Table of contents for Greenhouse Experiment
I am starting to read a few posts on bulletin boards about this experiment, and some points need to be made clearly.
1. The main purpose of the experiment is to verify that maximum G=1.5, the greenhouse factor, not to replicate the conditions in the real atmosphere. That is, it is to test the Su=3OLR/2 constraint, not the B=H(1+τ)/2 lapse rate constraint.
2. I am amazed that no-one has done this. Reading around the web, experiments show profound ignorance of the greenhouse effect, replicating only the spectroscopic absorption by increased CO2. That is they deal in the effect of increased optical depth τ and hence the relationship described by B=H(1+τ)/2 — not the real greenhouse effect constraint. Because CO2 absorbs more IR, they mistakenly assume that shows earth’s temperatures will increase.
3. Negative responses to date have seemed to say that the experiment is irrelevant to B=H(1+τ)/2. Yes it is, because that is not what it is looking at. This illustrates the general ignorance of the independent energetic constraint, the greenhouse maximum, that Ferenc Miskolczi asserts is imposed as Su=3OLR/2.
4. RealClimate published its toy model of the greenhouse effect here. This comes the closest it appears anyone has seen of a documented theoretical basis for the greenhouse effect. It results in a greenhouse factor of 2. That is, the maximum temperature is twice the blackbody temperature.
5. If the maximum greenhouse effect is 1.5, then RealClimate and the conventional understanding of greenhouse effect is wrong.
6. This is an easy issue to settle. Demonstrate the greenhouse effect exceeding 1.5 with an apparatus similar to the one I have shown. (Don’t raise spurious arguments.) I’m happy to be shown wrong on this.
7. RealClimate’s toy model assumes a temperature discontinuity between the atmosphere and the surface. This could only arise in a perfect vacuum at the earth’s surface. Hence it is an unrealistic model.
8. The factor of 1.5 can be derived in a number of ways. Ferenc has one in his paper, via eqn M7. But it can also be derived from simple heat flow equations across the gradient:
dH/dx =F/2 as the heat flows both in and out equally at any point.
H = Fx/2+c by integration.
When x=0 at the outside of the jar, F=c, the heat flow equal to solar input. When x=1, at sensor on the inside of the jar, H = Fx/2+F or H=3F/2, as required.
Note that the important difference from the RealClimate model is that there is no discontinuity between the sensor and the environment, a much more realistic assumption.
267 responses so far ↓
Jan, thanks again for the test/experiment and now the explanation. It should certainly raise some comments around the blogosphere.
CoRevs last blog post..16 Nov 2008 Articles & Interessting Finds/Posts
David, this comment relates to your previous post in this series, which doesn’t seem to have a comments section. I couldn’t understand this:
Is E a temperature or a flux (or an energy?)? Are you speaking of TOA (where maybe F=OLR)? Or is it an atmosphere of uniform temperature?
David, I don’t believe there is any natural greenhouse factor that you can derive by elementary means. The realclimate example was just to give a simple analogy of how obstructing IR increases temperatures. It is equivalent to what you previously referred to as the steel greenhouse, where a gas that totally blocks IR is compressed to a single narrow thin layer. The factor of 2 is arbitrary; if you compress to two separate layers, you would get a factor of 3.
For Venus, OLR is only a bit higher than Earth, but Su is huge; a factor much higher than 2.
Nick #3
Don’t you mean the OLR is a little higher on earth than for Venus? (about 20 K worth)
Venus fact sheet
Did you know that around about 50km altitude on Venus the temperature and pressure for Venus are similar to Earths ground level?
So pressure level : pressure level Venus us actually cooler than we expect it to be.
test
David: Fascinating experiment. When I was a graduate student I worked two summers with the Forest Service in Colo. doing research on solar drying of lumber. We used large greenhouses, clad with PE. I remember seeing empty closed greenhouse temperatures as high as 160 F, but I don’t remember what the ambient was. For the 3/2 relationship, it would have had to have been over 100 F. Very possible. Wish I still had all the data.
#3 Nick, I guess if you argue that abstractions, including the RealClimate one, are of no relevance then it will difficult to discuss this with you. RealClimate’s post says that this is their basic understanding of how the greenhouse effect works and we have known it for 100 years (dummy).
The atmosphere is the greenhouse gas layer they refer to. There is no suggestion that CO2 and H2O concentration start, stop, then start again anywhere in the atmosphere as a 3 layer model would require. These things are mixed in the atmosphere. If the maximum greenhouse effect is 1.5 and not 2 as their model says, then they have been wrong for 100 years. If its 2 and not 1.5 then I am. It can’t get much clearer.
#7 David
I’m quite comfortable with abstractions. But before taking RC’s toy example too far you should note what they say about it:
Now the single layer is a key issue. If you gather the GHG material into a single thin layer, as they say, you get a maximum G factor of 2. That doesn’t mean that it will actually be 2; just that that is the max for a huge amount of GHG. But if you gather it into two layers the max is three. Into three layers, the max is 4. And so on.
Continuing abstractly, when you gather into more layers, each layer is more permeable than the single layer. So the increase in G is not as great as the increase in the max would be. But it can increase, and is not limited by 2. And as you increase the layers, you approach more closely the real continuum atmosphere.
That’s just as well, because Venus is a hard test. the argument applies to it just as well as to Earth, but the G factor there is much greater than 2. So something is wrong with any argument that leads to 1.5 or 2.
Nick: I think the layers have to be separated by a vacumn to be proper ’steel greenhouse’ layers with a unit G factor. That simply is not the case. Venus is another issue, in that it is fully opaque to short wave. It is more truely semi-infinite. That is different from a setup transparent to short wave and opaque to IR, where the sun shines on the surface. Venus is also possibly emanating significant heat from the surface. It has to be solved for different conditions.
