As promised by the paper, condensation has reduced the pressure from 620 to 610 hPa. This effect would appear to be the basic underlying idea for the whole paper.
10. Ah, but wait. The latent heat of vaporization of water being 2260 J/g, condensation of 0.3 g releases 0.3 * 2260 = 678 joules. In order for the condensation to count as adiabatic we need to take the warming effect of this conversion of latent heat into sensible heat into account.
Not knowing whether condensation will happen immediately or over a long period, let’s consider each of these two possibilities in turn, as extreme cases: in a flash or extremely slowly.
A. In a flash.
11. The inertia of the surrounding air means that condensation takes place in a constant volume. We therefore need the constant-volume specific heat values for each of air, water vapor, and liquid water.
12. Consulting Wikipedia we find these to be respectively
air: c_v = 20.7643 J/mol/K
w.v: c_v = 28.03 J/mol/K
liq.w: c_v = 74.53 J/mol/K
12. We have a mole of air and 1/60 moles of each of the other two. Hence to raise the temperature of the parcel by one degree requires
20.7643 + (28.03 + 74.53)/60 = 22.47 joules
13. Our 678 J from condensation therefore raises the temperature of the parcel by 678/22.47 = 30.2 degrees, namely from 280 to 310 K.
14. The pressure which had dropped to 610 hPa now increases to 610*310/280 = 675 hPa. This is an increase from our original pressure of 620 hPa of 675/620 = 1.09 or 9%.
B. Extremely slowly.
15. Since liquid water weighs the same as the water vapor it condensed from, there is no change in the weight of the column of atmosphere above our parcel of air, whence the pressure remains unchanged, although the parcel may change (very slowly) in volume. We therefore need the constant-pressure specific heat values for each of air, water vapor, and liquid water. The first two are significantly different from their constant-volume counterparts (thermodynamics is subtle).
16. Consulting the same source as in 12, we find
air: c_p = 29.07 J/mol/K
w.v.: c_p = 37.47 J/mol/K
liq.w.: c_p = 75.327 J/mol/K
17. Repeating step 12 with these numbers, to raise the temperature of the parcel by one degree requires
29.07 + (37.47 + 75.327)/60 = 30.95 joules
18. Repeating step 13, the temperature increases by 678/30.95 = 22 degrees more or less, namely from 280 to 302 K.
19. The volume which would have decreased by a factor of 610/620 has therefore actually increased by a factor of 610/620 * 302/280 = 1.06 or 6%.
The challenge to the paper’s authors is to identify which step or steps if any does not satisfy the conditions from which they derive their conclusion that condensation reduces pressure.
Vaughan, welcome back to old school :) I am not sure if they are confusing barometric pressure or just using the saturation vapor pressure as a potential energy.
I looked at it as potential and it does tend to provide the right range of wind velocity though when water vapor actually decides to condense and precipitate is a bit iffy. Given the translation gaps, Russia, Aussie and Redneck I think there is a semantic situation going on.
It would be nice to get this resolved so we can move on to the fun stuff, the atmospheric window.
“Rosenfeld’s objection was simply that condensation increased pressure, contrary to the authors’ belief that it decreases pressure.”
And is a wet lapse rate a manifestation of this increased pressure.
Pressure is weight of atmosphere. But wet lapse rate isn’t about weight of atmosphere [in terms weight it’s slightly lighter due lower density gas] but it’s about an increase of energy [there is both kinetic and potential- but it’s concerning change phase of water from gas to liquid- so kinetic energy
which affects the pressure. Or the increased pressure is local.
So Stand under a wet lapse rate and one doesn’t have higher pressure, but the gas above you has higher pressure.
continuing
If we assume that condensation H2O gas increasing pressure, locally within the atmosphere though not changing the barometric pressure
of entire atmosphere.
What happens when you liquid water with just air above it.
In this situation the majority of H20 molecule are condensing in the water and being added from the water as H20 gas. In this situation one less pressure locally.
In other words if wet air is balanced in regard to condensation and evaporation- the net result should higher pressure of the air. But with air above the water, the condensation is mostly occurring in the liquid water and the liquid water is evaporating [adding H2O gas]. So in situation this should not have as much pressure.
Or whenever their is net increase of condensation in the air, this increases pressure, and conversely when whenever there is increase in net amount of evaporation in the air, pressure should lower.
This changing pressure is the same as changes in lapse rates.
So net condensation is higher pressure and lower lapse rate.
And net evaporation is lower pressure and higher lapse rate.
Rosenfeld’s objection was simply that condensation increased pressure, contrary to the authors’ belief that it decreases pressure.
Yet, looking out the window, if it is raining buckets, there is a low pressure zone centered on the region where it is raining buckets.
So the author’s belief agrees with common sense casual observation.
James, the authors’ theory is indeed a radical one. Before their groundbreaking paper it was customarily held that lows bring rain. We now know from their paper that rain brings lows.
At least that’s how I’m interpreting your statement. Or did you mean something else?
Rosenfeld expressed his conclusion thus:
“taking into consideration for an adiabatic parcel in a fixed volume both decompression due to condensation and compression due to the inevitable latent heating that must occur with the condensation, demonstrates that the latent heat induced compression overwhelms the condensational decompression. Therefore, accepting the reasoning of the authors, who assert that the condensational decompression energizes hurricanes, leads to a contradiction with the calculations presented here. This contradiction renders the hypothesis presented by the authors invalid”
But in my opinion that is not fatal for their hypothesis in itself because additional warmth causes reduced density, reduced weight and therefore increased buoyancy which would lead to a lower surface pressure.
What makes air rise from the surface when evaporation occurs is increased pressure horizontally which increases buoyancy of the less dense parcel by forcing it upward vertically for a reduction of pressure at the base of the column.
What is fatal for their hypothesis is that even despite the higher temperature from latent heat release the dry air parcel still fails to achieve the buoyancy that was previously achieved by the water vapour laden air parcel.
Thus the dry air must descend adiabatically which results in an increase in surface pressure beneath it because it has then become denser and heavier thant the rising vapour laden air was when it reached the same height.
The reality is this:
i) At the surface increased horizontal pressure from evaporation induced expansion accompanies uplift for decreased pressure vertically.
ii) At the height where condensation occurs decreased horizontal pressure from condensation induced contraction accompanies descent for increased pressure vertically.
It is like an opject being held aloft in an updraft. Decrease the weight of the object and it will be propelled higher but increase the weight and it will sink lower.
Stephen said, “What makes air rise from the surface when evaporation occurs is increased pressure horizontally which increases buoyancy of the less dense parcel by forcing it upward vertically for a reduction of pressure at the base of the column.”
Wait now, let’s specify “surfaces” You have dirt, the ground but you have a condensation layer at 1 km. The condensation layer is saturated or over saturated and isothermal horizontally. As condensation increases above the condensation layer, it would expand horizontally but the altitude is fixed by the temperature. There is no vertical motion unless the ground level temperature changes. With condensation, energy is released, some of which would warm the near saturated air above the condensation layer creating an upward motion not of the condensation layer base, but the air above which would expand the the cloud upward which would induce moist air into the region of condensation. The cloud builds just like clouds build with that flat base at the isothermal layer.
For the isothermal layer not to descend, there would need to be a balancing force to offset the expansion above the condensation layer.
So which way you want to go? All that energy released is not pushing the cloud down unless the surface is at the saturation temperature. There is a butt load of energy released above the condensation layer, so constant volume is shot. Pressure is changing above the condensation layer, so what happens below? Static pressure gets converted into velocity pressure. So it is low pressure above and high pressure below, the wind has to blow.
here ya go,
http://www.youtube.com/watch?v=DN0DlNM9ZQA
and
https://lh5.googleusercontent.com/-i0UApz1xKQc/UQ5rdbDFvYI/AAAAAAAAHD4/bslBj27ttN0/s729/moisture%2520ground%2520plane%2520surface%2520evaporation.png
Now if that cloud base remains as a constant altitude what has to happen?
But in my opinion that is not fatal for their hypothesis in itself because additional warmth causes reduced density, reduced weight and therefore increased buoyancy which would lead to a lower surface pressure.
According to the ideal gas law PV = nRT, additional warmth (increased T) reduces density when P is held constant, namely by increasing V while leaving n (the number of moles and hence the mass) unchanged.
The increased volume exactly offsets the decreased density, no need to change either the net mass or the pressure in order to account for changes in volume and density.
Thus the dry air must descend adiabatically which results in an increase in surface pressure beneath it because it has then become denser and heavier thant the rising vapour laden air was when it reached the same height.
Why would it become heavier if it consists of the same number of air molecules? According to PV = nRT, the increasing density of a descending air parcel can be entirely accounted for by a decrease in V exactly offsetting an increase in P. How could the net mass change in that process? No matter is created or destroyed.
What makes air rise from the surface when evaporation occurs is increased pressure horizontally which increases buoyancy of the less dense parcel by forcing it upward vertically for a reduction of pressure at the base of the column.
Interesting, hadn’t heard that one before. My understanding of how thermals form is that the air is pretty stable in the morning. In the afternoon the heated ground heats the air resting on it, which then rises like a hot air balloon (even shaped like one, i.e. pretty spherical). They tend to rise off equator-facing (Sun-facing) hillsides, with a preference for local hilltops.
Some descriptions of thermals depict them as a stream of rising warm air, others as a sequence of very large bubbles of warm air. It would make sense I guess that the former happens on hotter days, much like the difference between water from a slowly dripping tap vs. a running one. Not something I know a lot about.
I must say however that I’m impressed at the number of theories of reality that have been accumulating in this thread. While I don’t think of experimental physics as a religion (how do you argue with an apple landing on your head?), I’m starting to understand that theoretical physics has much more in common with religion than experimental work. They didn’t warn us about this when I majored in theoretical physics in 1966.
Vaughan, “The increased volume exactly offsets the decreased density, no need to change either the net mass or the pressure in order to account for changes in volume and density.”
Yeah but, that increasing volume would be pushing in all directions and need to be offset if the cloud base is to remain at a fixed altitude. Can’t have cloud bases jumping around if the base is isothermal.
It seems that the particular mechanism involves vapour to condensate transition (the ideal gas law is somewhat moot) and thus the volume change acting as a suction to amplify the upflow of warm air below. The condensate then spreads out horizontally as momentutum is conserved at the cloud height. Speaking as a humble hydrologist – this seems entirely feasible. Haven’t looked at the math in any detail – equations 1 and 2 seem entirely conventional.
Oh – and you should remember that Stephen is a lawyer by trade and may push the limits of the laws of physics unless actually enforced in court.
@captdallas: Yeah but, that increasing volume would be pushing in all directions and need to be offset if the cloud base is to remain at a fixed altitude. Can’t have cloud bases jumping around if the base is isothermal.
That’s exactly right. Furthermore what everyone has been dismissing as a tiny 0.17% implosion (a mere reduction from 620 to 610 hPa even with such a huge reduction from 200% RH to 100%) is now seen when latent heat is taken into account to be a massive 9% explosion!
Makarieva et al could now rewrite their paper from the point of view that their previous paper had elicited indignant protestations from everyone that they too had been modeling the same teeny tiny 0.17% implosion all along, when the reality is that this whole model is completely wrong and that the latent heat of condensation causes a massive 9% explosion.
Now that would be a genuinely groundbreaking bit of research.
They would of course have to come up with a whole new set of phenomena that this explained, such as condensation over Africa driving valuable nutrients from the Sahara to the Amazon. The difference would be a pressure from Africa that is 50 times the suction power from the Amazon in the previous theory.
“is now seen when latent heat is taken into account to be a massive 9% explosion!”
You’re right that it’s an expansion. But isn’t it more like 0.9%?
Opps factor of 10 issue – 9% is OK, but it should be 1.7% contraction.
@NS: Opps factor of 10 issue – 9% is OK, but it should be 1.7% contraction.
Oops, you’re completely right. 10 lashes with a wet noodle. Thanks for keeping me honest, Nick.
So the improvement is only from -1 to +5, not to +50. Sigh. Oh well, that still should move nutrients faster from Africa to the Amazon. Wonder if that condensation mechanism for pumping pressure is patentable…must be if Amazon’s 1-click ordering is patentable. ;)
Vaughan
i) The lower surface pressure arises once the air parcel begins to ascend. If it stays in position you are right but it doesn’t. Low pressure develops below rising air and high pressure below descending air.
ii) The mass per unit of volume changes when contraction occurs. That results in higher density and heavier air which must then fall.The same number of molecules in a smaller space. Note that dry air is always heavier than water vapour. It is not just a matter of the number of molecules.
iii) The air parcel with water vapour rises like a balloon because resistance from above is then less than resistance from the sides. The expansion laterally increases pressure laterally but pressure from above stays the same because even if the atmosphere were to expand there would be the same number of molecules above the air parcel. That doesn’t apply horizontally because the distance around a sphere is fixed so that distance cannot change in order to offset lateral pressure changes.
These are not new theories. They are facts implicit in the known physics.
Vaughan,
Here (Appendix 5) is Dr Rosenfeld’s similar calculation, made in rejecting the earlier paper. Much of Sec 2 of this paper M13 is a process of working through the frustration of this simple and clear refutation, completely of course missing the point. Here is Dr Meesters making the same point.
Nick,
From this point of view it’s difficult to say what the paper is really claiming and what not. They make errors at various points and have gaps in reasoning, but then they start again with reasonable discussion of further points.
They do discuss many relevant factors and not always erroneously, but what they do tries to cover a hugely wider range of issues than a single paper can cover.
The principal problem of the paper may thus be that they try to do an undoable. In that attempt they drop the normal stepwise approach to understanding the atmosphere and try to jump directly in the middle of a lengthy analysis. In that they make the simple technical error that we have been discussing.
Beyond that the argumentation gets qualitative, based mostly on reasonable ideas but ending up in various erroneous conclusions.
Pekka,
Yes, but some of this is amazing. Sec 2.2, for example, has 16 equations, lots of greek, dedicated to proving that under adiabatic conditions at constant volume, condensation cannot occur. But of course no-one has ever said such a thing. Constant volume, no heat in or out – nothing can happen. You don’t need fancy math to see that. It has nothing to do with whether in the free atmosphere condensation causes contraction.
Nick,
I agree. The early parts contain some strange sentences that one, which I have referred to as strawman argumentation.
Otherwise I don’t know about any serious errors up to equation (18), or actually up to equation (33). The discussion of 3.3 and 3.4 is irrelevant and misleading, but probably mathematically right.
There are some minor issues likes the point on equation (1) that L is normally enthalpy of vaporization and the words “heat of vaporization” refer to that, but that’s not serious. I haven’t checked any of the derivations of that part, but have no reason to suspect that mathematics.
Too many typing errors again, but hopefully understandable.
Vaughan, “That’s exactly right. Furthermore what everyone has been dismissing as a tiny 0.17% implosion (a mere reduction from 620 to 610 hPa even with such a huge reduction from 200% RH to 100%) is now seen when latent heat is taken into account to be a massive 9% explosion!”
Using the 30C at 100% RH it would be more like up to 4%, ( 0.9 specific volume initially with about 0.86 on the lower end but it would depend on the the area below the cloud base and the rate of energy transfer inside or above the cloud base). If the induced draft gets a twist, then conservation of momentum kicks in which does look like it would be limited by the delta (Pws- Pda).
This is where the paper need to make it or break it. There is nothing new about pressure differentials, but using molar mass which is more easily measured by satellites, gives it a neat twist IMO.
@NS: Yes, but some of this is amazing. Sec 2.2, for example, has 16 equations, lots of greek, dedicated to proving that under adiabatic conditions at constant volume, condensation cannot occur. But of course no-one has ever said such a thing. Constant volume, no heat in or out – nothing can happen.
Wait, what? If condensation were to occur suddenly (“in a flash” as I put it in my 200% RH example) in a super-saturated parcel, the inertia of the surrounding air would oblige constant volume. But if as “everyone” seems to be claiming “nothing can happen” then that would imply that at 200% RH condensation cannot occur suddenly, even if only to 190% RH. I don’t follow the reasoning behind that, please explain. Is there some assumption you’re making that is invalidated by the super-saturation assumption of 200% RH?
The passage from the first to the second paragraph of Section 3.3 of the paper (incorrectly referred to as Section 3.2 in my example) results in roughly 200% RH at 4 km altitude in the first column of air, which was the basis for my example. Unless I’ve misunderstood something, that second paragraph is talking about essentially my example.
What I can see is that, contrary to what I claimed in the example, condensation to 100% RH does not imply that half the water vapor condenses out (the value I used in my 19-step argument). This is because condensation raises temperature and hence the partial pressure for 100% RH must rise too!
I would guess only a quarter that amount will condense out, namely .075 g instead of 0.3 g, releasing 2260*.075 =170 J while lowering the partial pressure from 20 hPA only to 17.5 hPa, not to 10 hPa. Heating the resulting cocktail requires
20.7643 + 28.03*(17.5/600) + 74.53*(2.5/600) = 21.9 J/deg
whence the temperature increases by 170/21.9 = 7.75 C, raising 280 K to 287.75 K or almost 15 C, which according to my steam tables is the dewpoint at a w.v. partial pressure of 17 hPa. Only 0.5 hPa away from my guess of 17.5 hPa, pretty good guess eh?
So yet again I have proved myself wrong. Instantaneous condensation from 200% RH does raise pressure, but only a factor of (617.5/620)*(287.75/280) = 1.0235 or 2.35%, not the 9% I claimed.
On the other hand the same analysis shows that if the pressure increase due to temperature is neglected, the pressure decrease from 620 to 617.5 hPa is now only 0.4%. 2.35% is 5.9 times as big as 0.4%. So the ratio of the two hasn’t decreased, in fact it’s increased slightly over the ratio of 5.
A slight refinement of my guess would make all this exact, but shouldn’t change anything much.
Vaughan,
According to the textbook that I have (Wallace and Hobbs), the oversatutation in atmosphere is seldom more than a couple of percent and never more than about 10%. The oversaturation of 10% is enough to induce condensation in pure air without excessive delay, a small amount of aerosols is enough to limit the oversaturation to a few percent.
When the oversaturation is removed that doesn’t occur as a flash event over significant volumes. The real atmosphere stays close to local equilibrium and the processes considered are not only nearly adiabatic but mostly also nearly isentropic. All elementary derivations are based on the isentropic processes. That applies also to the present paper except for the chapter 3.3, which not essential for the remaining paper.
Vaughan, they key word was “adiabatic”, for which you should read “reversible”. 200% RH to 100% is not a reversible process. The moist adiabatic process is only reversible if it doesn’t supersaturate, or if cloud doesn’t persist below 100%. Everything occurs at the vapor-liquid equilibrium state of 100% RH.
Vaughan,
,“Wait, what?”
Nothing can happen because the situation is so constrained. A parcel of air at constant volume – well, gas law and all, must be at constant pressure, unless the temperature changes. But it’s adiabatic – can’t even exchange heat.
Even if it was supersaturated when put in the tank, if a little bit condensed the LH could not escape, so the temperature would rapidly rise until it wasn’t supersaturated any more.
@NS: Even if it was supersaturated when put in the tank, if a little bit condensed the LH could not escape, so the temperature would rapidly rise until it wasn’t supersaturated any more.
Yes, I fully agree with that in every detail, Nick. Not only that but my math shows precisely how much the temperature would rise by, namely 7.75 C or so, the point at which it ceases to be supersaturated. Do you see any error in my calculation?
I even took into account the different heat capacities of all three of dry air, water vapor, and liquid water. Usually people gloss over this by neglecting the latter two. While this is a perfectly reasonable approximation in practice, a stern tone is in order for this paper and handling all three together should leave no tone unsterned.
@PP: According to the textbook that I have (Wallace and Hobbs), the oversatutation in atmosphere is seldom more than a couple of percent and never more than about 10%. The oversaturation of 10% is enough to induce condensation in pure air without excessive delay, a small amount of aerosols is enough to limit the oversaturation to a few percent.
I’m fine with that, Pekka. In fact in my first comment on this thread on February 3 I pictured a gradual rise with RH remaining at or below 100%. My main error there was misinterpreting the paper’s claim as being that condensation reduced temperature rather than pressure. Neither is true of course.
In sharp contrast, Section 3.3 of the paper sets up a situation where RH is 100% throughout an isothermal column, and then in the 2nd par. drops the temperature to the MALR profile leaving the surface temperature T_0 unchanged. The RH is still 100% at altitude z = 0 but climbs with increasing altitude so as to reach 200% at somewhere around 4 km (my very rough estimate, not sure what theirs was if any).
My example dealt with that situation under the assumption that condensation reduces the 200% RH down to 100%. Whereas the paper claims this condensation depressurizes the column, I claim it raises PV by 1.5% to 2% depending on whether it happens quickly or slowly, but that ultimately PV will have risen by 1.5% because pressure P won’t have changed and volume V will have increased by 1.5% (the limiting slow case).
I have no problem with the point that RH = 200% (more or less depending on what they think the w.v. scale height should be) is ridiculous in a real atmosphere. Everyone’s been saying that, I just wanted to explore a different facet of that premise without simply dismissing it out of hand.
@PP: When the oversaturation is removed that doesn’t occur as a flash event over significant volumes. The real atmosphere stays close to local equilibrium and the processes considered are not only nearly adiabatic but mostly also nearly isentropic. All elementary derivations are based on the isentropic processes. That applies also to the present paper except for the chapter 3.3, which not essential for the remaining paper.
This is why I gave the instantaneous and very-slow cases separately: to serve as boundary cases for reality which will be somewhere in between (where exactly I don’t know but see the next paragraph). Sticking to the absurdly high (for a real atmosphere) value of 200%, the very-slow or constant-pressure case raises the temperature and hence volume by 6/4 = 1.5% (my first estimate of 6% requires the same factor-of-four correction as the instantaneous constant-volume case when taking into account that rising dewpoint raises the corresponding w.v. partial pressure, which I’d previously neglected). This extreme is somewhat lower than the 2.25% of the instantaneous or constant-volume case. Reality is somewhere in between.
This link gives a formula for condensation rate in clouds. The rate is proportional to RH − 1 (interpreting RH = 110% as RH = 1.1), e.g. 110% is an oversaturation of 0.1 while 200% is 1.0. The rate is also inversely proportional to the cross section of the cloud particles, and is therefore extremely low in an aerosol-free cloudless atmosphere, whence the slowness in starting condensation in the absence of catalytic nuclei.
With that formula in mind, an interesting experiment would be to mix a mole of very clean dry air at 276 K with 1/30 moles of very clean steam at 100 C in a 38.8 liter container with very low thermal conductivity and watch what happens.
While I haven’t tried this experiment myself, I foolhardedly predict that the resulting mixture will start out as supersaturated air at temperature 280 K and pressure 620 hPa with RH = 200% (anyone who cares will have no trouble verifying this using only the specific heat numbers I’ve given earlier, no thermodynamic needed), and (more importantly for thermodynamics) that it will in due course become air at T = 287.7 K and P = 634.5 (higher pressure) with RH = 100%.
Skeptics would then be duty bound to find theoretical objections to this little thermodynamic experiment in case it met my predictions. But if it didn’t meet them they would take that as license to return to this evidence of my incompetence whenever I claimed anything that they disagreed with in the future no matter how unrelated. Skepticism used to be more rational, but regrettably this is how it operates in those countries benefiting most from capitalism. (As a huge beneficiary of capitalism myself I’m only mumbling vaguely about its side effects, not its clear benefits. In that respect capitalism is rather like chemotherapy.)
Of course when rain starts falling the resulting downdraft changes all of the above analysis very significantly. Since the paper’s analysis did not go there I see no reason to do so myself in critiquing the paper.
Vaughan, “I even took into account the different heat capacities of all three of dry air, water vapor, and liquid water. Usually people gloss over this by neglecting the latter two. While this is a perfectly reasonable approximation in practice, a stern tone is in order for this paper and handling all three together should leave no tone unsterned.”
Yes they should, if that were what they were trying to do. Instead they are treating the water (liquid) as the stable base and not concerned about the latent energy per se, but the pressure in terms of molar density. The base of liquid water at a stable temperature and pressure would be like the blade of a fan, only the blade is fixed and the pressure is the induced.
Kinda neat really.
What they have completely ignored in this paper is the famous quasi-sawtooth.
@Jim D: Vaughan, they key word was “adiabatic”, for which you should read “reversible”. 200% RH to 100% is not a reversible process.
From the Wikipedia article on “adiabatic”: “An adiabatic process is any process occurring without gain or loss of heat within a system (i.e. during the process the system is thermodynamically isolated- there is no heat transfer with the surroundings). This is the opposite of a diabatic process, where there is heat transfer.”
(Not to be confused with a diabetic process, where there is sugar transfer to the kidneys.)
This is the definition I’ve been working with. It says nothing about reversibility, and moreover it should not given that it is not obvious how to reverse the conversion of say the kinetic energy of a meteor to thermal energy. Such conversion is so sudden as to be basically adiabatic. Same for your disc brakes getting hot when braking to a halt. These are adiabatic conversions, even though they are not reversible (though the Prius Hybrid does indeed aim at making braking more reversible).
@DM: What they have completely ignored in this paper is the famous quasi-sawtooth.
Unlike Don Waste-My-Time, for whom it has become an obsession…
Yo, doc
You are spending an awful lot of time attacking Annastasia, et al. Shouldn’t you be working on your ill-conceived parody, of Wood’s parody of a physics experiment? And your deep earthquake BS? The AGU meeting will be coming around again, before you know it. Get your little easel polished up.
All this time I’d been thinking that DM was trolling me, and now it turns out I’ve been trolling him.
Saves bait, now that he’s taken mine he can simply offer it back to me. A whole new ecosystem there, much more efficient than the old model…
Vaughan, I think thermodynamicists equate adiabaticity with reversibility because there is no increase (or decrease) of entropy in these processes, so in principle the same thing can happen backwards. This cannot be said of condensing from supersaturation because that heat can’t then be removed without reducing entropy to go back to 200%. In an opposite sense, rain evaporation is also not reversible because it can’t happen backwards either in subsaturated conditions.
Jim,
That’s perhaps a common bad practice, but it’s better to use isentropic for the reversible processes. That way the ambiguity is avoided. In thermodynamics isentropes are discussed commonly rather than adiabats. The term “adiabatic expansion” is, however, very common in practice.
In most cases the adiabatic processes discussed are isentropic, but not always. Condensation without supersaturation is essentially isentropic as long as the droplets don’t fall, but their descent under gravitation is neither isentropic nor adiabatic, as they carry heat across the boundaries of a parcel of gas. Condensation with supersaturation is adiabatic but not isentropic. Similar considerations apply to chemical reactions when the occur in a state off from equilibrium. The term adiabatic is used regularly in that connection for a process that’s not isentropic. Look, e.g., for “adiabatic flame temperature”.
Pekka, yes, again from Wikipedia, it seems to equate adiabatic and reversible applies mostly to perfect gas processes with only phase and/or pressure changes. In other systems they may not equate.
Stephen the warming does not immediately reduce density if you look at short time scales. First it increases the local pressure, then that drives air outwards from the heated volume, and that reduces density. It happens at the speed of sound, but that is the order of events. The density can’t change without some kind of expansion velocity to move the mass out.
Jim,
We are mostly interested in gradual processes, cases like flash condensation are very rare. Thus most of the calculations are done assuming that all quantities are at every moment very close to the local equilibrium value of the quantity. Thus the density follows warming without any delay that needs to be considered.
Looking at the dynamics more closely it’s not possible to say which variable moves first and which follows, they all change having their values just a little off from the local equilibrium value. Studying the small scale nonequilibrium dynamics is not done practically at all. On larger scales the dynamics is studied, but the issues are then different.
Pekka, yes I agree this is all gradual in natural condensation. However, the paper is about a pressure reduction due to vapor loss that is also more than canceled by the pressure increase due to latent heating, and both of these are part of the effectively instantaneous adjustment to a lower density. Their pressure reduction is lost immediately, just the way expansion occurs immediately as a result of the net pressure increase. The paper is therefore about a small component of the expansion process that is never seen in reality because it is too fast.
Pekka, “We are mostly interested in gradual processes, cases like flash condensation are very rare. Thus most of the calculations are done assuming that all quantities are at every moment very close to the local equilibrium value of the quantity. Thus the density follows warming without any delay that needs to be considered.”
With changes in the average energy and distribution of energy it is hard to say what should or should not be considered. SSW events appear to have a cyclic nature and are capable of transferring huge amounts of energy. Deep convection events not only transfer energy but mass that interacts with the stratospheric chemistry. There should be some caution applied before assuming something is insignificant.
In some sense it pointless to discuss, what the paper is about in that sense as it’s so terribly wrong.
It’s also in other ways a bit pointless, because they don’t give any unique answer on, what they consider. Looking at the the more substantive parts of the text they do consider the same physics as others, but their discussion becomes impossible to interpret fully because the huge errors make such considerations differ strongly that would not do that without their error.
CaptDallas,
This paper studies physics mainly trough differentials, i.e. it looks at local micro level physics. That’s always a useful step (as long as serious errors are avoided). It’s clear that looking at micro level does not answer all questions, and that’s also one problem of the paper in addition of it being wrong. They don’t even try to really bridge the micro level physics with large scale studies (which didn’t stop them from drawing conclusions). Building such a bridge takes a book, not an article.
Pekka, “They don’t even try to really bridge the micro level physics with large scale studies (which didn’t stop them from drawing conclusions). Building such a bridge takes a book, not an article.”
I don’t think they are even trying to get crazy with the micro physics.
http://journals.ametsoc.org/doi/pdf/10.1175/1520-0469(2003)060%3C2957%3ASOWVIC%3E2.0.CO%3B2
That paper gets into some of the issues and still has this, “A rigorous mathematical proof of Eq. (32) is not trivial and it goes beyond the scope of this paper. Numerical modeling indicates that Eq. (32) is valid both for liquid, ice and mixed clouds.”
The problem I think is trying to make it something it isn’t. The isothermal saturated layer in the x (and y) dimension would be liquid. As you increase in z above that liquid base, there would be condensation. Since the liquid base is saturated, liquid and lower in altitude than z, it would have a higher pressure, but since that layer is liquid, it is not going to evaporate to replace the condensation. You have a local low pressure where with the x-y fixed, the replacement air will have to pass over the x-y layer from some distance at some velocity to match the upward expansion of the cell. It is just saying that there is an induced draft that is limited to the x-y horizontal at some zo + some higher altitude. Condensation from that induced draft would sink to the zo layer and flow outward. I don’t think that is anything earth shattering. What they are claiming is that knowing the limit in z they can determine the width in x and the potential energy that will become wind or breeze.
They probably just used x since they are looking a lines like the ITCZ, but should have at least threw a bone in the y direction.
I don’t know if that is special or not. What it does though is recognize that the condensation base is a useful reference layer. It would be more than saturated with water vapor, it would be saturated with liquid water which could precipitate through that layer, but the layer would remain saturated until the system broke down.
What I think you and Nick are assuming is that the isothermal layer is the source of the condensation instead of the sink. If that is not what the authors are saying, then you and Nick would be right.
I’d like to provide my summary of the discussion in that part that concerns our exchange with Nick Stokes and how it relates to paper content and the broader picture of our work.
I. The “0=1” argument.
It was repeated by Nick here, here, here and elswhere.
This argument presumes that our system of equations (32)-(34) is mathematically inconsistent (contains a “mathematical fallacy”). This argument is false. The system is mathematically consistent. Those willing to consult an independent opinion please see the comment of Tomas Milanovic.
When challenged to derive the contradiction “1=0” explicitly from (32)-(34) Nick made two elementary algebraic errors (see here and here). At one place he seemingly implicitly conceded that there is no mathematical contradiction (but still made a later effort to derive one).
My take on the “mathematical fallacy” issue is here, here and here.
The argument “1=0” profoundly misinforms the reader about the actual issues that are worthy of discussion.
II. The derivation of Eq. (34).
It should be made clear that Eq. (34) in the paper was not derived from any other known equation. It was formulated based on several key physical propositions. These propositions are summarized here. They deserve a thoughtful physical consideration by anyone willing to understand the underlying physics.
1. The scenario of discussing these propositions developed as I suggested in an earlier comment here:
In line with this, Nick once called our propositions arm-waving, another one that they are of “no relevance”. His most specific claim was
When asked if any formula that contains something depending on gravity must explicitly contain g, Nick changed topic. To this I responded that “to see no relevance” and “find an error” is not the same.
2. Additionally, Nick produced an alternative derivation of S (the one also presented by Dr. Held in his review with a reference to Nick). This derivation was independently reproduced by Cees de Valk. See here for a relevant thread.
Note that Eq. (34) is S = wN∂gamma;/∂z, where γ = p_v/p is the relative pressure of water vapor and N is total air molar density. The alternative derivation of Nick et al. produces what we can call S_Nick = wN_dpart;gamma;_d/∂z, where N_d is dry air molar density and γ_d = p_v/p_d is the ratio of water vapor to dry air pressure. Since p_v << p, p_d and p are very close, so S (34) and S_d differ by a small magnitude.
Now important: we do not have any objections as to from what principles S_Nick was derived. I accept a priori that it can be correct (which would mean that our S (34) is wrong). That would be fine. But, as everybody agrees, when put into the continuity equations (32)-(33), S_Nick produces a non-sensical result:
u∂N/∂x = 0. (1C)
This result is mathematically valid (no contradiction in the equations), but it implies that generally, under any possible circumstances on an isothermal surface, winds cannot have a velocity component parallel to the pressure gradient ∂p/∂x = RT∂N/∂x.
So this result is invalidated by observational evidence (e.g. radial convergence in hurricanes). By inference, whatever were the physical grounds from which S_Nick was derived, they were incorrect.
In contrast, our S (34), despite different from S_Nick by a small relative magnitude, when fed into the system (32)-(33) produces instead of the above equation
-u∂N/∂x = S. (2C)
It is a classical case where a small detail matters (“devil in details” etc.) To understand why it matter requires a deeper insight into the underlying physics, not merely manipulating with the continuity equations.
3. There is an alternative physical derivation of S (34) as presented in the blog and here based on energy consideration. This complementary view, although explicitly present in the paper, might not have been sufficiently clearly articulated. This alternative derivation does not involve any small factors in deriving (2C) above. These considerations provide independent support for the physical arguments used in deriving (34). None of the commentators here, Nick included, has ever commented on that.
In the meantime, we now think that this second view on how (2C) (the main result) is obtained is more easy to understand from the physical viewpoint. We recommend people who just have their first look on the subject to evaluate it first.
4. The argument about S = CNv (see here, here, and here) is confusing.
Otherwise this argument is a distraction. When challenged during the ACPD discussion to better explain the physical foundations of Eq. (34), we showed that Eq. (34) can be interpreted as S = wkvNv, where kv is the degree to which water vapor partial pressure deviates from hydrostatic equilibrium. This proportionality is physically consistent with S being a first-order reaction in Nv, but certainly S = wkvNv cannot be formally derived from chemical kinetics.
From the empirical viewpoint, since C is not a constant (it does not depend on Nv, i.e. it is a constant with respect to Nv), the formula not presume rain from dry air. Since to appreciate this linearity argument requires an understanding of what kv is, this argument is of little help for those who want to get a first idea of the physics behind (34). Again, I recommend this account.
III. Issues of interest
1. Since S (34) is not formally derived from any pre-existing equations but formulated based on several plausible physical propositions (two independent sets of them), it cannot (and has not been) refuted mathematically. It has not been shown to be in conflict with any physical law either. The only way to falsify Eq. (34) consists in checking the result it yields
-u∂N/∂x = S. (2C)
against empirical evidence, as I outlined here. Nick made a few attempts, but they have so far been inconclusive.
I emphasize (2C) is an extraordinarily strong statement (see Eq. (4) in the post which is the same). It predicts that where condensation is absent, winds cannot blow along the pressure gradient or that the pressure gradient must be absent. It also predicts the reverse (lack of condensation where u∇p = 0), but the latter prediction is less informative as it does not specify the scale at which this lack of condensation should be manifested (it can be very narrow).
The meaning of the differential form of 2C (see Eq. 4 in the post) is that all potential energy released from condensation is locally converted into the power of the large-scale horizontal pressure gradient force u∇p. This is very strong. At what scale it is actually true remains to be seen. It will also help to discriminate between condensation-induced dynamics and other mechanisms at work in the atmosphere (e.g. forced convection can be different). As a bottom line, the integral form of (2C) has already produced meaningful results, so it can serve as an integral limitation on the dynamic power of circulation.
For us the main point is that our theory (unlike the existing models) yields empirically falsifiable predictions. It is a working theoretical concept for a moist atmosphere.
2. The last but one section in our blog is very important for future theoretical analyses.
3. How the theory can be empirically tested is outlined here and here in response to manacker (Max). It is said quite enough for anyone who got a basic physical idea to publish original papers based on observational analysis. There seem to be lots of relevant data around.
IV. Miscellaneous
1. My personal view on why the paper was accepted.
2. I would like to express our gratitude to Nick Stokes for his persistent attention to our work. My personal view: For people who like us have a clear picture of underlying physics Nick’s comments provide additional details and angles. For those people who do not still have a clear physical picture and make a first acquaintance with the idea, Nick’s comments are paralyzing and preventing any further understanding. Cannot be recommended for students.
Thank you very much for this exciting discussion.
Anastassia
Comedy!
• Strong physical theories can be derived by many mathematical paths (all leading to the same final expressions) — this vital mathematical property is called “naturality”.
• Weak/wrong physical theories cannot be derived by even one mathematical path — Anastassia’s post (in effect) stipulates that her article’s theory is of this latter (weak/wrong) kind.
Anastassia Makarieva, you and your coauthors urgently need to recruit skilled mathematical expositors to your research team! You will find no shortage of mathematicians who take pleasure in deriving (correct) results by multiple mathematical paths.
Conversely, if you cannot present your equations by multiple, independent, mathematically natural paths, then the question you ought to be asking yourself, is whether your mathematical manipulations are valid at all! As contrasted with “dubious manipulations you do to make yourself feel better about your physical ideas“.
FOMD, “Comedy!”
Yeah, you are pretty funny. If air is rising in a moist lapse rate and condensation forms a isothermal layer, the motion has to be conserved. Static pressure is converted to velocity pressure, pretty much like air over a wing converts velocity pressure in to static lift. The explosive release of latent energy increases the local pressure at the condensation layer forcing down on the condensation layer which is immobile due to the condensation temperature limit forcing, expanding air out in all directions while condensation cause a smaller scael implosion. This process is fed by the pressure differential which induces fuel, moist air from the same and higher pressure level and dry air from the lower pressure level.
It is a floating steam engine :)
So they have a grasp of the process that appears to be over your head, they just need to fine tune the limits in their math.
Fan, here is a pretty picture show to watch.
http://www.eldoradocountyweather.com/current/satellite/south-america-wv-satellite-loop.html
Speaking of naturality:
http://boole.stanford.edu/pub/coimbra.pdf
Pratt, Vaughan. 1999. Chu Spaces. School on Category Theory and Applications, University of Coimbra.