David Stockwells last blog post..Model of Global Warming
David #9
No, they don’t need to be separated by vacuum - just by IR transparent gas. The model just gathers the GHG in bands - everything else can remain in place. The mode of heat transfer is IR only between layers, then within, conduction of the heat intercepted by layers to keep the temperature on both sides the same (that’s why the layers are assumed thin, so conduction is perfect).
The greenhouse theory actually just applies to any heat source at the surface - incident sunlight has the same effect as volcanoes or whatever. Something warms the surface, and some fraction G of the resulting IR gets through. On venus, the only complication is that the incident sunlight warms the atmosphere directly, but that doesn’t explain the massive GE.
I thought the definitive model description for AGW was by Spencer Weart;
http://www.realclimate.org/index.php/archives/2007/06/a-saturated-gassy-argument/
This is the classical layered atmosphere with vertical CO2 barriers trapping the ascending IR. I don’t get it. Everyone agrees that CO2 does its absorbing at around 650cms as the Cabauw measurements confirmed; at this level SU effectively =ED, but it is the equation ED=SU [1-TA] which is pertinent; this is consistent with Stefan-Boltzman and the Pielke Snr et al paper: “Unresolved issues with the assessment of multidecadal global land surface temperature trends.” Basically at each ED=SU[1-TA] point there is a LTE; a Local Thermodynamic Equilibrium is defined as the distance an air molecule travels before encountering a temperature change; this is the mean free path of the molecule; with the GH effect at this low level there can be no discontinuity between the surface and the LTE above; the GH exchange is balanced and the CO2 molecules have been completely thermalised within the LTE air parcel; Miskolczi assumes at (c) on p3, that the atmosphere is in LTE but it is regionalised so that, as he says on p12, “At the top of the atmosphere the net IR radiative flux is equal to the global average outgoing long wave radiation.” This is entirely consistent with Pielke Snr and Stefan Boltzman and requires no anthromorphised gaia-like planet which knows what IR should go where to achieve the Kirchhoff equilibrium. What happens at each LTE has been described by Chilingar and to a lessor extent, Douglass and Christy; the LTE is convectively lifted to the CEL, which at 7-8km, is far below the stratosphere outward emission point (OLR) demanded by the Weartian model. At the CEL the LTE parcel is no longer temperature distinguished from the surrounding air and its CO2 molecules can therefore emit outside the no longer thermalised LTE; and here is the issue; this air has lost energy rising; the isotropic emissions, ED and EU are equal, but the classical model says the CEL ED adds to the CO2 absorbing level ED and therefore adds energy to the surface; but it can’t do this for number of reasons; the surface LTE will maintain itself because this is where the GH effect dominates; if there is more CEL ED then there will be more SU EU because the CO2 surface layer will cause it; no extra SU EU will be absorbed by the surface CO2 because that surface CO2 has been thermalised and saturated; any extra CEL ED will become SU EU passing through the saturated window. Alternatively, the Specific Humidity at the surface increases cooling of the effect of extra CO2 at that level. This appears to be happening. Is there anything happening which is consistent with the Weartian model?
The characterization of the RealClimate simple Greenhouse model is not accurate. It is meant as an conceptual explanation of a much more complicated calculation that requires computers, so that people can understand the concepts involved and get a rough idea of the physical principles that operate, while putting up some reasonable numbers.
For example, the real atmosphere has a complicated dependence of temperature versus height, while they use a single value of temperature in their calculation. A more complex simplified model is given in the July issue of the Physics and Society Newsletter at the AIP web site.
The maximum possible greenhouse effect parameter, which reflects the ratio of the upward radiation from the ground, to the incoming solar radiation using this simplified model is 2. The actual value, which was derived from actual experimental data on upward radiation from the ground is given as
1.62 if you carry out the calculation using the value obtained for lambda.
This is obviously less than the theoretical limit, but clearly greater than 1 and is clearly greater than the 1.5 that you claim is the alternative theory’s upper limit.
So it is the real data on upward radiation that is providing the value of the greenhouse parameter, not the simplified theory which illustrates how it works. The simplified theory is not meant to calculate the value of the greenhouse parameter itself, and cannot possibly do so.
The validity of the theory of of Ferenc Miskolczi seems in doubt to people with a knowledge of physics who have read it carefully. People over at the Eli Rabbet blog have pointed out that Miskolczi’s invocation of the virial theorem with regard to the earth’s atmosphere is invalid, and his interpretation of the concept of local thermal equilibrium, and the misuse of Kirchoff equations.
The paper itself was published in an obscure Hungarian journal because it would not pass peer review in a respectable international journal.
As has been pointed out, the simple experiment with the air and the bottle is trash, unless it is shown that the variation in radiation from the sun, and the IR and optical absorption of the container as well as the convection have been taken into account.
Eric: Is RC’s model “trash,” too?
Eric:
“The validity of the theory of of Ferenc Miskolczi seems in doubt to people with a knowledge of physics who have read it carefully. People over at the Eli Rabbet blog have pointed out that Miskolczi’s invocation of the virial theorem with regard to the earth’s atmosphere is invalid, and his interpretation of the concept of local thermal equilibrium, and the misuse of Kirchoff equations.”
The problem with all their criticisms about the Kirchoff Law and the Viral Theorem is that the empirical data seem to support M’s concepts, regardless of whether M might be using the wrong terminology. The empirical data does not seem to support any alternative theories, especially the surface discontinuity nonsense. And no apparent warming of the troposphere at 10 K in the tropics, as predicted by all models.
Jae,
Firstly, upper limits that are derived from equations that don’t apply cannot be relied on to be correct. If they are consistent with data, it could simply be an accident.
Second, as I have already mentioned, the empirical data shows a ratio of upward radiation from the surface to incoming solar radiation of 1.62, which is larger than the supposed upper limit of 1.5 which is claimed as the upper limit according to Mizkolczi. This upward radiation is on top of the other means by which heat is removed from the surface to the atmosphere, transpiration and convection.
It would seem that a flawed derivation, and contradiction by experiment should doom this paper to oblivion.
The apparent lack of warming of the upper troposphere in the tropics was a result of data analysis problems which have been corrected, by recent papers.
With these corrections it has been shown that the upper troposphere is warmer than the surface, and has been warming.
The supposed conflict was with model elements unrelated to the absorption and re-reradiation by GHG’s, but rather with the adiabitic approximation used to calculate the temperature of moist air which rises by convection.
This physical principle is alleged to be used by all climate models.
Jae said,
Eric: Is RC’s model “trash,” too?
It is not a real model. It is a conceptual framework to help non scientists understand the gist of what is done in a real computer calculation. It fills a need to transmit to lay people, such as you and I, what the more complex equations and computer calculations are doing. The real model is too complex to describe in detail.
The value of lambda, as I pointed out is not derived from the equation, but taken from real measurements of upward IR radiation, which average 390W/M2, versus 240 W/M2 for the actual incoming solar radiation not directly reflected into space.
The real model is too complex to describe in detail.
Now, that is really scary. How can you validate such a model?
“The value of lambda, as I pointed out is not derived from the equation, but taken from real measurements of upward IR radiation, which average 390W/M2, versus 240 W/M2 for the actual incoming solar radiation not directly reflected into space.”
Where can I get those measurements?
Jae,
If complex science scares you , maybe you should stop reading about it.
Complexity is a fact of life.
The measurements are from a classic paper published by Kiehl and Trenberth (1996).
See the diagram in the following link for a description of the earth’s energy balance.
http://oceanworld.tamu.edu/resources/oceanography-book/radiationbalance.htm
Eric: I’ve read that paper. It is certainly no more “convicning” than Miskolczi’s. Care to explain the discontinuity at the surface?
BTW,
“The paper itself was published in an obscure Hungarian journal because it would not pass peer review in a respectable international journal.”
That is a cheap shot, Eric, and I don’t need to tell you which logical error is involved with the statement. And Steve McIntyre is showing over and over that “peer review in international journals” doesn’t mean much in the “climate science” world, especially in cases where close buddies and former co-authors are the reviewers. Climate science has a black eye, IMHO.
My link, which you refer to as a paper, is an online text book in Oceanography for college students, by a research professor from Texas A & M.
The research you are quarreling with can be found here:
http://www.atmo.arizona.edu/students/courselinks/spring04/atmo451b/pdf/RadiationBudget.pdf
I wouldn’t have had a chance at the cheap shot, if the paper would have stood scrutiny by the atmospheric physics community. I admit that I am not a mind reader, but the mistakes found in the Miskolczi paper, are consistent with an expectation that its publication in a main stream journal would not be possible. A paper of such importance, if it were valid, would get more favorable attention in the climate science community, and citation than this one has.
If you have some specific flaws to point out in the research by Kiel and Trenberth, it would be of interest to know what it is, and why you find it unconvincing.
Proofs of errors in the Kiehl-Trenberth distribution
See Zagony: http://hps.elte.hu/zagoni/Proofs_of_the_Miskolczi_theory.htm
“My link, which you refer to as a paper, is an online text book in Oceanography for college students, by a research professor from Texas A & M.
The research you are quarreling with can be found here:”
I was referring to K&T, which doesn’t appear to me to be based on any better evidence than Miskcolczi’s papers. Still waiting to hear about the surface discontinuity.
BTW, p.200 in KT starts out with this sentence: “We must rely on model calculations to determine the surface radiative fluxes.” Hmmm, that is not empirical evidence. Miskolczi does better.
“5. If the maximum greenhouse effect is 1.5, then RealClimate and the conventional understanding of greenhouse effect is wrong.”
From the expert — Spencer Weart;
“We have to be careful here. One problem in the debate is people (engineers for example) who understand just enough math to get into trouble.”
Are engineers ignoring this silly talk ?
After all, the Earth needs terraforming
Eric:
LOL, that TAMU article is hardly “deep science.” Looks like it was written for a high school primer on the subject. Same-ol, same-0l stuff that’s everywhere. Why did you link it?
““We have to be careful here. One problem in the debate is people (engineers for example) who understand just enough math to get into trouble.”
Not only engineers, but chemists. I may be in that camp, but I am learning, and that is my goal
David - you are calculating a ratio that depends on the temperature achieved inside your greenhouse to “ambient” outside temperature. There are much better greenhouses than the one you have devised - thick double- or triple-paned glass for instance to let the light in and avoid conductive heat loss. Put your little experiment (with sufficient insulation) out in space, and you can easily achieve hundreds of degrees C, with an “ambient” temperature of 2 K or so, for a T^4 ratio in the millions. Run your experiment in the arctic or antarctic and see how it does. Or just wait for freezing winter-time temperatures: real greenhouses exist with no internal heat source that maintain 30 C inside with outside temperatures of -20C, which gives you a T^4 ratio over 2.
In other words, this experiment may tell you something about how physical greenhouses behave, but it has nothing to do with the way so-called “greenhouse” gases work, and your greenhouse ratios are meaningless in relation to the actual atmospheric GHE.
Hi Arthur, You can’t use the ratio of temperature of x/2 in space, you would have to use OLR. In that case, you would have a situation identical to the earth, and the result should be 3/2, as the earth demonstrably obtains. Being a home experiment (at home) I rely on the ambient as an approximation of the black body temperature for a given solar intensity. Thats why I think it is approximately valid to run it for a clear sky in the middle of the day, but not when its cloudy, sunrise, sunset etc. Cheers
It could be, thats what I want to find out. So far, in the configurations I have used, the expected maximum has almost been achieved, and increasing GHG has reduced the temperature. No doubt if the max is exceeded in a reasonable configuration without increasing isolation (mirrors etc) you would agree it is falsified. If its not, then would you concede?