It floats, like a cloud bounced off your hand.
=====
LOL … Google Ngrams (which conveniently tracks phrase usage) plainly shows us the ‘Hockey-Stick’ acceleration of mathematical naturality since 1900.
Conclusion Prior to (say) 1960, texts on (thermo)dynamics and/or fluid mechanics in general didn’t discuss notions of mathematical naturality. So it’s time to upgrade!
This comment was written as response to the above comment of Anastassia Makarieva.
Well, you’ve heard from me before. So I’ll just set out how the section with 32-34, which generates a the wild results, should be done properly.
Start with a proper 3D vector form of the mass conservation equations. Nd is the molar density of dry air, Nv of vapor.
v is 3D velocity and ∇ is the 3D derivative vector:
0 = ∇ .(Nd v)=v.∇Nd + Nd ∇ .v (32)
S = ∇ .(Nv v)=v.∇Nv + Nv ∇ .v (33)
Since N=Nd+Nv=total air
S = ∇ .(N v)=v.∇N + N ∇ .v
(33a)
Now the objective for the moment is to remove velocity gradients. Since these are all in the scalar divergence ∇ .v, that is just a matter of linear algebra (elimination) on 32, 33.
S = v.(∇Nv – (Nv/Nd) ∇Nd) = Ndv.∇(Nv/Nd) (34a)
to use the form preferred by Cees.
So far this is just a manipulation – everything is reversible. But it has the form of this paper M13’s (34). It is not a new equation; any pair of 32,33,33a and 34a can be used to regenerate the others.
Now we approximate. Originally the 2nd y dimension was dropped. The implications of this have not been well considered – it leads to a very artificial geometry, but we then have in components from 34a:
S/Nd= u ∂/∂x (Nv/Nd) +w ∂/∂z (Nv/Nd) (34b)
Then M13 contend that the problem can be reduced to a single vertical dimension. The assertion here is that the horizontal rates of change are negligible relative to the vertical. It is not a claim that they are zero; just that the vertical components will be little changed if they are omitted. That leads to
S/Nd= w ∂/∂z (Nv/Nd) (34n)
I’ve called it 34n to distinguish from the M13 34. It’s exactly the same except that Nd is replaced by N. The reason is that Anastassia in effect used 33a, but neglecting S, on the basis that air was a suitable non-condensable reference. I’ll refer the qn of non-condensable air to anyone who has ever been caught in the rain! If that equation had been correctly used with S, 34n would have also resulted.
Now Anastassia says 34n can’t be right because it leads to “nonsensical” u ∂N/∂x=0. But it only does this if you make the rookie error, as in M13, of not removing the original equation from the set when you provide an approximate replacement. Here’s how that happens. If you say 34a and 34n are both true, then you can subtract them:
u ∂/∂x (Nv/Nd) = 0, or
u ∂ Nd/∂x = u ∂ Nv/∂x
Since u∂ Nv/∂x=0 had been assumed, it can be added to both sides:
u ∂ N/∂x = 0
This isn’t nonsensical – it reflects what was assumed in going from 34b to 34n. It’s out of context; it was neglected relative to the vertical terms, but not assumed zero. That last came with that rookie error, which is exactly what Anastassia does with her 34.
And then you get a succession of nonsensicals. I saw a suggestion that her u ∂ N/∂x = S should be experimentally verified. Yes! Set up a big tank with dry air and a downwind density gradient. Just collect the water. You’ll be famous.
“If you say 34a and 34n are both true” should be 34b and 34n
Clouds form where the temperature is about equal to the dewpoint at a minimum of 7-% RH.
https://courseware.e-education.psu.edu/public/meteo/meteo101demo/Examples/Section6p03_5b.html
There are about 0.00025 L/m3 of water in stratus – condensed from some 0.425L of vapour. The question of the significance of this can only be resolved in an appropriate mathematical framework – which I have not got to yet. My gut tells me that %ages don’t do it for me.
Release of latent heat drives the expansion of clouds upwards in the development of thunderheads. I would doubt that they are additive processes operating in the same space.
Pingback: Summary of discussion at Climate Etc. | Stormy Science
Anastassia Makarieva | February 5, 2013 at 12:01 am | Reply
Your comment is awaiting moderation.
My comment with the discussion summary got stuck in moderation, so I duplicate it here.
Anastassia,
1) You have not presented any real independent derivation of equation (34). You have only some arm waving to support that.
2) The only credible basis for ending up with that equation is that it’s picked from some derivation of an approximate continuity equation. It’s exactly such an equation and you have not presented any alternative explanation for its form.
3) Whatever the origin of the equation of (34), combining it with equations (32) and (33) leads to series of results. Many of these results are so totally against real world situation that the set of equations (32), (33), and (34) is totally refuted by observation. The most obvious case is perhaps that it leads to the result that condensation is fully controlled by the horizontal pressure gradient. Without such a gradient there could not be any rain. That’s just absolutely false result at the extreme.
Thus you have no derivation for (34) and (34) leads to totally false results.
The only possible conclusion is that the equation (34) is totally wrong when it is presented as a third equation with equations (32) and (33).
Chierf Hydrologist said:
“Release of latent heat drives the expansion of clouds upwards in the development of thunderheads. I would doubt that they are additive processes operating in the same space.”
With respect, that is wrong.
The presence of latent heat in the form of water vapour drives the expansion of clouds adiabatically.
The release of latent heat via condensation terminates further expansion from that latent heat. In practice more comes up from below faster as long as the cloud keeps growing but once the resupply from below reduces below the rate of condensation the cloud will dissipate.
Condensation results in adiabatic descent of the drier air parcel which increases surface pressure. Usually the drier parcels are pushed away to one side by the continuing updraft so the higher pressure is adjacent to the thundercloud. If the continuing updraft fails to push it aside then the cloud will dissipate.
Condensation cannot reduce surface pressure, it can only increase surface pressure. In order to reduce pressure it would need to have the net effect of making a dry air parcel lighter and thus more buoyant than an air parcel containing water vapour.
It cannot do that once the effect of contraction (increased density and reduced buoyancy) has been offset against the release of latent heat (higher temperature and increased buoyancy) and upward radiation has taken another slice.
Stephen,
Condensation releases heat. That heat cancels part of the cooling that the adiabatic expansion would lead to without condensation. Therefore the moist adiabatic lapse rate is smaller than the dry adiabatic lapse rate. Uplift continues as long as the rising air is lighter than the surrounding air at the same altitude. Even with the moist adiabat that’s over at some altitude (the moist adiabat differs very little of the dry adiabat when the saturation value of the vapor partial pressure is as low as it is at high altitudes).
At no point the above leads to the situation where air that starts to subside. For that additional cooling is needed. That additional cooling is provided by the emission of IR by CO2 (and other GHG’s but CO2 is the most important in this) by the high altidude air. This is not a comprehensive description of circulation, but this is the main mechanism.
Condensation does not “release” latent heat to the surrounding air; it recaptures it as sensible heat in the condesate, which occupies a very much smaller volume than the vapor. There is no temperature change accompanying the phase change. This condensate is a much stronger radiator of IR than any of the trace GHGs on a per volume basis. It’s astonishing how deeply “climate science” has muddled this basic physical process.
http://www.youtube.com/watch?v=kapTREk0gXg
Some of these penetrate to the stratosphere. And we are talking specific types of cloud. There are obviously convection cells and perhaps some momentum. The exothermic process of condensation releases heat but until then the latent heat ‘lies hidden’ as the Latin derivation suggests. Warm, moisture laden air rises untll it loses sufficient heat to cool to the dew point when condensation occurs and the heat becomes sensible and the convection cells bubbles upward and outward. The question is whether the loss of volume on condensation creates a sufficiently strong vacuum effect to contribute to the observed affects of low pressure at the surface and especially large scale circulations such as Hadley Cells. You will forgive me if I do not reach conclusions but think about this for quite some time to come.
Chief, here is a study questioning the amount of tropospheric overshooting that was thought to occur. Not sure if this will help you in your thoughts or not but I found it interesting.
http://www.atmos-chem-phys-discuss.net/11/163/2011/acpd-11-163-2011-print.pdf
Thank you Steven.
Anastassia, I had a brief look at it again, trying to focus not too much on specific assumptions which may be doubted. If (34) is replaced by dividing it by a factor (1-γ), you get horizontally uniform fields. So to get something meaningful, the relative error in source term (34) should be at most o(γ) (cannot be O(γ)). I don’t know of that has been/can be shown. I think this type of problem asks for analysis of perturbations like nonlinear stability analysis.
Actually, in that spirit, approximating a vertical gradient from the adiabatic rate of change divided by velocity could work better: just like the vertical temp. profile cannot deviate much from the lapse rate.
Pekka said:
“Uplift continues as long as the rising air is lighter than the surrounding air at the same altitude”
Once condensation has occurred the air which has had its vapour removed can never be lighter than surrounding air containing water vapour since water vapour is lighter than air.
Remove the water vapour and the air which lacks it must descend.The heat released by condensation can never heat it enough to make it lighter than air containing water vapour.
If the updraft of vapour laden air continues then the heavier drier air will be pushed aside to fall elsewhere.
Have you never heard of downdrafts within convective clouds ?
Have you never wondered why high pressure cells are located alongside low pressure cells ?
Stephen,
I didn’t say anything about the moisture content of the surrounding air, but it might well be even less than in the uplift.
I didn’t say anything about the level where the uplift ends, only the condition that determines that level. In tropics the level may be very high up, in some other place much less.
There are pressure effects of all kind, but my point was essentially that the uplift stops when the balance has been reached. At that point the air is not heavier than surrounding, it’s equal. Therefore it doesn’t start descending, it stops at that level until something additional changes the balance and for larger scale phenomena that additional is typically radiative cooling.
Chief Hydrologist said:
“Warm, moisture laden air rises untll it loses sufficient heat to cool to the dew point when condensation occurs and the heat becomes sensible and the convection cells bubbles upward and outward”
The convection cell bubbles upward and outward as a result of air laden with water vapour rising within it.Water vapour is lighter than air.
When a particular parcel of air loses its water vapour via condensation it becomes heavier than the vapour rich air around it and begins to descend unless forced aside by the continuing uplift from more vapour rich air.
The sensible heat released can help the rising vapour rich air to rise a little further before condensing out but once air has lost its vapour it can no longer participate in the rising process since the sensible heat released can never make it as light as vapour rich air.
At the top of the column where the updraft can go no higher then all the remaining vapour condenses out and the air containing the remains of the cloud then spreads out horizontally before starting to fall back towards the surface.
In the proces of descending any remaining condensation gets reabsorbed by evaporation during the descending process.
People here are not distinguishing between localised pressure effects where one parcel of air abuts another and surface pressure effects as a result of a rising or descending column.
At the surface, increased pressure from injecting water vapour into a parcel of air via evaporation causes the parcel to rise so that surface pressure below it falls.
At the top of a convective cell decreased pressure from removal of water vapour via condensation causes the parcel to fall so that surface pressure beneath it rises.
The principle applies as much to the Hadley and Ferrel cells as to a single localised convective cell.
People here are not distinguishing between localised pressure effects where one parcel of air abuts another and surface pressure effects as a result of a rising or descending column.
Because they don’t have real gas atmosphere, but an imaginary one of empty space populated by ideal gas without weight or volume.., perhaps because they’ve missed out the water cycle they’ve never got around to thinking in terms of lighter than and heavier than in a fluid medium.
Is there anything intrinsically wrong with the concept of condensation driven winds if this was brought brought back to traditional meteorology? I don’t know what it does to the maths, but if Anastassia reversed the high and low pressure which as it stands could be thought of a pre condensation? So on condensation, under the still rising hotter lighter air carrying on doing its thing and perhaps adding new layers, there would be the accompanying volume and temperature decrease of that first previously rising lighter volume now heavier liquid water and added to by the adjacent volumes of heavier colder air flowing beneath the still rising lighter hotter and into the space now available on condensation of its neighbour, which all now being heavier will increase the pressure at the surface as they all sink together displacing the lighter. This new volume of heavier colder air, triggered as it were by condensation, perhaps speeded up?, is basic wind which is volumes of the fluid gas air on the move.
That ol’ hot muggy day that we wait for the building storm to break to clear the air..?
Reminding me of a day in Fatehpur Sikri after the three day build up of temps during the monsoon. We could see from horizon to horizon from the Buddha tower and watched the massive black cloud appearing in the far distance being driven by the wind towards us, which on reaching us poured rain into all the ingenious levels of gutters and channels designed to capture it as it passed over the city abandoned from lack of water.
Anyway, I think if she transposes her high and low pressure she comes back to the basic wind system.
Stephen – I eschew easy answers and I would suggest that you do too – if that wre not wasting my breath. Please note that I have given up trying to decipher your comments – so unless you are speaking to a wider audience save it.
Near as I can tell every breath of yours is a waste.
But hey, according to the Vaughn Pratt probability calculus with every wasted breath the odds rise that the next breath will not be wasted.
So keep blowing.
Well springer that’s real mature. It comes under the banner of relentless idiocy with no technical content – but perhaps Judith is busy. Pretty much business as usual for you then.
The odds of throwing n heads is 1/2^n – so the probability of continuing the run decreases as n increases. In other words – if you wait long enough it will rain.
I had thought to correct your inprobable probabilities earlier – but hell life is short and there are just too many idiots.
ROFLMAO
The result of each flip of a fair coin is independent of the result of any previous flips.
Casinos love people like you.
You will find that the chance of throwing one head is 1/2^n where n is 1. The chance of throwing 2 is 1/4, etc. You are confused as usual big dave. If you wait long enough it will rain. In fact I would bet the house on it. Willing to take it dig dave? Every day is a new day hey – I guess you have the attention span of a goldfish.
Unbelievable. This is statistics for toddlers.
Each individual flip of a fair coin is exactly 50% chance each for heads or tails. Previous flips have no influence at all on future flips. It doesn’t matter if you tossed a million heads in a row prior, if it’s a fair coin then toss 1 million + 1 is still 50-50.
It would give me real joy to see you plaster this silly mistake about probabilities all over the internet and make an even bigger fool of yourself.
You’re messing with me, aren’t you? The situation you describe is so commonly held among dimwits it has a name, The Gambler’s Fallacy.
http://en.wikipedia.org/wiki/Gambler's_fallacy
I’m being very magnanimous with you giving you an opportunity to say you were just trying to get a rise out of me. I suggest you avail yourself of my generosity.
You’re a total maniac springer. I will bet every cent that it will rain again. Or that a tail will be tossed. I don’t need to bet on each toss. Just on the cumulative probabilities.
Where did this stupid conversation start? “However there’s a higher probability of rain following a long dry spell than a short one.” from VP
This is fairly light hearted and inconsequential and perhaps not rigourously stated. However – it is apparent that a run of 5 heads is much more likely than a run of 10. Then there’s a tail by definition or it rains – whatever.
So all things being equal – there is a greater likelihood of a short dought than a long one. Thus we get into the statistics of extremes. But you decide that no – some sort of idiotic idea that a 50% chance of a head on each toss means anything at all in the statistics of cumulative probabilities.
And insist on this in a bombastic display of just why you are such a f_ckwit.
I am speaking to a wider audience.
There are many who do understand the basic physics and my style of exposition.
That’s funny – I could of sworn that the people who understand atmospheric physics and those who follow your expositions are disjunct sets.
One small example is that consdensation occurs when the air temperature is approximately the dew point. But there are others. The process is of a chain of events – and any false assumption in any of these steps invalidates the whole. There are many examples in this post of broken chains by you and others. What I think you need to do is – as Feynman put it – see the wiggle waggle of the cloud and proceed by reference to physical observations – one step at a time and confirming each step as you go.
yes and people speak to their plants
Steven Mosher
That’s OK, but when they start hearing the plants speak back…
Max
(Climatologists call it “positive feedback”.)
http://hint.fm/wind/
I see NINA!
========
Sometimes we get vertical mixing in an air column, thermals, producing an isoentropic air column. This, however, seldom results in storms.
More commonly,or at least more noticeably, we get large scale horizontal movements because the air column in one region is warmer than the air column in another region, for example the sea breeze, induced by the land becoming warmer than the sea. The sea breeze, however, is a breeze, not a storm.
We get storms when we get rain, which suggests that the storm is caused by the air being warmed by the latent heat of condensation, “warmed” in the sense of not cooling as it rises as fast as dry air would.
Water vapor
http://www.eldoradocountyweather.com/current/satellite/south-america-wv-satellite-loop.html
No NINA, but something like snowflakes.
========
Chief Hydrologist | February 5, 2013 at 12:24 am |
“I have experienced relativity as a flower opening in my mind over decades. ”
Wouldn’t that be fractal thinking i.e. a flower dreaming of a flower?
There is an open invitation to a Texas death match at the Grand Western. Thought you didn’t read it? I would rather you didn’t. And that you stop stalking me. If you have nothing to say – why do you sit there mumbling inanities like some mad crone on ice. Oh…wait…
I stalk opportunities to poke fun at blowhards. Nothing personal. You just blow more than the average bear.
At your usual standard of originality and wit big dave – something about the level of a four year old. Is this really the place?
See you at the Great Western instead – or we’ll all know you scream like a girl.
You mentioned Great Western and Grand Western.
Which is it? In either case I have no idea what the significance is.
Ah – a Texas death match
http://www.urbandictionary.com/define.php?term=texas%20death%20match
“A Texas Death match is a wrestling match”
Just goes to show I don’t know everything.
So you’re inviting me to roll around on the floor with you under hot lights while both of us are nearly naked.
Why am I not surprised?
And here I was thinking it was a no rules, fight to the death – behind the bull ring at the Great Western. A place where a man can swear, drink, spit and ride bulls at the same time. Instead it seems to be just typical American woos stuff. Your imagination going into overdrive is a little disturbing however.
Still the points were – this is not the place – your wit and wisdom is nonexistant – you have failed to rise to the level of substantive comment – you continue to stalk me with about the maturity of a schoolgirl bully. You are a sad case big dave.
Vaughan Pratt | February 3, 2013 at 2:11 am | Reply
“Hey, BC, did you know that rain follows the dance? People don’t enjoy dancing in the mud, they prefer to wait until the ground has thoroughly dried out. However there’s a higher probability of rain following a long dry spell than a short one. So after a while people started to notice that rain follows the dance, giving rise to rain dances as a way to encourage rain.
What’s more it worked, as long as you refrained from dancing in the mud.”
Ah, the “overdue” theory or probability. Casinos love people like you. The more you lose the more you think your next bet will be a winner.
Statistically speaking Pratt, the mostly likely time of seeing a raindrop is immediatly after seeing a raindrop. In any rainstorm there are fewer first and last raindrops than those which fall in between the first and last.
Not that you’d ever stop to think about anything you believe to be true…
I’ll double the chance you don’t miss it.
You’re a total maniac springer. I will bet every cent that it will rain again. Or that a tail will be tossed. I don’t need to bet on each toss. Just on the cumulative probabilities.
Where did this stupid conversation start? “However there’s a higher probability of rain following a long dry spell than a short one.” from VP
This is fairly light hearted and inconsequential and perhaps not rigourously stated. However – it is apparent that a run of 5 heads is much more likely than a run of 10. Then there’s a tail by definition or it rains – whatever.
So all things being equal – there is a greater likelihood of a short dought than a long one. Thus we get into the statistics of extremes. But you decide that no – some sort of idiotic idea that a 50% chance of a head on each toss means anything at all in the statistics of cumulative probabilities.
And insist on this in a bombastic display of just why you are such a f_ckwit.
Chief Hydrologist | February 6, 2013 at 2:51 am |
The problem is not that. The problem is that the dynamic effect of latent heat (and the associated pressure gradients) cannot be calculated considering the rising air parcel alone. Consider you have an ambient lapse rate of 6.5 K/km. Imagine that a saturated air parcel ascends (for some reason) and becomes warmer than the environment, because the moist adiabatic lapse rate is smaller than 6.5 K/km. This rising parcel acquires some CAPE that can be potentially converted to the kinetic energy.
However, in order to rise one air parcel, you need to make another air parcel to descend. It is circulation. Since the descent occurs dry adiabatically, the descending air parcel will too become warmer than the environment (because dry adiabatic lapse rate is higher than 6.5 K/km). So you actually need some extra force to push this warm dry air downward.
This positive buoyancy of the descending air acts as a brake on circulation. It does not allow the CAPE of the rising air to be converted to the kinetic energy. Basically under most conditions all rising CAPE will just go to push the warm dry air downward.
Because of the complications involves by this cancellation effect, currently no theoretical accounts exist of pressure gradients (or dynamic power) produced by latent heat. The only approach to study this is the slice method (developed in the 1930s), which however only addresses the issue of stability/instability but not the resulting dynamic power of the circulation.
The summary is that the common argument about a small pressure fall because of gas removal and a large pressure rise due to latent heat release are not relevant to the discussion of the condensation power. It is not possible to decide about latent heat importance based on 1-D considerations.
The gas removal effect that we describe do not suffer from any cancellation process in the descending branch of the circulation. The potential energy resulting from condensation in this way can be fully converted to the kinetic energy.
Hi,
This was to do with one of your critics that was linked to earlier – to the effect that sensible heat on condensation caused an increase in pressure at the cloud base – presumably by the ideal gas law – and therefore pressure increased at the cloud base and your physics were obviously incorrect on first principles. It is something echoed elsewhere in this post. And I am trying hard not to overstep comprehension.
But there are a number of processes in Hadley Cell circulation – for instance that may or may not include the condensation effect. For which – btw – the lapse rate on the dry arm seems irrelevant.
Thanks you for your diligence and patience. I wish you well.
However, in order to rise one air parcel, you need to make another air parcel to descend. It is circulation.
Still unsure if this is just your way of putting it, or if you’re really saying that a volume of air only rises if another volume decends, in effect, the decending volume driving the rising of the first.
Gases rise when they are or become lighter than air, as they become less dense they will spontaneously rise. Water vapour, as Stephen Wilde pointed out above, is anyway lighter than air, but heated will expand more in volume becoming even less dense and rise faster, as will air itself, nitrogen and oxygen. A volume of air heated will become less dense expanding in volume and rise because lighter than the air around it which is colder. The air around being colder therefore denser and heavier, with more condensed volume, will sink; gravity having less of a grip on the hotter less dense rising expanding lighter volume with less mass than it does on the denser colder heavier with more mass. The colder heavier denser air sinks displacing the rising lighter hotter under gravity, this is convection, this is circulation.
Gas and liquids are fluids, this sets up convection currents in the heavy fluid voluminous gas which is our atmosphere as they exist in the fluid liquid water which is the ocean, in the air these volumes of air on the move, packets as some call them.., are called winds.
Gases which are lighter than air, such as water vapour and methane, will always rise in air unless work is done to change that, just as, gases which are heavier than air, like carbon dioxide which is one and half times heavier, will always sink in air and will not spontaneously rise in air, unless work is done to change that.
Hot air rises because it is relatively lighter than the colder denser volume under gravity, not because cold air pushes it up. It doesn’t need a heavier volume to sink first to make it rise.
These discussions are inherently confused because there is deliberate confusion created about the basic properties and process of gases in our atmosphere. The AGW Greenhouse Effect has excised gravity as it has convection because it claims that carbon dioxide can accumulate in the atmosphere for hundreds and even thousands of years and for this it needed to take out any physics to do with relative weight. So temperature driven convection currents and relative weight between gases under gravity had to go and it brought in the fake fisics that “carbon dioxide is a non-condensing gas”.
Here is a clear explanation of this by R.Gates: http://wattsupwiththat.com/2011/05/21/happer-on-the-truth-about-greenhouse-gases/
“This is all about carbon and all about a long term cycle the controls the amount of carbon dioxide in the atmosphere. From the earth’s perspective, carbon dioxide is hardly just a minor trace gas, but because it is non-condensing, and is not taken out of the atmosphere by simply lowering the temperature, it become the master thermostat of the planet.”
Carbon dioxide expands when heated just as does air, nitrogen and oxygen, and condenses, becomes more dense when cooled as do all gases, but AGWScienceFiction doesn’t allow this because it has created a fictional narrative about “greenhouse gases”, example here:
http://www.giss.nasa.gov/research/briefs/lacis_01/
Carbon dioxide is also taken out of the atmosphere by the water cycle, all natural rain is carbonic acid, as well as being heavier than air.
So, that’s the reason there is so much confusion, and why I’m not sure what you’re saying here.
There is a SERIOUS need in this here fest for some fundamental knowledge about convection.
Start here dipthongs (you DON’T know who you are, by the way):
http://en.wikipedia.org/wiki/Atmospheric_convection
On cats and unicorns.
willard
This issue is well worth addressing. What is “correct”? And how can it be decided?
Let’s follow the logic. There are continuity equations (32)-(33) and there is an equation on condensation rate (34). Nick and Dr. Held as per your quote believe that (34) is not a valid equation containing independent physics, but just an invalid version of something in (32)-(33). Note that (32)-(34) give -u∂N/∂x = S, i.e. pressure gradient proportional to condensation.
In order to prove that, Nick proposes the following experiment: let us make the dry air circulate with a motor (such that u∂N/∂x is not zero), but it will not rain (S = 0). So we are wrong.
But there is a logical fallacy in this reasoning. Imagine that there IS independent physics in (34) (which neither Nick nor Dr. Held can see, while we can), which is pertinent to condensation-induced circulation. Then Nick’s dry air motor example is not relevant: it only proves that dry air circulated by motor is NOT a condensation-induced circulation. So actually, a direct test of (34) is to go where condensation occurs and calculate the observable variables in the equation to see if they match (we did several such examples). Note that any condensation-induced circulation has a dry part too (where the air descends), so the condition S=0 can be tested as well. As I said above, in the dry part of the condensation-induced circulation winds must be either geostrophic or pressure gradient be negligible. This is a testable prediction.
So, in terms of things discussed so far, what we propose is an equation that describes, say, a unihorn: -u∂N/∂x = S. At this point nobody knows how we derived this equation (but I am certain people will understand), but it is empirically testable. So if you test it and it is not fulfilled, you can be certain: it is NOT a unihorn.
We then propose that in the nearest forest most animals are unicorns (most circulation patterns are condensation-driven). When we integrate our equation over the entire forest, we obtain the right figure of the total animal biomass (our estimate of global circulation power matches the observation).
Now comes Nick and says, hey, I’ve tested your equation on my home pet (motor that cycles the dry air) and it does not work! My pet is a cat, not a unicorn! Go away with your absurd theory.
But while it is fine to have a cat as a pet, this does not in any way disprove our statement that in the nearest forest most animals are unicorns. E.g. hurricane is a unicorn.
Anastassia,
“but it is empirically testable.”
How? It seems to make a clear enough statement, that if you can make dry air move down a density gradient, then water (why water rather than some other liquid?) will condense. But you say we can’t test just that. We have to consult the insightful people who can see the invisible physics used in its derivation.
Laws of nature are supposed to be just that. They are properties of the real world, and should be testable by anyone. They don’t depend on what the person who derived them was thinking about at the time.
No, you do not need to consult anybody. As I said above:
Of course, you are free not to go but be satisfied with the dry air cycled by motor. But if other people do go and find the predictions to be correct, then other people will be more inclined to trust our statement that there is some independent physics in Eq. (34) which Nick has somehow overlooked. Because it would seem strange to them to believe that just a random combination of variables produces a sound quantitative result. Moreover, people will be inclined to take a closer look at Eqs. (1-4) in the post where the same result is derived without any reference to the continuity equation. If, on the other hand, they do not find the predictions to be correct, that will mean that the condensation-induced circulation as we have described it does not exist. So it’s all very simple and open to anyone’s check.
Anastassia,
Present the derivation of (34) from independent physics.
The formula has no sign of independent physics. It’s form is that of an approximate continuity equation. It has not any hint of including anything else. That you claim otherwise means nothing until you present the derivation of that equation from independent physics.
Making an error in derivation is not independent physics.
Anastassia,
“the dry part of the condensation-induced circulation winds must be either geostrophic or pressure gradient be negligible. This is a testable prediction.”
No it’s not. You have built a theory which claims to prove that winds are induced by condensation. S=u ∂N/∂x is part of that. But now your saying that we can only test the theory predicting condensation origins by finding winds that we know are induced by condensation. Then we can apply the theory.
I see nothing strange having designed a theory to describe winds induced by circulation then suggest to people
First of all, not induced but accompanied. Remember people believe that winds are induced by temperature gradients. Second, these winds — accompanied by condensation — may have nothing to do with our theoretical prediction. Quantitative observations of these winds can either support or disprove our theory.
On the other hand, it would be strange, having formulated a theory for condensation-driven winds, to suggest that people should go and test it on a planet where the air is dry.
Anastassia,
Where is the basis for equation (34). Tell that with formulas, not waving arms.
Please, Pekka, this kind of comment adds nothing and makes you sound like MiniMax.
Williard,
The issue has in many comments been the validity of the equation (34). The most important claim in support of the paper by Anastassia has in my view been that the equation is based on independent physics. If she could, indeed, show that to be the case, that would be significant for me and I would reconsider much that I have said.
What she has written is that the equation does represent independent physics. If that is true then she must be able to tell, how that’s the case. Only formulas count in that, because the outcome is a formula.
All that I made was to require her to back up her claims. That’s the only way I can imagine that the paper could regain credibility.
Of course, I don’t expect that she could succeed in that, as I believe that the contrary evidence is very strong. But if she can’t then she should admit the error. What’s the point in resiting for ever the truth.
it’s right to reserve some time before admitting errors to avoid admitting non-existing errors, but there’s a point where that should end.
It’s not a matter of opinion that the equation (34) lacks proper justification. That’s true for every equation for which no justification can be presented. Vague words are not a justification, more is required for it to count at all.
Pekka,
I know what you think the issue is. You’ve told us many times now. Do you want me to count?
Please hang in there. Things take time.
Pekka –
I second willard’s request (not assuming he agrees with my reasoning).
I will tell you that AFAIC, your accusation of handwaving is at best pointless, and at least in my opinion, (FWIW – which ain’t much, particularly as someone who can’t understand the technical issues being debated) only undermines your own argument and taints your credibility.
I see Anastassia offering her arguments in good faith, fully engaged in presenting her perspective. I don’t see how, in any way, that can be described as “hand-waving.” I understand that one might think, from a technical perspective, she is not directly addressing the technical criticisms offered – but it seems evident that she is not merely avoiding those criticisms. In the very least, she seems to clearly be attemptingt to address them head on. Even if you think that there are fatal logical flaws to her arguments, it doesn’t strike me that her arguments are irrational (one can be wrong w/o being irrational), intended only to distract, trying to gloss over something, etc. IMO, handwaving implies a poor faith engagement with the issues at hand – something that does not seem to apply here.
From what I’ve seen in blogospheric debates, the accusation of “handwaving” invariably discounts the question of subjectivity. It is like the accusation of “troll,” or “appeal to authority,” in that the veracity of those labels is , obviously, subjective. When I see, say, a Willis accuse a Fred Moolton of “handwaving” because Fred is of a differing opinion than he, it does little other (IMO) than signal someone (i.e., Willis) who doesn’t recognize the difference between a difference of opinion and an argument offered in bad faith.
Now perhaps, if I understood the technical issues, I would agree that Anastassia is simply avoiding addressing substantive criticisms, but even if that were true, what is the point of accusing her of hand-waving?:
Here is how I break down the ramifications of your accusation:
(1) someone who understands the technical issues already has formulated their opinion – thus, your accusation of hand-waving serves no purpose. It does not inform that person of anything new.
(2) it has no positive impact on Anastassia – and if anything, only inclines her to respond in-kind, thus lowering any chance that a respectful exchange of opinions will ensue.
(3) it means nothing of substance to anyone who doesn’t understand the issues – unless they hold onto some blind faith that whatever you say is necessarily true. I would suggest that the # of people who might fit that description is small.
(4) for someone who (like me) generally respects your opinion and uses your technical expertise as a touchstone for evaluating technical questions they are incapable of understanding (either by way of insufficient background or intellectual limitations – or both, as in my case), you undermine your own credibility by accusing someone who at least seems to be engaging in good faith, of engaging in bad faith.
I think Pekka’s point is that the only way to justify an equation is with other equations that show where it came from. Anything just done in words is no good (handwaving) from a mathematical viewpoint. Despite many requests, equations were not still forthcoming.
Joshua,
I tend to agree that I have written too much on this paper. Thus I don’t refer specifically to that here. I note only that to me every real argument of a theoretical paper of the nature of this paper is, is presented by a formula. Every formula must be based explicitly on something, i.e, on a well known physical principle or on other formulas that must satisfy the same requirement.
If the derivation is dependent on a formula that’s not justified in that way then it’s not a derivation. There are cases where the requirements are a little relaxed with respect to accuracy, but even then every input must be justified.
The only comment that links the above to my earlier comments and this paper is that the formula (34) lacks totally such basis. It’s not justified at all, it comes from thin air. That’s not acceptable. By hand-waving I refer to loose statements that don’t resolve that defect at all satisfactorily, or actually not at all. I insist only that scientists present arguments as scientists should do that in all their articles.
Jim –
Your point notwithstanding, I think that the four points I finished with stand.
I get the gist of how you are summarizing Pekka’s point – but I am incapable of evaluating whether or not Anastassia showed where her equations came from (let alone that it is a mathematical requirement). I see someone like Tomas – (from what I can tell someone who possess much expertise in mathematics and physics) – who seems to in essence support Anastassia’s perspective. I see Anastassia – from my ability to judge as someone w/o the ability to understand the technical argument – who seems to be engaging in good faith, saying that she has addressed the criticism.
So again, I see no upside to the assessment of “handwaving.” Simply offering the opinion that she has failed to address the substantive criticisms is enough. The accusation of handwaving seems to run against reality (I cannot see how she is being irrational, attempting to distract, etc.).
Pekka –
In response to your 4:28 – I’d offer that instead of using “handwaving” as a shortcut, your second paragraph is the sort of statement that works.
it is clear and understandable to me. “Handwaving” has a larger connotation. Since you (and Anastassia) are discussing, at least in part, what belongs in scientific arguments “handwaving” doesn’t belong. It isn’t scientific syntax. It is rhetorical syntax, and rhetorical syntax that is over-used in the blogosphere in the stead of solid arguments.
Joshua,
Perhaps you can imagine what has made to write more. I have tried to convince others that what you write is in error, that the issue is uniquely resolved beyond reasonable doubt, and that I feel frustrated,when I fail in my attempt to get that more widely accepted.
I feel that all my points are supported by facts, not opinions. The facts are on the process without need to consider the substance. By facts on the process I mean that it’s possible to look at the paper and notice that the equation (34) is just given without justification. It’s possible to read Anastassia’s comments and notice that no concrete support for that formula is given there, etc. Very little expertize is needed to find out all this. Tomas has not either provided any evidence that the formula (34) would find support from anywhere. He made a philosophical remark that is not strong enough to counter the above.
I consider the above to be objective facts, verifiable by almost anyone.
An important further point that requires a little more substance understanding is that the paper fails totally without the equation (34).
I wasn’t the first to use that expression. I took it from an earlier message that referred to the same lengthy comment of AM in precisely the same meaning. AM referred in a later comment to that “h-w” comment stating that the justification is there, but there was only “h-w” in that comment. I requested as response to that for something better. She probably understood this whole chain.
We have also the other result that Anastassia has herself presented and others have also derived from the formulas of the paper:
The paper has proven that an uplift cannot have condensation at a point without horizontal density gradient. That’s as obviously wrong as can be.
Paper’s results have been tested empirically and found wrong. This result is very strongly linked with everything that they derive, it’s not a minor detail.
Anastassia,
Just caught your comment, don’t have much time. You ask: how can we decide correctness? In physics, I don’t know. Vaughan would be better placed to tell you that.
Associating the proper types of terms usually help. See for instance:
http://judithcurry.com/2012/12/04/multidecadal-climate-to-within-a-millikelvin/#comment-282517
This example shows that typing the terms and the functions might be useful in some contexts.
That’s what I did at Eli’s btw, when I distinguished that Held’s statement was factual, not deontological.
There are more techniques like that, but I have to go.
PS: Your last comment at Eli’s was quite good, btw. By sticking to evidential-based reasoning, the lines of attack from your adversaries have been reduced.
I think that all of this discussion comes down to the difference in two values and what their relative magnitudes are.
On one hand we have the balance in density that comes from temperature differences. This ‘warm air rises’/‘cold air falls’ drives circulation in the mass of air in fairly well understood ways.
The second comes from evaporation and condensation of water. As water vapour occupies about 1000 times the volume of the water/ice it comes from that has the possibility to create very large forces (witness old time condensing steam engines).
The question is then about how these two effects are combined and their relative importance.
If the air contains a large percentage of water vapour then that vapour may well end up being more responsible for any air motion that dry air of a similar volume given the expansion ratios.
If the air is low in water vapour then the overall motion will likely be dictated by the classical temperature/density balance.
At what point the two factors are equal is the real question.
If considered as a classical engine then the work done (in dry air velocity) will be proportional just to the difference in the hot and cold.
The addition of water vapour to the air will allow more work to be done (higher air/vapour velocity) for similar temperatures.
So this all comes down to the ratio between dry air temperature driven density and water evaporation/condensation injection/removal.
Anastassia et al. claim that the second factor can become significant in overall circulation. I see no reason to disbelieve this.
N.Stokes wrote
The dispute isn’t much over the correctness of 34, but whether it is independent from 32 and 33.
We have been over it in detail already 2 years ago. The answer is a clear YES even if Pekka didn’t still understand that.
What should be said clearly is that all possible results derived from 32 + 33 + 34 AND the question of independence of 34 are 2 totally different worlds.
One can derive for eternity any kind of consequences of 32+33+34 without changing a iota on the mathematical consistence of the system of equations. That’s what most people did on this thread but because the system isn’t closed I find it without merit and not very interesting.
On the other hand the latter question is well defined and has a crisp unique answer and that answer is “Yes 34 is independent from 32 and 33”.
Even if I have shown 2 years ago that 32 + 33 lead to a similar form as 34, it is not 34. Hence independence is demonstrated and there can be no more any discussion about that.
But I would like to repeat and stress the N.Stokes (equations) argument.
Btw you have an annnoying name because I have always to specify whether I talk about you or about Navier Stokes equations :)
It is because this approach justifies Anastassia’s approach that I was always pretty sure that the mathematical side of her paper had to be fundamentally sound.
Indeed a mixture of water vapour and dry air is fully described by N-S equations.
Moreover we know that there is a unique solution in the 2D case (3D is a mess but we don’t consider that here).
Now you consider that there is condensation and have to add an unknown new function S(x,z,t) to the continuity equation(s).
The system is no more closed, you have an infinity of solutions and have to add an equation S=whatever you feel appropriate.