Black body temperature for normal (clear-sky) solar intensity of at least 1000 W/m^2 is found on the surface of the Moon in the middle of the lunar day. Earth’s surface doesn’t achieve those temperatures because convection quickly wicks the energy away from the surface to warm the atmosphere (and at the ocean surface convection works similarly in a downward direction), and the associated heat capacity is sufficient to maintain a relatively stable atmospheric temperature between day and night. A physical greenhouse (like your experiment, or like the inside of a parked car) gets warm under the sun because it prevents that wicking-away process, confining almost all the absorbed heat locally. As Nick has been explaining to you. Suitably insulated, your experiment could easily achieve temperatures of 90 C at least, well over your 1.5 limit even with normal ambient temperatures - and that 90 C is accessible even when the outside temperature is freezing too.
You have a clear position, except for the “suitably insulated” part. I think the outside of the vessel must not be insulated so it can approach ambient.
David Stockwells last blog post..Model of Global Warming
Arthur Smith and Eric,
You apparently do not have a problem with the result of lowered temp with the ADDITION of GHG??
Jae said,
Eric:
LOL, that TAMU article is hardly “deep science.” Looks like it was written for a high school primer on the subject. Same-ol, same-0l stuff that’s everywhere. Why did you link it?
I linked to it, because it has an energy balance diagram for the earth showing the values of upward radiation and solar radiation absorbed by the earths surface. You asked where those numbers in the RealClimate article came from. I also linked the original source of those numbers which is a scientific journal article.
The diagram has appeared in many links which you can find on this subject on the internet. If a piece of research is cited so often, it must be an authoritative and accepted piece of work. In the field of science that counts for a lot.
In the denier blogosphere, it apparently doesn’t mean anything.
Do you claim that the scientists and professors don’t know what they are talking about? Do you really have enough knowledge to make that judgement?
“Do you claim that the scientists and professors don’t know what they are talking about? Do you really have enough knowledge to make that judgement?”
Come on Eric, gimme a break, already. There are serious disagreements among serious “scientists and professors” on these matters. So, yes, I think they generally know what they are talking about, but they cannot all be right. I am also a scientist and I was trained in the (maybe old-school?) tradition of not accepting everything that is written, not even if it is in a peer-reviewed journal.
I am very well aware of all these “radiation cartoons.” Miskolczi, for one, doesn’t agree with them. Especially with regard to the surface discontinuity (a subject which you seem to be avoiding, altogether). What I’m trying to discover is who’s right.
Arthur:
“Earth’s surface doesn’t achieve those temperatures because convection quickly wicks the energy away from the surface to warm the atmosphere (and at the ocean surface convection works similarly in a downward direction), and the associated heat capacity is sufficient to maintain a relatively stable atmospheric temperature between day and night.”
Right on! Then why do the radiation cartoons suggest that convection accounts for only 24 W/m^2? BTW, you are getting very close to saying that thermal storage has something to do with the “greenhouse effect.”
kuhnkat: The response to that will be that “the optical path is too long for the GHGs in the little greenhouse to have an effect.”
But the optical path is certainly long enough in the tropics. However, temperatures there never go over about 32-33 C. Why not, if additional GHGs cause more heating? Surely, we should see some 40 C readings in the heat of the day, jsut like in dry areas, but we don’t. And where is all the “positive water vapor feedback” in the tropics. Heck, using the IPCC concept, it should be 50-60 C during the day in the tropics, no? It seems to me that is that there is an over-supply of GHGs, and all the water feedback is negative, effectively creating a thermostat. I’ll bet that 32-33 C is the maximum attainable average temperature for the Earth, and it would be a very, very cloudy place, indeed.
David #30
A point certain folks around here seem to be missing, and that ambient was constant throughout the experiment and in fact it was a controlled trial.
31 kuhnkat // Nov 17, 2008 at 10:14 pm
“Arthur Smith and Eric,
You apparently do not have a problem with the result of lowered temp with the ADDITION of GHG??”
I don’t know what you are talking about.
I have said that the results of the experiment have nothing to do with absorption and emission properties of GHG’s.
The GHG’s in the glass, as Nick has pointed out, absorb a miniscule fraction of the incident radiation because the total number of absorbers is so small. It takes Km of air to absorb a sizeable amount of radiation.
They molecules get their energy from the walls of the container by collisions.
Eric, 37
“They molecules get their energy from the walls of the container by collisions.”
Are you sure? I don’t think you are viewing this correctly. The glass is transparent to the SW radiation. As I understand the workings of a greenhouse, the SW is absorbed by the “floor” and contents of the container, reemitted as IR, absorbed by ghgs, and is then transferred to the rest of the air by THERMALIZATION . SOME IR energy is also absorbed by the glass and thermalizes the air inside. And some IR energy “leaks” by conduction through the glass and provides energy to the outside air by (this is why the container’s temperature doesn’t quite get to the black line, and it is a minor deal). Have I got this wrong?
IR Radiation has to pass through many KM of atmosphere for significatnt
fraction to be asbosbed. The upper atmosphere has no walls with which to collide, so radiatin is the princple way it gains energy. Collisions with the walls are more frequent for gas molecules in the jar. All the molecules participate, not just the GHG’s. I am quite sure that the absorption of IR radiation by GJG’s is not significant in determinint the temperature of the gas in the glass.
It is the job of texperimental scientist to support the conclusions of his experiment by showing that the mechanims I have mentioned are negligible.
Why not direct your questions at him?
I forgot to include thermalization by contact with the “floor” and other materials in the greenhouse.
It would be interesting to fill the greenhouse with pure oxygen and see how fast it warmed up. In that case, all the thermalization would have to occur by contact with the surfaces. There would be no absorption of IR by the “air.” I’ll bet that would be much less effective in heating the air, but maybe not?
“IR Radiation has to pass through many KM of atmosphere for significant fraction to be absorbed.”
CO2 exhausts its absorption at around 650/cm; the KT model shows this, as does HARTCODE and the Cabauw verifications.
Eric:
“IR Radiation has to pass through many KM of atmosphere for significatnt
fraction to be asbosbed.”