The system is closed again and you restore the unicity of the solution.
The only necessary and sufficient condition is that your new equation has to be independent from N-S.
Once you made sure of that, you can apply the usual heavy artillery (Fourrier, DNS etc) and derive all kinds of consequences from your system.
Because unicity is warranted, the results can easily be verified or falsified by observation.
From the mathematical (N-S) point of view, this is exactly what Anastassia is doing.
However because she didn’t close the system I am agnostic as far as the mathematically derived results are concerned.
Tomas,
You are wrong.
If it would be based on independent physics, then some specific physical input must be introduced to that of continuity equations. That extra input must be expressed in a way that’s justified explicitly from the additional physics. Nothing like that has been presented. The arm waving that has been given is not a way to introduce physics to a set of equations. An unjustified equation out of thin air is not additional physics input.
I have asked Anastassia several times to justify explicitly the equation. She hasn’t done anything like that. She has answered to Nick, but failing to present anything of significance.
As I have emphasized, it’s also certain that the set of equations is in total contradiction with empirically confirmed knowledge of physics. Thus we have a totally unjustified equation out of nowhere and we get totally wrong results. Isn’t that bad enough.
I forgot.
That wasn’t all. In addition a dependence of one variable on x was derived from the assumption that certain other variables do not depend on x for a system where we know actually that either all are independent of x or all depend on x, the alternative assumed in the paper is not just unjustified, it’s wrong.
They derive again strong results from an assumption taken from thin air and not only some assumption but an assumption that’s explicitly contradictory.
So what would v1 and v2 be in the below experiment?
Richard,
It’s impossible to say much from so little information.
What can be said easily is that the dry case leads to a circulation with hot air rising and cold subsiding as long as cold means colder than the temperature that the hot air has when it has reached the top and cooled according to the dry adiabat. To close the circulation there would be horizontal wind from right to left at the top and from left to right at the bottom.
The wet case raises the question: can it exist? If the top is at high altitude, it’s cold also on the wet side and cold air is in absolute terms always dry, even when the relative humidity is 100%. When such air subsides, it’s dry by all measures. Thus the case cannot be maintained but changes rapidly to one with rising moist and subsiding dry column. That leads further to the situation that the lapse rate is larger in the subsiding column than in the rising column. How the temperatures, densities and pressures differ at each altitude in that case requires more information on the external conditions.
Tomas,
I have shown here, spelt out explicitly for this case, why you can’t use dependent equations. You can derive from 32/33 an equation in 2 or 3 dimensions with the form of 34. At that stage, any mathematician would be aware that no new equation had been created.
Then a 1D version is made, by ignoring the horizontally varying terms. This goes not assert those terms are zero, just that they are small relative to the vertically varying terms. That is 34.
Having created an approximate form, you should then remove the original from the set. Everyone knows that. If you simplify the Nav-S eqns by removing viscosity, you solve the Euler equations. You do not include the viscous momentum equation as well.
If you do not remove it, then you can subtract the original and exact equations to say that what you omitted in the approximation is exactly zero. This was never intended.
In our case, ironically AM says that our deduced form of 32/33 can’t be true, because it leads to a horizontal derivative being zero. I point out that the math is perfectly valid. The alleged contradiction is only because of retaining both forms. And the “contradiction” is exactly the difference between them, being set exactly to zero.
Pekka
You are probably wrong but it is hard to see in your posts which are exceptionally confused and inconsistent.
But what is sure is that you didn’t understand one word of my post and are totally off topic.
Tomas,
Simply. In physics it’s not enough that equations are independent. They must also be physically meaningful, not just some random combinations of symbols.
If the equation (34) is not only a continuity equation, it must be based on something well specified. There must be physical content in it that justifies the given form for the equation.
Anastassia has listed other physical phenomena, but has not at all explained, how they would give that equation.
This is exactly the case that Mosher has been ironizing with his unicorn example.
And, on top of that the outcome is in strong contradiction with empirical knowledge.
Pekka, is the unihorn still 0=1? Like when Pws-Pwv=0 S=1? But Pwv can be greater than Pws since RH is imperfect, when Pws-Pwv<0, there is more water molecules (unicorns) in the volume than ought to be. How much energy is in each of those unicorns that aren't supposed to be there?
Capt.Dallas,
I’m not trying to solve everything. In this thread my point is only that the paper being discussed does not have any results that would add to our understanding, because everything is affected so strongly by the error (except the results of the chapter 3 that are irrelevant and misleading and agreed as irrelevant also by Anastassia Makarieva in one of her comments.)
Pekka, that would depend on how the dynamics of moist air are handled currently. The CAPE model from what I understand requires tweaking “virtual” temperature. The C-C equations don’t seem to handle super saturation all that well. So what it may add depends on how accurate what exists actually is.
As I mentioned before, with higher temperature and saturated air, condensation is initiated at a lower altitude (higher temperature). If the amount of super saturation increases, there is more “explosive” condensation, and increase in Deep Convection, more stratospheric penetration. Not a neat tidy increase in lapse rate, but more lid rattling.
As far as the confusion in the derivation, the -N whatever term is just the degree of supersaturation. If you compare that to RH, a non-physical limit, then you would be missing the point.
Something that’s totally wrong is not inaccurate, it’s wrong.
Pekka, “Something that’s totally wrong is not inaccurate, it’s wrong.”
I agree, let’s rewrite all the psychrometrics so that 100% relative humidity is relative to local conditions. Let’s fire all the scientists that assume a constant RH. And let’s pull this paper since they neglected to include the y axis in their derivation :) I believe an isothermal layer in gravity would be a plane in the x-y, but that is beside the point.
Since the “nonphysical” part of the derivation is that small puzzle where the number of water molecules can exceed the “limits” imposed by RH, the paper would be meaningless anyway :)
Capt.Dallas,
I wrote some time ago a response to Vaughan where I commented on the difference of between two types of being “wrong”.
The first type is the one that means that nothing related to the error has any relevance any more as the the error may have changed by any amount.
The second is the opposite of “right” in the following simplifying sentence: “No model is right, but some are useful.” When we study systems like the atmosphere, no calculation is exactly right, but many calculations are useful. How useful varies from case to case.
In the case of the first type of wrong, we just don’t look at the results at all, because we don’t know at all, how they are related to the reality. This paper is wrong in that way.
Pekka, “In the case of the first type of wrong, we just don’t look at the results at all, because we don’t know at all, how they are related to the reality. This paper is wrong in that way.”
I know, I just ain’t that smart.;
Pekka, as a BTW, the paper “implies” (in quotes because smart people don’t think I am allowed to do that) that dirty air reduces over saturation which would “effectively” reduce the lapse rate. Odd that :)
Capt.Dallas,
Even without understanding what the equations tell, everyone interested may find out that the authors have nowhere justified the form of equation (34). At best Anastassia Makarieva has listed physical phenomena not included in equations (32) and (33), but nowhere has anyone explained why the (34) would represent those phenomena.
There are only two alternatives available.
– One is the one that Nick and I claim: the equation is also a continuity equation as (32) and (33), but an approximate one and using the approximate equation at the same time the exact is used is a serious error.
– The other one is that no explanation is given for it’s form – its an “unicorn theory”.
The above can be verified without any understanding of the details.
Pekka, I agree that the paper doesn’t adequately describe the phenomena they “thought” they were trying to describe. They appear to have accidentally noticed something and attempted to describe what was going on with tools not up to the task. This would be where Tomas enters the picture.
The fact though, is that “averages” in a nonlinear systems are close to useless and RH is nothing more than an average. So if you are going to trash the paper, don’t throw the accidental insight away.
This debate is a fine example of a scientific debate at the theoretical level. Some of the claims seem overly strong but that is how it goes.
In particular Pekka’s uncivil claims seem overly strong. The authors are clearly not fools, which he seems to make them out to be.
David,
I’m so uncivil, because they don’t answer any explicit question. The reason for not answering should be obvious, but still they don’t answer and still we have others who are ready to accept the situation that they cannot contest my explicit claims.
David,
It would be more valuable, if the disagreements were an problems in interpreting and understanding alternative views on science in areas where genuine and legitimate uncertainties persists. I would say that there are many such issues in climate models, but these issues may be too difficult to all of us, or at least almost all.
it’s not equally valuable when it’s on an explicit error in the paper where all the rest falls when the error is acknowledged.
We are not really discussing at all what goes on in the atmosphere, we are discussing one detail of one paper.
Pekka, the authors have responded many times to you so it sounds like you are talking past one another, which as Kuhn pointed out is a common feature of theory discourse in science. I do not have time to study these exchanges but it is possible that your questions embody assumptions about their work which the authors do not share. In that case they cannot answer your questions they can only object to them. I think I have seen several responses of that nature.
The proof is in the pudding, and several trial batches have been in and out of the oven, baked to a turn.
===========
David,
They did answer at the beginning, when my comments were of more general nature, but they haven’t answered to any of the specific comments that I have written when I didn’t only believe that the paper is wrong, but knew the way it’s wrong. As soon as I started to discuss in technical terms the faults they have not answered any of these comments.
These technical comments are much stronger as they prove that the paper is all wrong rather than express the personal judgment that it must be wrong. I have alternated with Nick in making the statements as specific as possible and describing as fully as possible, why those errors make results of the remaining paper worthless. These comments have been both on the theory and on the fact that the results contradict very strongly also empirically confirmed knowledge.
The errors are so obvious that I can’t really understand that they don’t admit them.
Saying that totally wrong is totally wrong is perhaps not polite, but what should I say?
Sorry – previous version too wide.
So what are v1 and v2 in the below?
My above answer is based on the same figure. My browser (Firefox) had no problems with the wider one.
It is perfectly possible to create just this experiment. The wet and dry can be created using a ‘swamp cooler’ type arrangement of pads. Ideally the water would be be at the same temperaure as the space ‘behind’ them.
My ascii art left out the middle layer – I’ll try again.
Yes, it’s possible to do that, but
– what are the dimensions?
– what are the temperatures at hot and cold?
– how effective is the heat transfer at hot and cold point?
– how wet is wet and how it’s level is maintained?
– should we consider dynamic development or only the stationary case?
The answer depends on all these or depending on some of the answers perhaps not so much on all the rest.
Last time I promise!
Sir Richard of the Latent Heat, I think there is already some ASHREA performance tests that might show the effect. Comparing a swamp cooler with HEPA filtered air to air with various dust contents. The filtered air should produce a higher efficiency by allowing greater saturation of the air.
Apples and Oranges. Nick, Pekka and Steven all have their panties in a wad because they like apples and the other gang likes oranges.
Here is the core of the paper, “but the Clausius-Clapeyron equation
(Eq. 3) shows that the vapor pressure remains unchanged being purely a function of temperature.)”
Saturation vapor pressure is not a function purely dependent on temperature. Aerosols change saturation vapor pressure. Clausius-Clapeyron is the unicorn. It is an estimation that does not provide sufficient accuracy when compared to at tiny change in atmospheric composition due to a doubling of CO2. That makes CO2 another unicorn.
0.8 +/- 0.2 is the baseline, not 3.0
Pekka
Simply. In physics it’s not enough that equations are independent. They must also be physically meaningful, not just some random combinations of symbols.
That is why I think that you are confused about the issues. As for Mosher, I don’t remember having ever read something that would make even half physical sense and as far as fluid dynamics are concerned, he clearly doesn’t box in the same category as Anastassi does.
Now I generally don’t do just empty statements about unicorns or random symbols but use the relevant maths whenever I can.
So here for you some random combinations of symbols (wome would call it equations) which will allow me to make some claims.
– div(Ri.V) = Fi
– curl (V) ^V) + grad (V²/2) = – grad (Ep) – grad (P/R) + µ/R.Laplacian (V)
Where :
i is an index 1 or 2
– R is density
– V is the velocity field of components (u,w) in 2D
– Fi are some functions of (x,z) that I don’t specify for the moment
– Ep is the potential energy per mass unit (here gravity)
– P is the pressure
– µ is viscosity
Now the claims I make are :
1) The equations are independent and the system of random symbols has a unique solution
2) This solution uniquely determines the velocity, pressure and density fields in the steady 2 D case
3) If one of the 2 fluids condenses then I obtain the dependences of the fields on the condensation dynamics (rate, spatial distribution etc) what is actually what this whole thread is about.
As this is not more and not less than what I already wrote in the post above, I expect that you consider all this as a nonsensical random combination of symbols and that you of course dismiss all 3 claims as wrong.
Right ?
Tomas.
“Now I generally don’t do just empty statements about unicorns or random symbols but use the relevant maths whenever I can.”
The point about unicorns was not directed at the math of the paper, but rather of the logic of Douglas. Douglas’s logic was that science operates by people publishing theories and then other people looking for evidence.
To illustrate the silliness of his unscientific characterization of the scientific process I constructed an example. Do you know what kind of argument this is called. You should. Having followed this for a long time I’m well aware of the issues. Well aware that Nick is more than capable of handling himself. Well aware that the author may not appreciate 16 people throwing the same math argument at her. So, rather than piling on, I’ll focus on some of the silli things that the co authors are saying about the scientific process. I’ll expect you to stand back and watch or ignore it. Since you didnt get how the unicorn case was addressed at Douglas’ silliness, I think standing aside is your best option.
Tomas,
What’s the connection to the set of equations (32), (33) and (34)?
How can equation (34) be derived correctly as an equation that’s not dependent on the continuity equations?
Why does it lead to nonsensical results?
Tomas,
div(Ri.V) = Fi
curl (V) ^V) + grad (V²/2) = – grad (Ep) – grad (P/R) + µ/R.Laplacian (V)
Those equations are in the system. The approximation is that you can ignore the acceleration and viscous stress, so that
0 = – grad (Ep) – grad (P/R)
That is the hydrostatic equation; the equation is simplified, but it’s there.
Our objection is that an extra continuity equation, 34 has been added. It incorporates the simplification of no horizontal change, but no new physics.
I gave above my x=0.9999, x=1 example. I think this trivial example does display the issue. But for those who prefer something more continuum:
A black membrane in space, subject to variable sunlight. T?
Write, balancing heat
ρC dT/dt = Sun flux -σT^4 + k*Laplace(T) (1)
Now we say, conduction is really small, we can leave the last term:
ρC dT/dt = Sun flux -σT^4 (2)
But if these are still both equations of our system, we can subtract:
k*Laplace(T) = 0 (3)
Now that is new, substantive physics, and almost certainly totally inconsistent with (2), which does genuinely express the solution we want.
Nick,
The equation (34) is supposed to tell the sink term (or source term). That should come from Clausius-Clapeyron equation and adiabatic expansion of ascending air under gravity or something equivalent to those, but such an equation should include additional parameters and not have a form so close to the continuity equation.
Perhaps they have started to add such an equation, but manipulated the formulas until they were left again with the continuity equation, but in an approximate form due to simplifications made during the manipulation. This is, of course, pure speculation, but a plausible explanation anyway. Being left with an equation that has lost all those physics parameters seems impossible to explain in any other way.
kim
Your “proof of the pudding” remark hits the mark.
The ONLY way to REALLY find out whether or not the “condensation driven winds” hypothesis proposed by Makarieva et al. is vaild is TO TEST IT with empirical data.
We have already agreed that it is not a “crackpot theory”.
What I am reading here are all sorts of a) rationalizations that “it can’t be correct because I do not agree with…”, b) arguments that it “violates basic physics” or c) other theoretical objections to equations used or suggested, but these are all utterly meaningless, unless the authors of the paper concede that these arguments have invalidated their conclusions (which has not occurred as yet.) So it is simply the opinion of one scientist versus that of another.
Hey folks, this is not the first time that one scientist is convinced of a hypothesis while another equally qualified scientist is convinced of exactly the opposite, is it?
So I really do not see why those like Pekka Pirilä or Vaughan Pratt are so opposed to simply TESTING the hypothesis, to see if it is falsified or corroborated by empirical evidence.
That’s the way “science” is supposed to work, isn’t it?
Max
Joshua could weigh in about ‘motivated reasoning’. moshe could chat with willard.
============
Cap’n done cut the bait and has gone fishin’.
==========
Kim, some light reading.
http://earthobservatory.nasa.gov/Features/UrbanRain/urbanrain3.php
““The extra boost makes the storm act like a vacuum cleaner,” says Bell. “Something needs to replace the rising air, so more moist air is sucked up.” This vacuum-cleaner effect allows the storm to pull in more material to work with than it would have without the urban aerosols, and more rain results. “The effect only happens when fast-rising air would form a thunderstorm anyway, and when the air near the surface is moist,” Bell adds.”
Heh, phase change. Tricky stuff that H20. CO2 not so tricky.
=======
@manacker: The ONLY way to REALLY find out whether or not the “condensation driven winds” hypothesis proposed by Makarieva et al. is vaild is TO TEST IT with empirical data.
Sorry, what’s the hypothesis exactly, Max? Is it that condensation drives winds by pulling them or pushing them? Marakieva claims pulling, but everyone who’s attempted to actually calculate the effect claims that the effect if it were at all significant (very unlikely since it’s so tiny) would have to be pushing.
Without knowing which, how would you even begin to test such a hypothesis, especially when the effect is miniscule compare to the other forces that drive wind?
Marakieva –> Makarieva. I keep making that mistake and usually manage to correct it in time.
A test, which seems too obvious, is to take a chamber at 200% saturation for example, like the nuclear physics cloud chambers. Let that condense, and see if the pressure rises or falls. My bet is it will rise because of the latent heat release and thermodynamic principles outlined here.
The only way to really find out if unicorns cause wind is to test it.
The question is if you cant tell WHAT the system is suggesting you cant test it.
Or if you show a math error you know it can never be true.
Steven Mosher “if you cant tell WHAT the system is suggesting”
You can ask …
There are in fact three interesting states. The other is where the Cold pads are Wet and the Hot pads are Dry (raises the humidity but does not add Evaporation).
v1 is the air velocity in the top box – A
v2 is the air velocity in the middle box – B
v3 is the air velocity in the lower box – C
Hot and Cold are held at constant temperature by external sources.
Water (when present ) is at the same temperature as the heat source/sink.
The values of v1, v2 and v3 will be bounded by the relative energy transfers occuring between Hot and Cold.
v1 will be driven by the denisty differences of dry air at the Hot and Cold ends.
v2 will be driven by the denisty differences of mixed air and water vapour at the Hot and Cold ends, with the overall humidity being controlled by the temperature of the Cold end.
v3 will be driven in addition to the above by the Evporation and Condensation taking place in the pads from Hot to Cold.
If the traditional view is correct adding water to the Hot pads will only change the flow rates slightly. If Anastassi et al. are right then the flow rate changes will be large.
The empirical content of this theory is that winds are primarily driven by condensation. Obviously the afternoon sea breeze is not driven by condensation. Equally obviously, storm winds look very much as if they are driven by condensation.
Conventional theory is that big winds cause big rain. This theory is that big rain causes big winds.
Because of the messiness and complexity of the atmosphere, we can never resolve this by modeling.
Surely we can check this by looking for delay between big winds and big rain: If mostly positive, this theory true, if mostly negative, other theory true.
We know winds related to thunderstorms owe their origin to processes related to condensation, but not by this vapor-loss suction method. Latent heating is needed, otherwise thunderstorms don’t happen. This paper omits mention of its importance, and goes as far as denying it by saying the pressure reduction dominates any increase from heating which is provably wrong.
I think there might be quite a few mentions of evaporation – and therefore latent heat – without which there are no clouds and no circulation.
Condensation far exceeds evaporation in a thunderstorm. The net difference is in the rainfall and clouds.
Jim – really? In the air there is both condensation and evaporation – depending on local conditions.
Ultimately precipitation equals evaporation from the surface over a very short period.
We all know that evaporation leads to reduced surface pressure and that condensation is the opposite of evaporation.
It is perverse to try to argue that condensation also causes a surface pressure reduction.
Doesn’t, cannot and never will.
The local pressure reduction pointed to when condensation reduces volume is instantly offset by mass flowing into the original volume from the surroundings and the energy released is not enough to make the air parcel and the liquid contents lighter than air containing water vapour.
The parcel of air containing liquid must descend which raises surface pressure beneath it.
We see that in every convective cloud where the precipitation is mostly within local downdrafts.
An extreme example is microbursts which can be a threat to air traffic.
All else is chaff.
The paper is fatally flawed which is to my regret because we do need to focus on such matters in order to see the flaws in AGW theory.
It is sad that this paper is wrongly directed.
Here is the weather forecast for the next four days for Australia. – http://www.bom.gov.au/australia/charts/4day_col.shtml – It shows a coulpe of things. To the south a low pressure system spinning of the circum polar system brings a bit of rain to southern Australia before being again blocked by a high in the bight. The negative values of the SAM more recently bodes well for decent autumn and winter rainfall in southern Australia.
To the north there are the lows of the monsoon season. Rainfall over quite a good portion of northern Australia and far south on the east coast. The lows are traditionally seen as the result of rising warm air which pulls in air behind it on the surface creating a self sustaining system. Anastassia is entirely correct in that the rising air must be balanced by descending air creating high pressure in the descent arm. The condensation effect is an amplifying effect at cloud height that must exist.
Dear All,
As we have just passed the 1,000 comments mark, I believe it is time for another summary (the previous one was here; it remains relevant). Please those willing to read something that has NOT been yet discussed in this thread, skip directly to point B below.
Point A. Let me remind you what all the discussion is about. The main result of our paper and our theory obtained for a horizontally isothermal atmosphere (∂T/∂x = 0) is this
–u.∇p = SRT
It says that condensation rate S determines the horizontal pressure gradient provided a horizontal velocity component parallel to the pressure gradient is known.
Mathematically, this result, as shown in the paper, follows unambiguously from the following equations: the continuity equations (32)-(33), the stipulation that water vapor molar density is horizontally uniform (∂Nv/∂x = 0), the ideal gas law p = NRT, and Equation (34) that specifies condensation rate S.
Let me emphasize that specifying S is an important physical problem. Without an independent knowledge of S, the continuity (mass conservation) equation is uninformative (e.g., referred to as a “tautologism” by Truesdell and Toupin (1960) The classical field theories, Handbuch der Physik Band III/1, 226-793).
We specified S by equation (34). As I said above, this equation is not derived from any other pre-existing equations. It was formulated by us on the basis of several physical considerations concerning the peculiarities of condensation rate in hydrostatic equilibrium. These considerations are summarized here.
In physics, unlike in mathematics, new equations and new statements cannot be always proved to follow from a formal set of axioms that everyone can find in a textbook. In order to say something new, one needs an insight — a more or less plausible proposition. Once you get one, it must be mathematically formulated in a consistent way. Next, it must be seen if it is not in conflict with the fundamental laws of physics. If these initial tests have been passed, you can start formulating your theory based on this proposition and see how the results yielded by the theory agree with observations and if they are not in conflict with the laws of physics.
Our equation (34) is such a proposition. I see the attacks on this equation that have happened so far as being largely pointless. This equation per se can be falsified in only two ways: (a) by showing that it contradicts some other well-established physical equation and (b) by producing an alternative independent expression for condensation rate that fits the reality better. The discussion has never come close to these topics. Indeed, given how fundamental the problem of finding S is, if someone were able to produce a competing alternative to (34), they would be now writing a paper, not a blog comment. Note that in the meteorological science until now no theoretical treatment of S has ever existed.
The point that Eq. (34) is based on arm-waving is, in my view, counter-productive. As I said before, it leads to nowhere, because there is no formal definition of “arm-waving”. This perception depends much on the knowledge and physical intuition of the individual discussion participants. If person A gives her arguments (presumably thinking that it is not arm-waving but plausible arguments), and person B perceives these arguments as arm-waving, the discussion discontinues.
Point B. This is what this blog post was actually about. And what has been totally neglected so far in this discussion. The point is that our main result, see the equation above, can be obtained from independent considerations that do not involve the continuity equation or Eq. (34) at all.
Those willing to understand what it is all about are welcome to read the post. There are only four equations, and the main result is Equation (4). There is also some additional coverage here. For those unwilling to read the post, I’ll now make a two-sentence summary. Our main result follows from two propositions. The first one is that the dynamic power of circulation (variable q in the post) is equal to potential energy associated with the non-equilibrium vertical gradient of water vapor. The second one is that in hydrostatic equilibrium, where the vertical gradient of total air is balanced by gravity, the dynamic power is only produced by horizontal pressure gradients (i.e. q = –u.∇p). This independent set of physical considerations that yield the same result provides additional theoretical support to the validity and soundness of those physical propositions on which the theory of condensation-induced dynamics had been built. The ultimate check of the validity of our theory will come from empirical evidence.
Thank you again for this exciting discussion.
Anastassia
A correction to one of the statements above:
“The first one is that the dynamic power of circulation (variable q in the post) is equal to potential energy released per unit time that is associated with the non-equilibrium vertical gradient of water vapor. “
The issue of the role of new hypotheses in physics papers has come explicitly up in the comments of Tomas and in the above comment of Anastassia. It’s certainly true that new ideas are commonly introduced trough a new equation, not derivable from earlier knowledge. An the other hand a very large majority of papers is built on earlier knowledge. All input formulas are taken from the current set of physical formulas. Some assumptions are made to specify the problem to be studied and derivation done on that basis.
That leads to the questions:
1) When is it appropriate to introduce new equations?
2) To what extent should even the new equations be justified?
3) How explicitly should it be stated that what follows is based on a new hypothesis and not results derived from existing knowledge
4) When a new hypothesis is presented, how much effort should be spent to assure that it does not contradict well known physics.
In the case of the equation (34) it is stated in the paper to be the source term. In the above comment Anastassia tells that the equation gives the pressure gradient from the source term. The logic is reversed, but the equation is an equation in both directions, so is this fine? No, it’s not fine the consequences of the source term are given by the continuity equations. Her latest statement is in line with that, but that interpretation has been vehemently denied, because then it would contradict (32) and (33).
Could it still be a equation for the source term in spite of the fact that it looks an approximate continuity equation and in spite of the fact that AM tells that it serves as a continuity equation?
No it’s not a source term equation. The paper discusses the nature of the source term in chapter 2 and presents many formulas there that could be used to derive a source term equation. Such an equation would, however, include parameters related to latent heat or Clausius-Clapeyron equation. The nature of the source term equations is known, it’s known to the authors of this paper. There’s no way that equation (34) would come from there.
Could it be a genuinely additional physics equation? The continuity equations and the thermodynamic equations related to isentropic processes of rising air and condensation determine the whole process. Adding one more equation of that type would make the system overdetermined. That’s not allowable. The extra equation must be dependent on the existing ones. From it’s form, it follows that it must be dependent on the continuity equations – and it is up to the approximations.
There seems to be no way around the conclusion: It is a approximate continuity equation and using that together with the exact ones may produce anything with no connection to reality.
That’s what happens in the paper. Results that contradict totally the reality are derived.
This is not a case of making correctly a new hypothesis and proceeding to test it.
That the equation (34) would not be essential for the paper is wrong.
The chapter 2 has nothing new except a few misleading sentences. The chapter 3 discusses an irrelevant case admitted by Anastassia as such. The rest is either dependent on the equation (34) or pure speculation without quantitative basis.
@Pekka”The issue of the role of new hypotheses in physics papers has come explicitly up in the comments of Tomas and in the above comment of Anastassia. It’s certainly true that new ideas are commonly introduced trough a new equation, not derivable from earlier knowledge. An the other hand a very large majority of papers is built on earlier knowledge.”
An issue with scientific method IMO is the basic rational, linear, one step at a time, logic that underpins it. New hypotheses and new ideas rarely spring from this type of thinking.
Peter,
We seem to agree. The exceptions are, however, important. Planck modified the formula for intensity of emission that way, Einstein introduced his equations, more or less all major steps proceed through a hypothesis that’s not derivable from what’s known before. Very often that’s done to explain a serious contradiction between earlier theories and observations. Another common situation is the case where a new hypothesis provides an unified basis for earlier disjoint theories and adds in that also new predictions.
The above applies best to fundamental theories, but very similar processes occur also in theories of complex systems supposedly controlled by fundamental theories but too complex for direct solution. The atmosphere is a good example. Simplifying hypotheses help in solving problems related to the behavior of the atmosphere. There’s still space for new simplifying assumptions and approximate “laws” in the physics of the atmosphere. Much of the physics of the atmosphere is, however, well known and also tested empirically. That knowledge must be taken into account when developing new approaches. Some additions to the set of equations make sense, some others are obviously wrong, because they lead to internal contradictions within the theory and/or severe contradictions with empirical data.
It has been demonstrated in many different ways that the present paper fails on both counts. It’s addition is of the type that it involves internal contradictions within the mathematical framework (like the assumptions on the independence on horizontal coordinate that leads to such dependence on that coordinate that contradicts the assumption). It has also been emphasized by Nick and me that it leads to predictions that contradict very severely the empirically known physics.
There are various ways to check whether some particular new assumption is plausible. It’s form and set of parameters may tell, whether it’s just an new formulation of the same laws that have already been taken into account. If that’s the case then either the old way of including the laws must be taken off or the new law must be left out. This test applies to the present case and tells about a severe error.
Physics is basically an exact science. Even when approximations and new hypotheses are introduced there are laws of procedure that must not be violated. It’s not as free as one could interpret what Tomas wrote, and how AM did interpret his writing. That’s what I meant when I wrote that he is wrong. Nick did explain nicely, how that error occurred in Tomas’ specific example.
Thanks for your thoughtful response Pekka. I understand where you and some of the others are coming from but feel that intense resistance to change of the orthodox science seems to be a feature of many of the paradigm shifts that have occurred in recent, and, of course, not so recent history. These changes seem to have only occurred at considerable cost to the progenitators of such changes and that is to me a matter of regret.
Anastassia said:
“The first one is that the dynamic power of circulation (variable q in the post) is equal to potential energy released per unit time that is associated with the non-equilibrium vertical gradient of water vapor.”
That seems clear enough but isn’t it well known ?
The amount of potential energy released per unit of time is dependent both on the rate of the initial evaporation AND the rate of condensation and the net outturn will influence the winds via the pressure gradient and thus the dynamic power of the circulation.
It is true that different factors can affect the rates of evaporation and condensation differently leading to changes in the pressure field and winds but the evaporation has to occur first.
Condensation being a mirror image of evaporation and evaporation causing the initial surface pressure reduction then condensation must at the very least cancel out the pressure reduction caused by that initial evaporation.
Looking at a single convection cell after precipitation has begun one can see by looking at the base which section is being uplifted and which section is condensing and descending. The latter has precipitation falling from it and represents net dissipation of that part of the cloud with higher pressure at the ground than beneath the ascending section.
The recent series of papers by Makarieva, Gorshkov, et al has in common that they invoke the principle that adiabatic condensation reduces pressure due to loss of water vapor. This has led them to the notion of a “biotic pump,” a new theory of hurricanes, the present paper, and a brand new paper adding a sixth coauthor, Peter Bunyard, at arXiv:1301.3083 [physics.ao-ph] titled “Why does air passage over forest yield more rain?”
An objection has been raised repeatedly to this principle, in 2008 by Antoon Meesters, in 2009 by Dan Rosenfeld, and in this thread.
The objection is that the loss of water vapor is more than offset by the gain in temperature. In every objection that has quantified this difference, the temperature gain as a percentage has been about 5 times the gas loss as a percentage.
As all these quantitative estimates are a bit tedious to work through, it is natural to ask whether they could be simplified.
This can be done as follows using only the ideal gas law, the heat of vaporization of water, and the specific heat of air.
Consider a large parcel of n moles of wet air at absolute temperature T degrees K. By the ideal gas law we have
PV = nRT
Knowing n, R (a constant), and T, we therefore know PV.
Now suppose 1 mole of water vapor condenses.
Clearly n decreases, namely by 1/n of its original value, e.g. 0.1% when n = 1000.
The only tricky question now is by how much T increases due to latent heat becoming sensible heat.
The molar heat of vaporization of a mole of water is 40680 joules, which is therefore the amount of sensible heat added here.
The specific heat of air is between 21 and 30 joules per mole depending on whether the pressure or volume is increased by this heat, say 25 to pick an intermediate value. Neglecting the water in the parcel, it therefore takes 25n joules to raise the parcel’s temperature by one degree. The temperature therefore rises by 40680/25n = 1630/n degrees. As a fraction of T this is 1630/Tn.
When T = 300 this fraction is 1630/(300*n) = 5.4/n.
So although n went down by 1/n of its value, T went up by 5.4/n of its value, overwhelming the vapor loss by a factor of more than 5.
For example if the parcel originally consisted of n = 1000 moles then n decreased by 0.1% while T increased by 0.54%.
From this we conclude that condensation increases nT and therefore increases PV.
That’s all there is to it.
Why not just set n = 1000 and T = 300 in the beginning, instead of cluttering up the algebra with 1/n and T? If we were looking for the simplest calculation surely that would be even simpler?
Good question. Doing it as above shows that the ratio between 1/n and 5.4/n is independent of n, equivalently that the 5.4 ratio remains fixed throughout the process of condensation. This would not have been apparent if we’d set n to 1000 in the beginning. It also shows that this ratio increases with decreasing T, as happens at higher altitudes.
Vaughan, I think the question is still which volume gets heated. With cooling/condensation, the volume of air is releasing heat.; If that volume regains its own heat, nothing happens. Since the altitude of the condensation remains fixed, or relatively close to fixed, combined forces on the layer have to remain close to equal. If the storm is to build or maintain strength, it has to be fed fresh moist air, not sinking dry air or “wet” air.
Vaughan and Capt.Dallas,
The altitude of condensation is determined by the moist adiabatic lapse rate and the moist adiabatic lapse rate is as low as it is, because of the warming that Vaughan described.
All the variables are related. That’s the content of every derivation of the moist adiabatic lapse rate.
It’s not particularly useful to pick one variable and look, how condensation affects that, when the better approach is well known and results can be found from various sources.
The formulas get a bit complex in the moist case as can be seen from the Caballero lecture notes as one easily available source. Otherwise no surprising issue come out from the derivation.
The above tells naturally also the most straightforward argument for knowing that nothing like the paper discussed here is needed. The problem has been solved in full, and the solution can be found from textbooks and lecture notes. Nothing significant is left out from those problems that this paper tries to calculate.
Pekka, “It’s not particularly useful to pick one variable and look, how condensation affects that, when the better approach is well known and results can be found from various sources.”
The last I read though, clouds and indirect aerosol effect uncertainties are large and getting larger. As I have mentioned before, dense water/water vapor can behave as a ground plane for radiant heat. A dense cloud base is obviously stable and super saturated not only with water vapor but liquid water which would restricted airflow as well. Clausius-Clapeyron doesn’t deal well with super saturated conditions and is obviously missing something or the models wouldn’t be going belly up. Likely the low altitude impact of aerosols on water droplet size. Clean air causing greater supersaturate larger droplets size, more ice and mixed phase cloud formation etc. etc.
The interesting point to me is not the basics of condensation and latent heat, but how the depth and density of the condensation layer impacts the distribution of that heat.
@cd: If that volume regains its own heat, nothing happens.
I don’t buy that, capn. What’s to stop a volume warming up as a result of some of its water vapor condensing? That’s what both Meesters and Rosenfeld describe, with which I fully concur. The whole warming process can happen without the RH dropping to 100%, e.g. it might decline from 105% to 101%. Throughout that decline temperature will outpace vapor loss 5 to 1.
Vaughan Pratt, “I don’t buy that, capn. What’s to stop a volume warming up as a result of some of its water vapor condensing?” Nothing, we have puffy clouds with no rain all that time. But to have a storm, the cycle cannot scavenge its own energy, it has to direct that energy to a sink.
Vaughan,
“As all these quantitative estimates are a bit tedious to work through, it is natural to ask whether they could be simplified.”
My version is here.
Pekka, “Pressure changes all the time.” Of course it does. Micro bursts of 70 knots happen all the time. A heavy storm releases energy pushing the cloud top up to sometimes 20km in deep convection and downdrafts increase with evaporative cooling with the downward flow converting to horizontal flow. It is call dynamics I believe.
http://www-das.uwyo.edu/~geerts/papers/waf_gusts/
WINDEX is a strongly empirically based program to estimate gusts and micro bursts. Since I know that the density and average particle size of water droplets impact ventilation of a condensation layer I am curious why you keep saying that “we got it all covered. Nothing to see here folks.”
@NS: My version is here.
Thanks for that, Nick. Amazing similarities with my numbers: n = 1000 moles of air, T = 300 K, etc. Great minds think alike. :)
One difference is that you kept the Clausius-Clapeyron relation, which I was able to dispense with in my third attempt on the ground that it is irrelevant. (My first two attempts were more complicated because at that point it’s irrelevance had not yet become clear to me.)
The simplest way, perhaps too simple, is this.
1. 1g/kg of condensation releases enough latent heat to warm the air by 2.5 C.
2. 1 g/kg has a partial pressure of 0.001*p/0.622 which is 1.6 mb at sea level, less higher up.
3. 2.5 degrees of warming raises the pressure by 2.5/T(Kelvin), close to 1%, which is 10 mb at sea level. (This is all at constant volume because that won’t change until the pressure has changed.)
4. 10 >> 1.6 QED heating >> vapor loss.
5. Pressure increases, volume expands, density drops, buoyancy ensues, air rises.
Actually we would replace the 2.5 with something like 3.5 if we considered constant volume, then we get an even bigger ratio, but I think the general idea is like Vaughan’s.
Concerning presentations of both Vaughan and Jim I don’t see why the constant volume case is discussed at all. Constant pressure formulas are the only natural ones for discussing the atmospheric processes. There are no walls around volumes there and so fast phenomena that the initial phase would resemble constant volume are extremely rare, if they exist at all.
3. 2.5 degrees of warming raises the pressure by 2.5/T(Kelvin), close to 1%, which is 10 mb at sea level. (This is all at constant volume because that won’t change until the pressure has changed.)
1% about 10 mb would produce what velocity of winds?
The pressure changes all the time. Nearly simultaneous flash condensation in a large volume might be an exception, but such phenomena require special reasons like shock waves. A lightning could cause such a chock wave. Jet fighters and supersonic meteorites might do that as well, but perhaps we are not considering such events.
For all other purposes the changes are simultaneous, not one following the other with a delay. When one extra molecule enters a droplet the void that it leaves disappears very soon without perceptible drop in pressure.
@PP: Concerning presentations of both Vaughan and Jim I don’t see why the constant volume case is discussed at all.
Only because the constant volume case brackets one extreme end of a range that the constant pressure case brackets at the other end.
In my first attempt at answering the question of whether condensation increased or decreased nT (before I saw Meesters’ and Rosenfeld’s analyses answering this question years ago in reviews of this ongoing series of papers from Makarieva and Gorshkov) I considered both cases in order to bracket the whole range of possibilities. That analysis required one look-up of the steam tables aka dewpoint calculation aka the Clausius-Clapeyron relation.
My second attempt worried about undershooting RH = 100% which therefore required two look-ups to see where the undershoot happened.