That just doesn’t make sense. The air at the surface in the real world is not just heated by contacting the surface (if the air was over cold water, it would never heat). It is heated primarily by thermalization reactions, wherein the IR-excited GHG molecules collide with the N2 and O2 molecules (and they with each other, and so on). And this could not be a major mechanism, if the path length were “many kilometers.” In your world, you could not have LTE!
David: does your “greenhouse” have a black floor?
Eric # 32:
Once again, we find the only reason people can find to believe something is that everyone else believes it. I like this quote.
David, you noted
in response to my comment that, properly insulated, you should reach 90 C. Of course the outside of the vessel is in contact with the ambient atmosphere, that defines “outside” - the question is the degree of insulation between outside and inside - i.e. a well-sealed container (no convective heat flow), with thick double-paned glass (to reduce conductive heat flow) would be better controlled. Just like a well-insulated house, except with transparent walls and ceiling to allow radiative energy flows. That should be the ideal set-up for your experiment (to the extent you are actually testing any hypothesis here), and under those conditions you should be able to reach quite a bit higher temperatures, pretty much independent of the external “ambient” temperature.
Kuhnkat asks about the “lowered” temp you observed - well, you didn’t exactly add greenhouse gases, you added liquid water. That has to evaporate (thanks to the low humidity), which consumes energy, lowering temperatures. Not exactly unexpected. The radiative effects of the added water vapor will be small as Nick has mentioned - but in fact they also would likely be negative (leading to cooling), because the entire inside of your container is at a pretty uniform higher temperature. This is what happens with GHG’s in the stratosphere: they radiate at a higher level when at a higher concentration, resulting in a *cooling* of the stratosphere. That cooling is at least partially responsible for the imbalance between incoming and outgoing radiative energy that results in warming of the planet beneath. There’s certainly no reason to believe, based on the standard understanding of the atmospheric greenhouse effect, that additional water vapor in such a container will leave it warmer rather than cooler. A lowered temperature is quite unsurprising for both of these reasons, though I expect the evaporative latent heat effect is much more significant.
Jae asks, quite reasonably:
Remember this is the *net* *average* heat flow from the surface to the atmosphere caused by convection. During a hot clear summer day, at low latitudes, the heat flow is quite high. At night, and as the atmosphere circulates between low and high latitudes, the heat flow drops or even goes negative (the atmosphere warming the surface). The net is much less than the peak flow rates. The latent heat flow is also quite variable, but I don’t think it goes negative anywhere (evaporating water always cools the surface and warms the atmosphere where it condenses).
Once again, what David’s experiment demonstrates is the traditional physical greenhouse effect. If we actually wanted to test an analog of the atmospheric greenhouse effect on a laboratory scale you would probably need to meet the following conditions:
* A system capable of holding a large temperature differential from top to bottom (i.e. not a gas or liquid, they can’t hold large temperature differentials on a laboratory scale thanks to convection - some sort of solid material with low thermal conductivity perhaps)
* Several versions of the system with differing bulk absorption levels in the thermal spectral region, but otherwise with identical heat capacity, thermal conductivity, etc.
* A steady source of heat on the bottom, perhaps an electrical element (the fact that incoming solar energy is radiative is irrelevant to the basic physics of the atmospheric greenhouse effect - as long as the surface is heated some way or other, that’s all that matters).
* A bottom surface that is close to a black body (perfect absorber and radiator) at thermal wavelengths.
* All surfaces except the heated side of the bottom free to radiate into a cold vacuum (surround the laboratory system by a cooled enclosure).
Then the relevant hypotheses are (a) the temperature of the lower, heated, surface will be higher the higher the bulk infrared absorption level is, and (b) it will rise with no particular limit on its relation to the ambient cold enclosure as that absorption is increased.
Arthur:
“Remember this is the *net* *average* heat flow from the surface to the atmosphere caused by convection. During a hot clear summer day, at low latitudes, the heat flow is quite high. At night, and as the atmosphere circulates between low and high latitudes, the heat flow drops or even goes negative (the atmosphere warming the surface). The net is much less than the peak flow rates. The latent heat flow is also quite variable, but I don’t think it goes negative anywhere (evaporating water always cools the surface and warms the atmosphere where it condenses).”
Baloney, methinks. Again, the surface of the planet during the day would be exactly like a closed greenhouse if it were not for convection. You can average that effect anyway you want to, and you can’t get a measly 24 w/m^2 for convection. As Miskolczi demonstrates, the transfer of energy from the surface occurs by non-radiative means. Radiation is small potatoes, when convection occurs (think, again, about why you have a fan in your computer, instead of a simple heat sink). And I again pose the question: What about the disconuity at the surface? It doesn’t happen in real life, only in radiation cartoons (which is why the term “cartoon” fits so well). None of you true-believers* have addressed that very important question, to my knowledge.
*I label you thus, because you are just too damn certain of yourself. Like you know it all. You end up looking like a preacher! If you were being a straightforward scientist, you would find at least SOMETHING that is not absolutely certain. Hell, even IPPC didn’t come out with 100 % certainty.
Arthur: Thanks for your lengthy description of your ideal experiment. Which illustrates why I think people need to be patient in dialogue while they work out what each other is saying. I think you misunderstand.
You can’t discharge into a cold vacuum. It has to be at the same temperature as the black body temperature of in incident radiation. What you perhaps could do is sit the whole experiment on an insulating sheet, and heat the inside dark body with the same energy level as what it would take to bring it to the ambient level. Alternatively, if you wanted to sit it at 2K, then you could only supply the energy equivalent of 2K to it. The temperature then could at most be 4K, unless you live in RealClimate world, where it could be 4K, 6K, 8K, 10K whatever you like. In practise I think its going to be at 3K for reasons mentioned.
David Stockwells last blog post..Model of Global Warming
I think I understand, David. Brilliant! Boundary conditions apply.
Yes, Arthur is a confident chap; and another thing he hasn’t addressed is the lack of a THS or Stratosphere cooling independent of volcanic residual effects.
David #30
“Should not be insulated”? But why not? Or to put it another way, what kind of boundary are you trying to achieve? You started with plastic, which has not only a membrane, but on each side a boundary layer of near-stationary air, which is a good insulator. You then went from plastic to glass, which is a bigger barrier to conduction. Why would adding more insulation alter the principle you are testing? I was going to suggest adding a layer of bubblewrap, which would probably increase the temperature substantially.
You are supposed to be testing a formula based on IR transmission. Is the wall supposed to be IR transparent? And is it? Why would the thermal conductivity matter?
It needs to meet the ambient boundary value. Perfect insulation would not do that, or it would take a long time to equilibrate. I am working on a post about this, so more later about insulation layers. Sorry I can’t answer all your previous tricky questions right away. Just too many right now.
Nick:
“You are supposed to be testing a formula based on IR transmission. Is the wall supposed to be IR transparent? And is it? Why would the thermal conductivity matter?”
No, the wall is not IR transparent for glass, but it doesn’t matter. I think the Wood Experiment back in the 20’s or 30’s showed that this doesn’t matter wrt the greenhouse effect. (He compared a NaCl greenhouse, which is transparent to IR to one with glass. Same temperatures inside, as long as the sun’s NIR was blocked by glass. Long ago, I did experiments with polyethylene greenhouses, which are also quite transparent to IR; same results as David got. David’s experiment makes sense to me. The outside of the greenhouse is at ambient temperature (or very close). That simulates the lower boundary condition. Move the greenhouse higher and higher in the atmosphere and the same thing happens. I think?
Eric
“IR Radiation has to pass through many KM of atmosphere for significatnt fraction to be asbosbed.”
Steady state fog above a swamp. (cloud chamber)
How to design a thin layer of fog into the experiment ?
Zagoni states that high altitude clouds insulate, equivalent to closing the 10 micron window. So if we get a semi-transpatent steady state fog. Can we expect Su=3OLR/2 ? Measure with a couple of photocells, up and down looking.
Nick #50
It’s Arthur that want’s him to build a solar oven in order to prove that things can only get hotter.
This is what the walls need to made of
My PC has been off-line for 5 weeks, so you can all rejoice at the unexpected break. But I’m back.
Philosophically, a thought experiment has troubled me for some time. It concerns the difference between heat and temperature in a gas, particularly in a rarified atmosphere.
The gas atoms/molecules have an energy which is related to heat. The faster they move, as a group, the more heat there is. But, something has to detect this heat and we use devices like mercury thermometers. Problem. If we have sparse gas molecules, they have a relatively large mass of mercury to interact with. Although they might be full of energy and moving fast and hot, there simply might not be a large enough number of them to intersect the mercury and change it significantly.
For related reasons, way out in space, the temperature might be the 3K micromave temp, but if you place an object like a satellite there, it can get hot from radiation. Radiation is waves not particles, but it simply complicates my thought experiment, which is to devise a heat measurement device that produces an answer correct in physics. I suspect that some of the above arguments touch on this thought experiment, especially when one is dealing with trace GHG and trying to determine how their real temperature is measured before being passed on to surroundings.
#55 Geoff
The concept which is relevant here, often misunderstood, is local thermodynamic equilibrium. This basically means that a temperature can be measured. It is often quoted as a requirement for Kirchhoff’s Law (the real one). It means that gas collisions are so frequent that processes involving them proceed faster than anything else you can measure. This includes the averaging of collision impulses that constitute pressure, or of kinetic energy exchanges that constitute temperature. LTE applies in the atmosphere except for the most rarefied levels.
Jan #54
ZnSe won’t work - it’s transparent to IR but not sunlight. But David is building a solar furnace. He added glass and it got warmer. Bubble-wrap would make it warmer again. He’s aiming, apparently, to show a max of about 60C, which insulation gets you towards, and more insulationwould exceed it. But what type of wall is wanted? Should it be transparent to all radiation? If so, how is it a test of anything except the ability to block convection?
Nick #57
“ZnSe won’t work - it’s transparent to IR but not sunlight”
That’s the general idea so it won’t be a solar furnace. It only excludes blue green light from the sun so it shouldn’t be a problem for some solar heat to get in.
#56 Nick
Thank you, i appreciate what you say. But my concern is not to assume LTE at some given density, but to work out the actual heat taken in by dispersed GHG molecules before LTE can be achieved. From lab experiments with GHG at high concentrations, we can work out how much heat and light is found, but it’s a rather more subtle exercise when the candidate gas is a few ppb in the atmosphere. My mind cannot make the leap from the GHG being a major component in an experiment, to it being a trace component. Why, other gases like nitrogen might partake of the incoming energy before the trace GHG even sees it. LTE is still established, but without any greenhouse effect in this scenario. So how much GHG in an atmosphere does one have to have to measure its temperature change in rection to energy input? Is it valid to extrapolate from high concentrations? What is the lowest concentration of methane, for example, whose presence can unequivocally account for a quantitative effect when light is passed through it?
Geoff #59
LTE is a property of gas in general at a particular temperature and pressure - related to mean free path of molecules. It’s not particularly associated with IR or GHG or other individual gas components, trace or otherwise. And its effect, as above, is to ensure that kinetic energy, for example, is evened out locally faster than any process you’re likely to try to measure. The practical effect here is that you don’t need to worry about GHG molecules being individually heated. They absorb IR as individual molecules, but the heat immediately becomes a property of the gas mixture locally.
“ZnSe won’t work - it’s transparent to IR but not sunlight. But David is building a solar furnace. He added glass and it got warmer. Bubble-wrap would make it warmer again. He’s aiming, apparently, to show a max of about 60C, which insulation gets you towards, and more insulationwould exceed it. But what type of wall is wanted? Should it be transparent to all radiation? If so, how is it a test of anything except the ability to block convection?”