My third attempt decided that RH = 100% was a red herring, reducing the number of look-ups to zero. So what if condensation reduces RH to below 100%? The 5:1 result remains valid nonetheless. Clausius-Clapeyron, which merely plots the liquid-gas boundary for RH = 100%, has nothing to do with it. This makes the analysis way simpler!
That last attempt took the middle ground of allowing both P and V to increase, which I associated with a specific heat for air of 25 joules/mole/K, intermediate between cv = 21 and cp = 29 for air. Given that temperature trumps vapor loss 5 to 1 this difference between 21 and 29 surely doesn’t matter.
Pekka, we can assume constant pressure too, but in the real case the pressure is also decreasing as the air ascends. Changing from constant volume to one of constant pressure does not change the basic result that adding latent heat has a bigger effect than removing vapor. I was viewing it as, if this was the only thing happening, the pressure would change first which results in an expansion of volume and reduced density, but all this is very fast.
captdallas, the velocity from this pressure rise depends on the volume over which this is occurring. But the volume finally expands by 1% with this velocity to compensate, so this is a temporary situation, and also 1 g/kg is not condensed suddenly in the real case, because supersaturation remains low, so it is released gradually keeping the pressure gradient small and the expansion gradual.
I agree 100% with Jim D. I don’t understand Pekka’s concern but it doesn’t seem like a big deal.
The reason for my comment is simple. One way is correct, the other one is not. The processes at the local level where they occur are follow very closely the constant pressure equations, where constant volume equations give different results they give wrong results.
What I write is not totally exact, but very close to that.
Constant volume formulas apply, when the space has solid walls, constant pressure formulas apply when expansion is counteracted by the pressure of gas outside the small volume being considered.
There are cases where neither applies much better than the other, but those are complex in other ways too, and usually need non-equilibrium thermodynamics. Fortunately much can be done without. The the above choice of approach applies.
JimD, “captdallas, the velocity from this pressure rise depends on the volume over which this is occurring. But the volume finally expands by 1% with this velocity to compensate, so this is a temporary situation, and also 1 g/kg is not condensed suddenly in the real case, because supersaturation remains low, so it is released gradually keeping the pressure gradient small and the expansion gradual.”
The 1% decrease in volume is like starting the turbine spinning on a jet. If you keep fuel coming, you can maintain that compression, about 75 knots of wind worth. You just need to direct the exhaust.;
If you look at a hurricane, squall lines form in a radial pattern channeling the induce surface winds toward the eye wall. The Pws gives you the potential of the Hurricane, but the configuration has to be near perfect to realize that potential. That’s dynamics. Thermo with Cv or Cp just describes parts of the cycle.
https://lh5.googleusercontent.com/-6kOECaf5sPQ/URTZvwXBDBI/AAAAAAAAHIo/IdsG1W8AlMw/s800/condensation%2520cycle%2520psych.png
That should be roughly the ground or compression part of the cycle. The efficiency would depend on the minimum pressure at the exhaust and the fuel air mixture at the surface. Instead of metal blades, you have liquid water.
When you have saturated air at a high velocity across a liquid water surface or surfaces in this case, you get a higher degree of super saturation which would increase the compression and combustion efficiency. Then again, too high blows the engine apart :)
To continue from CD.
What happens in atmosphere is determined by a combination of thermodynamics and fluid dynamics. We must have equations from both, but at the minimum we don’t need anything else than the continuity equation from fluid dynamics. That’s enough when we don’t try to study dynamic phenomena where time is one independent variable, but are satisfied in comparing two states that differ from each other by an infinitesimal amount. They are states at two times separated by an infinitesimal dt, but we may fix the two states without any reference to the intervening time step. Alternatively we can refer to the intercal but keep some velocities fixed at externally defined values, that could tell, e.g., the velocity of a vertically ascending parcel of air.
If we want to determine the velocities from physics for a realistic case then we need much more complicated fluid dynamics, the Navier-Stokes equation and rather complex boundary conditions.
All the considerations described by formulas in the subject paper of this thread and the first part of understanding ascending moist air is done having the continuity equation as the only fluid dynamics equation. Therefore the set of questions that can be considered is limited. All these questions are answered fully by the standard theory of moist adiabats found from Caballero lecture notes as an example of a suitable source. It takes a few pages loaded by equations. The subject paper of this thread presents a fair part of this theory in chapter 2 as well, but nothing that’s not part of the standard theory and not all that belongs to the standard theory.
Pekka, “If we want to determine the velocities from physics for a realistic case then we need much more complicated fluid dynamics, the Navier-Stokes equation and rather complex boundary conditions.”
True, the paper only considers one of the boundaries, the condensation plane or plate. As Tomas mentions, the other boundaries would all need to be considered, but the condensation plane would limit those boundaries.
I don’t see any great new “insight” in the paper, just a solid frame of reference with Pws or Nv as a potential based on that reference. The degree of super saturation that can be generated with high velocity air over a liquid water boundary layer is the indicator of the efficiency. There is still plenty of problem left, just a different starting point.
@PP: Constant volume formulas apply, when the space has solid walls, constant pressure formulas apply when expansion is counteracted by the pressure of gas outside the small volume being considered.
As long as condensation isn’t instantaneous I’d be ok with this.
But in any case the difference between constant pressure and constant volume is only a factor of 7/5 in the case of air, whereas latent heat increases PV about five times as fast as vapor loss contracts it at any realistic temperature.
More precisely, at T = 300 this ratio is 40680/(300*20.76) = 6.53 in the case of constant volume and 40680/(300*29.07) = 4.66 for constant pressure. In either case latent heat is raising PV much faster than vapor loss is lowering it, whence it’s academic which case we have.
captaindallas, not sure if you understood what I said, but basically if you heat air by 1%, its volume increases by 1% at constant pressure redcuing the density. Any pressure increase would be temporary as would be the velocity associated with expansion. If you somehow removed vapor without latent heat release, the pressure reduction causes the volume to quickly fill in with dry air, which than increases the density and causes negative buoyancy, somewhat opposite to what the paper expects a negative pressure anomaly to do.
JimD, ” Any pressure increase would be temporary as would be the velocity associated with expansion. ” How temporary? I can mix up a fuel chemicals and make some sugar fuel. If I limit the rate of expansion and direct it with a nozzle, make a fair rocket. Most of my rockets were just temporary, rarely more than 10 seconds of burn. Some were even more temporary though spectacular launch pad explosions.
Same way with condensation, it releases energy in all directions, but if you direct the energy and control the burn, you can move stuff.
captaindallas, the volume adjustment to pressure changes occurs at the speed of sound. This is the way sound waves work. In fact, low-frequency acoustic modes are radiated in the process of heating (see Pielke’s comment).
JimD, “captaindallas, the volume adjustment to pressure changes occurs at the speed of sound. ”
And the speed of sound is dependent on the temperature, pressure and density of the medium it is traveling through.
http://en.wikipedia.org/wiki/File:Comparison_US_standard_atmosphere_1962.svg
So if the base of the condensation is fixed by a solid surface, it can “fill” more quickly. If the solid surface can “fuel” the condensation process, the rate of “fill” can be maintained or even accelerated. The surface allows the reaction to induce, fuels creating a flow aka wind.
captaindallas, a column fills in from around just as quickly, not that condensation would ever occur in a column down to the surface. Anyway you neglected the opposing latent heat effect that dominates.
JimD, “captaindallas, a column fills in from around just as quickly, not that condensation would ever occur in a column down to the surface. Anyway you neglected the opposing latent heat effect that dominates.”
Why not condensation down to the surface? Look at the micro climate of a tropical rain forest. The surface temperature can be 7 to 10 degrees cooler than the canopy top. Once you put a column or columns in motion, there is a lot to consider.
Wind is nothing more than velocity pressure. If a cloud builds, obviously the latent heat is creating a velocity pressure. The heat released would be isotropic. Buoyancy forces a larger portion of the heat upward. Cooler more dense air will replace that volume, but it doesn’t have to be all cold dry air from some higher altitude. There is evaporative cooling which is increased with surface winds, aka velocity pressure that also can do the “filling”.
Location, location, location. If the location of the condensation layer allows more saturated or near saturated air into the “combustion” chamber, there is a greater dynamic flow of energy.
captaindallas, so you’re departing from Anastassia in allowing for buoyancy and heating pressure at all. You should say this part louder. The rest doesn’t fit together well. Expansion and sound waves respond far more quickly to heating than buoyancy ascent (see atom bombs). Once this has happened there isn’t much pressure gradient left except to fill in where the buoyancy has lifted the warmed air so the wind is responding to that buoyancy-driven motion. This is just the conventional view, however, no biggy.
JimD, “captaindallas, so you’re departing from Anastassia in allowing for buoyancy and heating pressure at all.”
Yes, as a way to explain. What Anastassia appears to be doing is saying that the change in moles of water vapor, the fuel, can be used to describe the process, pretty much like fuel/air mixture is important in determining the process of any engine. I think it is rather elegant myself.
Getting people to think outside of their boxes is a bit of a challenge, but a change in a frame of reference can greatly simplify solving problems.
In their case outside the box, is outside thermodynamics too.Thermodynamics was such a promising science, Clausius-Clapeyron, Boyle’s Law, entropy, Gibbs free energy, enthalpy. I’ll be sad to see them all go, or, maybe I will just stay inside the box.
Vaughan Pratt | February 7, 2013 at 2:20 pm | Reply
Vaughan, I appreciate your interest in our work. If you are certain that your calculations of the significance of latent heat are correct, I’m fine with that too. But if you still have some doubts (e.g. why this so obvious reasoning apparently did not convince some people who are presumably not complete ignorants) I recommend that you read this comment and another one made at the blog of Eli Rabett. My personal opinion is that as soon as you at least vaguely understand what these comments are about, you will immediately stop entertaining yourself with the above calculations and will lose interest to similar exercises of other people.
@AM: If you are certain that your calculations of the significance of latent heat are correct, I’m fine with that too.
Sorry, not following. Are you saying you don’t believe latent heat is significant during condensation?
why this so obvious reasoning apparently did not convince some people who are presumably not complete ignorants
Those most clearly not convinced here would seem to include Max Manacker, David Springer, Don Monfort, and Peter Davies. Did you have others in mind?
I recommend that you read this comment and another one made at the blog of Eli Rabett.
The logic in your first comment can be used to prove that air bubbles cannot rise in water, since that would oblige some water to move down. Having personally observed air bubbles rise in water on many occasions, I find this argument unconvincing. The logic in your comment at Rabett Run would appear to depend on the same reasoning.
Anastassia I appreciate the work you and the other members of your group have put into this project. I am moved to comment on your paper as a lay person who is not a scientist because Vaughan Pratt seems to have read somewhere that I am not convinced that their criticismsof your paper have any basis in science and I have been lumped together with some other commenters who are “non scientists”.
I am a fan of Tomas Milanovich and he has made some comments on your paper both during the original peer review stage and now, and I believe that his point that your system of equations should have been extended so as to ensure that most of the relevant variables have been specified and for a sufficient number of equations so that each variable may be solved mathematically.
The model that you have used for demonstrating the principle of the biotic pump probably needs to more clearly bridge the gap between the macro effects of this principle and the molecular behaviour of the GH gases that affect pressure above and below the clouds in each chimney.
@PD: Vaughan Pratt seems to have read somewhere that I am not convinced that their criticismsof your paper have any basis in science
My apologies, Peter, I misinterpreted your concerns about my criticisms of the paper. I would gladly remove your name from my list if WordPress would allow it.
Yes, Vaughan, but what about additional mass being added to the original volume from the surroundings ?
That will add weight and mitigate the effect of the heat release by sharing it out amongst more mass.
And after all that the parcel will still not be as light as the rising water vapour laden air rising up from beneath it so it must descend and in doing so raise surface pressure.
The point that is being missed here is that condensation is the mirror image of evaporation.
Evaporation draws most if not all of the energy it requires from the water surface which then cools.
Condensation puts most if not all of the energy it releases into the water droplets.
The reason in both cases is that, for the same temperature, liquid water has a far higher thermal capacity than air.
A phase change involves no change in temperature so for the condensate to be at the same temperature as the surrounding air then due to its higher thermal capacity it must retain most if not all the latent heat released from the vapour form.
The latent heat being a form of potential energy goes straight to gravitational potential energy in the droplet which is heavier than the vapour and which must therefore carry more gravitational PE.than vapour.
In both forms of water the mass is the same, only the volume changes.
It is all very well calling it sensible heat and so it is but it is used in keeping the sensible temperature of the condensate at the temperature of the air at that height.
It is the pressure at that height which determines the temperature of both water droplets and air just as it is pressure at the surface which determines the temperature of both water and air at the surface.
Turning now to the matter of down drafts and micro bursts.
At the surface, evaporation takes energy from the water surface or from the surface on land and reduces the density of the air parcel above so that the air parcel becomes lighter and can rise. As the parcel rises it cools adiabatically until condensate is produced.
When condensate is produced everything stays at the same temperature because all the released latent heat goes to the condensate in order to keep it at the same temperature as the surrounding air.
That increases the density of the air parcel because dry air is heavier than air containing water vapour so the air parcel must start to fall with its droplets.
In the process of falling it warms adiabatically and begins to reabsorb some of the droplets by renewed evaporation. That evaporation cools the air parcel which makes it yet denser so the downdraft accelerates taking its water droplets with it and if not all the water droplets are reabsorbed then the surface gets rain.
The point being that evaporation starts the initial ascent and adiabatic cooling keeps it going until condensation occurs.
When condensation occurs the descent begins and adiabatic heating occurs which is NOT offset by evaporative cooling.
The reason for the difference is that when air is rising it goes with the gravitational field so that the reducing pressure gradient with height ADDS adiabatic cooling to any further evaporative cooling within the air parcel
whereas
when air is falling it goes against the gravitational field so that the increasing pressure gradient with height results in adiabatic warming that more than OFFSETS any evaporative cooling within the air parcel.
It is pressure at a given height which determines the sensible temperature of the air and of any water in it whether it be in vapour or liquid form and it is the movement of the air up or down that then determines whether and when that water changes from liquid to vapour or vice versa.
The key to the whole process is relative densities and pressures at any given height as per the Ideal Gas Laws.
If a parcel of air is lighter than it should be for its height then it will rise.
If it is heavier than it should be for its height then it will fall.
Air containing water in vapour form will rise higher than dry air because it is lighter so when the vapour is removed it must fall back to its ‘correct’ height but because of the air around it becoming warmer as it descends it will remain too dense for its height until it reaches the ground and receives more energy from the irradiated surface.
Then it picks up more water vapour and starts off on the cycle again.
Before signing off I want to highlight that “conventional ideas” should also be subjected to scrutiny. I recommend this illustration http://arxiv.org/abs/1212.3100 Unlike the high-level theory we have been discussing here it is very simple.
Thanks again to Judy for allowing us to the opportunity to be here, and to everyone who has participated in this discussion.
The water cycle should be described as follows:
i) KE at the surface causes evaporation which converts KE to latent heat which is a form of potential energy that does not register as sensible heat.
ii) The rising vapour rich air gradually cools as KE is converted to gravitational potential energy which does not register as sensible heat.
iii) When the air has been cooled to its dew point condensation will occur but a phase change involves no change in temperature.
iv) Condensation converts the latent heat to gravitational potential energy in the condensate.
v) The condensate falls to the ground and as it does so its gravitational potential energy is converted to KE
vi) At a later time the air also sinks back to the ground without its vapour load and as it does so its gravitational potential energy is converted to KE such that the dry air is then able to take on another load of water vapour.
The point being that there really is no sudden surge of sensible heat when condensation occurs.
The lifting of the water vapour to the height at which the dew point is reached ensures that all the latent heat goes to gravitational potential energy within the condensate.
The reason must be that at the dew point the energy value of the latent heat of evaporation becomes exactly equal to the amount of gravitational potential energy required by the condensate at that height.
It is the strength of the gravitational field acting with atmospheric mass and total system energy input that determines the relevant pressures, densities heights and temperatures for the phase changes.
This debate is a microcosm of the AGW debate in that thermodynamics is a well accepted science, and yet as soon as someone comes along and questions it, certain individuals who seem to have a deep resentment of established science, jump onto the bandwagon even when they don’t understand the established science first, and prefer to follow the rhetoric of contrarians. It is a state of mind that I would not call skepticism, because they are biased towards disbelief in conventional science and need only a small excuse to battle against the consensus.
It is a state of mind that I would not call skepticism, because they are biased towards disbelief in conventional science
It’s more complicated than that. They view themselves as being in perfect alignment with “true science.” Any scientist with opinions different from theirs has clearly taken the blue pill and fallen under the spell of the The Machine, which they view as the enemy of true science.
Had the scientists taken the red pill they would have seen the error of their ways. Eventually truth will out and the artificial science foisted on an innocent public by The Machine will be swept away and replaced by true science.
I participated in discussion on conspiratory ideation in another thread.
There’s is something familiar in this discussion on, what skeptics think and what we think about skeptics.
I like you guys and enjoy reading your thoughts, but if you think bias around here is a one way street you have taken the blue pill.
+1
In science there shouldn’t be bias, just considered opinion taking into account all the facts and your own previous knowledge. That leads where it may regardless of what you thought before. The bias comes in when you don’t have the background to make your own scientific evaluations, e.g. just reading a few blogs and Web links.
Jim D, the person who is not susceptible to biases and prejudices has not yet been born. Hence the need for rigour in science.
Rigor equals well founded basic knowledge. If you build a theory it has to go back to the roots of basic knowledge.
Jim D, you know how many long-standing theories over the ages have subsequently been found to be wrong?
And what’s the point of science when you can just read it from a book?
Jim, I am the person that posts here with the least amount of bias and even I am incapable of preventing any bias at all from seeping through.*
* This is opinion and subject to the author’s bias.
@steven: I like you guys and enjoy reading your thoughts, but if you think bias around here is a one way street you have taken the blue pill.
Who said anything about bias? Who here thinks they’ve taken the blue pill?
When I was growing up in Australia, only foreigners had accents.
Vaughan, people with an accent may not talk funny. There is always the possibility you just hear funny.
@steven: Vaughan, people with an accent may not talk funny. There is always the possibility you just hear funny.
Right, I believe that’s another way of expressing my point.
My mistake. I thought your point was that you couldn’t understand me due to my accent.
Jim D: This debate is a microcosm of the AGW debate in that thermodynamics is a well accepted science, and yet as soon as someone comes along and questions it, certain individuals who seem to have a deep resentment of established science, jump onto the bandwagon even when they don’t understand the established science first, and prefer to follow the rhetoric of contrarians.
That is true of some, perhaps, but not all. I and others have pointed out that the classical thermodynamics assumes equilibrium or “local” thermal equilibrium because they permit the solution of tractable equations, or at least approximate solutions. As described in Kondepudi and Prigogine, Modern Thermodynamics attempts to go beyond the restriction of equilibrium and describe high-dimensional dynamic dissipative systems “far” from equilibrium. (Readers will appreciate that “far” and “local” require some concepts of “nearness” .) As soon as you appreciate that the (local) equilibrium constraint is an approximation, you appreciate that it has an inaccuracy, and the question arises whether that inaccuracy is sufficiently great as to matter in particular cases: here, the particular cases of heat transfer within the Earth climate system.
Makarieva et al. have attempted an improvement, and have calculated a rather small effect, appx 2 W/m^2. The principle objection to her work, enunciated by Pekka Pirilla and Isaac Held, among others, is that they have explored an effect already concluded by experts to be too small to be important. How small is too small to be important? Their result is close to the estimated equilibrium effect of doubling the concentration of CO2. Is it consistent to maintain that up to 4 W/m^2 is extremely important but that 2 W/m^2 is so small as to be negligible? I don’t think so.
I have reread all of Pekka Pirila’s comments, and he seems to change and revise his arguments so much that I can’t tell which of them he still endorses. (Some of this is the effect of the threading; admittedly, some might be my cognitive limitations,) It seems to me that a published interchange between him and Makarieva et al would be a good contribution to the literature. If there are specific technical deficiencies in Makarieva et al (as there were in Einstein’s first 5 papers on general relativity — he finally “nailed it” in paper number 6), those specific deficiencies can be remedied by better approximations, and the effects on the subsequent derivations can be clearly displayed (Einstein did this in correcting his own published works.) I hope that the editor of the journal will encourage such a debate in the journal.
Equilibrium depends on your point of view. Viewed in the frequency domain a sine wave has a constant frequency, leading one to describe it as in equilibrium. But in the time domain it is constantly oscillating back and forth and therefore is clearly not in equilibrium.
Matt,
Nothing has changed in my views on correct physics. What has changed a little, is my guesses of the reason for the errors of the paper. Reading more carefully and checking the equations more carefully, I have got a more precise view of where they fail.
There are also several independent problems in the article, one is making a totally irrelevant comparison that’s presented in a misleading way. In that the mathematics is correct, but the results are used to support claims that they cannot support.
The other problems relate mainly to the use of the approximate continuity equation and the accurate continuity equation together to derive results that are nothing else than spurious consequences of the the error done. This error can be considered as a two independent errors that act together, one of the errors relates to the horizontal dependence of various variables, the other to handling of total molecular density and molecular density excluding vapor.
Results that are totally based on an error cannot be corrected, they can only be dismissed. That’s true for everything that is based on equations (34)-(37).
Looking at the rest of the paper, every formula is based either on the totally irrelevant case of chapter 3 or on the equations (34)-(37). Nothing concrete is left, when these erroneous parts are taken off.
Matt,
On the value of the debate. Anatomy of errors of a paper, whose results are based on trivial errors is not a relevant contribution to literature.
It’s even less when similar arguments were presented and put publicly available by others already before the publication, and when the editor tells that publishing the paper was opposed by the reviewers.
Jim D, it’s just good ol’ belief in the ignorance of experts, you know science.
Thermodynamics is one of those odd sciences that is not intuitive when you learn it as an undergrad, especially entropy. You only really start to understand it if you need to explain it to someone else, or if you need it for your job, which is why exercises like this thread are so useful, even when you think you know thermodynamics.
For some reason the formula dQ = TdS, which I learned as a physics honours student, really stuck in my mind even after several decades of having no use for it. Maybe the word-play with “tedious” had something to do with that.
In recent years it occurred to me to interpret temperature as noise. The interpretation of entropy as (negative) information then makes sense of the formula: each bit dS of information requires an amount dQ of energy to communicate it in the presence of an amount T of noise. The more noise, the more energy needed to communicate information.
Suddenly the Cray Model 2 cooling tower made perfect sense: reducing T reduced the energy needed to move information around.
Vaughan Pratt: dQ = TdS
I hope that is not all you remember. It only applies to a closed system. More generally, dQ – TdS >= 0. Because the Earth accumulates energy from the sun (photosynthesis makes carbon bonds later converted to coal and oil), dQ – TdS > 0 in the Earth climate system. The difference is “small”. I do not know whether it is (locally or everywhere) negligibly small.
@MM: More generally, dQ – TdS >= 0.
Thanks, Matthew. In my information theoretic interpretation this would correspond to saying that TdS is a lower bound on the energy dQ required to communicate dS information in the presence of T noise.
I’m glad the sign went that way. I’d have been deeply troubled if TdS had turned out to be an upper bound. ;)
Your point is that there may be other sources of noise than T, for example a cocktail party in the adjoining apartment, making it clear that additional energy may be needed to shout over those other sources.
Jim,
Thermodynamics is surprisingly difficult. We had a few decades at my university ago a very smart professor of thermodynamics, who joked:
“There are three people in the world, who understand thermodynamics. We meet annually in Paris.”
Having worked with people who use thermodynamics regularly in research, I have found out that very few of them understand the issues well. The reasons for having constant volume and constant pressure quantities and equations is perhaps the first obstacle, and the one that we have discussed here in recent messages. Entropy, exergy, etc. bring a second set of issues, Phase transitions, chemical reactions, and electrochemistry is one more layer.
All the above is within equilibrium thermodynamics. If we want to understand the atmosphere, fluid dynamics is equally important, and that’s unfortunately still much more difficult. The reasons for not being able to model the atmosphere accurately are mostly problems of fluid dynamics. The pressure derivatives that the present paper tries to calculate and the weather phenomena that Capt.Dallas commented on can be determined only when fluid dynamics is included. The potential for additional new ideas and approximate constraint equations comes mainly from fluid dynamics part. When that part is in better order, handling thermodynamics well enough is likely to be relatively easy.
@PP: Phase transitions, chemical reactions, and electrochemistry is one more layer.
And then comes Maxwell’s Demon, about whom my 11th Ph.D. student Paul Fahn wrote his thesis.
Maxwell’s demon is, however, not part of classical thermodynamics but realted to the statistical version of thermodynamics, which is certainly better in the way that it goes one step deeper and explains, why classical thermodynamics works from more fundamental principles.
The problems we have discussed here are within classical thermodynamics.
You forgot heat transfer Pekka. However, I always found thermodynamics surprisingly easy and never understood people who found it difficult.
Edim,
By heat transfer you mean probably radiative heat transfer. That’s not a significant factor in these phenomena as has been noted in this thread before.
How has your understanding of thermodynamics been tested?
Let me count the ways.
=============
Pekka Pirila: The pressure derivatives that the present paper tries to calculate and the weather phenomena that Capt.Dallas commented on can be determined only when fluid dynamics is included.
I think much good would follow if you or other critics of Makarieva et al would follow the example of their paper, replace the incorrect assumptions with assumptions that you think are correct, or at least more accurate, and derive the consequences. My guess (less reliable than Isaac Held’s “guess”) is that convincingly more accurate approaches than theirs will be hard to come by.
“By heat transfer you mean probably radiative heat transfer.”
Absolutely not Pekka, I mean Heat Transfer, as it is taught in mechanical engineering for instance. it includes ALL forms of heat transfer.
Edim,
The other forms of heat transfer are conduction that’s still far less significant than radiative heat transfer and convection that’s included as it latent heat transfer if you consider that a form of heat transfer as well.
Nothing significant is left out.
@MM: My guess (less reliable than Isaac Held’s “guess”) is that convincingly more accurate approaches than theirs will be hard to come by.
Just to put this in perspective, their claim in a series of papers since around 2006 or so has been that condensation reduces pressure (better stated as reducing PV, pressure times volume, since air is going to move in pretty quickly to restore P).
Whereas Held merely guessed, starting at the latest in 2008 various people have explicitly calculated that latent heat offsets that effect several times over.
One would conclude from this that a less accurate approach than theirs would be hard to come by.
Vaughan Pratt: Whereas Held merely guessed, starting at the latest in 2008 various people have explicitly calculated that latent heat offsets that effect several times over.
Are you saying that Held guessed that various people have explicitly calculated that latent heat offsets that effect several time over? What’s that guess, that people have explicitly calculated, or that latent heat doesn’t conserve energy but offsets “that effect” several times over.
Is there a reference you can supply to these explicit calculations?
Matthew Marler, one mistake was when they cooled the dry and moist columns without reducing the pressure in them. If they had, the columns would have remained comparable, but the dry column would have had a lower pressure because it was cooled more, and this would have dominated the pressure effect they eventually looked at making the dry column have a lower pressure. As it was, they subtracted different heat from each column, then you may as well choose two random columns and compare their pressures which is quite pointless.
Matt,
In the chapter 4, which is that of main interest here, the paper studies how condensation occurs in a persistent and essentially stationary settings. That problem has been studied very widely and described in all textbooks and sources like the Caballero lecture notes. There’s no need for me to repeat all that here. Many of these sources do it without any additional approximation or simplification. It’s all well known. The textbook descriptions seem very different of the paper, because they approach the problem logically rather than jump in to the middle of the analysis.
The claim of the paper that they bring to the discussion a phenomenon that has been wrongly dismissed by others is a red herring, there’s no basis for that claim. The only thing they add is an explicitly erroneous equation that leads immediately to totally unphysical results. That equation is the approximate continuity equation, which is an error, when the exact one is already included. It’s not new physical input, it’s only an error.
Condensation occurs in a stationary way in an moist uplift. When air rises, it’s pressure drops. The process is essentially adiabatic as heat does not have time to cross to a significant extent the boundaries of the parcel of air considered. Most discussions are on a smooth enough process to allow it to be considered isentropic (reversible) as well. One part of the approximate isentropy is that the level of supersaturation is low, as condensation of supersaturated vapor is not isentropic.
Adiabatic expansion leads always to cooling. In the isentropic case the rate of cooling can be calculated exactly, otherwise the cooling will be less. A sudden condensation of supersaturated vapor may release so much heat that the parcel of air warms more from that than it has cooled over an extended period, but even in this case the main trend is cooling.
The reason for condensation is the cooling, and the reason of cooling the reduction in pressure. The removal of vapor in condensation adds to the reduction in pressure, but the heat released in the same condensation influences the temperature more, and the effect trough the temperature is roughly five times stronger than the effect of the reduction of vapor. The derivation of this factor of five has been presented in this thread by Vaughan. A link to a similar derivation by Nick Stokes has also been given. It’s an elementary calculation found certainly in very many sources.
The chain is:
uplift motion ->
pressure drop ->
cooling ->
condensation ->
reduction in vapor content and reduction in the rate of cooling ->
reduction in the pressure drop
The whole chain from pressure drop to reduction in pressure drop must be closed by an equation that requires the pressure drop have one value. Solving the set of equations of the whole chain gives as an outcome the formula for the moist adiabatic lapse rate as well as formulas for all the variables involved as function of the altitude.
That’s the uplift part calculated for the isentropic hydrostatic case. That’s a useful idealization as that’s calculable without large models. To calculate the whole circulation, fluid dynamics must be involved in an essential way. That gets much more complicated, nothing like that is discussed in the paper. Atmospheric scientists have learned over years, how many issues related to that can be understood without reference to full GCM’s. An elementary description of that takes tens of pages in introductory text books and much more at more advanced level.
Pekka Pirila: Condensation occurs in a stationary way in an moist uplift. When air rises, it’s pressure drops. The process is essentially adiabatic as heat does not have time to cross to a significant extent the boundaries of the parcel of air considered. Most discussions are on a smooth enough process to allow it to be considered isentropic (reversible) as well. One part of the approximate isentropy is that the level of supersaturation is low, as condensation of supersaturated vapor is not isentropic.
It’s not stationary in fact, and “essentially adiabatic” means “not quite adiabatic”. For “a significant” extent it would be nice to have accurate measurements. (In God we trust. All others bring data.) “to be considered” is another statement of an approximation whose approximation error is unknown, as is “approximate isentropy”.
All the above is within equilibrium thermodynamics. If we want to understand the atmosphere, fluid dynamics is equally important, and that’s unfortunately still much more difficult
You seem to be debating within yourself, but in writing, whether you agree with you or not. I have referred to non-equilibrium thermodynamics, and other high-dimensional nonlinear dissipative systems, through my reference to the textbook “Modern Thermodynamics” by Kondepudi and Prigogine.
Matt,
In physics the real situations are almost never exactly in agreement with equations used in doing calculations, but it’s very often known that they are close enough.
If you wish to have absolute statements, it’s better that you forget everything that tells about real world. If you wish to learn about real world, you must learn to understand, when approximations are good enough and when not.
There’s nothing problematic in what I have written in that comment.
Pekka Pirila: If you wish to have absolute statements, it’s better that you forget everything that tells about real world. If you wish to learn about real world, you must learn to understand, when approximations are good enough and when not.
On that we agree, sort of. I would say that you have to demonstrate that approximations are good enough .
Matt,
Naturally it must be assured that the approximations are good enough.
In this case I started be defining a rather idealized case of uplift that’s adiabatic and not far from isentropic. I did comment on one factor that causes irreversibility or deviation from isentropic, namely oversaturation.
Naturally all turbulence does the same as does all mixing of air with properties that differ in any way and many other phenomena.
The real atmospheric processes do certainly deviate significantly from the idealized case, and determining more realistically what is going on is important for meteorology. Fortunately meteorologist have very many measurements from the real atmosphere at their disposal and can use that to improve their models, both the simplified conceptual models and the large models that they use in forecasting. That’s the part of the atmospheric science where these issues are developed and tested. Climate science cannot test it’s models equally well, but these issues are not really their immediate concern.
The paper discussed in this paper made equally strong idealized assumptions on certain issues, and where they didn’t, they didn’t propose anything more realistic either. Thus this paper is not one that even tried to improve on this point, they tried something else and failed.
It was always intuitive to me as it gets. It’s just energy balance for thermodynamic systems plus other stuff (entropy for example). Piece of cake.
Did you meet annually with the other three in Paris?
Then it would be intuitive that condensation increases pressure at constant volume, or increases volume at constant pressure, or did you not see that mistake they made.
JimD, Condensation does not increase pressure in a constant volume unless the latent heat warms that volume, as in adiabatic then if the volume is adiabatic it would be warmer so there would be evaporation which would return everything to its initial state. You are stuck in an idealized situation. They covered that in the adiabatic and non-adiabatic comparisons.
“Condensation can be accompanied by a pressure increase only if dS<0. This requires that work is performed on the gas such as occurs if it is isothermally compressed. (We note too, that if pure saturated water vapor is isothermally compressed condensation occurs, but the Clausius-Clapeyron equation
(Eq. 3) shows that the vapor pressure remains unchanged being purely a function of temperature .)”
The latent heat has to be removed from the system by conduction, very slow or radiation pretty quick. .Which Pekka said was not part of the problem when it actually is the problem, evaporation has to cool something and condensation has to warm something, outside of the little adiabatic box, or the box explodes or gets crushed. .
Why did they exclude dT? Seems that would be where the latent heat goes.
JimD, I don’t think exclude dT, just you can’t contain all the dT. Some energy has to leave the volume if it is to remain a constant volume. Soit is the amount of loss, dS which they are considering which is about 1% or 4 Wm-2.
The confusion is that they are mainly concerned with the potential energy to estimate limits. In a hurricane dS would be greater, a tornado would be the greatest. So they used the isothermal and saturated in the horizontal versus lapse rate in the vertical in an attempt to bound the model.
My problem is that the isothermal would be in the x-y plane and both dimensions would need to be considered eventually, but for limits, what they have is reasonably close.
In any case, there is energy lost moving mass horizontally in the atmosphere that is not included in the models, i.e. SSW events are not included in the budget. That advection has to be considered.
capt.d., in their own words “Excluding dT/T from Eq. (17)…” They just do things like this with no explanation. They assumed non-adiabatic means isothermal. They go on to say that this requires loss of dS for condensation, to which I would say “well, duh…”. Condensation is adding heat, and to keep it isothermal you have to remove it just as fast so dS less than zero for sure. Condensation is sure not isothermal, and if they want to say it is, they need more than “Excluding dT/T…” to explain what they are assuming.
That is for the isothermal x dimension, since they are making the saturation level isothermal, which is very physical lots of clouds have flat bottoms, there would be no dT in x. That is just saying that the cloud base doesn’t move up or down. Had they been more specific saying that the plane of condensation in the x and y dimension is assumed to be stable in altitude and isothermal, then…. it would have been clear.
capt.d., no you have to follow the air. It doesn’t make sense to use thermodynamics if air is flowing through the system. Following the air, it is gaining latent heat. Their assumption of isothermal condensation is completely non-physical in nature, and also not explained in any way, which is a major shame because their whole result depends on it.
Capt.
I wrote a lengthier answer to Matt a little higher up in this thread. The point is that overall trends are
– pressure drops
– temperature drops
– volume expands
– vapor content goes down
– latent heat is released
The signs of changes are these. Reduction in vapor content makes the expansion on volume a little less than it would be keeping everything else fixed, but the latent heat release has a stronger effect trough temperature and adds to the volume expansion. Thus condensation makes the volume expand more when the pressure change is kept fixed.
I have written these already, but I do it again in a modified form.
Studying complex systems like the atmosphere benefit from new ideas where new simplifications are proposed to better understand dynamics beyond the local scale thermodynamics, which is the only part known well enough for most purposes from first principles.
Even on local scale the dynamics of condensation is not understood well. In contrast the local consequences of condensation are understood well. It was an unfortunate starting point for the paper discussed here that it presents new derivations in a way that would change exactly that small part of the theory that does not need new ideas. From that it did also follow that only two outcomes where possible for that proposal, either the results agree with the existing theory or they are wrong. They didn’t agree and they were wrong. They were wrong by being internally contradictory, differing essentially from a correct theory and by leading to totally wrong predictions in contradiction with experiments.
Anastassia Makarieva has written many papers on a variety of fields. She must be a creative scientist. We need creative scientists who make often errors, because they make also new discoveries. When new discoveries meet resistance the authors should not give up too easily, but there are situations where they should give up. This paper is certainly one that should be given up. The errors are really impossible to get over.
Many of the counterarguments are not strong. They don’t go to the heart of the problem and can be largely dismissed, but the equation (34) together with assumptions on horizontal dependencies of the variables is just totally unjustified and even more importantly leads to totally wrong results. It’s not an allowable new hypothesis, it’s totally wrong. The following equations cannot be used for any purpose, they are part of the nonsensical outcome that the wrong equation leads to.
We need creative scientists who make often errors, because they make also new discoveries.
AI has the concepts of generate and filter. AM is great at generating a wide variety of theories but terrible at filtering them for reasonableness. She urgently needs a collaborator who can filter well. So far there is no sign that any of her coauthors (six at last count) can help her with that, which makes her very much a loose cannon.
How is the curve fitting going, doc? Found anything physical, yet?
Woods? It’s been about three years.
Woods cant test what it purports to test.
Steven,
Tell that to the doc. He needs some help. It’s been three years. He must be spending a fortune on Scotch tape and that premium Saran Wrap.
What’s the connection to the set of equations (32), (33) and (34)?
How can equation (34) be derived correctly as an equation that’s not dependent on the continuity equations?
Pekka you disappoint me. But I think that N.Stokes got my point – must be the name :)
First, and you apparently got this part, we are dealing here with a problem of dynamics. As you correctly wrote, thermodynamics deals with equilibriums and that’s why it can’t give any useful answer on this question.
It may give some local constraints (because we are in LTE) but that’s about it.
That’s why the equations I wrote are those that correctly deal with the dynamics, spatially variable fields etc.
They are correct, describe the physical reality and most importantly have a unique solution in the considered frame (2D steady state).
If you deny the above statement, I will prove it.
What is the connection to 32,33,34? It should have been obvious! 32,33,34 are just a particular subset of the equations I wrote.
I wrote that I didn’t specify the Fi so I will do so now.
Let F1(x,z)= 0 and F2(x,z)=w.dRv/dz.
And here you have 32,33 and 34 embedded in the system.
It is trivial to see that I didn’t change the mathematical properties which were valid for all Fi(x,z).
So the system is still consistent, correct and allows a unique solution.
The difference to what Anastassia did is that she didn’t write the second equation and paid for it by loosing unicity.
Another difference is that I didn’t mention T (temperature field)-I hope that everybody sees why this is not a problem.for the discussion we have.