No, I think more insulation would just raise the boundary temperature, and the 3/2 relationship would still hold.
Jae writes:
- yes, but think of the other consequences. Imagine our air was replaced by a medium with very high viscosity so it could not move about easily, and a density independent of temperature to prevent convective instabilities. Without convection hot “air” would stay in the tropics. The poles would be considerably colder, with no warm lower-latitude air coming in during the summer, and heading towards absolute zero during their long winter nights - the convective greenhouse does not help if the sun is not shining. So mid-latitude day-time convective flow is not the typical situation and the average convective greenhouse (removing net convective flow) does not actually warm the surface very much on average.
How would you get precise numbers on the average convective heat flux? Looking at a climate model’s convection would be a good place to start - in fact, I believe that’s what the Kiehl-Trenberth diagram is based on. Feel free to suggest a better way to measure or estimate it!
Jae also asks:
- actually, I have no idea what you’re talking about: the Kiehl-Trenberth diagram doesn’t claim, mention, or as far as I know depend on any discontinuity (in temperature?) at the surface. On the other hand, a (near) discontinuity in temperatures at the surface isn’t unheard of - have you ever touched your hand to a metal object sitting in the sun, or walked barefoot on a sandy beach on a hot day?
And yes, I am a “true-believer” in physics
David Stockwell writes:
There is no incident radiation in my proposed experiment - the heating comes from an electrical element on the bottom surface, which replaces the heating of Earth’s surface by incoming solar energy. The purpose of the cold vacuum is to try to closely mimic the situation for Earth as a whole, where there is essentially no incoming thermal radiation from surrounding space to our planet. You could do my proposed experiment at a variety of enclosure temperatures to see what effect that has - obviously, warmer enclosures will add some incoming thermal radiation to the system, resulting in slightly warmer temperatures. But even with a near-absolute-zero enclosure, infrared absorption in the system will keep the bottom surface warmer than the temperature you get from Stefan-Boltzmann, by a ratio that can be arbitrarily large.
cohenite writes:
I am not familiar with the acronym “THS”. Pinatubo warmed the stratosphere, but overall it’s been cooling - you seem to have some argument here with which I’m completely unfamiliar.
More on “discontinuity in temperatures at the surface” - in general, the temperature of the air (a few feet) above water can be quite different from the water temperature: if you don’t believe this, I challenge you to a swim in the Atlantic on a warm day in March! In late fall, or on summer nights, the water temperature can be quite a bit higher than that of the nearby air. Having continuous temperatures along some curve means maintaining local thermodynamic equilibrium along it, but if the curve crosses a material boundary, there is no good reason such equilibrium should be maintained: the constituents on one side of the boundary do not easily cross to the other, the boundary itself is a material discontinuity. Temperature may be the same on either side if the two sides have had enough time to equilibrate, but it’s a much slower process than actual mixture of materials within one side or the other. If it’s not instantaneously continuous, there’s certainly no reason it should be continuous on average either.
Anyway, what brings this up at all? Is this one of Miscolczi’s bugaboos?
Anyway, what brings this up at all? Is this one of Miscolczi’s bugaboos?
Yes. Please read M’s papers and Zagoni’s explanations. Then we could discuss. Easy to google.
The microlayer of water next to the air is probably in thermal equilibrium with the air. Otherwise, the absolute humidity of the air over moist areas would not be as close to saturation as it is (average of about 75% saturation at 20 C).
Arthur, you say:
“How would you get precise numbers on the average convective heat flux? Looking at a climate model’s convection would be a good place to start - in fact, I believe that’s what the Kiehl-Trenberth diagram is based on. Feel free to suggest a better way to measure or estimate it!”
I don’t know how to estimate the total effect of convection, but it’s got to be more than 24 wm-2. The temperature difference between the outside and inside of a greenhouse proves that to me, as I said before. All that extra heat is removed by convection. I think that Miskolczi says that virtually ALL energy from the surface is transferred to the atmosphere by non-radiative means (convection, latent heat). I think that is right. As any engineer knows, thermal radiation is a very poor way to get rid of heat. And Mother Nature is at least as smart as the engineers.
Alex Harvey // Nov 18, 2008 at 1:07 am
“Eric # 32:
If a piece of research is cited so often, it must be an authoritative and accepted piece of work. In the field of science that counts for a lot.
Once again, we find the only reason people can find to believe something is that everyone else believes it. I like this quote.”
It is cited because the experts in the field find it useful. That is a good recommendation in my book, if I am unable to find the time to read it and evaluate it for myself.
Of course one doesn’t take advice on things from people who are ignorant of a subject, no matter how many of them repeat the same thing.
A cited reference is not the same as an old wives tale.
In designing the experiment one must be aware that under ideal conditions, a temperature change of about 0.017C is the most that could be expected under the best of circumstances as a result of changes in absorption reemission in atmosphere inside the vessel that is used.
Check out my calculation in the Home Science experiment post no 70.
http://landshape.org/enm/science-experiment-at-home-to-disprove-global-warming-theory/#comment-174565
I’m fascinated byArthur’s stratospheric cooling explanation:
“The radiative effects of the added water vapor will be small — they also would likely be negative (leading to cooling), because the entire inside of your container is at a pretty uniform higher temperature. This is what happens with GHG’s in the stratosphere: they radiate at a higher level when at a higher concentration, resulting in a *cooling* of the stratosphere.”
I assumed that stratosphere cooling by AGW is mainly due to more heat energy being trapped farther down: ie energy balance. Then again, on reading the IPCC scientific basis report they seem to conclude that it’s ozone depletion causing most, if not all of the cooling. Does that put Arthur’s mechanism in a poor 3rd place among the best guess mechanisms?
Arthur:
When M talks about discontinuity at the surface, it is the average global temperature, not a transitory temperature of surfaces of varying emmissivity. The standard theory predicts a considerable discontinuity at the surface in the global average temperature, which does not occur. Hence, M’s theory is more accurate in that (and other )respects.
I still suspect you are missing the point. The greenhouse factor G being looked at G is a ratio OLR/Su or Tearth/Tblackbody if you like. The Tblackbody is a product of certain flux level, heat or radiation. The Tearth or Tgreenhouse is a result of that same flux level entering into a bandpass diode-like enclosure . The flux can’t be arbitrary, but must be the same as the background of the apparatus.
jae // Nov 18, 2008 at 4:52 pm 65
“Arthur, you say:
“How would you get precise numbers on the average convective heat flux? Looking at a climate model’s convection would be a good place to start - in fact, I believe that’s what the Kiehl-Trenberth diagram is based on. Feel free to suggest a better way to measure or estimate it!”
I don’t know how to estimate the total effect of convection, but it’s got to be more than 24 wm-2. The temperature difference between the outside and inside of a greenhouse proves that to me, as I said before. All that extra heat is removed by convection. I think that Miskolczi says that virtually ALL energy from the surface is transferred to the atmosphere by non-radiative means (convection, latent heat). I think that is right. As any engineer knows, thermal radiation is a very poor way to get rid of heat. And Mother Nature is at least as smart as the engineers. :)”
Why should we care about what Miskolzci says?
Measurements say the average upward radiation flux from the earth’s surface is 390W/M2, way more than the average energy conveyed by convection. If are quoting Miscolzci it shows he is full of sh–. As any scientist knows the results of measurements are the last word, assuming they are done properly.
You are deceiving yourself on 2 counts.
1) The convection is higher than 24W/M2 during the day time so the greenhouse gets significantly warmer at that time.
2) The net LW radiation lost to the atmosphere, after you subtract the downwelling from the upward radiation is actually about 65W/M2.
Arthur #63
Yes indeed and the diurnal cycle is a heat engine that pushes day sailors out to sea in the morning and back to shore in the evening. It’s the average over the day the year etc. that creates the static model that this equilibrium between surface and atmosphere is seen.
Eric #69
Are you sure you understand what is being said? Miskolczi’s paper says that a large portion of that 390w is compensated by downward atmospheric radiation so that the bulk of transport of heat is convective. It is a measured result. He states it specifically in the paper and that a study of convective transport (K) was beyond the scope of the paper (p26).
Radiation however is the only way that the earth can get rid of it’s heat and that is the process that he has been studying ( and calibrating the instruments that measure it ) for the past 30 years.
Eric:
I could probably say that you are deceiving yourself on only one count, that is absolute, positive, belief in the KT radiation cartoon.
eric and arthur: you might want to look at this paper, too. It also demonstrates that convection is far more important in transferring heat from the surface than is radiation: http://www.lpl.arizona.edu/~rlorenz/MEPRG.pdf
#60 Nick
You write “They absorb IR as individual molecules, but the heat immediately becomes a property of the gas mixture locally.”
I am questioning this. It seems more likely to me that most heat acts on the major local gas mixture before the GHGs get a look in. I also think we are generalising too much about the relative process speeds of molecular absorption, kinetic transfer etc especially in sparse atmospheres and the difficulties of using thermometry to explain them (as say, opposed to spectroscopy/photon counting). I’ll cogitate some more.
Meantime, if Admin does not object, or if there is interest in a new thread, I would like to mention some absurd generalisations of the Uncerainty Principle to everyday TV life and to note that the term “black body radiation” is heading in the same direction. It would be interesting to set a time and date for contributors to post a half dozen paragraphs on their understanding of black body radiation and its context in climate. I suspect that a comparison of responses would show enough diversity to need correction.
In this fruit-salad post here, I would also note that the near surface climate processes on land are being generalised too much. Many land areas have water tables that can be many metres down. Some evaporation, latent heat etc happens at these depths and the consequences can be but slowly transported to surface measurements. This observation impinges on questions of whether there should or should not be a land/air temperature discontinuity at a scale of close examination of detail and include factors such as relative humidity, night/day changes etc.
On a related topic, it is hardly fair to talk without qualifiers about surface sea temperatures when the temperature gradient in the sea varies so much with depth.
Sorry for the omnibus.
Geoff #75
You’re right about sparse atmospheres - that’s exactly when LTE doesn’t apply. It’s a measure of non-sparsity. It’s often described as a condition for temperature to be defined. Another version - there is just one temperature for all molecules (in a local region).
Geoff:
“Some evaporation, latent heat etc happens at these depths and the consequences can be but slowly transported to surface measurements. This observation impinges on questions of whether there should or should not be a land/air temperature discontinuity at a scale of close examination of detail and include factors such as relative humidity, night/day changes etc.”
It appears that the humid areas on land have the same absolute humidities as the ocean areas, on average of 30 years. The vegetation evidently takes care of this seeming inconsistency, by pulling that water out of the soil/water table and giving it to the air. ALL humid areas have an absolute humidity that is relatively close to (75% or so) saturation absolute humidity (on average over 30 years). It is the same as the tropics, where there is no shortage of water. See here, Hotterdryer.xls, figure 1. http://www.esnips.com/web/climate
Jan Pompe says
// Nov 19, 2008 at 12:03 am
“Eric #69
Measurements say the average upward radiation flux from the earth’s surface is 390W/M2, way more than the average energy conveyed by convection. If are quoting Miscolzci it shows he is full of sh–. As any scientist knows the results of measurements are the last word, assuming they are done properly.
Are you sure you understand what is being said? Miskolczi’s paper says that a large portion of that 390w is compensated by downward atmospheric radiation so that the bulk of transport of heat is convective. It is a measured result. He states it specifically in the paper and that a study of convective transport (K) was beyond the scope of the paper (p26).”
According to Trenberth’s figures even the net upward radiation into the atmosphere after subtraction of the atmospheric down welling is 66W/M2.
That is much larger than the 24W/M2 for convection given by Trenberth.
Does Miskolzci give a reference for his statement that contains the measurements to support what he says? It seems that he isn’t aware