Now the last point left is why I would specify F2 like I did.
Well I could answer “And why not” and leave it at that.
But I will add that a dependence of the condensation rate on the vertical velocity component and on the humidity gradient doesn’t strike me as specially contradicting anything I know. It even sounds rather plausible.
Tomas,
No way can that be Navier-Stokes.
The source term is included trough the continuity equation, it cannot be reintroduced trough Navier-Stokes. That would be over-determination. The other side of the source term comes from Clausius-Clapeyron, not Navier-Stokes.
And whatever it is, if it leads to obviously unphysical results, it’s wrong.
I could reformulate the above as saying that the continuity equation can be derived from Navier-Stokes under specific assumptions, but then it gives again the exact continuity equation and is not independent of (32) and (33) but the same. Thus Navier-Stokes cannot provide an independent extra equation like that.
“.. under specific assumptions ..” was superfluous in the above.
The point is that the equation (34) has come from somewhere. Perhaps it’s derived from Navier-Stokes as the continuity equation making approximations in the derivation, but then it’s just the continuity equation. That’s a plausible explanation, but then it cannot be used like it has been used. It’s not an independent equation due to the error terms introduced in the approximations.
If there’s a real alternative basis for that equation, no hint has been given by anyone on, how that leads to this specific equation. They have got it from somewhere through some manipulation. Why haven’t they told the origin in detail for us to see, whether it’s indeed an approximate continuity equation?
I wrote the above a little too hastily to have everything right. The basic message stands.
The set of equations used, when Navier-Stokes equations are applied includes the continuity equation. The source term comes trough that equation. Otherwise it’s external to the set of equations, and given by some other mechanism described by other equations. In this case that other mechanism is the condensation described by thermodynamic equations like the Clausius-Clapeyron equation.
The whole set of fluid dynamics equations cannot say anything more about the source term. In the case of the paper it’s included in the equation (33). Nothing more on that can come from Navier-Stokes.
If the calculation would be a full fluid dynamic calculation, Navier-Stokes would be essential, but that’s not the case. What I wrote of the form of the equation (34) remains valid. Navier-Stokes cannot produce terms that correspond to the difference between the two equations for S.
Tomas should think a little more about the physics of this particular case. Navier-Stokes cannot be the explanation.
Pekka,
I noticed another contradiction. As you’ve noted, our version of 34 follows by simple algebra from 32/3. But could they still both be right?
Only if 1/N ∂N/∂z = 1/N_d ∂N_d/∂z. This is equivalent to
∂/∂z (N_d/N) = 0.
ie N_d/N is constant in the vertical. That’s mole fraction of dry air. Then mole fraction of water vapor is also constant. Specific humidity is constant in the vartical.
But air starts out saturated at the bottom, and then it gets colder…
Nick Stokes | February 8, 2013 at 9:48 am |
Nick, imagine what a splash you would have made if you showed people here that a similar contradiction follows from (32)-(33) and OUR (34). Finding contradictions following from YOUR version of 34 is a form of self-criticism.
Nick,
First I remind that results obtained from the equations of the paper are empirically totally wrong, and that the assumptions related to horizontal dependence are contradictory.
I come back to the role of different equations.
1) Basic thermodynamics tells dependencies like that between pressure and temperature
2) Clausius-Clapeyron tells the amount of condensation. It’s the equation that tells the size of the source. It’s not possible to determine that without it.
3) Continuity equation tells, how the source term affects masses or numbers of moles of each substance. It’s the connection between these quantities.
4) Navier-Stokes tells about effects of viscosity. Without viscosity we have Euler’s equation.
Both Clausius-Clapeyron and Navier-Stokes add new parameters to the formulas. These parameters are external for the theory discussed in the paper. We don’t see those parameters. Thus these equations are not included. Setting viscosity to zero leads to Euler’s equation. That’s the only additional equation that could be considered, when new parameters don’t enter, but that doesn’t solve the problems listed at the top.
Anastassia,
“Nick, imagine what a splash you would have made if you showed people here that a similar contradiction follows from (32)-(33) and OUR (34).”
That is exactly what I have done. For our form follows from simple algebra from 32/33.
Actually, Nick, where YOUR (34) follows from, does not matter for OURS. Take (32)-(33) and OUR (34) and produce a contradiction. Is it so difficult?
That is exactly what I have done. For our form follows by simple algebraic manipulation from 32/33. And then there is a contradiction with your 34.
Nick Stokes | February 8, 2013 at 10:43 am |
No, Nick, these are not “simple algebraic operations”. It is an additional physical assumption that u.∇ N = 0 (where u is horizontal velocity). Under this additional assumption our star result -u.∇ N = S gives S = 0. It is not a contradiction, but a prediction for the descending branch of condensation-induced circulation. I discussed it several times, see, e.g., here.
Nick,
One more comment. They assume that T and N_v do not depend on x, and derive the result that pressure does depend on x. That means that N and N_d do depend on x by equation of state but their difference does not. Thus their ratio does depend on x.
Further we see that the pressure must fall with different rates at different values of x (the alternative would be a wall of equal pressure gradient at all altitudes, which is clearly not a physical possibility for several reasons including the decrease of vapor partial pressure by altitude). That means that also the temperature must fall with different rate at different values of x. Thus it can be independent on x at one altitude only and that altitude is likely to be the surface as a boundary condition. (Or it could alternate, if the sign of the difference would alternate, but even so the temperature would depend an x at almost all altitudes).
It’s as contradictory as before.
Pekka,said, “In this case that other mechanism is the condensation described by thermodynamic equations like the Clausius-Clapeyron equation.”
Condensation is one of the mechanisms and C-C does not deal well with super saturation. Moles of H2O per volume, deals well with supersaturation.
Capt.
If we are discussing the paper we can notice that they write in connection to the equation (34):
.. Assuming that vapor is saturated at isothermal surface ..
They discuss explicitly the case of no supersaturation.
If you discuss that point more generally, then it’s true that some error is introduced when supersaturation is not taken into account, but the level of supersaturation is not normally so high that the error would be large.
@cd: C-C does not deal well with super saturation.
That’s like saying Prohibition did not deal well with Al Capone. C-C is the RH=100% line, with liquid on the left and gas on the right. It’s merely a law, it doesn’t describe reality. The illiquid crossing of that line to the left is what defines super-saturation. Think of condensation nuclei as cops. If there aren’t enough cops Capone can get away with it.
Concerning the additional point
That’s just one of points that is contradictory. The formulas are used to define horizontal derivatives. The continuity equations couple strongly the horizontal and vertical derivatives. That’s the whole nature of them. What changes vertically must be compensated horizontally to maintain the validity of the equation. When a horizontal pressure derivative is introduced, it’s unavoidable that the horizontal dependence of the temperature changes. Here an assumption is used to derive a result that’s contradictory with the assumption. What could be more incorrect.
What you describe as plausible is contradictory and thus must be wrong.
There are clearly two errors involved in (34). One is mixing N and N_d, another is in making assumptions about independence on x. Both are just errors, nothing more.
Tomas,
“But I will add that a dependence of the condensation rate on the vertical velocity component and on the humidity gradient doesn’t strike me as specially contradicting anything I know. It even sounds rather plausible.”
Well, maybe. But where do you see that in this paper?
The star result is now said to be:
-v.∇ p = SRT.
“It says that condensation rate S determines the horizontal pressure gradient provided a horizontal velocity component parallel to the pressure gradient is known.”
My complaint there has been that on the left side, nothing implies the presence of water. Rain out of dry air.
But the bizarre thing about that form is that the hydrostatic assumption says
∇ p = -g
So now precipitation rate is proportional simply to the vertical velocity (again, dry or not).
∇ p = -ρ g
Nick, before you criticize our star result, take notice of what it actually is.
It is not -v.∇ p = SRT. It is -u.∇ p = SRT, where u is horizontal velocity.
The hydrostatic assumption is not ∇ p = -ρg. It is ∂p\∂z = -ρg.
Anastassia,
You do have the opportunity of telling us, why you have written exactly that formula for equation (34). That would give some more basis for discussing it. I have proposed a plausible history of that. If that’s not correct, then why don’t you tell the correct history. Such formulas do not come from nowhere and they don’t come from words only, there must be a history for the formula.
Anastassia,
Now i notice that you tell that you describe your position as having made an assumption on condensation derived circulation including the descending flow. For this process you get -u.∇p = SRT. In vertically ascending part of the circulation you have no condensation, it happens only, when you have horizontal velocity.
That’s, of course, possible only, if there are no places where saturated air ascends vertically, because condensation can most certainly not be prevented in such ascending flow. You tell that vertically rising flows are forbidden.
On the other hand in the subsiding flow, the temperature is rising and no condensation is possible. Thus no horizontal pressure gradients are allowed in the subsiding flow. Thus the saturated ascending flows are always non-vertical and have always a horizontal pressure gradient, but all your subsiding flows are without any horizontal pressure gradients.
It would be interesting to get a description of the circulation that has such strange properties.
No – it’s still as contradictory as ever.
Pekka
You are still trying to wiggle to escape some very specific claims I made.
This can either be because you don’t understand what I am saying or because you are trying to obfuscate.
I prefer to believe the former.
In my first post I have written a system of equations and made 3 claims.
Do you agree both with the equations and the claims?
If no then I will show why you are wrong.
If yes then in the second post I explained that the Anastassia’s 33,34,32 were just a particular subset of the equations I wrote.
In this case I hope that you will agree that a proposition cannot be simultaneously true and wrong.
Btw what you wrote about the “horizontal” and “vertical” derivatives is mathematically not even wrong.
Tomas,
Three points have not been contested properly by you or anybody else.
The equation (34) has not been justified as an independent equation.
The assumptions concerning dependence on x are contradictory.
The set of equations (32), (33), and (34) lead to results that contradict essentially well verified physics.
The first is most difficult to make clear to everybody, but the two others should be very obvious. I have explained them several times. Thus I don’t repeat the details any further.
I would add that it incorporates the assumption that condensation rate is proportional to specific humidity. I haven’t been able to get anyone to affirm that one.
@TM: This can either be because you don’t understand what I am saying or because you are trying to obfuscate.
Tomas, if Pekka were trying to obfuscate he’d have written twice as many dels as you did.
Ding dong del,
Nobody can tell
When their brain is all a whirl
Cause of grad, div and curl.
Quote of the day:
Pekka Pirilä | February 8, 2013 at 6:43 am |
Anastassia Makarieva,
With such a lovely name, I don’t see how you could be wrong. If you get tired of the critics, you could go into tennis :)
Pekka Pirila? Alliterative, but not so lyrical. Finn? Maybe that explains the hostility.
@DM: Maybe that explains the hostility.
No one would ever call DM hostile. It goes without saying.
I used to think the Koch brothers paid him to do it, but lately I’m starting to think he pays them to let him keep doing it. I picture Steve McIntyre holding his face in his hands every time DM posts yet another of his inanities.
DM is god’s gift to warmistas.
Why do you worry so much about what I do, doc?
You are not being collegial to your fellow scientists here, doc. And your little sniffing, yapping sidekick willie is not scolding you for it. Where is that little hypocritical varmint? Somebody step on him?
I did comment on that myself.
There seems to be strange behavior of eqns (4, 5 & 6).
If the right-hand sides of (4) and (5) are equated, the limit gamma = 0 leads to dV/V = 0, which is unphysical. If the right hand sides of either (4) and (6) or (5) and (6) are equated, the limit gamma = 0 leads to DV/V not = 0.
The equations appear to be consistent only if all quantities are invariant as gamma goes to zero. Should not solutions smoothly approach those for dry air as gamma goes to zero?
Have I erred?
Pat,
I think there are algebraic errors there. I think the RHS denominators of 4,5,6 should be:
4. Not 1+μγξ^2 but 1+γξ^2
5. Not 1+γξ but μ+γξ
6. Not 1-μ+μγξ(ξ-1) but ξγ(1-ξ)+μ-1
I don’t know if it affects anything.
In that last denominator, I reversed the overall sign – should be
1-μ+ξγ(ξ-1)
Nick, thanks, that looks better.
I’m late to the conversation and haven’t read the entire thread, so I apologize if what I say here is redundant.
As I understand the situation, Nick, Pekka, et al. regard eqn. (34) as an approximation to the continuity eqn. in which certain horizontal gradients are taken to be much smaller than corresponding vertical gradients. In that case it is not an independent eqn. and adds no new information. No, no says Anastassia, eqn. (34) is an independent relation for S, justified heuristically on the basis of physical arguments. In that case, the vanishing of horizontal gradients is a consequence of eqn. (34), and eqn. (12) for the horizontal pressure gradient follows.
But one can derive an expression for the horizontal pressure gradient from the system eqns (4, 5 & 6) (with Nick’s corrections) without introducing any new hypotheses about S, which is, after all, already determined in terms of other state variables. The resulting eqn. gives the horizontal pressure gradient in terms of p, T, their vertical gradients, gamma and the other fixed parameters. It bears little resemblance to the simple expression given by eqn. (12); a hasty evaluation indicates that the two expressions cannot be generally reconciled. (I have house guests and cannot spend any more time on this!). If Anastassia can show that the two expressions are equivalent, eqn. (34) is justified; otherwise, her hypothesis is invalidated. I expect the latter.
Pat,
Those equations are exactly what should be added to the set of equations. They represent the source term that feeds the flows trough the continuity equation. The problem of the paper is that it attempts to close the set of equations without the introduction of the physics described by these equations.
Yes, exactly.
I had somehow overlooked the first sentences of the chapter 4.2. There the paper confirms that the equation (42) is just another continuity equation. For me at least that’s told in no uncertain way.
Thus they do actually confirm in the paper itself that the argument of Nick and me is correct. They reintroduce the same equation in a little approximate form and derive all their results from the contradiction between the accurate and approximate continuity equation.
They are using Nv as a non-condensable gas If water vapor were non-condensable and well mixed, how would that impact the ideal gas law and the specific heat of a volume of air including H20 and a non condensable. So for that special case as Nv approaches N there would be no condensation, since it H2O is hypothetically non-condensable. They are going out of their way to explain they are trying to use what would be a real gas law as a reference instead of the ideal gas laws, plus an approximation of water vapor based sole on T which produces an estimated saturation pressure that can be super saturated or have a difficult to estimate RH or specific humidity.
What you call a contradiction, is a fairly well known INaccuracy. That INaccuracy, is only a few percent which is the same order of magnitude as the CO2 estimated impact.
You could do about the same thing by using an inert atmosphere, but I imagine that would create even more questions.
Capt.Dallas,
The point is that approximations should not be, the point is they use a set of equations that includes the exact continuity equation as (32) and (33) together with the approximate (34). Requiring that both are true simultaneously leads to the equation that the small error made in an approximation must be exactly zero. Such an equation may produce any results what so ever, and that’s certainly forbidden.
They understand that and therefore Anastassia has repeatedly assured that the equation (34) is not an approximate continuity equation or continuity equation at all, but some independent physics. I have asked her to explain in detail, what it’s supposed to be getting no such answer, only that it’s something else.
Only now I realized that the paper tells in the later chapter, what the equation is supposed to be, and it tells it with the description of an continuity equation without any independent physics.
All the lengthy argumentation that we have had has been unnecessary, the paper confirms that our suspicions have been exactly on point and the counterclaims not true.
@cd: They are using Nv as a non-condensable gas
How is Nv a gas? I’d been reading it as the molar density of water vapor, which when I was a student in the 1960’s was 18 grams per mole, regardless of whether the water was a gas, a liquid, or a solid. Did I miss a memo in the intervening half century?
Vaughan., “How is Nv a gas? I’d been reading it as the molar density of water vapor, which when I was a student in the 1960′s was 18 grams per mole, regardless of whether the water was a gas, a liquid, or a solid. Did I miss a memo in the intervening half century?”
Poor typing on my part. They are using Nv as if H2O were a non-condensable gas. What kind of velocity would the release of an 18 grams per mole gas create trying to become well mixed in 35 gram per mole volume?
Pekka, I think their point is that the derivation is not from the same continuity equation. I would have done it different, then that is a luxury in thermo where you can choose you own frame of reference.
I would do something like use Boyles law of partial pressures and say that the partial pressure of H2O can be considered independently if the remaining volume is not changed. Then the difference in H2O partial pressure in two volumes would be the diffusional pressure differential.
Then I would use that to make something work and not worry about convincing folks why.
Capt.
I thought that I answered already earlier, but the answer is not visible. Perhaps I have exceeded my quota.
They consider a gas mixture of two components, dry air and water vapor. For that two continuity equations can be given. Liquid water is taken into account by the source term. Adding to that a third continuity equation without an additional gas component is either superfluous or a serious error. It’s superfluous, if the equations are not independent and the third equation adds nothing, in all other cases it’s a serious error. That’s why they have insisted that it’s not a continuity equation, but now I see that they tell in the next chapter 4.2 that it’s, indeed, a third continuity equation as I have thought throughout this argumentation.
Thus every derivation based on the set (32), (33), and (34) is just wrong.
Pekka, what is a continuity equation? It is an equation that describes the transport of a conserved quantity. Equation 34 describes the conservation of water as a non-condensable gas. That is being compared to conservation of what in 32 and what in 33? So all 34 is doing is defining a different frame of reference.
Since the Pws at 300K and sea level is considerably different than Pws at 200K and sea level or 200K at 300 mb, what would the difference be is H2O were a well mixed gas? That is all there is to it.
I do the same sort of thing with my static model. The static model determines limits or potentials. If I wanted to get fancy, I would say they are defining a Herbert space or a Chuckie Cheese space then I can use Gibbs, Helmholtz or Chuckie’s free energy to determine dissipation. They, the author’s, are just defining a moist air envelope and meeting just as much resistance as I have.
Right,
The only conserved quantities present in these equations are the amounts of two gas components. That allows for two continuity equations.
Pekka, “That allows for two continuity equations.”
different directions. Continuity in Z and Continuity in X, which really should be the x-y plane. Z is not a problem, the X requires more detail I think since they don’t consider the impact of the ratio of x to y. They could use x as a radius, but since that is not all that versatile, something with a bit more substance would be nice.
Captn,
Number of moles or mass has no directions. They are scalars. Scalar quantities have one continuity equation, conserved vector quantities like momentum have three.
What’s coming from above and below may go to all sides. All this must be considered together, not each direction separately.
Pekka, “What’s coming from above and below may go to all sides. All this must be considered together, not each direction separately.”
Generally. Equation 34 though is describing a space or volume and the isothermal and saturated in x and/or x-y plane, a section of that volume. When I use my moist air envelope, is use dry air as the volume and an envelope at -1.9C instead of a plane surface. More fun :)
Capt.
It’s described using expressions that fit exactly with continuity equation. The description does not hint to anything else.
The first term is the prototype vertical part of the continuity equation, when no corrections are needed to that. The reason for adding the second term is explained in the text. That is a description of one correction that is needed to the basic term. The horizontal terms have been excluded by assumptions presented below the equation. Everything is in full agreement with a description of a continuity equation when the horizontal contributions are assumed zero and when a little lacking care is applied for the vertical part (the second term is only approximately correct).
Pekka, ” Everything is in full agreement with a description of a continuity equation when the horizontal contributions are assumed zero and when a little lacking care is applied for the vertical part (the second term is only approximately correct).”
Yep. Like I said that needs a bit of work. Since 34 though is describing the space, that is fine in my opinion, Z is fine, any issue would emerge.
Now if they are assuming the issues in x and/or x-y are emergent, then they should have clarified that. So there is a bit of confusion. She is though Female and Russian, which is likely a bit more enigmatic than most.
Has no one else noticed that a pressure change within an air parcel has an opposite sign pressure effect at the surface below ?
Thus:
i) If pressure within an air parcel falls as is proposed here then of course the contents do contract to occupy a smaller space but that increases density and weight which increases pressure at the surface below.
ii) If pressure within an air parcel rises then the contents expand to occupy a larger space but that decreases density and weight which reduces pressure at the surface below.
This paper and everyone else commenting here seem to think that a decrease in local pressure causing contraction somehow reduces surface pressure which is not the case.
It all depends.
If a parcel of air loses pressure due to its air escaping to an adjoining parcel, then it makes sense that pressure will increase in the latter.
But if it loses pressure in some other way then it will suck air from its adjoining parcels to compensate, which will cause them to lose pressure.
It all depends.
Vaughan.
In the paper above it is proposed that shrinking of the parcel containing condensing water vapour reduces surface pressure.
It clearly doesn’t because the original contents become denser and heavier and as you say more mass is introduced from the surrounding areas.
The surrounding areas are effectively the entire rest of the global atmosphere so there is not going to be any significant pressure change outside the original air parcel.
The proposed surface pressure change is the opposite of what actually happens.
I have mentioned this previously but there has been no intelligent response.
Some have suggested that the release of sensible energy during the condensation process should be more than enough to prevent the contraction but such an injection of new sensible energy would re evaporate the condensate again which clearly doesn’t happen.
The proposals in the paper are contrary to observation and common sense.
@SW: I have mentioned this previously but there has been no intelligent response.
Why the past tense in the second half of the sentence?
Evaporation initially reduces the density of the air within a convective cell below the global average and condensation simply restores the density within the cell to the global average so it is trite to suggest that the condensation process additionally reduces density and surface pressure outside the parcel.
If it could do that then the entire water cycle would be in reverse with air rising outside the convective cell and descending within the convective cell which is just plain daft.
The problem with this kind of comparisons is that the alternative can be defined in many different ways.
When we compare a moist case with a dry case we must state the surface temperatures as well as the moisture contents. When other factors are the same, it’s likely that the surface is warmer in case of the dry column than in case of the moist column. That need not be the case, but that’s likely. Usually it’s so much warmer that near the surface the dry air is actually lighter than the cooler moist air. The power needed to warm the dry air is typically much less than that needed for evaporation.
When ascending the dry air cools much faster and low altitudes, at high altitudes the difference is small. The faster cooling at low altitudes makes it possible and even likely that the initial difference is reversed.
To make significant conclusions on these points, the cases must be specified in more detail than just stating that one case is dry and the other moist. Furthermore the further specifications should reflect a comparison of real interest.
I’ve realised why some of my earlier points are wrong or partly so but nonetheless others remain valid.
However I’ll save the corrections and refinements for another day.
We wont wait.
Well, silly us. It turns out Makarieva et al’s theory is true after all. The section on causes of wind in the Wikipedia article on wind says “atmospheric gradients are caused by forest induced water condensation resulting in a positive feedback cycle of forests drawing moist air from the coastline.” The article says this a “new, controversial theory.” Roger that.
Jean Giono-1954.
=============
Very good, kim. Except that instead of planting 100 acorns a day they’re planting two papers a year. Both methods seem to work.
How many trees in the Sahara desert does it take to give birth to a hurricane?
The UNtopia theory (Sahel reforestation and bass pond) is similar the condensation driven winds, only the trees to be planted by the scientists provides a natural surface temperature inversion with cool moist air, the wind bags, er scientists, provide the initial kinetic as they rush to publish.
VP –
Considering all the crap you gracefully put up with following your own recent head-post, it must feel good to be throwing (instead of catching) rotten tomatoes. All the more reason to admire and complement Makarieva, et al for their own grace and civility. And to hold the rotten eggs in reserve.
As my own harshest critic (so I like to think), I’m just as prepared to save a few rotten eggs to throw at my own theories. Which I had to do in connection with my 15-year estimate of Hansen delay, whose testability I greatly overestimated.
It’s a question of whether you attach more value to people’s feelings or the theory. Some manage to find a nice balance there—I’m among those who lean more towards the theory.
VP –
I pretty much agree with being one’s own harshest critic. When attempting to write bullet proof code, for example, it’s best to always presume there are undiscovered bugs and to relentless search and destroy. That way when the testers have the code, one is still convinced of remaining bugs and is nevertheless proven wrong. :)
No we know why there is no wind on mars. Oh wait. there is wind on mars, therefore there are trees on mars.
The concomitant quaternary glaciations on mars suggest response to changes to an external forcing,such as solar obliquity or precession,the concomitant changes to the martian prescape as of earth, to say obliquity would be remarkable coincidence of chance.
Steven Mosher, Funny. Remember that the abstract states:, “The water vapor delivered to the atmosphere via evaporation represents a store of potential energy available to accelerate air and thus drive winds. Our estimates suggest that the global mean power at which this potential energy is released by condensation is around one per cent of the global solar power – this is similar to the known stationary dissipative power of general atmospheric circulation. We conclude that condensation and evaporation merit attention as major, if previously overlooked, factors in driving atmospheric dynamics.”
1% approximately 4Wm-2. As Dave would say, “Write that down.”
capt. that 1% if it exists is not magically added to the system.
Steven, there is no magic nor is there a free lunch. It takes energy to move stuff. That movement inside the ERL and above the ERL has to be considered.
So the “surface” used to estimate radiant impact is constantly in motion. Using a moving target for a frame of reference is a great way to screw up. Pick a more stable reference, life gets simpler.
The simplest “surface” frame of reference is the average energy of the oceans, which happens to be in the ball park of 334.5Wm-2. Since that “surface” only covers 70% of the actual “surface” You have an average net at the ERL surface of 0.7*334.5 or 234 Wm-2. The actual surface that is sandwiched between the 334.5 and the 234 Wm-2 has to move energy to the land and ice regions. That takes work which would have a less than perfect efficiency.
Instead of near ideal radiant shells, you have more like concentric radiant Wiffle balls.
https://lh6.googleusercontent.com/-EJRZz5idMVg/URMRvgO0NHI/AAAAAAAAHG4/ab3ULgzYDVs/s512/willis%2520fix.png
CO2 condenses in the martian winter. Wanna start a fight at a planetary science conference, point out that Mars really does not have an atmosphere in the winter.
By “Martian winter” you presumably mean for the hemisphere currently in winter.
This fight has yet to spill over into either the Wikipedia article or its talk page. Who’s claiming this?
“new, controversial theory.” Roger that.
And links to APCD. They could have said “new peer-reviewed science”
The underlying theory ieThat a redistribution of mass in an adiabatic system becomes energetically favored in a gravitational field( eg velocity increases) is well described in the literature eg Landau–Lifshitz,The application of heuristic arguments to phenomenological equations( of which all fluid equations are)is a limiting constraint on ALL experiments in fluid mechanics ie Do not ask too much from the equations (Gallavotti)
Someone should add to the Wikipedia article that AM et al’s theory fills a much-needed gap.
Certainly condensation halves DALR to create MALR. In doing so it raises the temperature at higher altitudes by an amount equivalent to adding 80 W/m2 (averaged over the Earth’s surface) to the thermal energy of clouds. That’s even more than the 66 W/m2 of net radiation upwards from the surface (390 up minus 324 down according to Kiehl and Trenberth).
AM et al’s theory is that condensation has an additional effect. It sucks (the condensation, that is).
So what is this gap that their theory fills? Nowhere is the need for this gap felt more strongly than in this thread.
It is widely held that chaos can cause hurricanes (and vice versa): a butterfly flaps its wings in the Amazon and destroys New Orleans, and gets away with it.
But far less chaotically the Hadley cells steadily blow north and south depending on the cell, the hemisphere, and (as kim rightly observed a while back) the altitude.
In turn Coriolis forces have the strong secondary effect of bending these primary effects east or west accordingly.
And as a tertiary effect the resulting winds collide with mountains and each other to create even more complexity.
What this new theory is up against here is that any empirical test of its validity is like testing whether feeding pigeons in the park has any impact on world pigeon population. The most you might hope for is to measure an impact on park pigeon population, and even that presents challenges. Maybe pigeons simply find parks more hospitable than other city blocks with or without humans feeding them.
What’s needed right now is not yet deeper analyses of the veils and dels in the paper’s theory but a crystal clear account of the extant theories of wind, in particular Hadley cells and the Coriolis effect, so we have something concrete to look at when judging what gaps remain in those theories that this “new, controversial theory” might actually fill.
Volunteers?
Vaughan, “In doing so it raises the temperature at higher altitudes by an amount equivalent to adding 80 W/m2 (averaged over the Earth’s surface) to the thermal energy of clouds. That’s even more than the 66 W/m2 of net radiation upwards from the surface (390 up minus 324 down according to Kiehl and Trenberth).”
Closer to 88Wm-2 latent and 24 Wm-2 related to latent surface cooling with an average surface energy of 400Wm-2 +/- 17 and a “DWLR” of 334 to 345 Wm-2 including 18Wm-2 of “surface” window energy that interacts with the clouds instead of beelining into space. K&T had a “minor adjustment” :)
Figure 7 of K&T 1997 showed 78 W/m2 for “evapotranspiration”. In his 2009 paper “An imperative for climate change planning: tracking Earth’s global energy” Trenberth ups that from 78 to 80 leaving the rest largely unchanged except for making sure there was about 1 W/m2 of disequilibrium to represent global warming. (The 1997 paper was in perfect equilibrium!)
What’s your source for these numbers?
(One complaint I had about the 1997 paper was that it completely neglected the cooling effect that evaporation has on rain, which is strongest at the onset of a rain storm because the humidity is lowest then. Scienceofdoom disagreed with me about that on the ground that the hydrological cycle is closed, but it seemed obvious to me that if the rain is cooler at the beginning of a storm than in the middle it will cool the ground, which K&T don’t take into account. From that perspective I’d be fine with 88 W/m2 instead of 80 if that’s what this additional effect turned out to be.)
Some of the K&T numbers are certainly contested.
The most accurately known part of them is that of TOA energy balance. There we have only radiation, for net SW down and LW up. even these values are uncertain at the level of a few W/m^2 when only empirical measurements are used. Adding GCM level models, the inaccuracies can be reduced somewhat as that helps in using additional satellite observations to reduce uncertainties. Without the model these additional measurements would not have sufficient coverage.
Some of the parts of the surface energy balance are much less known. A recent paper by Stevens and Schwartz Observing and Modeling Earth’s Energy Flows tells about the uncertainties and deviations from the numbers of K&T.
One of the worst known numbers and one that’s badly wrong in the K&T papers is the direct radiation from the surface to space. For most purposes this number is not important as it doesn’t really matter much, whether the radiation goes to space directly or by being absorbed and reemitted, while the OLR at TOA is kept unchanged.
Vaughan, “The 1997 paper was in perfect equilibrium!”
Nope, the 1997 paper was “closed” into a perfect equilibrium. I noticed the missing cloud interaction and dug out the mistake. Luckily, non-rednecks also noticed the error,
http://judithcurry.com/2012/11/05/uncertainty-in-observations-of-the-earths-energy-balance/
Pekka, “For most purposes this number is not important as it doesn’t really matter much, whether the radiation goes to space directly or by being absorbed and reemitted, while the OLR at TOA is kept unchanged.”
Well, it is only important if you wish to determine sensitivity at the real surface or determine what is the real surface :)
Other than that, what is 20 Wm-2 between friends.
BTW, with the atmospheric window split between two “surfaces” The “surface” is a lens of moist air.
VP –
You have nominated the winds paper as describing an effect possibly stronger than butterflies. And you are looking for volunteers to promote your idea. Geez.
No need to throw rotten eggs when rotten tomatoes will suffice.
@bi2hc: you are looking for volunteers to promote your idea.
I truly wish “the extant theories of wind, in particular Hadley cells and the Coriolis effect” had been “my idea,” bi2hc. I am therefore humbled by your generous suggestion that this could have been even a remote possibility.
In the hope of getting any credibility whatsoever here, let me at least take a shot at explaining the Coriolis effect in terms that even Max should be ok with. That is, I’m volunteering to volunteer.
We all know that hot air rises, right? Look at hot air balloons.
The primary effect of the two tropical Hadley cells (one for each hemisphere) is for the rising hot air at the equator to suck surface air from the higher latitudes (north and south) along the surface towards the equator, pump it vertically at the equator, and at a suitable height push it polewards, one pole per cell, up where the jet planes fly.
In each of these two cells it descends at 30 degrees from the equator, about 6.5 degrees polewards of the Tropic of Cancer and the Tropic of Capricorn respectively. (Because of condensation during the equatorial ascent, which hot wet tropical monsoons tend to dump on New Guinea, Bangkok, etc., the air descending at 30 degrees ends up being pretty dry, but that’s not super-important here.)
Closer to the poles, each hemisphere has two more (counter-rotating) Hadley cells, at 30-60 and 60-90 degrees, but they’re irrelevant to the sequel so we can forget about them.
Ok, so what’s Coriolis force then?
Coriolis force is a secondary effect resulting from this motion in the two tropical Hadley cells. It is standardly described mathematically in terms of the curl or ∇☓ operator.
But you don’t need to understand it that way.
At any given latitude the whole atmosphere at that latitude is traveling in a circle around the Earth’s axis.
At higher latitudes (i.e. further from the equator) the radius of that circle is smaller.
Hence the atmosphere is moving more slowly at higher latitudes.
Hence the ordinary momentum of any given parcel at any given instant is less at higher latitudes.
Which can also be expressed as saying that the angular momentum of a whole circle of atmosphere at that latitude is less. These are equivalent points of view provided the atmosphere resists centripetal force by continuing to travel in a circle.
Now consider what happens to any given stationary parcel of air near the surface that the tropical Hadley cell sucks towards the equator.
If it is to stay on the same meridian (longitude) as it loses latitude, it must gain velocity and hence momentum because the equator is traveling faster (in meters per second) than higher latitudes.
But for any given parcel that would violate conservation of momentum. (I’m speaking here of ordinary momentum; angular momentum is simply the sum over all parcels of air of their respective ordinary momenta when constrained to move in a circle.)
So instead the surface air that is moving towards the equator starts to fall behind the rotation of the Earth.
Since the Earth is rotating from west to east, for the surface air to fall behind means that is must start traveling west relative to the rotating Earth.
Which is to say, blowing from the east, the nautical convention in referring to wind direction. (This is because when you’re sitting in the crow’s nest looking for action at sea, disturbances in the sea’s surface in the east means that a wind is coming towards you from the east and so you naturally call it an easterly, having slept through Linear Algebra 101 where they draw the arrows pointing west.)
In the Northern Hemisphere (NH) this wind started out as northerly, which the Coriolis force bent into northeasterly. In the SH it is bent into a southeasterly.
This is exactly what the Coriolis force is all about. If you’re lying on a spinning disc in an amusement park and you sit up and your inner ear complains that this was a very bad idea, what I wrote above describes exactly what happened to your inner ear.
Hopefully that explanation of the Coriolis force gets me at least epsilon credibility.
While there is way more to wind than that, there is no more to Coriolis force than that. Unless you like algebra, that is, in which case there are lots of useful equations for it.
Thanks VP even I could follow your explanation. You said, however, that Earth’s rotation from west to east affects wind speed (and direction) in both hemispheres, especially near the equator.
I was under the impression that gravity keep the atmosphere and stratosphere pretty close to the pace of Earth’s rotation and that airlines, for example, would not expect to gain much traction from this rotation when flying from east to west.
Peter,
The idea is that the velocity at some longitude at the equator is higher than at the same longitude at higher latitudes. Someone at the equator has to travel further in a day than someone in Melbourne. Because of this anything moving north or south is deflected east or west. Ice bergs from the Arctic decscribe a spiral as they move south, cyclones spin up because of this, toilets have a clockwise vortex in the south and anti-clockwise in the north. Naw – only kidding about the toilets. .
Cheers
Vaughan,
Science of Doom is a site, where the emphasis is on trying to understand and explain various atmospheric processes at a level accessible to as many and possible without excessive simplifications. SoD has had several posts on winds and circulation as well. I like the concept of the site very much, which has led also to participation in the discussion trying to present my own ideas on, how we could get as close as possible to the best scientific understanding, present the arguments in a more and more widely understandable way, and answer also the questions that arise for a variety of reasons.
The site is not for a quick look by someone, who has a weak background. Having a grasp of physics is certainly required very often and some effort may also be required from the readers to get the best of the posts and discussion. I do think that the site is more accessible than textbooks that cover the same issues, certainly it’s more interactive through discussion. It goes sometimes a bit deeper on specific issues like radiative calculations in a recent series of posts. On the other hand it’s certainly not as systematic and balanced in coverage as a good textbook.
@PD: airlines, for example, would not expect to gain much traction from this rotation when flying from east to west.
Actually they might if they flew under the radar. It’s unclear who or what would scream louder, the jets flying right over the soccer fields or the soccer moms right underneath. However the proximity to the ground (much drag between the wind and the ground) would greatly reduce this effect compared to the efficacy of the jet stream at altitude where there’s much less drag.
What you’re overlooking here is where I wrote:
at a suitable height push it polewards, one pole per cell, up where the jet planes fly.
Since this is the opposite direction, what you have up there in the tropical jet stream is excess momentum as the air flows polewards, therefore carrying the wind east instead of west. This is the dominant effect with plane travel in general since planes fly at about the tropopause for maximum engine cooling. (Any higher and they start to suffer stratospheric heating due to the vertical temperature profile reversing at the tropopause—the lapse rate changes sign and the air gets warmer instead of colder with altitude.)
This jet stream seems to have enough momentum to penetrate into the temperate 30-60 degree zone. I don’t have a clear picture of what happens to planes flying between San Francisco and Moscow for example, which fly over Iceland and therefore might be expected to experience the opposite jet stream in the 30-60 or Ferrel cell. However the Ferrel cell rotates the “wrong way” which tends to make its upper portion too vaguely defined to have much impact, so basically the tropical Hadley cell prevails in general.
Vaughan,
Picking up the discussion of Coriolis force from your comment. That is also an issue that has been discussed at SoD. There and often elsewhere more emphasis is given to the high altitude winds that blow off from the equator in the Hadley cell and feed the jet stream from west to east.
In this connection we had a discussion on the Coriolis effect and of using the concept of angular momentum conservation in that connection. Due to the additional forces related to stationary pressure gradients the effect is stronger than what one would get simply from conservation of linear momentum. I wrote first a couple of badly formulated comments in the discussion thread, but perhaps that led to a more comprehensive explanation of the phenomena in later comments.
Vaughan,
The prevailing winds are taken into account in optimizing routes of the jets. The latitude of choice varies by season and also on shorter term. The changes can be substantial.
http://en.wikipedia.org/wiki/North_Atlantic_Tracks
In particular the route from east to west may differ substantially from that from west to east.
Pekka, I agree with everything you say about SoD. My thought about a failure of closure in the hydrological cycle, namely rain evaporation cooling the ground during the start of a rainstorm, was too much even for SoD.
As John S would put it, I’m willing to take full possession of this heretical thought, without however getting exercised about negative reactions. You might call me passive-heretical.
@PP: There and often elsewhere more emphasis is given to the high altitude winds that blow off from the equator in the Hadley cell and feed the jet stream from west to east.
It depends on whether wind speed or wind impact counts more. One can expect higher wind speed at altitude due to less resistance from the surface. Which is fortunate as it would be disastrous to have 100-200 mph jet streams going through your back yard every day!
However those who are recording wind speeds with ground based anemometers (like me in my house for example) are impacted mainly by what happens at the bottom few km. In my neighborhood, at 37 N, clouds generally move west to east, though my anemometer shows much more variability despite my efforts at mounting it up high well clear of the house.
Vaughan,
New ideas that are not easily disprovable, but which have not been discussed in textbooks or scientific literature, may be problematic for a site like SoD. If neither he or any of the regular commenters can say anything substantive on the issue, not much can come out of that.
A different site would be needed to discuss in an uncommitted way ideas that are perhaps not very likely but worth some pondering.
I could easily add such a site open to guest posts as a third parallel site on top of the Finnish and English sites of a different profile. There has been discussion of that nature under the title “random topics”, but more structure would be needed to continue along that line. I have, however, doubt’s on the level of activity that such an additional site could maintain. It would probably die off as so many sites hove done.
Vaughan,
The main reason for my reference to the discussion of high altitude winds at SoD, was not that they are high altitude or that their direction is different from the low altitude winds, but the clean way the pressure gradients enter to strengthen the consequences of the Coriolis effect. I.e. pushing the high altitude air flow off from the equator does not lead only to the fact that the air moves faster than surface below but to the fact that it’s absolute speed increases maintaining the same angular momentum around the Earth axis with a smaller radius.
Forcing a jetstream like latitudinal circulation to a higher latitude takes work. That work is transferred to the kinetic energy of the jetstream. This happens only when the latitude changes, not when it’s kept constant, but the properties of the stationary state are due to the same physics.
Thanks VP and Pekka for an interesting dialogue about jet streams and airplanes (presumedly no connection between the two at all) and intuitively I understand how currents and cyclones and wash basin vortexes move anti-clockwise in the NH and clockwise in the SH (hope that’s right?) so it must be true for winds as well.
Pekka mentioned the Science of Doom website as a good source of basic information on science for lay readers and I agree with this. I was also interested in Pekka’s comments on the utility of a site that permits the discussion of new ideas in a non-committed way.
I was rather hoping that Judith’s open threads could allow people to float ideas and discuss them in an unbiased way but it seems that CE has distinct battle lines facing one another re AGW.
Perhaps Judith might see fit to have a regular monthly open thread for intuitive and lateral thinkers (like me) to throw in what their thoughts are on climate/weather phenomena and where some basic research along the lines of what Capn Dallas and some others already do can be put forward without the risk of being flamed.
There is also the problem we have with some of our more passionate and committed souls who would seize every opportunity to drive home their POV’s, at the expense of rational discourse.
VP –
I saw mosher’s Martian tree mockery, the opening double entendre, the wikipedia stuff and somehow misread the last paragraph. Sorry about that.
Yes. Improved info on Hadley Cell and Coriolis info outside of wikipedia is much needed. A link from wikipedia to SoD or Pekka’s site would be great. Unfortunately, a wikipedia effort might become degenerate.
For example, Widening of the tropical belt in a changing climate claims that model predictions cannot explain the rapid expansion of the tropical belt. It calls for a more nuanced understanding of the causes. Finally, the concluding paragraphs catalog the dire implications. No, not a listing of potential policy failures as would be revealed by additional model failures. No mention of benefits in the tropical belt and no mention of desert expansion/compression. Nevertheless, ok, I guess.
However, in the Hadley cell entry, the cited article is the reference for this: “Scientists fear that the ongoing presence of global warming might bring drastic changes to the ecosystems in the deep tropics and that the deserts will become drier and expand.”. Really.
You are a good teacher, doc.
Thanks, Don. You made my day. :)
(Note the difference from here — the colon makes it genuine.)
@bi2hc: and somehow misread the last paragraph. Sorry about that.
Thanks for going back and rereading that last paragraph, bi2hc. I presume you’re referring to my request,
What’s needed right now is not yet deeper analyses of the veils and dels in the paper’s theory but a crystal clear account of the extant theories of wind, in particular Hadley cells and the Coriolis effect, so we have something concrete to look at when judging what gaps remain in those theories that this “new, controversial theory” might actually fill.
Algebra greatly simplifies logic, so much so that when used properly one can get by with much less logic than claimed by logicians. Physics textbooks have way more equals signs than existential quantifiers for exactly this reason.
I love dels when used in that role. In my sophomore year, 1963, we used them to predict the behavior of a ping-pong ball dropped at an arbitrary location on a vinyl record spinning at 33 rpm, with great agreement between theory and observation.
But they can also be used as veils to hide what’s really going on. I would feel much more comfortable if both Anastassia and Tomas would come out from behind their veils of dels and argue about the physics in naive physical terms.
Sometimes they do, but as soon as the going gets tough the tough take cover behind their veils of dels. This strategy wears down those of us trying to understand what’s actually going on physically.
I’m not saying either Anastassia or Tomas is wrong, only that they put way too much faith in algebra as it applies to physical phenomena. Algebra is as effective as dynamite when used appropriately, but like dynamite it is easy to misuse.
Please don’t interpret this as meaning that I’m an algebraphobe. If you’re in any doubt on that score, see the very last line of this article on algebra, which I wrote with an audience of philosophers in mind.
I call them like I see them, doc. I am just as impressed with your abilities, as I am amused by your foibles :) You may have missed it, but I similarly praised your explanation of the GHE, on the recent Berkeley Earth thread. Very elegant and enlightening.
If I were in charge of the world, as I should be, I would reward you for your distinguished career by promoting you from Emeritus to a teaching position at the worst ghetto school I could find, where you would do the most good.
Vaughan Pratt: Sometimes they do, but as soon as the going gets tough the tough take cover behind their veils of dels. This strategy wears down those of us trying to understand what’s actually going on physically.
The French probabilist Levy said: (appx) “The first time a theory is proposed the presentation is incorrect and obscure. Then someone comes along with a proof that is technically correct but impenetrable. [for example, Schwinger’s Nobel prize winning work.] Eventually someone presents a proof that is both understandable and technically correct. [for this you would have Feynman, who shared the Nobel Prize with Schwinger.]”
Another possible example is Einstein who published his general relativity in 6 papers, 5 of them correcting errors in the precursors, and whose work was shown to be accurate before most people took it seriously. Modern quantum mechanics (and its sequels like quantum chromodynamics) is full of mathematical expressions about whom people have diverse and unclear understandings.
If their math is correct (which has been disputed), then the physical understanding can be developed.
VP –
Thanks for your link. Math is beautiful isn’t it? One equation is worth a thousand words. It made me abandon plans for a law career and switch to electrical engineering. I will say it is a hard row to hoe when one’s first exposure to physics at any level is a 3rd quarter calc-prereq while simultaneously enrolled in trigonometry. :) Eventually, I caught on.
As far as Makarieva et al, I find their ideas to be astonishingly beautiful. Unfortunately, I cannot determine the truth and beauty of their math and physics. For that I must rely on others, such as you. Thank you.
And I notice now that Pekka agrees with Tomas that the eqs. 32, 33, & 34 are independent. It’s not over until it’s over.
It was all the time clear, and I have always said that I agree on that.
I’m, however, as certain as ever that they are independent only, because one of them is approximate. If (34) would be exact, it would not be independent. Thus the independence is totally true to a fatal error in the paper.
Pekka –
Yes, I had thought it was N, Stokes that pressed the idea of eq. 34 being derivative. However, I do appreciate your point that the independence is achieved by its inexactedness. Thanks.
I agreed with Nick from the beginning, but I realized only much later that the paper does actually confirm directly in the beginning of chapter (34) that our suspicion was right. They did, indeed write the equation (34) as an approximate continuity equation. As they consider two gas components, only two continuity equations can be presented for the variables N_d, N_v, and N=N_d+N_v but they wrote three. The paper leaves no doubt on that. Thus it’s explicitly wrong. No speculation is needed.
Pekka –
Hope you read this. You and Captain Dallas have carried much of the load in the discussions. Your exchanges have been of the sort that I would pay for while sitting back with my mouth (keyboard) shut and witnessing.
It is exposure to those like you that raise the intelligence of everybody in the room. I imagine you are purposively doing this. But even if I am wrong you nevertheless have my appreciation.
@DM: You may have missed it, but I similarly praised your explanation of the GHE, on the recent Berkeley Earth thread. Very elegant and enlightening.
I did indeed miss that. :( Thanks yet again. :)
If I were in charge of the world, as I should be, I would reward you for your distinguished career by promoting you from Emeritus to a teaching position at the worst ghetto school I could find, where you would do the most good.
If that’s your assessment of the audience for Judith’s blog, Don, then it’s fortunate for you they don’t know where you live. ;)
Not that you’d suffer much: twits only get mildly beaten. If you’d had the intellectual pretensions of a Richard Lindzen they might have gotten more creative.
Pekka and Vaughan,
https://lh4.googleusercontent.com/-DNmIBBQ5v_g/URejjPfgTXI/AAAAAAAAHI8/dGL4Y8HIaNg/s746/Forest%2520or%2520trees.png
Pictures can be useful. If you have ever played with venturis or had to deal with static reqain and velocity pressures, you might have an idea of how much energy can be converted from potential to kinetic in such a situation.
One of my favorite sayings is “energy is fungible, the work done is not.” That dang dS changes on ya.
I see you are all going at it still. I have a question. Sometimes there is an assertion that a statement “is not physical” or “is physical”. Is “adiabatic” physical? As in “adiabatic lapse rate”? Surely a rising air mass dissipates some of its energy in friction, the most dramatic occasion of which is lightning and thunder. Equally noteworthy are the spirals of thermals and tornadoes.
Matt,
I wrote an answer to you here. I add the link here as that’s already about 100 comments up in the thread.
If I say that something is “unphysical” or “not physical”, I mean that it’s totally different from reality on the aspect that’s being discussed. It’s not a little questionable approximation, it must be totally wrong to earn those attributes.
@MM: Is “adiabatic” physical?
A physicist, a meteorologist, and a philosopher walk into a bar. The bartender asks, “Is `adiabatic’ physical?” Five minutes later they’re out on the street fighting like cats.
The physicist says it means “no energy exchange with the environment.” The meteorologist adds the requirement of reversibility. The philosopher faults every definition of “no energy exchange” proposed by the physicist and every definition of “reversible” proposed by the meteorologist.
The short answer is “no,” the long answer is “yes.”
Vaughan Pratt: The short answer is “no,” the long answer is “yes.”
Maybe. I think the idea of “adiabatic” (and congeners) is an approximation whose approximation error is generally not known. If air masses rose adiabatically, there wouldn’t be lightning and thunder. In the totality of atmospheric energy flows, that is not a lot of energy, and the approximation error in the adiabatic assumption probably does not matter much. In understanding the genesis of cyclones (dust devils, cyclones, hurricanes and typhoons) and some other processes, a better approximation is needed.
In this thread, “non-physical” means “different from previous approximations.”
The statistician watching the fight you described says that the observations and analysis have not yet been worked out in sufficient detail to answer the question.
@MM: In understanding the genesis of cyclones (dust devils, cyclones, hurricanes and typhoons) and some other processes, a better approximation is needed.
Sure but understanding hurricanes is not the focus of the present paper, although it was for their previous (2008) submission to ACP, “On the validity of representing hurricanes as Carnot heat engine” which proposes that the decrease in pressure resulting from condensation is a source of energy for hurricanes.
The present paper asks “Where do winds come from?” and seems to be focusing on winds in general. If say 20% or more of total wind energy is in hurricanes then they’d be worth including in the equations. However if 90% or more of the wind energy budget accounts for calmer winds then simply ignoring storms altogether should be an adequate approximation to get started with. At least that’s how I’ve been interpreting the equations of the present paper, and I didn’t see anything in it to contradict that.
Incidentally the wind energy of the atmosphere, with or without hurricanes, must be way less than one percent of the total energy of the atmosphere, which should be about 2.57 GJ/m2, namely 1.82 for kinetic energy (including vibrational) and 0.75 for potential energy, where both are measured relative to the (rotating) Earth’s surface. The relevant calculations can be seen in my reply to Miskolczi. (Had we used the nonrotating surface we’d have had to add another ½ I ω²/A = 0.73 GJ/m2 to KE for the rotational energy of the atmosphere where I is from Sidorenkov, NS and Stekhnovskii, DI, Moment of Inertia of the Earth’s Atmosphere, ω = 2π/86400 sec‾¹, and A = 510 Mm² is the area of the Earth, but no one expects the atmosphere to stop spinning. The dependence of energy on frame of reference gives the philosopher some leverage in arguing with the physicist in the bar.)
Vaughan Pratt: However if 90% or more of the wind energy budget accounts for calmer winds then simply ignoring storms altogether should be an adequate approximation to get started with.
No disagreement here. It is getting past the “getting started” phase where better approximations become necessary. The effect of CO2 is small. It may be completely obliterated by a bunch of other processes that have previously been ignored because they have up til now been *assumed* to be negligible without having been *shown* to be negligible.
@MM: The effect of CO2 is small.
True enough, only about 1 W/m2 for the disequilibrium resulting from CO2 forcing according to Trenberth et al. That’s smaller than the 4 W/m2 of “weather power” AM was attributing to Lorenz (which Pekka probably knows more about than me), or the 9.4 w/m2 of power transported polewards by the Hadley circulation.
It may be completely obliterated by a bunch of other processes that have previously been ignored because they have up til now been *assumed* to be negligible without having been *shown* to be negligible.
The reason the climate skeptics wanted to disembowel me for my “millikelvin” post is that I’d split the low frequency portion of HadCRUT3 into just two easily described components, my cute little “quasisawtooth” (which they really hated) plus a monotonically rising curve. That separation of low-frequency HadCRUT3 into just two components pretty much blew away any hope that a rise well correlated with known CO2 emissions had no significant effect on temperature. That was like telling all religions subscribing to the Nicene Creed that the Trinity is rubbish. The Spanish Inquisition burnt people at the stake for that sort of thing.
Climate skeptics are dedicated to the proposition that the ocean oscillations rose sharply after 1980, accounting for the bulk of the observed rise in HadCRUT3. I’d be fine with that if I could find any theoretical support for it beyond the mere assertion that that’s how it is.
Vaughan Pratt: The reason the climate skeptics wanted to disembowel me for my “millikelvin” post is that I’d split the low frequency portion of HadCRUT3 into just two easily described components, my cute little “quasisawtooth” (which they really hated) plus a monotonically rising curve.
Now we are getting away from my main point, which is that approximations have been assumed to be “accurate enough” without a quantitative estimate of how inaccurate they are or how accurate they need to be.
You know already that I disputed some of the criticisms of your millikelvin work.
@MM: Now we are getting away from my main point, which is that approximations have been assumed to be “accurate enough” without a quantitative estimate of how inaccurate they are or how accurate they need to be.
Your main point if I understood it was “I think the idea of “adiabatic” (and congeners) is an approximation whose approximation error is generally not known.” However your very next sentence was “If air masses rose adiabatically, there wouldn’t be lightning and thunder.” This got us away from your main point immediately because in the context of more than 90% of weather this is a non sequitur.
If you’re willing to stick to “typical weather” one could make a start on estimating the associated approximation error. One way to approach this would be for you to cite some assumed approximation you’re dubious about and challenge it by saying it could well be off by 50% or more. If anyone disagrees let them argue why it should be less. If not then your point would remain unchallenged and hence valid.
You know already that I disputed some of the criticisms of your millikelvin work.
Indeed, and I appreciated that support, many thanks.
Vaughan Pratt: One way to approach this would be for you to cite some assumed approximation you’re dubious about and challenge it by saying it could well be off by 50% or more. If anyone disagrees let them argue why it should be less. If not then your point would remain unchallenged and hence valid.
I already wrote that the square root of the mean square error of the equilibrium approximation to the Earth temperature is 5% (taking the present mean temp as the approximate equilibrium temperature) and the estimated effect of CO2 on the equilibrium temperature is under 1%, if the Earth is even capable of equilibrium, which has not been established.
V.Pratt
I’m not saying either Anastassia or Tomas is wrong, only that they put way too much faith in algebra as it applies to physical phenomena. Algebra is as effective as dynamite when used appropriately, but like dynamite it is easy to misuse.
,
can explain you the question of “faith”.
First it is not about “algebra” but about mathematics and in this particular case about “differential calculus” if one wants to be accurate.
Physics without mathematics is little more than astrology or alchemy.
Every single physical law is expressed mathematically and one of the most impressive pieces of physics ever are Noether’s theorems which show that purely physical laws (conservation of energy and momentum) are equivalent to symmetry properties of the equations.
That’s why I generally ignore any “physical” statement which is not expressed mathematically.
This has a reason – a mathematical statement is either true or wrong and that absolutely (we’ll avoid here a few Gödelian sophistications which would be off topic).
Therefore any physical statement which would happen to be mathematically wrong is sure to be wrong physically too.
The opposite is not true – a true mathematical statement is not necesarily relevant for the physical world despite staying true.
For example Pekka’s statement that “32,33,34 are not mathematically independent equations” is wrong and I have already shown it above.
Follows that all and any “physical” or “mathematical” statements that build on this (wrong) proposition are wrong too and can be ignored.
Indeed I have shown that :
div(Rhoi.V) = Fi(x,z) with i an index 1 or 2 is :
– true
– independent
– admits as a particular case Anastassia’s 32,33,34
– when joined to Navier Stokes closes the system which admits then a unique solution for all fields
Conclusion: it passes the mathematical test because there is neither contradiction nor inconsistence and the equations have a very specific physical meaning – mass conservation.
However as I wrote above this doesn’t mean that it will pass the physical test too. Mathematical truth is a necessary but not sufficient condition.
What I miss for the physical test is simply to solve the system because we know that it is closed and admits a unique solution.
Then by looking at the fields or their derived properties I would say whether they are plausible or not.
At this stage by ignoring the fundamental key to the dynamics, Navier Stokes, I stay agnostic but have a problem with the continuity in 0.
Tomas,
Of course they are mathematically independent. That’s not the question. The question is, do the describe physics that allows for three equations or only two.
@Tomas: Physics without mathematics is little more than astrology or alchemy.
Isn’t that a bit extreme? Suppose mathematics didn’t exist but bicycle pumps and black plates did. Why would it be “astrology” to assert that temperature increases when you pump up a bike tire or leave a black plate out in the sun? A guild of premathematical physicists could find these statements perfectly satisfactory while agreeing that astrology and alchemy are nonsense. Postnuclear physicists still dismiss astrology but could only object to alchemy on grounds such as not being a cost-effective way of turning lead into gold as well as posing serious health and environmental hazards.
@Tomas: purely physical laws (conservation of energy and momentum) are equivalent to symmetry properties of the equations.
Is there (or might there be) a purely physical law governing condensation?
Unless I’ve misunderstood you (always possible), you’re concerned that (34) is not a purely physical law, but that still leaves open the possibility that one exists for condensation.
The paper’s statement “S (Eq. 34) is the sink term describing the non-conservation of the condensable component (water vapor)” would by your definition (again assuming I’ve understood it) constitute an admission that (34) is not “purely physical”.
This is an informative source The general circulation of the atmosphere. Schneider, T. (2006) Annu. Rev. Earth Planet. Sci. 34: 655–88
The dynamic power of global atmospheric circulation (i.e. the rate at which kinetic energy is generated and, in the steady state, dissipated) is about 4 W/m2. This power is generated by large-scale pressure gradients and dissipated via smaller-scale turbulent eddies. Before our “new, controversial” theory came in there has been no theoretical account of why the observed power of atmospheric circulation has this magnitude. Our theory relates this power to the intensity of condensation and produces estimates of circulation power that are in agreement with observations on a variety of spatial scales.
Anastassia,
As far as I can judge you got the efficiency estimate from the equation (40) that’s derived from (27) and thus from the comparison with the isothermal atmosphere. This comparison is totally irrelevant for case of real atmosphere. It tells nothing about the efficiency of the atmosphere.
@AM: This is an informative source The general circulation of the atmosphere. Schneider, T. (2006) Annu. Rev. Earth Planet. Sci. 34: 655–88
It is indeed. Thanks very much for that, Anastassia, I hadn’t seen it before. It’s very well written. It also fills in some gaps in my account above of the interaction of the Coriolis force with the Hadley cells, especially in the Ferrell cell which I’d glossed over for want of the sort of insight in Schneider’s survey. And the fourth and last of its concluding issues, namely
“4. How do theories for the Hadley circulation and for atmospheric macroturbulence based on dry dynamics need to be modified in the presence of moist processes, which alter, among other things, the effective static stability of the atmosphere? What is the structure of water vapor fluxes and of the global distribution of water vapor in the troposphere, given statistics of atmospheric macroturbulence?”
define one gap that your approach could be viewed as intending to fill.
The dynamic power of global atmospheric circulation (i.e. the rate at which kinetic energy is generated and, in the steady state, dissipated) is about 4 W/m2.
I don’t even know what that refers to. Figure 5 of the Schneider paper you kindly referred me to (adapted from Walker & Schneider, 2006) cites a poleward energy flux in Hadley circulations of 4.8 PW or 4.8/0.510 = 9.4 W/m2. I would have thought the mechanism there is very clear: the Hadley cells transport heat from the warmer lower latitudes to the higher colder latitudes as per the second law of thermodynamics. The lower latitudes get their excess heat from the Sun, while the higher latitudes radiate it to space more readily by virtue of being colder.
Is this 4 W/m2 you refer to some fraction of this 9.4 W/m2, or in addition to it, or what?
Condensation actually works against that 9.4 W/m2 flow of heat, by dumping much of the latent heat of condensation back on the tropical surface in the form of monsoonal rains. Do you mean that there should have been 13.4 W/m2 but the condensation has removed 4 W/m2 of it, perhaps?
This leaves relatively dry but still (potentially) warm air to find its way polewards to 30 degrees from the equator, thereafter swinging down (thereby turning the potential temperature back into real warmth by adiabatic compression) and back up at 60 degrees to complete its journey by carrying that warmth to the polar troposphere which radiates it to space (not so effectively since Stefan-Boltzmann doesn’t consider potential temperature to be real temperature at high altitude).
As I wrote in my first comment in this thread, “condensation simply is not a driver.”
Instead condensation comes and goes as the point (p,T) wanders back and forth between the liquid and vapor regions of the phase diagram for water tracking environmental temperature and partial vapor pressure. This “wandering” results not from any actual air movement but merely from our choice of time and position in the atmosphere where we inquire about (p,T) at that time and position. We can pick any such at random, or we can move it smoothly around which tends to be more interesting, especially given that (p,T) is allowed to vary with time while holding the position in the atmosphere fixed!
This wandering is entirely reversible: when there are large droplets to condense on, temperature equilibrates to the value of T for which water molecules are condensing on and evaporating from droplets in equal numbers, just as in any reversible chemical reaction.
Water vapor has an elevated chemical potential which condensation reduces exothermally but entirely reversibly. (This reversibility weakens my meteor and disc brake analogies somewhat: those processes aren’t reversible because meteors and brake pads aren’t reconstituted by heating their fragments.)
Conceivably there could be a Lotka-Volterra type situation on the liquid side of the phase boundary, thinking of the droplets as the foxes and the vapor molecules as the rabbits. The droplets might capture so many molecules as to reduce their population, thereby favoring evaporation over condensation and so going round in the characteristic Lottka-Volterra cycle.
But there is an added complication: these rabbits being highly energetic are vicious. The foxes, lacking the requisite armour, lose a little bit with every rabbit they eat, in that they gain temperature, which is not good for foxes/droplets. Without that added detail foxes and rabbits are simple; with it this situation could be more chaotic.
But we’re talking here about very very small molecules that are way up high up in the sky performing these steps in time on the order of 100 picoseconds (10 GHz for you nerds out there). That’s not exactly the sort of thing your average ABC or NBS meteorologist is going to be thinking about while getting their 5-and-9 makeup on. And pilots who have to fly into this stuff at 500 mph aren’t in a great position to stop and smell the molecules along the way either.
We don’t have a huge amount of experience or intuition with molecular scale events happening at 10 GHz several km over our heads, whence their connection with the various shapes of the possible solutions to any proposed Lotka-Volterra-type equation is going to be pretty speculative, yes?
captdallas2 0.8 +/- 0.2 | February 4, 2013 at 8:33 am | Reply
I missed this comment. The horizontal dimension is of course important. We have delta p determined by surface water vapor pressure which we can “spread” over the horizontal dimension by assuming the atmosphere being vertically in hydrostatic equilibrium. For large X the mean pressure gradient delta p/X will be small and at some value of X the circulation becomes impossible because of friction losses.
Conversely if X is diminished at some value of X friction becomes insignificant and velocities V develop that correspond to full pressure potential delta p = rho V^2/2. This happens in hurricanes where there is an additional effect of spatial concentration of kinetic energy from the entire condensation area and where friction becomes significant near the eye region only.
In the ACP paper we did not estimate X but simply checked if our result is applicable to the observed Hadley cell.
Anastasia, right, but the missing y seems to be creating confusion.
http://redneckphysics.blogspot.com/2013/02/just-puzzle-for-sunny-sunday.html
I made a little puzzle yesterday for anyone curious about how to set up the problem. You can start with several assumptions and end up with roughly the same results making it as simple or complex as you like. For you I imagine comparing the forest special case to a desert case where there would be no local temperature inversion would be a nice start.
As mentioned here, the hypothesis of Makarieva et al. can be tested by comparing their expression for the horizontal pressure gradient (eqn. (12)) with that derived from the set of equations 4, 5 & 6 that determine relations among the state variables p, T, N and p_v, for the adiabatic behavior of saturated air. The derived equation for dp/dx is:
u dp/dx = w [ (p/T) (dT/dz) f – dp/dz ]
The corrected factors given by Nick Stokes have been used. The derivative dT/dx has been set = 0. (Here, all derivatives denote partial derivatives.) The factor f is:
f = ( 1 + γ ξ^2 )/( μ + γ ξ)
All of the quantities within the square brackets are larger than those in Makarieva et al.’s expression by a factor of 1/γ , which indicates that the vertical temperature and pressure gradients dominate the contribution of condensation to the horizontal pressure gradient. In any event, the expression eqn. (12) cannot be reconciled with that derived above. This implies that the disputed eqn. 34, from which eqn. (12) was derived, is not a valid independent relation.
Thus it appears that the hypothesis that horizontal winds are driven mainly by condensation is not justified by the analysis in this paper.
Pat Cassen, “Thus it appears that the hypothesis that horizontal winds are driven mainly by condensation is not justified by the analysis in this paper.”
I think we all agree with that. I think we all agree that condensation and evaporation can enhance horizontal winds, which can in turn can enhance evaporation and condensation. The degree of enhancement would depend on a number of things that would require special cases.
captdallas2 0.8 +/- 0.2 –
Glad to hear about the agreement – I apologize for not reading all the comments.
“The degree of enhancement would depend on a number of things…”
Yes indeed, much is left out of the physics considered in the analyses so far. Besides feedbacks of the kind you mention, one surely cannot ignore changes in radiative transport with changes in γ, as has been done so far (for instance). But the equation derived above suggests that the latent heat effect always dominates condensation-related volume effects; all condensation effects vanish with ξ, even for non-zero γ.
Pat I agree with that too. I must be coming down with something :)
I was looking at their forest hypothesis and considering a rectangular orifice or venturi. I can see how latent heat could be directed which would stimulate condensation and induce surface winds, but like most things in nature, it requires a jump start, possibly the natural temperature inversion between forest floor and canopy. That would be different dynamics than a hurricane though.
@cd: I think we all agree that condensation and evaporation can enhance horizontal winds,
The only mechanism I’ve seen so far for doing so is precipitation resulting from condensing forming droplets sufficiently large as to fall, thereby drag the air down with it. Even though the terminal velocity of the drops might be only 5 m/s, that’s only relative to the air. As the air that’s being dragged downwards accelerates, so do the drops. The drops can thereby fall at up to 35 m/s even though (or because) the air is only 5 m/s behind them in speed.
This pushes the stationary air below to one side (horizontal motion) but there is also suction at the top of the rain column. The pressure at the bottom combined with the suction at the top results in a cylinder of air being drawn up at very high speed around the falling column of air.
I don’t buy any of this stuff about condensation either decreasing or increasing pressure. Condensation can’t have that sort of mechanical effect.
Vaughan, “I don’t buy any of this stuff about condensation either decreasing or increasing pressure. Condensation can’t have that sort of mechanical effect.”
All by its lonesome, no, but a stable condensation “sink” could direct energy and pump fuel into a cycle. Then saturation vapor pressure would tend to indicate an limit of efficiency. It like a chicken egg situation. The number of moles of water vapor would indicate a magnitude, but not a “cause”. Once you get the ball rolling, then you would consider the boundaries to optimize the cycle.
Their “results” could be right for the wrong reason.
a stable condensation “sink” could direct energy and pump fuel into a cycle.
I bet Tomas would be asking for the equation for that.
Come to think of it, so would I. Or at least a reasonably detailed physical account of the sink, the pump, the fuel, and the cycle.
Vaughan, I am playing around with it. If the cloud and cloud base is stationary, I may be able to model the “feed” as a rectangular venturi, which would put a sustainable surface wind at around 6 to 10 times the prevailing (~800 meter “throat” and 6000 meter or so inlet). Then at 30C saturated, you would have a 25 J/g to 40 J/g, “exhaust”. Because of the lower pressure in the “throat”, the “fuel” would be super saturated and pre-cooled. Not something to build in your back yard, but a neat description.
A tropical forest floor is already cooled to near saturation with a built in temperature inversion. The rough lower surface of the throat would make a neat turbulent to laminar transition.
http://www.eldoradocountyweather.com/current/satellite/goeseast-wv.php
Vaughan Pratt, yes condensation drives motion, but the main reason is that latent heating generates buoyancy allowing for conditional instability of certain atmospheric profiles (CAPE) to be converted into kinetic energy in updrafts. This part works even without rainfall. Latent heating generates buoyancy by the thermodynamical processes we have discussed here. Namely, compared to dry air at the same temperature, a lifted saturated air mass condenses water and releases latent heat, and this heating expands the volume (decreases the density) increasing buoyancy relative to the dry air around it that has no such energy source. No one disputes that warm air rises, so it is surprising we have to explain that warming from condensation is no different than warming with a flame in creating expansion and buoyancy relative to the unwarmed air around it.
@Jim D: it is surprising we have to explain that warming from condensation is no different than warming with a flame in creating expansion and buoyancy relative to the unwarmed air around it.
Except that the flame is deriving its energy from an irreversible chemical reaction. In contrast the warming from condensation is entirely reversible, which facilitates equilibrium.
The energy Steve Jobs contributed to Apple was more that of a flame than condensation. You did what Steve said, there was no equilibrium.
Pat Cassen | February 11, 2013 at 6:02 pm | Reply
Pat, thank you for your inputs. All the above has been already discussed by us (the authors) and can be expressed much more laconically, e.g. see Eq. (6) in the post (Eq. (A7) in the paper). It says that
-u.∇N = (S – S_d)/γ_d
while our result is
-u.∇N = S.
As you can see, in agreement with your observation, all the terms in the right hand side of the first equation are larger than those in Makarieva et al.’s expression by a factor of 1/γ (recall that γ ≈ γ_d)
But this does not
Once again, but in a nicer format.
Pat Cassen | February 11, 2013 at 6:02 pm | Reply
Pat, thank you for your inputs. All the above has been already discussed by us (the authors) and can be expressed much more laconically, e.g. see Eq. (6) in the post (Eq. (A7) in the paper). It says that
-u.∇N = (S – S_d)/γ_d
while our result is
-u.∇N = S.
As you can see, in agreement with your observation, all the terms in the right hand side of the first equation are larger than those in Makarieva et al.’s expression by a factor of 1/γ (recall that γ ≈ γ_d)
But this does not
nor that
On the contrary, Eq. (34) put into the above general equation produces our result -u.∇N = S.
This result does show however that, as we emphasized in the post, S must be defined with a high precision (exceeding γ) to correctly describe the dynamics.
Here it is proper to recall that the second law of thermodynamics is an approximate equation that neglects the kinetic energy of the gas (this is what equilibrium thermodynamics is about). In terms of temporal changes, it neglects the rate at which the kinetic energy is produced (i.e. precisely the rate we are concerned about). In simple words, it is not an energy conservation equation.
Thus, information on dynamics cannot be retrieved from this equation which for that reason is never used in circulation models in its classical form. In reality, in current models, again as we discussed in the blog post with examples, one adds empirically fitted parameters into this equation such that the dynamics produced by the entire system of equations matches observations (including the observed u.∇p). In such an approach no independent theoretical specification of S is at all needed. One can just fit whatever one wants by minor modifications of parameters for which no independent stipulations exist.
If we however turn to theory, all the second law of thermodynamics is valid for is to determine the scale of the vertical temperature gradient (and this is how we used it).
Anastassia –
Thanks for your response. However I do not see how it addresses the issue I raised. Either I am missing your point, or you are missing mine.
I will try to clarify here, and then let you have the last word.
You have neatly summarized the thermodynamic constraints in the set of equations 4, 5 & 6. These equations do not include the purely dynamical constraints of continuity or hydrostatics, nor do they invoke the Second Law, but nevertheless define certain relations that must hold between the state variables. From these equations, which are exact in the present context, on can derive an expression for the horizontal pressure gradient:
u ∂p/∂x = w [ (p/T) (∂T/dz) f – ∂p/dz ]
where
f = ( 1 + γ ξ^2 )/( μ + γ ξ)
As previously stated, the corrected factors given by Nick Stokes have been used, and the derivative ∂T/∂x has been set = 0. This expression holds for the steady state, adiabatic motion of saturated air, if the temperature is constant on horizontal planes. It is to be compared with the expression that you have derived:
u ∂p/∂x = w [ ∂p_v/∂z – (p_v/p)(∂p/∂z)
This equation relies on your equation (34), which is introduced either as an independent physical constraint or as an approximation to a full continuity equation; at this point, I am not concerned with which. The point is that your expression for ∂p/∂x is incompatible with that derived from the exact thermodynamic constraints, i.e., the first equation in this comment. That this is the case is readily apparent in the limit γ → 0, for which your pressure gradient vanishes, but that given in the first equation goes to a ‘dry’ limit.
From these considerations, I conclude that your prescription for S, whether derived from an independent hypothesis or an approximation, is invalid.
Furthermore, it can be seen from the thermodynamic constraints embodied in the first equation that the effects of condensation on the horizontal pressure gradient (in the particular case under consideration) are always tied to the latent heat term ξ, not the volume change term γ alone; that is, γ is always multiplied by ξ. Thus I remain unconvinced that volume changes due to condensation are more important than the latent heat effects.
Anastassia –
Thanks for your response. However I do not see how it addresses the issue I raised. Either I am missing your point, or you are missing mine.
I will try to clarify here, and then let you have the last word.
You have neatly summarized the thermodynamic constraints in the set of equations 4, 5 & 6. These equations do not include the purely dynamical constraints of continuity or hydrostatics, nor do they invoke the Second Law, but nevertheless define certain relations that must hold between the state variables. From these equations, which are exact in the present context, on can derive an expression for the horizontal pressure gradient:
u ∂p/∂x = w [ (p/T) (∂T/dz) f – ∂p/dz ]
where
f = ( 1 + γ ξ^2 )/( μ + γ ξ)
As previously stated, the corrected factors given by Nick Stokes have been used, and the derivative ∂T/∂x has been set = 0. This expression holds for the steady state, adiabatic motion of saturated air, if the temperature is constant on horizontal planes. It is to be compared with the expression that you have derived:
u ∂p/∂x = w [ ∂p_v/∂z – (p_v/p)(∂p/∂z)
This equation relies on your equation (34), which is introduced either as an independent physical constraint or as an approximation to a full continuity equation; at this point, I am not concerned with which. The point is that your expression for ∂p/∂x is incompatible with that derived from the exact thermodynamic constraints, i.e., the first equation in this comment. That this is the case is readily apparent in the limit γ → 0, for which your pressure gradient vanishes, but that given in the first equation goes to a ‘dry’ limit.
From these considerations, I conclude that your prescription for S, whether derived from an independent hypothesis or an approximation, is invalid.
Furthermore, it can be seen from the thermodynamic constraints embodied in the first equation that the effects of condensation on the horizontal pressure gradient (in the particular case under consideration) are always tied to the latent heat term ξ, not the volume change term γ alone; that is, γ is always multiplied by ξ. Thus I remain unconvinced that volume changes due to condensation are more important than the latent heat effects.
Pat Cassen | February 12, 2013 at 1:36 pm | Reply
I replied below, but I’d like to discuss this separately:
You are right that “γ is always multiplied by ξ”, but once again, it is for equilibrium thermodynamics. In equilibrium thermodynamics there is no pressure rise because of the alleged latent heat “warming” — because latent heat is released only when the temperature drops. But this is not the point here.
The point is that when we turn to the real world, we immediately notice that the fate of heat and number of molecules are fundamentally different. E.g. if the ascent is very slow, most latent heat can be radiated to space producing a non-adiabatic lapse rate. But you will agree that this radiative transfer process will not impact the change in the number of molecules that occurred upon condensation. So there will be zero latent heat, but the change in γ will be there.
Or even simpler. Consider two air columns, one warm and another cold, one with little gas and another with a lot of gas. The temperature difference between the columns can equilibrate without any work performed, just at the expense of heat transfer. Meanwhile in order to equate the number of molecules in the two columns and to add some molecules where there is a deficit, you must perform some work against the existing gas pressure. It is possible to change temperature without performing work. But it is not possible to change the number of molecules without performing work, i.e. without applying a force. In our paper we quantify this effect for the case of condensation and show that it is significant in global context.
Having reviewed this thread one comes to the conclusion that there may be little or no additional convective uplift from the release of latent heat during the process of condensation for the following reasons:
i) The contraction involved in the conversion of vapour to liquid pulls additional mass into the original volume from the surroundings so that the air parcel is heavier overall and must descend causing an increase in surface pressure beneath it.
ii) Liquid having a much greater thermal capacity than air the condensate will absorb the bulk of the latent heat release for a very small increase in temperature and no reduction in weight.
iii) The latent heat that is absorbed by the air without vapour will not heat it enough to make it as buoyant as air with vapour at the same height so again it must descend.
On the face of it the Makarieva paper would seem to be implausible on those grounds so I invite Anastassia or her colleagues to suggest how they would deal with those problems.
One could suggest that the pulling in of mass from outside the original parcel reduces pressure around the parcel but that doesn’t work because the mass redistribution is simply a return to the average distribution that existed before the initial evaporation at the surface and so the inflow of mass from the surroundings would be neutral.
Jim D said:
“Latent heating generates buoyancy”
Latent heat has no thermal effect (being latent) and so does not generate buoyancy.
The buoyancy of air containing water vapour is simply due to he fact that water vapour is lighter than air.
The question then is whether the release of latent heat upon condensation adds to buoyancy or not and for the reasons set out in my post at 12.40 am I think such an effect has been overstated if it exists at all.
Would you prefer it if I said condensational heating adds buoyancy? There is no disputing that condensation releases the latent heat. Where does it go? Into the air (with a tiny fraction into the condensed water). Water vapor is, of course, lost in this process, so your idea goes in the wrong direction, and its loss leads to negative buoyancy, which we see does not happen in thunderstorm updrafts where the condensation is occurring.
Jim D.
What makes you think that most of the latent heat release goes to air rather than the condensate ?
Upon evaporation most of the latent heat comes from the water since it has a much higher heat capacity than air. It makes sense that upon condensation most of the latent heat goes back to the liquid condensate for the same reason.
Note that evaporation is a net cooling process but it still causes uplift because water vapour is lighter than air.
Similarly condensation is a net warming process but it still causes descent because dry air is heavier than water vapour.
It is not a matter of temperature but of relative densities.
In order for the heat released by condensation to cause renewed uplift it has to make the air which doesn’t contain water vapour as light as air that does contain water vapour.
It simply does not do so.
How hot would such air need to be ? Has air at that temperature ever been observed within a cloud ?
Thunderstorm updrafts continue despite condensation because the supply of new vapour is faster than the rate of condensation. As soon as the supply of new vapour falls behind the rate of condensation the cloud collapses.
Evaporation gives rising air and reduced surface pressure.
Condensation gives falling air and raised surface pressure.
The opposite of the Makarieva proposal.
Pekka has the right idea in pointing out that what she says is umphysical.
Stephen,
The specific heat of liquid water is about four times that of air when calculated per unit of mass or 2.5 times when calculated per molecule. The amount of liquid water in air is always very small, typically around 0.1% of mass inside a cloud. Therefore almost all heat goes to heat air.
Precipitation out of the volume does not change that. It does not change the number significantly, larger raindrops falling through the volume don’t have time to take of much of the heat. A rather heavy rain of 10 mm/hour represents 0.1% of the mass of atmosphere in a full hour.
Tomas Milanovic | February 11, 2013 at 8:23 am | Reply
Tomas, thank you very much for taking time to comment on our work.
The equation for condensation rate (34) is not physically independent of the Euler (N-S) equations because it assumes hydrostatic equilibrium. In this sense it is an approximation. Likewise the main result -u.∇p = SRT (as can be clearly seen from the blog post) is also based on hydrostatic equilibrium. From this perspective S (34) and the result it yields, -u.∇p = S, is not just an arbitrary specification of condensation rate from which the flow properties can be derived, but it a priori takes into account some flow properties. So it is part of the solution.
It describes steady-state circulation systems in hydrostatic equilibrium that use the potential energy of saturated water vapor for generation of kinetic energy.
What type of velocities can be produced becomes especially clear in the simplest case when turbulent friction is insignificant. Taking care of the a priori hydrostatic assumption, the system can be closed and solved approximately with some additional assumptions like the smallness of vertical velocity w. This is what we did in our paper on hurricanes. It produces a meaningful and physically transparent result. In 0 we have the eye where S = 0. (For the Hadley cell that would mean a feature something like double ITCZ albeit of a so-far unknown width. It could for example provide a clue to the narrow equatorial minimum of precipitation (e.g., as per Fig. 4c of Odell et al. 2008))
In our opinion, the ultimate test of the validity of our propositions will come from empirical evidence rather than numerical simulations. The theory gives unambiguous predictions for relationships between observable magnitudes. E.g. for hurricanes a straightforward check would be to analyze hurricane precipitation together with the available information on pressure gradients and radial velocity.
What is continuity equation on the number of molecules in gas?
It’s by definition an equation, that tells the change in the density of particles in a volume from the following factors:
1) difference in the number of particles flowing in and flowing out of the volume. That can be calculated as a sum of differences in each direction.
2) change in the volume occupied by a parcel of gas
3) source or sink term that tells, how many molecules are added or taken off from the volume
What is the equation (34)?
i) It’s first term is a major part of the point 1) applied in z-direction that is assumed to be the only important in the text below the equation
ii) It’s second term is explained in section 4.2 to represent point 2). Again it’s not exactly as it should be, but the explanation confirms that this it what it’s supposed to be.
iii) The other side of the equation is the source term exactly as it is in a continuity equation.
The equation is not an exactly correct continuity equation, but it’s what can easily be written, when the continuity equation is written based on the reasoning of the three parts of the continuity equation. The paper leaves no doubt that this is, what the authors have done.
For the number of molecules of a two-component gas two continuity equations are independent. In this case those are taken applying directly formulas from literature as equations (32) and (33). But then the authors derive a third continuity equation for the same quantities. They are not careful enough and get an approximation with some errors. Due to these errors the equation is mathematically independent, but without errors it would not be independent.
They have used physics of two equations to write three, two correct and one spurious. Then they go on to derive results from their error. Everything starting from (35) to as far the equation (34) affects the results is nothing but random results from an explicit error. The value of those results is zero.
To have a correct third equation they should have gone along the lines of Pat Cassen. Then their third equation would have parameters that describe the strength of the condensation from the physics of Clausius-Clapeyron equation and the release of latent heat. They discussed that physics in chapter 2, but somehow forgot it in chapter 4.
Anastassia mentions hydrostatic equilibrium in the above comment. It’s also an additional factor that has to be taken into account in going further in the calculation, but it does not lead to equation (34), it’s used in the correct derivation of the moist adiabatic lapse rate in literature like the Caballero lecture notes, and it’s use requires that also the physics that they discuss in chapter 2 is included.
I am really amazed that Anastassia cannot admit the facts. She was eager to comment on one of my comments. That was one where I wrote a stupid sentence. I wrote immediately another comment telling that I made the error. I try to admit and correct all significant errors that I make as soon as I notice them or learn of them. My own immediate admission and correction did not stop Anastassia from picking that sentence up and ridiculing that much later. That’s the only response that I have got from her during most of this discussion (there were others at very early state). She has not responded to any of specific and substantiated comments, but from that one we see that she’s not completely boycotting me. She just doesn’t answer, when she has no answer.
Concerning my argument with Tomas. I could have chosen better wording, but I really disliked his opening comment as it seemed to open up the idea that any equation can be written in a physics paper and by that would become worth consideration. From his latest comment we can read that he’s telling that the equations should be physically meaningful as well. I’m sure we agree that the equations should be both
– mathematically correct
– physically meaningful and justified for the role they have in the paper.
For me it was at that time already obvious for several reasons that the equation (34) in combination of the related assumptions on horizontal dependencies could not be physically meaningful. Therefore I wrote that it’s wrong to tell that it might have a change based only on the fact that it was mathematically independent. As we know now the independence resulted only from errors in it’s derivation.
@PP: What is continuity equation on the number of molecules in gas?
There are 42 equations in the AM et al paper not counting those in the appendix.
For me the problem comes with equation (5):
pV = RT (5)
My problem with this equation is that it assumes fixed n (number of moles of gas), even though n varies during condensation.
It should therefore read
pV = nRt. (5′)
And correspondingly equation (6), which currently reads
dp/p + dV/V = dT/T (6)
justified by equation (5), should read
dp/p + dV/V = dn/n + dT/T. (6′)
justified by equation (5′).
@PP: She has not responded to any of specific and substantiated comments, but from that one we see that she’s not completely boycotting me. She just doesn’t answer, when she has no answer.
Patience, Anastassia has lots of customers to serve. At least she had the good sense not to use bonus word “millikelvin,” which would have doubled your wait time.
When Kurzweill’s singularity comes and we’re all automated by robots, waiting will be a thing of the past.
Vaughan,
The paper discusses two chains of arguments. One is formed by chapters 2 and 3, and section 4.4. My main comment on that is that it’s totally irrelevant telling nothing new about the real atmosphere, because it’s in an essential way based on the comparison with the isothermal atmosphere. Whether it has some small errors in addition is therefore also irrelevant.
The other chain of arguments starts with equations (32), (33), and (34) and this is the thread that’s totally false. It might be a little more relevant, it it were correct at all, but it’s just totally wrong without any validity on any point. Here the error is not irrelevant, it leads to total failure of the whole thread.
Sorry Pekka, I didn’t mean to imply I disagreed with you on any of your points, which I’m completely fine with. But hell will condense to water before you persuade AM of any of this. She has the mindset of a mathematical physicist.
Mathematical physics is not physics informed by mathematics, it is mathematics motivated by physics. The latter is as irrefutable as any other branch of mathematics provided it is soundly argued. Whether it informs physics however is a matter of chance. Its chances of success are inversely proportional to how hard it tries to succeed unless considerable imagination is injected.
Pekka, first of all, I respect your opinions about our paper. Second, my own opinion is that you do not understand the problem, so your repeated criticisms are confused and mostly off-topic. You are making a lot of comments with basically one and the same content. But since people here seem to appreciate your discourse, I am not going to answer every your comment just to say that, in my opinion, you continue to be confused and continue to ignore what I’ve said so far. My “quote of the day” remark was just an indication for future readers that, in my opinion, you do not understand the problem. Your self-correction did not change that.
1. Let me take this your claim.
In my opinion, this statement, making absolutely no sense, nicely summarizes the confusion.
You continue
Please compare these claims with what I said about two hundred comments ago:
You may somehow believe that it is an error, while we have given clear physical grounds why S (34) is as it is. Your statements about some error that somehow produced S (34) remain unsubstantiated until you do either (a) or (b) above.
2. I mentioned several times in this thread that there are independent physical considerations that produce our main result -u.∇p = SRT without any reference to the continuity equation and which are consistent with Eq. (34). As before, this point continues to be ignored.
3. In your comment that you made in response to mine, you once again totally ignored what I said. Namely that the first law of thermodynamics, being an equation of equilibrium thermodynamics, cannot give information about the dynamics of the system. So derivations of S, which directly determines dynamics as per Eq. (6) in the post, cannot be made (and are never made) from the first law of thermodynamics. I also explained how it is made in models. I emphasize once again that no theoretical expression exists for S except for (34).
Once again, I fully respect your opinions and your right to repeat them as many times as you find plausible. But, please, appreciate that as I believe that your vision of the relevant physical problems is incorrect, and you do not seem to be interested in changing your view but are satisfied with your current vision, there is no point for me to answer your comments unless you say something new.
Anastassia,
In your comment you are discussing the physical process that leads to condensation, but the equation (34) is not about that, it’s derived by you to describe the outcome of the condensation, i.e., it’s derived as a superfluous continuity equation, as I have explained several times.
And you say that I repeat the same argument.
Yes I do, because that’s the single essential point, where your paper goes fully astray. As long as you don’t see the point, I can only repeat that and try to explain it in a little more clear way.
The only way you could counter my point is to present a step-by-step explicit justification of he equation (34), and not one that’s a derivation of a continuity equation. It would be very useful to do that without any approximations on the way. Justifiable approximations can be introduced later. My view is that you would then see directly that it’s not independent after all.
The two faults in the derivation are:
– dismissal of horizontal derivatives
– In the argument at the beginning of 4.2 N is used, but it should be N_d. As N_d is conserved in the condensation, it’s the right one to consider in that argument that tells about the change in volume of the parcel. The removal of part of vapor makes the argument slightly wrong with N. That slight error turns out to be essential, when put aside (32) and (33).
Anastassia,
As I have now told explicitly what are the errors/approximations that have made the equation (34) deviate from a similar equation derived from (32) and (33), I hope that you finally admit, that it’s just an approximate and by that a little erroneous version of a continuity equations.
You claim that I don’t accept your further arguments. How could I, when they don’t address the real question:
Does the equation (34) represent new physics?
As I have now proven to the smallest detail, it doesn’t. Earlier I thought that explaining what’s wrong in physical terms would be enough. Now you have also the full mathematical proof that the correctly derived (34) agrees exactly with what can be derived from (32) and (33) up to the horizontal derivatives that you have assumed to be zero. This tells that it’s not a physically independent equation, but independent only due to errors in derivation.
You may discuss additional physics, but that’s irrelevant as you haven’t put that in the equations.
To make all complete, I repeat the argument with formulas. Others have done that already, but I add some small comments.
The equations (32) and (33) are
Expanding both we have
Multiplying the upper by
and subtracting from the lower we have
Here we have first horizontal partial derivatives. They were assumed to be zero in the paper. In that we are left with (34) except that we have there N_d rather than N. As a explained above, the justification of section 4.2 tells that N_d is the right one.
We have derived the corrected (34) from (32) and (33). Dropping horizontal derivatives was justified only, when they are zero. They cannot be dropped otherwise. Thus the original (34) is not even supposed to be true otherwise.
Just to continue a little.
The physics involves the vertical motion or more relevantly the motion in the direction of the pressure gradient, that’s not necessarily vertical. The component of the motion along the pressure gradient leads to expansion and cooling and further to condensation. All this goes into the continuity equation. It’s important to notice that the pressure gradient is not necessarily vertical. Therefore the isotherm is not necessarily horizontal. One of the errors you make is to assume that the isotherm is horizontal. That makes the equation (34) inaccurate, too inaccurate for use in studying any horizontal phenomena as you do in (35)-(37).
If you had not done this unphysical assumption and if you had been also in other ways a little more careful in the derivation of (34) you would have seen that it’s not independent, but derivable from (32) and (33) in it’s precise form.
The other independent equation is the that tells from other physical laws, how much condensation the drop in temperature causes. That’s explained by the Clausius-Clapeyron equation.
Pekka, ” That’s explained by the Clausius-Clapeyron equation.”
C-C is one of the weaker estimates and the efficiency of the combination of process is not constant.
Think of a simple steam cycle that is 35% efficient where the waste 65% is not wasted but powers a second stage with 35% or so efficiency, you are combining cycles. She is attempting to describe condition for a combined cycle. The problem is that different configurations have different efficiencies per stage.
In the tropics you could have two or three stages and in the polar regions only one.
So I agree with Tomas, they could either try to solve N-S, not likely or refine special cases.
Clausius-Clapeyron is not that bad in this context. I don’t think that anyone would use it in calculation of a steam cycle, that’s done using steam tables that represent empirically measured values or parametrized formulas that agree with the steam tables (i have implemented years ago that approach in a small application used mainly for educational purposes as are certainly very many others).
The steam cycle is so inefficient mainly, because the maximum temperature of the steam is so low. Other factors add to that, but that’s the principal reason. The combined cycle can take advantage of higher temperatures of the flue gases as well and trough that reach a higher efficiency.
I cannot find any connection between the paper and combined cycles.
Pekka, “I cannot find any connection between the paper and combined cycles.”
That’s because they are attempting a more “universal” derivation without setting up specific boundaries. Separating out N-v though is considering the steam cycle. The fun part with the low efficiency steam cycle is there is more waste to be recovered. The tough part is which is waste and which is work?
“Pekka Pirilä | February 12, 2013 at 4:17 am | ”
Thank you Pekka, that is clear. There is not enough in the way of water droplets to attract a large portion of the heat released when condensation occurs.
That narrows down the issue to whether the heat released is enough to heat dry air to a point where it is more buoyant than air containing water vapour.
It would seem not because dry air descends when the vapour is removed.
The equation for condensation rate (34) is not physically independent of the Euler (N-S) equations because it assumes hydrostatic equilibrium.
Anastassia
Yes I understood this since the beginning.
32,33,34 are just mass conservation equations with a specified sink.
You could have multiplied 34 by an empirical constant to indicate proportionality instead of equality without changing a iota the validity of the derivation. But this is a detail anyway.
As such you of course realized the closure problem because mass conservation can’t yield alone the dynamics.
And there are only 2 strategies to solve that.
Either N-S what gives the accurate and rigorous description.
Or a simplifying hypothesis which allows analytical treatment.
The former is difficult, analytically untractable even in the “simple” 2D steady state case and you would have needed access to computers and a good DNS code.
The simplest example of the latter is the hydrostatic assumption in which a very simple equation is substituted to the full N-S allowing farther analytics.
You choose the latter.
Now it is known how much havoc can wreak this assumption both in numerical models and in the results on meso-scales. Gerald Browning has written much about exponential growth of instabilities and artefacts appearing when a hydrostatic assumption is taken.
This is a serious problem where dozens of papers have been written.
Clearly even if w/u was “small” it wouldn’t justify the hydrostatic assumption per se and especially not on small spatial scales.
This is the reason why I called myself “agnostic” – I confirm that your equations are mathematically consistent and closed but I am not at all convinced that your “hydrostatic solution” converges to the full N-S solution.
I would think that on meso scale (say 100 km and below) it doesn’t and above the jury is still out even if you show some experimental evidence.
Tomas
I basically agree with all your points. Yes, we chose the second strategy because it corresponded to our opportunities. We shall see what we will be able to do beyond that.
Exactly. Before throwing ourselves into solving N-S (I am not very optimistic about that) we need to get a more general expression for S, which among other things should also account for horizontal temperature gradients. There is still something to be done at the level of theoretical concepts and, as captdallas2 0.8 +/- 0.2 put it, refining some special cases.
Regarding evidence, I do think it will play a major role although I agree with you that formally it will not “prove” that everything occurs as we describe. However, it is noteworthy that the current paradigm with its heat-driven circulation is exactly in the same position — but additionally lacking working theoretical estimates of total circulation power that we already have.
Suppose mathematics didn’t exist but bicycle pumps and black plates did. Why would it be “astrology” to assert that temperature increases when you pump up a bike tire or leave a black plate out in the sun?
Well it would be astrology because it uses exactly the same method that astrology uses.
Namely collecting random correlations and putting them on a list.
What you obtain very fast is a vast list mixing coincidences, semi regularities and true regularities.
First ancillary problem is that you would run out of paper because you would see “new” correlations every day.
Second and this one deadly problem is that your list would be very confusing. Not only you would have to remove entries every day because they would appear non regular but the same actions (f.ex “blowing air”) would sometimes cause warmth and sometimes cold.
Physics without mathematics gives you a vision of the universe which is mostly an unpredictable and confusing chaos. Things that you thought sure suddenly stop happening and things that you thought impossible happen.
Quite a hell of a life :)
I would say you and I have irreconcilable differences on that point. (I’m assuming you publish primarily in mathematical physics whereas my publications been primarily in mathematical logic since 1976, so I’m not exactly a mathphobe.)
What purely physical principle did Walker and Schneider appeal to in estimating the poleward energy flux of the Hadley circulation at 4.8 PW in Figure 5 of http://www.clidyn.ethz.ch/papers/annrev06.pdf (the paper Anastassia referenced above)? According to you this would count merely as a random correlation to be added to a list of such. Likewise for this estimate of the moment of inertia of Earths’ atmosphere: not purely physical and relegated by you to a list, creating “a vision of the universe that is unpredictable and confusing chaos.”
What about the 23 (or whatever the number has grown to) fundamental parameters of physics? They’re not purely physical either.
You and I see these supposedly chaotic lists from very different points of view.
Up to a point I agree with Tomas, but only up to a point.
Physics is mostly a quantitative science, and a quantitative science gets its value from combining observations with a quantitative theory. Most of physics requires equations to express the theory, without that it’s very limited in comparison with the physics we have.
But physics is also a theory of the world where we live and that world did exist before anybody had written the first equation. Many properties of the physical world can be listed without equations. Collecting such information could be science, but it would be a much less powerful science than the physics we now have.
I believe that it’s fair to say that equations have been used as long as activities that could be classified as science of physics have been practiced. That’s probably true, but that’s not a logical imperative.
I use equations all the time to reason about physics. However I stick to my analogy with dynamite. Equations are tremendously powerful, but they are neither foolproof (you can be hoist with your own petard by using the wrong equation) nor a panacea (not every physics problem has an applicable equation).
If equations were foolproof you could judge physics papers solely on the basis of whether every step of their calculations was sound.
And if equations were a panacea you would never need to conduct physics experiments. SLAC, Fermilab, CERN, etc. could be dispensed with and the money saved spent on supercomputers and pencils.
Equations deliver quantitative precision, but only when they’re (a) available and (b) applied appropriately.
I should add that one of my hats is that of logician. From that perspective algebra is equational logic, that is, logic in equational form with no existential quantifiers. In physics the terms on each side of an equation typically evaluate in some numerical domain, but algebra as logic in equational form covers many other domains besides numerical ones, just as does logic itself.
Whereas I don’t (as yet anyway) see any difference between my position on this and Pekka’s, I see a huge difference with Tomas’s position. Merely because Emmy Noether has made a profound connection between symmetry and conservation principles, one that I have enormous respect for myself, is no reason to generalize to the claim that she has reduced physics to mathematics.
Physics can motivate mathematics, and mathematics can serve physics, but neither subsumes the other.
Pekka and Vaughan, On the efficiency theme, I was comparing an ideal radiant system to a general steam system. The efficiency per stage in a radiant system should be 50% and for steam roughly 38% per stage. Since water appears to be responsible for the atmospheric window energy, 20 Wm-2 true surface and 20 Wm-2 cloud surface would likely be the “normal”. With the steam cycle 38% surface and 38% for the cloud “surface”. This could provide the delicious chaos as the competing efficiencies try to find a happy place.
Energy is fungible, but the work done is not.
@VP: 23 fundamental parameters of physics
should have read “fundamental constants.”
Wow. Never has thinking about the 1% volume loss associated with water vapor condensation caused so much confusion and wasted time. Yes there is a tiny error in climate models because that volume loss is ignored. But it is a tiny error, and one that has long been recognized. There is nothing in the paper which is novel, it has serious problems with equations that are not justified by physical reasoning, and I am quite sure it is never going to have much impact.
@SF: I am quite sure it is never going to have much impact.
I would say an undisputed sentence in the Wikipedia article on wind is an impact. Whether its impact will expand from there or shrink remains to be seen.
Vaughan Pratt | February 12, 2013 at 2:04 am | Reply
4 W/m2 is the dynamic power of global circulation. It has nothing to do with heat transfer or radiative transfer.
Imagine that you switch on a fan in your room. It will create some wind. The kinetic energy of this wind will first grow rapidly from zero and then stabilize at some value. In this steady-state the fan creates as much kinetic energy of wind per unit time as the frictional dissipation destroys.
If your room is isolated from the environment, frictional dissipation of kinetic energy of wind (ultimately to internal energy of the air) will lead to a rise in room temperature. From the rate of temperature increase you will be able to determine the power of your fan (e.g. is it 1 kW or 10 kW etc.) In the steady-state (when temperature is constant) you will need to apply other methods to estimate the power of your fan. You can consider peculiarities of eddies and their temporal and spatial characteristics. More straightforwardly, if you have a linear flow over some distance X which has a pressure difference delta p, then, if there were no friction, the characteristic flow velocity would be given by rho V_max^2/2 = delta p. By comparing V_max with the actually observed velocity you can judge about the rate of frictional dissipation. When V << V_max, frictional dissipation (i.e. the fan power) per unit volume is just u.∇p.
Anyway, the bottomline is that it can be estimated and people did it quite a while ago. They found that the power of the fan that drives global winds is around 4 W/m2 (it is not our result), i.e. it is in the order of 1% of solar power. The question perceived by many distinguished minds (like Lorenz) as being of crucial importance for the theory of atmospheric circulation was: Why it is 1% but not 10% or 0.1% or whatever? This question has remained unanswered so far. We provide a clue.
@Anastassia: They found that the power of the fan that drives global winds is around 4 W/m2 (it is not our result), i.e. it is in the order of 1% of solar power. The question perceived by many distinguished minds (like Lorenz) as being of crucial importance for the theory of atmospheric circulation was: Why it is 1% but not 10% or 0.1% or whatever?
Thank you, Anastassia, that poses very nicely the problem you’re addressing. If one of your papers has already posed it in that form I apologize for overlooking it.
Correct me if I’m wrong, but my understanding of condensation is that it is a completely reversible process. The sensible heat liberated from latent heat by condensation is converted back to latent heat by evaporation with 100% efficiency. In either direction 2260 joules is exchanged in the conversion between the liquid and gas phases of water.
What I’m having difficulty grasping here is, how is any power at all needed to drive a 100% efficient process such as condensation? Is this addressed somewhere in your paper? Is there a microcosm of some kind giving an easily understood example of where condensation requires any power at all?
In either direction 2260 joules is exchanged in the conversion between the liquid and gas phases of water.
Joules per mole, of course.
Vaughan
Condensation and evaporation are reversible when the occur at saturation. Evaporation to air that’s not saturated is not reversible, neither is condensation from supersaturation. Removal of liquid water from air through precipitation is also irreversible.
The easiest component of dissipation to calculate semiquantitatively may be that related to rain where water lifted to some altitude falls down losing the gravitational energy as heat during the fall and hitting the surface. Lifting 1 ton of water to the altitude of one kilometer takes 10 MJ. Doing that in one year represents the power of 0.3 W. This is not a major factor, but not totally negligible. The mass of 1 ton corresponds to 1 m of rainfall on one m^2. This includes only the precipitation that reaches the surface. Repeated evaporation and condensation within the atmosphere is not included.
This drawing from a book by Sorensen sets the old estimate of Lorenz in wider connection. There we see that 1200 TW goes through winds. They are maintained by that power and they lose that power through dissipation.
It’s certainly debatable, whether the power that goes through winds is the power of atmosphere, but that’s the definition that Lorenz used in stating that the efficiency is about 1%. 1200 TW is about 1% of the energy flows into the atmosphere when reflected SW and direct radiation from surface to space is subtracted from the incoming radiation.
The typical temperature for energy input is about 290 K. An estimate for the cold side temperature is given by the effective radiative temperature of the Earth of about 255 K. Thus the Carnot efficiency, i.e. the efficiency without any dissipation would be about 35/290 = 12%. It’s not surprising that we have a lot of dissipation to bring that down, but giving a more quantitative estimate is not that easy. As I wrote above, I don’t think that the paper discussed in this thread helps in that at all, as their estimate is based on comparison with a totally irrelevant case.
@PP: Condensation and evaporation are reversible when the occur at saturation. Evaporation to air that’s not saturated is not reversible, neither is condensation from supersaturation. Removal of liquid water from air through precipitation is also irreversible.
Great, that’s more like it. Getting the rid of the equations helped (we can put them back once it’s clear which ones apply).
This drawing from a book by Sorensen sets the old estimate of Lorenz in wider connection. There we see that 1200 TW goes through winds. They are maintained by that power and they lose that power through dissipation.
Why didn’t you bring that up earlier?
The Hadley circulation transports exactly four times that much power (1.2 PW x 4 = 4.8 PW) polewards. Somehow Anastassia sees these as having nothing to do with each other.
As I wrote above, I don’t think that the paper discussed in this thread helps in that at all, as their estimate is based on comparison with a totally irrelevant case.
The paper loses itself in its equations. That anyone believes it at all is mute testimony to equations as an art form: “I don’t know equations, but I know what I like.”
Vaughan said, “Great, that’s more like it. Getting the rid of the equations helped (we can put them back once it’s clear which ones apply).”
Exactly.
“The Hadley circulation transports exactly four times that much power (1.2 PW x 4 = 4.8 PW) polewards.”
Not exactly and not equally to each pole. A minor shift in the “Thermal equator” or ITCZ has a major impact on internal energy distribution. So a small tug can be a big deal.
Vaughan,
Condensation is not reversible. To form droplets the water vapor must diffuse in, and the latent heat must diffuse out.
Vaughan, Just for fun think of this.
http://redneckphysics.blogspot.com/2013/02/combining-cycles.html
Instead of temperature think of this, 1/T=dS/dE
With about 210 Wm-2 to work with, we can move some air.
http://kestrel.nmt.edu/~raymond/classes/ph536/notes/thermo.pdf
One of the issues with Pekka’s example is Th of ~290K. The average ocean surface temperature is closer to 294K with the Th closer to 303K and Tc is roughly 200K in the Tropics.
Condensation per se does not require any power, it only requires a temperature gradient and saturated vapor. How condensation makes the gas move can be best illustrated on the example of heat pipes.
In these devices the hot end is where the liquid evaporates (and the vapor pressure is high) and the cold end is where the vapor condenses (and its pressure is low — by Clausius-Clapeyron law). Because of this pressure gradient, the vapor inside the pipe accelerates to enormous velocities (up to molecular velocities), which make the pipes very efficient in transferring heat and in cooling things that are difficult to cool by usual means. There is a very strong “wind” inside these pipes. Note that all latent heat that is released at the cold end is removed from the pipe by an external device (e.g., a fan). So it is not a reversible process at all.
@AM: How condensation makes the gas move can be best illustrated on the example of heat pipes.
Ironically the first time I used a heat pipe to cool a CPU was to make it fanless so people could have much quieter computers. This was in 2000, the year I went emeritus in order to focus on tiny ubiquitous technology. One such device was a fanless PC the size of a brick which we called the Subbook. The cold end of the heat pipe was embedded in a huge convection-cooled heatsink external to the enclosure. Katya Puzyrko did much of the detailed design. But my company was also working on a small handheld computer and we decided we needed to focus on the latter and didn’t have the resources to pursue both.
Because of this pressure gradient, the vapor inside the pipe accelerates to enormous velocities (up to molecular velocities), which make the pipes very efficient in transferring heat and in cooling things that are difficult to cool by usual means. There is a very strong “wind” inside these pipes.
Something like this (without the word “wind”) is indeed what tends to be taught in physics classes. In mechanical engineering classes one learns instead that the speed of transport of heat by the vapor is a red herring: if you slowed it down by a factor of ten or even a hundred it would make no difference.
The limiting factors are the rate of condensation and the rate of return of the condensed liquid being wicked (one syllable) back to the hot end. In practice the former tends to be the more important one. An underdesigned heat pipe will get so hot at the cold end that condensation can no longer occur, and in that case even if the vapor molecules sped up by another factor of ten essentially no heat would be drawn from the CPU. One fix (besides scaling everything up) is simply to use a condensate with a slightly higher boiling point and put up with the slightly higher operating temperature.
So it is not a reversible process at all.
This is clearly true at the hot end, where evaporation is occurring far from adiabatically. Condensation can occur all along the rest of the pipe, most strongly at the coldest points. It would be interesting to plot the efficiency of condensation as a function of temperature along the pipe, i.e. how far from adiabatic it is, equivalently how much heat is transferred by a given amount of condensation to the exterior of the pipe. (Since the goal here is to extract heat, efficiency is the complement of how one would define it when the goal is to retain heat, with 30% for the former becoming 70% for the latter.)
The situation with the environmental lapse rate is a very extreme case of the relatively adiabatic middle of a heat pipe. In stable air (between midnight and noon, very roughly) everything happens very slowly, and the evaporation and condensation processes that maintain the ELR of saturated air close to the MALR can be arbitrarily gentle and hence arbitrarily close to 100% efficient (defined as heat retained).
Very large angles are possible. In calm weather at 5 am in the bottom 1-2 km of air above the surface the ELR can be as strongly negative (temperature increasing with altitude) as the MALR is positive, an ultrastable situation. This happens because the ground has cooled to below the daily average (which is kept high by the Sun during the day) but the air above 2 km or so remains stuck at the daily average minus adiabatic lapse rate throughout the day and night. This in turn is because it is not sufficiently strongly coupled thermally to either the Sun, the surface, or space to lose or gain much heat over the course of 24 hours. In particular the diurnal oscillation of IR from the surface that GHGs capture is too rapid to make the air temperature oscillate appreciably.
In the afternoon the ELR straightens out to match the MALR, and then overshoots it, creating unstable air that then lifts off the ground as thermals, large spherical (droplike) or cylindrical (streamlike) parcels depending on strength.
The heat pipe analogy is more appropriate to the whole hydrological cycle, the essential energetics of which occur at the surface (the hot evaporating end), in the clouds where the larger water droplets accelerate condensation, and in precipitation, which converts very slowly acquired potential energy back into kinetic and thermal energy very quickly.
Transport of moisture from the surface to the clouds is a very slow and gentle process, and is therefore an implausible candidate for contributing a significant amount of energy to weather. No matter how accurately one models it with equations there is very little to learn thereby about where winds come from.
@NS: Condensation is not reversible. To form droplets the water vapor must diffuse in, and the latent heat must diffuse out.
Presumably you’re considering a non-equilibrium situation where one of condensation or evaporation dominates. At the surface evaporation dominates, and in clouds condensation dominates, and there I would agree the situation is less reversible locally.
Two comments on that.
1. I’ve been focusing (perhaps unduly) on the atmosphere between the surface and the clouds. Latent heat of vaporization of water plays a major role there in determining the MALR, and is governed by AM’s equations (the correct ones, that is). However that role can be played in an arbitrarily efficient, i.e. reversible, manner. Irreversibility happens at the endpoints, not in the middle where the environmental lapse rate is compared to the prevailing adiabatic lapse rate for that level of RH.
2. Even when work is done by thermally driven atmospheric circulation so as to create kinetic and potential energy, essentially all that work is soon turned back into thermal energy. Any variations in heating of the planet or of space by work resulting from atmospheric circulation is so slow as to be effectively in equilibrium.
That said, the 4 W/m2 that AM is referring to would still make sense if understood as the rate of conversion in each direction between mechanical (kinetic plus potential) energy and thermal energy, since it’s the kinetic energy of storms that is relevant to their impact. Is that how Lorenz defines it, or appears to be thinking of it as? Pekka?
Conversions of that kind must play some role in the 9.4 W/m2 of total energy carried polewards by the Hadley circulations, though I’m not clear as to what. I would be surprised if they turned out to be entirely independent, on the other hand I don’t have an argument either way.
@cd: Not exactly and not equally to each pole. A minor shift in the “Thermal equator” or ITCZ has a major impact on internal energy distribution. So a small tug can be a big deal.
I don’t have any quarrel with that.
Vaughan, “No matter how accurately one models it with equations there is very little to learn thereby about where winds come from.”
Good, then we can talk about fishing :)
Like the heat pipe, the efficiency depends on what you want. 70-30 is a pretty standard split between work and entropy, if it is entropy you want, you would be a consistent winner, provided you limit the game to a single stage.
@cd: Good, then we can talk about fishing :)
You go first, cd. I’m more familiar with phishing.
No matter how hard it turns out to be to factor large numbers, there is very little we can learn about phishing.
Vaughan, “No matter how hard it turns out to be to factor large numbers, there is very little we can learn about phishing.”
I am not much good at that kind. The type with an “f” though I do all right. Heck, sometimes I hit it just right and look like a fishing genius. It is all about location, location, location plus being there at the right time.
Kinda like this paper, for T to be isothermal and Pws to be saturated requires the right location and conditions. The biggest being a butt load of wet thermal mass. You still can’t get something out of nothing unless it has a “p” in front.
So i think they jumped the shark when they should have focused on their bait :)
Now had they focused on the location, near equatorial where there is a lot of potential but not much guidance, their little tug would have made a bigger splash.
http://www.eldoradocountyweather.com/current/satellite/goeseast-wv.php
Merely because Emmy Noether has made a profound connection between symmetry and conservation principles, one that I have enormous respect for myself, is no reason to generalize to the claim that she has reduced physics to mathematics
V.Pratt
I find this subject completely off topic so only a short comment (apologies Anastassia) because I don’t like misrepresentations.
Such a claim is indeed stupid and I don’t think that any scientist would make it.
As for me I even explicitely explained why such or similar claim would be stupid.
I merely made a trivially true logical statement :”physics implies mathematics”
Of couse as every logician knows, the truth of this statement is not equivalent to the truth of the statement “mathematics implies physics” what is what would mean the stupid claim you mentioned.
Physics without mathematics, as I said, would only be a mix of random collection of perceived qualitative correlations with no possibility to generalize or predict anything. Actually you couldn’t even tell how warm something is – because even for such a trivial observation you need basic geometry and metrics. For a significantly more useful observation you will need natural numbers, order relation, abelian group structure etc etc.
A discussion about Archimedes principle among “physicists” who ignore mathematics would look like the legendary Monty Python scene:
– Physicist 1 : What floats?
– Physicist 2 : Apples!
– Physicist 3 : Gravy!
– Physicist 4 : Stones!
(puzzled looks of the crowd)
– Physicist 4 : errr … very small stones?
– Senior physicist making a dramatic pause to get everybody’s attention : Ducks!
– Crowd in admiration: Ooooh!
Oh come one, Tomas, you’re not being serious. It is trivial to demonstrate Archimedes principle empirically with no mathematics at all. Simply place various shapes of objects in the liquid in question (water, honey, whatever), variously oriented, and observed that the resulting level of liquid always rises in direct proportion to the object’s weight.
To establish that the ratio is unity just place unit weight of liquid itself in the liquid for calibration.
Furthermore the principle can be demonstrated just as well with honey as with water. Honey being viscous dissipates and hence is not within the scope of Noether’s theorem, whence Archimedes’ principle transcends what you’ve been calling a “purely physical” principle.
Mathematics doesn’t give any sense of what warmth or weight really are, and it has no meaningful units. It is a useful tool to enable physics to be done, however. A mathematician would have no concept that links an equation to the real world, and nor do they care if their equations have a practical application. This is where physics comes in. It puts relevant parts of mathematics to use, and finds parts of the vast field of mathematics that are useful, such as Riemann geometries that Einstein used for general relativity, or eigenvalues and matrix operators used by various other physicists for quantum mechanics.
I have tried three times but for some reason my reply to Pat Cassen does not show up.
So I posted it there. Please comment in this thread, thank you.
Another try
Pat Cassen | February 12, 2013 at 1:36 pm | Reply
Pat, thank you for your clarifications.
I still think I fully addressed your concern in my first reply. Your expression for u∂p/∂x is derived from the first law of thermodynamics, Clausius-Clapeyron law and the ideal gas law, plus assuming horizontal isothermy ∂T/∂x = 0. As I said before, the first law of thermodynamics, being an equation of equlibrium thermodynamics, is inexact when applied to real-time dynamic systems. It errs precisely about the effect that we are trying to estimate — the rate of generation of kinetic energy. This rate is neglected in the first law — any work performed on the gas goes to increase its internal energy, not kinetic energy. Therefore, any conclusion about dynamics inferred from the 1st law must be different from the correct one. So the fact that you derive something from the 1st law that differs from our result does not in any way undermine our conclusions.
That is what I said before. But I can go further and be more specific. Your equation for u∂p/∂x is not merely derived from the 1st law, it is derived from the 1st law for an adiabatic case, i.e. for dQ = 0. Thus you somehow presume that the air changes pressure adiabatically in both x and z directions, but temperature changes in only one direction. This is unphysical, but I leave it to you to analyze why it is so. The main point is that horizontal isothermy in the real world (e.g. in Hadley cell, where it is a very good approximation as admitted even by Dr. Held) is certainly not adiabatic. It is ensured by efficient horizontal mixing that ensures a poleward flux of heat. I.e. for every air parcel dQ is not zero.
Therefore, whatever one derives from the 1st law assuming that the motion is adiabatic is incorrect for two reasons (a) the motion is known not to be adiabatic and (b) we are not considering a system in equilibrium. Therefore, the difference between your u. ∂p/∂x and ours does not in any way invalidate the latter.
I hope this might be is clear now, but I’d welcome further comments from you. Regarding the last word, as you might have noticed I am not keen about having the last word. E.g. until Pekka explicitly complained that I was boycotting him, I had not been interfering with his numerous last words about our work. So if you are interested in my feedback, it’d be a pleasure for me to provide one. If not, it is my pleasure to leave the last word to you whatever it is. In any case, thanks again for your interest.
Now we should believe in the next miracle. We have a paper that uses throughout equations taken from equilibrium thermodynamics, but with minor approximations and errors that grow hugely in significance when used as they are used in paper. Now we should take that as a discussion of non-equilibrium states. We should believe that in spite of the fact that it does not present any input from theories of non-equilibrium thermodynamics or fluid dynamics applicable to non-equilibrium states, while the equations become exactly the equilibrium physics equations when explicit technical errors are fixed.
If you wish to study some area of physics, you should define your system in a way applicable for that and you should take into account, what’s known about such systems.
An erroneous paper of equilibrium thermodynamics does not become a paper of non-equilibrium physics just by a declaration by it’s author. Errors remain errors and empty words bring nothing.
Again the only way of proving that I’m overstating your errors is presenting explicitly, how the equations presented in the paper can be justified, and how they do represent some particular physical phenomena. This is still lacking. The lengthy comments never go to the required level of specificity. They remain empty words.
That a problem is difficult for others as well is neither any evidence that what you write is of any relevance. To provide something of value, you must fill the gaps and avoid explicit errors.
Anastassia –
Thanks again for your response, which does clarify our differences.
The point iof my original comment was to explore the validity of your postulate about S by examining the consistency (or lack thereof) of your equation for the horizontal pressure gradient with other thermodynamic constraints. This approach was motivated by, among other things, the rather dramatic result that, under the conditions considered, horizontal pressure gradients were independent of latent heat.
Yes, indeed, I did consider the equilibrium, adiabatic, ∂T/∂x = 0 case.
Non-adiabaticity is readily included by adding an unspecified energy source dQ in the first law. The effect of this is to add a term proportional to dQ to the expression for ∂p/∂z:
u ∂p/∂x = w [ (p/T) (∂T/dz) f – ∂p/dz ] + g dQ
where g = g(μ,γ,ξ)
This factor does not affect my previous comments regarding inconsistency with your expression; the essential differences remain. (Yes, if all the latent heat is immediately radiated away, it probably has no effect on the local horizontal pressure gradient.)
The neglect of the horizontal temperature gradient is common to both of our derivations, is it not? You state “…[the condensation rate] is proportional to vertical velocity w (because condensation is due to cooling, hence it is proportional to the velocity of movement along the temperature gradient)…” (my emphasis). In any event, inclusion of a horizontal temperature gradient only adds more terms to the pressure gradient expression without altering its basic form.
As for non-equilibrium processes, well yes, I suppose anything can happen! Gas is accelerated, radiative opacities change, phase change fronts propagate, etc. But it is enlightening to analyze simpler, perhaps limiting, cases. After all, in your derivation of eqn. (12) for ∂p/∂z you too have found it expeditious to neglect ∂/∂t’s, assume hydrostatic vertical profiles, and so forth. The nature of these nonequilibrium processes, whatever they are, do not appear in your equation.
To repeat my main point: I sought some evidence that your equation for ∂p/∂z could be reconciled with other thermodynamic constraints, at least under some conditions. Had I found such evidence, other criticisms (e.g., by Pekka and Nick) might carry less weight. But I found none.
I hope these discussions will motivate you to seek a firmer theoretical basis for your hypothesis.
@PC: I hope these discussions will motivate you to seek a firmer theoretical basis for your hypothesis.
She’d be in great company if she didn’t. The Heisenberg-Schroedinger approach to quantum mechanics had been pretty well integrated into core quantum physics by the end of the 1920s. Einstein died in 1955 at the age of 76. At no point in the quarter century from 1930 to 1955 did Einstein throw in the towel on his “God does not play dice” thesis.
AM has a shot at beating the quarter-century mark.
I would be nowhere near as unkind if AM had shown any sign of paying genuine attention to any of her many critics. So far she’s paid them no more than lip service. She’s a perfect example of what can go wrong when a mathematician claims to have the insights of a physicist. See my reply to TM for more in that vein.
(I should add that I didn’t intend that to reflect badly on Einstein. Going way out on a limb here, I believe he was right but lacked the requisite tools to argue his case, which are only just now coming to light. I would estimate the odds of AM being similarly vindicated in the future as miniscule by comparison.)
Vaughan,
I wrote the following comment, but sending it failed (I hadn’t signed in to WordPress). Then I thought that it might be better to leave it, but after reading your above comment I decided to send it anyway.
—
I agree fully with your attitude and approach. Sometimes I cannot figure out what you really have in mind on the substance, but that’s a different issue and is certainly related to the differences in our background.
I have always had some problems in figuring out, what a mathematical physicist is. During the part of my career I was active physicist, I worked at theory division of CERN, elementary particle theory group of Argonne National Laboratory near Chicago, and Research Institute of Theoretical Physics in Helsinki. Thus I believe that I know, what theoretical physics is. Theoretical physicists do their practical work using mathematics, but they are physicists and use mathematics as a tool. They must all the time have the physical context clear in their mind to succeed.
I know also that there are mathematicians who get inspiration from physics and who do sometimes make crucial steps in the development of physics, but as mathematicians they study mathematics and have their criteria of excellence from mathematics.
But where is the place for mathematical physicists?
They may have an important role in checking the often sloppy mathematics that theoretical physicists sometimes create and making it more correct and rigorous, but that doesn’t sound particularly grandiose. Is that the reason that we meet sometimes people who describe themselves as mathematical physicists doing very questionable claims about physics. Have they found the role of a mathematical physicist too limiting and therefore gone over to theoretical physics without the required understanding of the physics?
> But where is the place for mathematical physicists?
Sometimes, they get quarantined into philosophy departments.
Just a kind word for Einstein here, it is his POV that has motivated everything on quantum entanglement. To some degree quantum computing is simply EPR actuated. Be nice to the clever fellow.
@PP: I have always had some problems in figuring out, what a mathematical physicist is. …Theoretical physicists do their practical work using mathematics, but they are physicists and use mathematics as a tool. … there are mathematicians who get inspiration from physics … they … have their criteria of excellence from mathematics. But where is the place for mathematical physicists?
Given that we’re both retired, Pekka, this might be an outmoded view of mathematical physics as a subject. We might have missed a maturing of it over the past few decades.
The Journal of Mathematical Physics states its purpose as “the publication of papers in mathematical physics–that is, the application of mathematics to problems in physics and the development of mathematical methods suitable for such applications and for the formulation of physical theories.”
The mathematics department at UC Davis has 13 faculty in mathematical physics and 18 in applied mathematics covering 18 topics half of which could be considered physics. The applied mathematicians there say “The question of what is applied mathematics does not answer to logical classification so much as to the sociology of professionals who use mathematics.”
One might ask a few highly respected mainstream mathematical physicists how they classify those who make “very questionable claims about physics.” And also whether “mathematical physics” used to have a more pejorative connotation.
Highly respected mainstream AI researchers would answer “old guard AI” to the former and “definitely yes” to the latter. You would have a better idea than me whether that analogy works for mathematical physics.
But it does raise the interesting question, who first conceived the idea of the biotic pump theory, Gorshkov or Makarieva?
I don’t know whether the journal Foundations of Physics has improved in the meantime, but my take on it 15 years ago was that it was where questionable claims could be sent without harsh review.
Vaughan,
We get here back to semantics and to what certain expressions mean intuitively to various persons.
Much of the work you list is what I described as physics inspired mathematics. The main point is that such science is mathematics, not physics. Having it in the mathematics department tells the same message.
It’s clear the some excellent work has been done on the borderline between mathematics and physics. One name that comes to my mind immediately is Penrose. Checking what Wikipedia says about him, the first description is “mathematical physicist” while I have always thought that he is mathematician, not a physicist with any epithet. One of my colleges and friends was professor of physics and was interested in the theories of Penrose and that lead one of his students to become professor of mathematics at the University of Helsinki. This guy started as a mathematically oriented theoretical physicist and is now a mathematician, not a mathematical physicist.
@PP: We get here back to semantics and to what certain expressions mean intuitively to various persons.
Exactly so. Computer graphics, computer systems, AI, and theoretical computer science are very different in character, yet most at Stanford doing them are in the computer science department and call themselves computer scientists. However at least half of the theoretical computer scientists would fit in perfectly well in a mathematics department, where they’d then be considered applied mathematicians.
Much of the work you list is what I described as physics inspired mathematics. The main point is that such science is mathematics, not physics.
That’s a boundary that could be moved more freely if the physicists were ok with the likes of Penrose calling themselves physicists. To the extent that they’re not that might be a difference from computer science.
Do you see a boundary between foundations of quantum mechanics and operators on Hilbert space? Would it make sense for a university to put the former in the physics department and the latter in the mathematics department, thereby preventing their respective practitioners from having neighboring offices and obliging them to walk a long distance to visit each other?
Mathematics describes physical phenomena given various assumptions but quantum mechanics could describe physical phenomena at the molecular level much more specifically.
Non-linear and non-ergodic dynamic systems are not mathematically tractible in classical science, so the challenge is for scientists to become more conversant with QM and chaos theory so as to more realistically model observed phenomena and to test various hypotheses as they are developed.
In particular, the present gap between climate models and empirical data is problematic.
Vaughan,
We are all a bit myopic, we see clearly small differences in what’s familiar to us and lump together what’s less familiar.
Having once been a professional theoretical physicists I see clear limits for what’s physics research (but I couldn’t exclude modern chemistry in any other way than saying that by definition, chemistry is not a subfield of physics). The line between physics and mathematics is drawn in water and moving around that line should be made easy. Both sciences have benefited greatly from work done in this bordering area. The great polymaths of the past, like Newton, were both mathematicians and physicists.
In my case the problem was that an expression formed as epithet + physicists means intuitively that we are discussing about a physicists, not a mathematician. Perhaps that expression is used only because physical mathematicians doesn’t sound good and because it’s not easy to find a better one.
Irrespectively of the above, I’m not happy, when I notice that a person calling her or himself a physicist with or without the epithet mathematical shows severe lack of understanding of physics. On several occasions that has happened with people claiming to be mathematical physicists. Thus there must be a significant population of bad mathematical physicists who fail on this point. A number of them have written ridiculous papers as climate skeptics.
physical mathematicians doesn’t sound good
Yet physical chemist sounds better than chemical physicist. Maybe starting a department of physical mathematics would help.
I’ve been reading the page http://www.bioticregulation.ru/pump/pump.php with considerable interest. As empirical evidence in support of the “biotic pump theory” the page cites a 2007 paper by Saleska et al as follows.
For example, in October 2007 an interesting prediction of the biotic pump theory was confirmed, namely that natural forests should increase transpiration during droughts (Saleska et al. 2007 Science 318: 612). Increased evaporation leads to intensification of the upwelling fluxes of moist air and of horizontal influx of moist air from the ocean, to offset the adverse effects of the drought. Forests that do regulate the water cycle are expected to behave like this. This behaviour was confirmed with satellite data on leaf area index in Amazon forests during the 2005 drought.
“Interesting prediction” is an understatement. I was shocked! For trees to increase transpiration during droughts would on the face of it seem suicidal: one would expect them to cut back sharply on transpiration so as to conserve what limited moisture was available to them.
Instead we have trees that have evolved to the point where they somehow “knew” that, by giving up their moisture to the atmosphere during a drought, the resulting condensation would bring them a huge ROI by sucking moisture in from the ocean air.
This was a mind-bogglingly new concept for me in the evolutionary biology of the plant kingdom. Trees, we find, are as smart as triffids. Who knew?
Figuring that Saleska et al must have rocked biology with this discovery, I went to their paper to confirm their findings.
What I found was a real let-down. Here in summary is what they said.
1. [Models predict that] water-limited vegetation responds promptly to initial drought by reducing transpiration (and photosynthesis), which in turn exacerbates the drought by interrupting the supply of water that would otherwise contribute to the recycled component of precipitation.
Ok, so Saleska et al must have checked this out and found that the models were wrong: droughts must have increased transpiration and photosynthesis. With bated breath I read on.
2. We used satellites to observe whether an Amazon drought in fact reduced whole-canopy photosynthesis.
What? They weren’t observing transpiration? But the biotic pump theory page says they were.
Instead of transpiration, what the satellite data showed was increased greenness, resulting from increased photosynthesis.
Nowhere does the paper say anything about increased transpiration, in fact it assumes decreased transpiration, in agreement with everything in the literature, not to mention common sense.
Regarding greenness they wrote,
3. Increased greenness is inconsistent with expectation if trees are limited by water but follows from increased availability of sunlight (due to decreased cloudiness) when water is not limiting—if, for example, trees are able to use deep roots and hydrologic redistribution to access and sustain water availability during dry extremes.
That is, in drought conditions trees don’t increase their transpiration to summon water from afar. Instead they shut down transpiration, just as every biology text with anything to say on the matter will tell you, and rely on “deep roots and hydrologic redistribution” for their water.
What drought does is to reduce clouds, thereby increasing sunlight, hence photosynthesis, hence greenness.
That is what Saleska et al found, by examination of the July-September 2005 data on the enhanced vegetation index (EVI) from the Terra satellite’s Moderate Resolution Imaging Spectroradiometer (MODIS).
Incidentally, for the benefit of any trees reading this thread, I’m not claiming you aren’t smarter than triffids, just that I’m sure you’ll be able to come up with more convincing evidence that you are than we’ve been shown so far. I wouldn’t want the tree mafia knocking on my door.
Vaughan Pratt — if I understand your objections above you assert that forest can increase photosynthesis while they decrease transpiration. Please check that idea online or with any trusted ecologist friends. You will find that within any given forest (with minor nuances) the two measures are tightly and positively coupled.
How do the trees manage? Well in Eastern Amazonia study shows some trees have roots to 18m and are thus accessing deep soil water reserves. The key prediction is that the presence of this productive forest impacts local weather patterns (not a typical monsoon swing).
So givent aht the point was misunderstood let me try again:
much of tropical South America experiences a prolonged drier season—but without a clear switching of air currents flowing to and from the coast (Zhou and Lau 1998). The forests provide an explanation as they remain green through the dry season by accessing deep soil moisture reserves (Juarez et al. 2007, Myneni et al. 2007). The resulting dry-season evaporation does not wholly overcome the seasonal influence of lower air pressure at sea, but it keeps the difference small and increases the likelihood of ocean-interior transport and terrestrial rain.
Hope that helps
Sorry (small screen) – that should have said “So given that the point was misunderstood …”
Vaughans in the game but 0 for 1 – triffids don’t have roots.
@DiA: If I understand your objections above you assert that forest can increase photosynthesis while they decrease transpiration. Please check that idea online or with any trusted ecologist friends. You will find that within any given forest (with minor nuances) the two measures are tightly and positively coupled.
Well, certainly, otherwise why would that assumption have been built into the models? It makes complete sense that increased transpiration would permit increased photosynthesis. Hence whenever one observed an increase in photosynthesis one would expect an increase in transpiration, with the latter explaining the former.
In hindsight we can see that the model builders never thought to ask whether anything besides increased transpiration could cause increased photosynthesis. Everyone simply assumed they were locked together.
The combination of deployment of the MODIS satellites starting in 1998 and a widespread drought in the Amazon during July-September 2005 would appear to have provided the first opportunity for ecologists to observe an alternative cause of increased photosynthesis besides increased transpiration. I’m not an ecologist (you and my wife both have the advantage over me of being ecologists, she was advised by Peter Raven on the fourth floor of the Herrin Labs at Stanford where the ecologists hung out back then), so you tell me whether there had been any earlier such opportunity.
Saleska, Didan, et al theorize that drought increases sunlight which increases photosynthesis. You seem to be claiming that when photosynthesis increases so does transpiration. That would be understandable when increased transpiration is the cause of increased photosynthesis. But why do you believe it is still true when increased sunlight is the cause of increased photosynthesis?
Saleska, Didan, et al claim no such thing. Moreover one would not expect it in drought conditions unless it served some purpose.
What do you propose as that purpose?
@CH: Vaughans in the game but 0 for 1 – triffids don’t have roots.
Robert, since neither you nor I have a clue about ecology, we could happily spend all Saturday contradicting each other on this point.
In this case John Wyndham is the go-to ecologist. Wikipedia interprets his novel as follows.
The base of a triffid is a large muscle-like root mass comprising three blunt appendages. When dormant/docile, these appendages are rooted into the ground and are used to draw nutrients, as with a normal plant. When active, triffids use these appendages to propel themselves along at a moderate walking pace.
You’re welcome to argue with either Wyndham (if you have the requisite contacts) or the relevant Wikipedia talk page as to whether triffids have roots, but I’m going to stay out of that one. 1 for 0, sorry.
Strike 3 – he’s out but has appealed to his wife’s authority.
Photosynthesis requires carbon dioxide, water and sunlight. Carbon dioxide is not commonly limitted – and plants adjust the size and density of stomata in carbon rich environments to reduce water loss. I have often speculated that might the most insidious threat to the hydrological cycle.
@CH: Photosynthesis requires carbon dioxide, water and sunlight.
First true word you’ve uttered lately, Robert.
But could you kindly explain how the poor plant is now obliged to surrender some of its precious water to the atmosphere?
Wouldn’t that be insanely stupid of the plant to do such a thing? What possible purpose could that serve?
Good questions. Indeed it’s been a huge ecological and biophysical mystery. Now we have an answer — the poor forest plant wants to run the biotic pump. This is how this seemingly insane water “wasting” evolved and was maiintained by natural selection. It just worked the right way.
http://judithcurry.com/2013/01/31/condensation-driven-winds-an-update-new-version/#comment-296070
was me.
In this case John Wyndham is the go-to ecologist.
In the trouble for Lichen he suggested
“This is not the age of reason, this is the age of flummery, and the day of the devious approach. Reason’s gone into the backrooms where it works to devise means by which people can be induced to emote in the desired direction.”
Mrs Pratt seems to have changed vocation from a botanist to and ecologist
Well Vaughan – I would really like to discuss in tedious detail the root system of an imaginary plant with you. But let’s just say they do not have 18 metre deep roots.
But my qualifications include both hydrology and environmental science. I have spent quite a lot of my life modelling and thinking about the hydrological cycle in all it’s gory details. This includes what happens in the root zone and in leaves believe it or not.
To be immodest Vaughan – my words are often both true and beautiful.
Plants in Australia can close stomata to reduce water loss. Groundwater is very deep across much of the country and so not easily available to the poor little tree. Not always the case of course. I am working on an inland site with groundwater only metres deep. It may have something to do eith the nearby impoundment.
What was the question? Oh yes – you are still thinking that the tree is water limited – but the 18m roots access the groundwater table.
@maksimovich: Mrs Pratt seems to have changed vocation from a botanist to and ecologist
Had you been either one you’d have known that they’re not mutually exclusive. Her work under Peter Raven was on the ecology of fritillaria. In 1970-72 we would drive around California in our beaten-up 1963 Volvo looking for the very rare stands of fritillaria that could be found on serpentine scree slopes that we’d have to scramble up. We’d put little bags around them to keep pollinating insects out and come back a year later to see if the plants had managed to pollinate themselves, as Peter had conjectured they would based on their pistil geometry. Unfortunately they hadn’t. Lot of work for nothing.
@CH: But my qualifications include both hydrology and environmental science. I have spent quite a lot of my life modelling and thinking about the hydrological cycle in all it’s gory details. This includes what happens in the root zone and in leaves believe it or not.
So let’s get serious and exploit these qualifications. You drive.
The triffid roots was the clue. You assumed that the plant is water limited because of a drought. Too much time in Australia – but even here the root systems of big gum trees are 10 or so metres deep.
With much deeper roots and shallower groundwater – water limitation may be less of a problem. Light limitation in dense rainforest is always an issue – with more light providing more scope for photosynthesis. Transpiration can be reduced when leaves dry out – but if there is no water limit then transpiration reduces heat stress.
@CH: if there is no water limit then transpiration reduces heat stress.
Excellent point. So you’re making the Coolgardie safe argument that plants are motivated to evaporate moisture because it keeps them cooler. According to the biotic pump theory the evaporated moisture subsequently condenses and draws in more moisture.
That could conceivably turn out to be harder to refute quantitatively than some of the other arguments that have been proposed here. Pekka, your thoughts on this?
Vaughan,
Perhaps I should not go into my views on the plant physiology. Some points are, however, clear.
– Under dry sunny conditions at low latitudes the low altitude relative moisture remains low with all the evaporation the plants may provide. During the night the situation may perhaps change.
– The condensation of the moisture higher up in the atmosphere does not draw in moisture. That’s purely an error by the proponents of the theory. What may, however, have some influence of that nature is the influence of the moisture on the density near surface. This effect is, however, counteracted by the lowering influence of the evaporation on the temperature. Therefore condensation leads usually to less uplift and drawing in less moisture from oceans, but there may be conditions where the effect is the opposite.
I haven’t spent much time thinking about the biotic pump. The extreme errors the authors make in the paper discussed here leads me to suspect everything they say and which is not accepted more widely. They have lost their credibility by the paper and by their stubborn refusal to admit obvious errors.
@PP: Some points are, however, clear.
All three are excellent.
1. Under dry sunny conditions at low latitudes the low altitude relative moisture remains low with all the evaporation the plants may provide.
In fact the plants might be providing most of the moisture in the air, since they’re presumably the principal transport between the atmosphere and the underground water. However the source would presumably be immaterial if the low level of humidity (whatever its origin) is insufficient to drive the hypothetical biotic pump effectively.
During the night the situation may perhaps change.
I’d guess less moisture from plants at night — less need for cooling, turgor pressure http://www.youtube.com/watch?v=44igB-0PVqA etc.
2. The condensation of the moisture higher up in the atmosphere does not draw in moisture. That’s purely an error by the proponents of the theory. What may, however, have some influence of that nature is the influence of the moisture on the density near surface. This effect is, however, counteracted by the lowering influence of the evaporation on the temperature.
So the water vapor makes the air more buoyant (by reducing its molar mass, water vapor having only 18/29 the molar mass of air) but at the same time cools the air by converting sensible heat to latent heat thereby reducing buoyancy. Which is stronger? Now we need Tomas.
We know by reversing earlier calculations (done by several of the paper’s critics starting in 2008, see here for a straightforward version) that evaporating 1 mole of water in 1000 moles of air (an increase in number n of molecules of 0.1%) lowers the air temperature by 0.5% when T is around 300K. However to compute the impact on buoyancy as opposed to volume, that 5x factor must be further increased to 5*29/11 = 13X because the buoyancy of water vapor (m.w. 18) in air (m.w. 29) is only 1 − 18/29 = 11/29. (The buoyancy of air in air is 1 − 29/29 = 0.)
Therefore condensation leads usually to less uplift and drawing in less moisture from oceans, but there may be conditions where the effect is the opposite.
Given that the reduction in buoyancy due to cooling is 13x the increase due to reduced density, it is hard to imagine what those conditions might be.
3. I haven’t spent much time thinking about the biotic pump. The extreme errors the authors make in the paper discussed here leads me to suspect everything they say and which is not accepted more widely. They have lost their credibility by the paper and by their stubborn refusal to admit obvious errors.
Indeed. Their many equations overwhelm the reader and make it hard to see the wood for the trees. (Apologies to anyone I’ve overwhelmed with my calculation of 13X above.)
@VP: evaporating 1 mole of water in 1000 moles of air (an increase in number n of molecules of 0.1%) lowers the air temperature by 0.5% when T is around 300K.
On second thoughts I don’t think this is right. It is the foliage that is cooled by evaporation, which in turn cools the air. One must therefore reduce my factor of 13 to reflect the relative heat capacities of the foliage and the air. For example if in equilibrium 2/3 the heat remains in the leaves and the other 1/3 enters the atmosphere then the ratio would be reduced to 13/3 = 4.3.
My guess would be that the heat capacity of the air would dominate (surely the volume of air in a forest is more than 1000x the volume of the leaves), in which case the cooling effect would still be an order of magnitude greater than the buoyancy of water vapor effect (but no more than 13X).
@VP (call this Q1): [In a drought] What possible purpose could [increased transpiration] serve?
@MM: Good questions. Indeed it’s been a huge ecological and biophysical mystery.
Q2: Is there anyone outside your group who has claimed that reduced moisture (e.g. drought) increases transpiration?
Now we have an answer — the poor forest plant wants to run the biotic pump.
Q3: Assuming a positive answer to Q2, wouldn’t a simpler and more plausible answer to Q1 be that the plant was merely “sweating” more in response to the increased heat of a drought?
Fans of Occam’s Razor would argue against the need for a more complicated explanation.
Interesting post VP. Now that you have “retired” from paid work I truly appreciate your posts,excepting only when it gets personal – and that applies to all contributors ;) and thank you for your time and effort.
Thanks, Peter. I’ll try to be less personal in future. I gave it a shot in my reply to Tomas Milanovic citing “On Floating Bodies.”
Anyone interested in Saleska et al, could start here
Thanks for that, Eli.
However it seems to be mainly a defense of the greenness finding itself. The closest Saleska comes to saying anything that might bear on moisture (other than the word “drought” itself) that I could see was
Even if satellite “green up” does in fact represent an increase in photosynthesis (as we think), could this in fact be a symptom of the trees compensating for the increased stress of the drought? The bottom line “carbon balance” of a tree depends on both photosynthetic uptake and respiratory losses, and it is almost certainly the case that those losses (which were not seen by the satellite) increased under the hotter and drier conditions of the drought as well.
The broken link to “Scientists speak: Amazon “myths” are not debunked” at the bottom refers to this response by 18 scientists conducting “research on Amazon forest, climate, and/or fires.” While they’re as unhappy as Saleska with the BU press release about the Samanta et al paper, they’re less concerned about the paper itself, which as pointed out by both Lewis and Saleska also demonstrates greenness as seen from satellites.
The 18 “pairs of boots on the ground” so to speak do say however:
The new study [Samanta, and by implication Saleska since the observations agree] contributes to our understanding of interpretations of data retrieved from satellites, but it does not prove or disprove anything about what is really happening on the ground.
So not only can we infer little or nothing about transpiration from the satellite data, this group is telling us we can’t infer much at all about what’s happening on the ground from satellite images.
This makes the Saleska paper an even less compelling point of support for the biotic pump hypothesis. We really need to ask those on the ground whether they’ve observed increased transpiration resulting from drought.
Simply place various shapes of objects in the liquid in question (water, honey, whatever), variously oriented, and observed that the resulting level of liquid always rises in direct proportion to the object’s weight.
My God V.Pratt!
But thank you to prove my point that your hypothetical mathematics ignoring “physicists” are able to say any stupidity and not notice that it is wrong.
Let’s take a bath of liquid volume V1 and an object of volume V2.
Assume V1=Sxh – your “physicists” cannot assume that because they don’t know mathematics and can’t suspect how easy it is to compute any volume..
Putting the object in the bath leads to a new level H = (V1+V2)/S
So the level rises by H-h = V2/S
To falsify your “physicists’s” absurd claim one will need a piece of advanced mathematics that is indeed far beyond their abilities.
We have a claim that V2/S = k.W where k is some universal constant and W is the weight of the object.
Follows that W/V2 = K/S where K = 1/k is still constant.
But W/V2 is the specific weight of the object and the claim means that the specific weight of ANY object is inversely proportional to the surface of any bath I may have a mind to put it in.
So your “physicists” have actually proven the impossibility to put lead in the ocean because the ocean’s surface is too large. We will pardon them though because nobody taught them how to add or substract. God forbid higher mathematics like dividing or multiplying.
That beats Monty Python :)
@TM: So your “physicists” have actually proven the impossibility to put lead in the ocean because the ocean’s surface is too large.
Those familiar with Volume 1 of Archimedes book “On Floating Bodies” will have had no difficulty seeing that I was proving Proposition 5, “Any solid lighter than a fluid will, if placed in the fluid, be so far immersed that the weight of the solid will be equal to the weight of the fluid displaced.” (T.L. Heath’s 1897 translation.) That is, floating bodies displace their weight in fluid.
The case of lead in water that you cite is covered separately in Proposition 7, “A solid heavier than a fluid will, if placed in it, descend to the bottom of the fluid, and the solid will, when weighed in the fluid, be lighter than its true weight by the weight of the fluid displaced.” Archimedes proves Proposition 7 differently from Proposition 5, though today we would treat 7 as a generalization of 5 and ignore whether it floats or sinks.
If your claim is that the greater generality of Proposition 7 prevents an empirical proof like the one I offered for 5, it would be interesting to run your claim up the flagpole and see how many physicists salute it.
As a practical matter one should conduct the empirical demonstration of Proposition 5 in a small vessel since accuracy falls off with increasing size of vessel, the ocean being an extreme case.
Vaughan’s in there batting but is 0 for 2 – you get that sinking feeling if your relative density is greater than 1 and you displace your entire volume. Archmimedes was last seen wandering naked in the street muttering wtf?
@CH: Vaughan’s in there batting but is 0 for 2
Ok, I get that you don’t like me, CH, but could you please pause just a moment and explain your problem with my empirical proof of Proposition 5 of Volume 1 of Archimedes’ “On Floating Bodies?”
If your point is merely that Archimedes’ Proposition 7 is the one that drove him to streaking then I don’t understand the problem. I wasn’t proposing to prove that proposition in the first place.
@CH: Archmimedes was last seen wandering naked in the street muttering wtf?
Projecting again, are we, Chief?
Projecting is a trifle derivative – lost any of the novelty that is the key to humour. You should try to be witty and original – failing that verbose and eccentric will keep the punters confused.
Simply place various shapes of objects in the liquid in question (water, honey, whatever), variously oriented, and observed that the resulting level of liquid always rises in direct proportion to the object’s weight.
I have nothing against Arcimeded, although it seems that your original proposition is not strictly true if the density of the object is greater than water.
Vaughan old buddy – I am devestated that you think I don’t like you.
@CH: I have nothing against Arcimeded, although it seems that your original proposition is not strictly true if the density of the object is greater than water.
Point taken. My only excuse is that I’d written something longer than I thought was needed, but in the process of editing it for brevity I chose to delete the statement of the proposition I was proving (namely that a floating body displaces its weight in fluid) thinking it was obvious. In 20/20 hindsight, bad choice.
Projecting is a trifle derivative – lost any of the novelty that is the key to humour.
There is novelty in “derivative?”
Vaughan, you have probably realized by now that due to your past contributions, many here will jump on construing something you say in such a way, however absurdly, as to try to discredit you. It is part of the mud-slinging process to see what they can make stick and diminish your standing, but a lot of it just makes them look small.
‘1. Resulting from or employing derivation: a derivative word; a derivative process.
2. Copied or adapted from others: a highly derivative prose style.’
Projection is used too often – not novel and therefore not amusing. Archimedes muttering wtf instead of euraka might be.
Honestly – such a minor mistatement doesn’t diminish you in my eyes at all. Would that even be possible?
No need to make a Federal case out of it – in fact your honesty is refreshing and welcome. If it weren’t for FOMBS – I could make a smilie now.
@Jim D: many here will jump on construing something you say in such a way, however absurdly, as to try to discredit you.
That’s all well and good, Jim, but what about when I’m wrong, which I am some of the time?
On the other hand why should I complain when they’re the ones crying wolf when I’m right?
Very noisy environment.
If it weren’t for FOMBS
Friends of Barton Middle School?
I was struck by the way a statement of Archimedes Principle made by you, and that everyone understands the meaning of, was then extrapolated to something Archimedes didn’t mean and attributed to you. That takes a special effort of some kind, and deserves a prize.
Oh, fan of …. Sorry to be so slow..
I can spell Archimedes right – I just don’t know why I don’t.
Robert, had I signed the Hippocratic Oath I’d probably have felt obliged to make you feel better with a complete list of my own typos. That may have something to do with why I didn’t follow in my parents’ footsteps.
It’s very amusing to read Cicero’s description of his discovery of A’s tomb. Re: H, one swears and follows, not signs after.
===========================
Every day in every way, I find Cicero more and more useful.
==================
If bein’ wrongs a crime – I’m servin’ time.
Vaughan Pratt | February 16, 2013 at 5:09 am |
If it weren’t for FOMBS
“Friends of Barton Middle School?”
Dyslexic?
Chief Hydrologist | February 16, 2013 at 4:12 am | Reply
“If bein’ wrongs a crime – I’m servin’
timeconsecutive life sentences.”Fixed that for ya!
Vaughan Pratt | February 16, 2013 at 3:37 am |
@maksimovich: Mrs Pratt seems to have changed vocation from a botanist to and ecologist
“We’d put little bags around them to keep pollinating insects out and come back a year later to see if the plants had managed to pollinate themselves, as Peter had conjectured they would based on their pistil geometry. Unfortunately they hadn’t. Lot of work for nothing.”
I knew a girl once that had trouble getting pregnant. Told her to put a bag over her head during sex and see if that helped. Sure enough it did.
Vaughan Pratt | February 16, 2013 at 3:21 am |
@CH: Photosynthesis requires carbon dioxide,
watera *hydrogen source, andsunlight.“First true word you’ve uttered lately, Robert.”
Not quite. Fixed it for ya!
* http://link.springer.com/article/10.1007%2Fs11120-006-9040-5
Gee I wonder where the hydrogen comes form and what the light is for. How much plant growth happens under artificial light? Dope comes to mind – and seems an apt metaphor for springer.
Chief Hydrologist | February 16, 2013 at 3:35 am |
“To be
immodestdelusional Vaughan – my words are often both true and beautiful. ”“Plants in Australia can close stomata to reduce water loss.”
Fixed the first sentence.
Second sentence. Interesting. I didn’t know Australia had radically different plant species that could accomplish gas exchange without stomal opening. This is equivalent to saying humans can conserve water loss by not breathing. While that’s technically true the methodology is fatal. It’ll kill the plant in just the same way too.
Botany is obviously not your strong suit.
Higher CO2 level in the atmosphere makes gas exchange go faster and limts the time and/or diameter of stomal opening. Less water is lost that way. That is how higher CO2 bestows greater drought tolerance. Now you know.
Chief Hydrologist | February 16, 2013 at 3:12 am |
“Photosynthesis requires carbon dioxide, water and sunlight. Carbon dioxide is not commonly limitted – and plants adjust the size and density of stomata in carbon rich environments to reduce water loss. I have often speculated that might the most insidious threat to the hydrological cycle.”
Wrong again. Reduced water use is coincidental. Higher rate of gas exchange in higher CO2 atmosphere requires less stomal opening. This is normally accomplished on the fly by irising the stoma open and closed as required. It is said that the stomata are the closest thing that plants have to a muscle. Signaling molecules are also generated in response to CO2 concentration such that a generation of leaves that experiences higher or lower than normal CO2 generates more/less of the signaling molecule in effect passing the history along to the next generation of leaves which then optimize stomal size/density accordingly.
It might help if you stopped guessing and dreaming about what goes on inside plants and actually read something on the subject instead.
As long as I don’t have to share a cell with springer.
One adaptation to higher CO2 is reduced size and density of stomata. The carbon needed for growth and energy can be obtained more easily – the loss of water involved in the gas exchange process can be reduced in this biological response to changing CO2. Water is a critical limiting factor for many land plants. This obviously doesn’t happen ‘on the fly’ as it is a morphological change to leaf structure that occurs with new growth.
‘The carbon cycle is important, but so is the water cycle,” Dilcher said. “If transpiration decreases, there may be more moisture in the ground at first, but if there’s less rainfall that may mean there’s less moisture in ground eventually. This is part of the hyrdrogeologic cycle. Land plants are a crucially important part of it.’
Certainly less moisture for shallow rooted and germinating plants. Changing CO2 changes the hydrology and ecology of global systems.
http://www.sciencedaily.com/releases/2011/03/110303111624.htm
There are a number of plant responses to desiccation. Only one of them is opening and closing of the stoma by guard cells. The adaptation includes closing at night, opening in the early morning, closing for extended periods, etc. There are plants that are much better at this than others – these are plants adapted to low rainfall. You can plant rain forest plants in the desert – but they won’t survive. It all depends on what plants are adapted to breed to. No simple one size fits all description is rational.
http://plantsinaction.science.uq.edu.au/edition1/?q=content/15-3-5-drought-stress-and-adaptive-responses
http://www.intechopen.com/books/abiotic-stress-in-plants-mechanisms-and-adaptations/stomatal-responses-to-drought-stress-and-air-humidity
Water is the key and there is a trade off between gas exchange to obtain carbon and water loss. You need to go to reputable sources rather than springer to get a balanced view.
2/17/13 demarcation comment.
Pingback: AWED Energy & Environment Newsletter | Turn 180
Pingback: Accepted at Atmospheric Chemistry and Physics after nearly two and half years : eloquentscience.com
I’m not sure if Stanford is a reputable source but some may think it is. My emphasis.
http://www.sciencedaily.com/releases/2008/11/081125113112.htm
Key Link In How Plants Adapt To Climate Discovered
Nov. 28, 2008
What plants do in response to heat stress:
How Plants Chill Out: Plants Elongate Their Stems to Cool Their Leaves
http://www.sciencedaily.com/releases/2012/05/120521132758.htm
Wow. They grow longer stems to space leaves further apart to facilitate airflow between them when hot. When colder they pull it all in closer together much as a person curls up into a ball when cold. Who knew?
Ellison is right about one thing. I does pay to go to reputable sources. But I already had. He needs to take his own advice. Physician, heal thyself.
@DS: Wow. They grow longer stems to space leaves further apart to facilitate airflow between them when hot. When colder they pull it all in closer together much as a person curls up into a ball when cold. Who knew?
Indeed. And when Warren Buffett’s investments go south he seeks out the nearest soup kitchen to avoid starvation. What’s your point?
Somewhat late, but if condensation drives winds by a decreasing the number density of water vapor in the air, then as clouds form they should shrink in size due to the negative pressure. OTOH, if the release of the heat of condensation is the primary driver, they should expand. Guess what?
Eli got video
When we see a cloud increasing in size this does not mean that the air is expanding. It is the area of convergence (shrinking) that is growing. Some details of the argument can be found here here.
Eli, they tend to do both. There is a general expansion of the convective entrainment zone during the morning to early afternoon when there is addition solar energy being absorbed at the surface and in the clouds. The average albedo of clouds is ~50% which is considerable energy. In the afternoon/evening the convective entrainment zone shrinks with reduce solar producing the effect noted by Anastassia. I suggest you read up on the marine atmospheric boundary layer which is more of a “global” situation than just a few puffy clouds. I think Anastassia may focus a bit more on that “special” case in the future :)
Get your stadium wave placards here. Two sides, many colors.
===========================
Pingback: Convergence, Shrinking and Implosion versus Divergence, Expansion and Explosion during Condensation | Stormy Science
Pingback: Forest climate and condensation | Climate Etc.