" I've stated many times on BH my belief that an unvalidated model is no more
than an illustration of a hypothesis "

Not really a statement but obviously a fact isn't it??

"Isn't the lapse rate derived from (or supported by) ideal gas laws? Isn't the real-life situation less than ideal. For instance, no conservation of energy within the gas law where there is radiation in and out."

Gases don't follow the ideal gas law exactly, but the differences are small enough that you wouldn't notice, at least on Earth. (The CO2 near the surface of Venus might be a different matter.) The fact that there is heat entering or leaving due to radiation is a violation of the other assumption - that the change is 'adiabatic'. The explanation is that we are assuming for this calculation that everything has settled down close to equilibrium, which in absolute terms it has.

"What does the earth look like from space using a thermal imaging system? A fuzzy ball of relatively uniform temp? A glittering jelly of large differences?"

The apparent temperature varies quite a lot from equator to poles, and at different wavebands. But 'fuzzy ball' is probably the more accurate description. There is plenty of IR imagery around - at some wavebands you can see the clouds, at others you can't.

"NiV, why do you say that level is at 5Km. I thought it to be at the tropopause. Above which there is no lapse rate. Do the gases stop obeying the law at 36000 feet?"

The lapse rate is fairly constant up to the tropopause, but you get emission to space from all altitudes from the ground up to about 10 km, and the *average* emission altitude is about 5 km. (Manabe and Strickler 1964 give some graphs - it varies quite a bit with altitude.)

"Here's another. What's the range of measured lapse rates from the balloons? How much do they vary from the ideal?"

Generally pretty close, apart from when you are close to the ground. There's a daily cycle where the night-time temperature drops at the surface and the cooling spreads upwards generally about a km, before the day comes again. http://commons.wikimedia.org/wiki/File:Atmospheric_boundary_layer.svg Above this in the free atmosphere, significant departures from the MALR are rare. (As I understand it - I haven't checked in detail.)

"NiV: Isn't calculating what radiates to which inside the atmosphere just examining how greenhouse gasses distribute energy throughout the system and have in fact no bearing at all on average radiating temperature?"

In a convective atmosphere, yes, because convection cancels out the radiative effects up to the adiabatic limit. The internal radiative effects are important in the sense that without them the energy flows don't balance, and without them there would be nothing to push the atmosphere up against the adiabatic limit, but they're not what controls the temperature. If I turn the gas up on a pot of boiling water on the stove, the temperature is 100 C, and that can be explained either because that's the boiling point of water, or because the heat entering the bottom of the pot, the heat radiated from the sides and top, and the heat lost through evaporation all add up to balance when the temperature is 100 C. Both explanations are 'correct', and all of those factors are relevant to the physics - if you turn the gas off it will obviously have an effect. But one explanation is more 'useful' than the other.

"As confirmation of my view: If I were incorrect, surely there would be a downloadable pdf from the Met Office with a title along the lines "A Detailed Explanation of the Greenhouse Effect" that would answer all of Dung's questions?"

I think as far as they're concerned, it's quite useful for them if we waste our time going round in circles. There's no advantage to them helping us get our story straight. And I rather suspect many of them are a bit confused about the physics, too.

But the climate system is complicated, and precisely quantified answers are hard to calculate. Qualitative approximations can still be useful, though, and you can still have a correct understanding of the principle of some physical phenomenon without being able to model all the details. If you stand out in the rain, you'll get wet. You don't have to know the trajectory of every raindrop.

Dung's questions have answers, but they're not going to be perfect - certainly not without a huge amount of work. The only actual laboratory-scale experimental model of the greenhouse effect I know of is a solar pond - those CO2-in-a-bottle experiments are all bogus. There's no maximum GHE when all the IR is absorbed because it's due to the average altitude of emission to space, not the back-radiation. The contribution of CO2 to glacial cycles is unknown - the CO2-temperature record doesn't tell you that CO2 contributes to temperature rise but it doesn't rule it out, either, and either way it's impossible to quantify. In particular, ice cores don't record the humidity, which would have a bigger effect. And there are plenty of physical processes and reactions that can act on 17 year timescales, or indeed, any timescale. There are natural variations in climate at *all* scales. Bob Watson missed out the "other things being equal" caveat.

The answer to a lot of questions is "We don't know." We don't know for certain that the world will warm very much, but we don't know for certain that it won't, either. My view is that it's worth investigating further, but it's not worth acting on just yet. Other people disagree, and they're entitled to. This is supposed to be a democracy.

But I think the main problems are over feedbacks and impacts, and with model validation and scientific integrity. While there are aspects to even the best explanations of the GHE itself that one can quibble with, and the most common ones are simply abysmal, in the long run it's not a productive line of enquiry and it usually makes sceptics look bad. It's fair enough for people to ask questions trying to understand it, but the people who spam every thread with their certainty that it's all a crock do huge damage to sceptic credibility. And again, they have every right to their opinion and I wouldn't want to shut them up, but it really makes me wince every time.

Real life is messy and imperfect, though. We have to deal with what we've got.

rhoda,

The back-radiation argument, in my view, is a major impediment to understanding. I can sort of understand why the climate scientists took that line - it's more to do with the pedagogy of how they teach it. They start with Stefan-Boltzmann, they calculate how much the ground radiates at 15 C, and then they point out that this is more energy than it receives from the sun. Convective losses from the surface only make the problem worse. If energy is to be conserved, there has to be another input, and that turns out to be back-radiation.

But they're confusing an accounting necessity with cause, and making a perfectly familiar bit of physics seem more exotic than it is. Put an object in a box, such as a room in your house, and it will radiate IR according to the Stefan-Boltzmann law, and if the energy it emits is greater than it would receive out in the sunlight, it is most certainly greater than the sunlight it receives in the shade of your house. So how does the energy balance? Well obviously, the surrounding walls are emitting radiation as well, and if the walls and the object are at the same temperature, the radiation going each way balances and cancels out. All we are talking about is the tendency of enclosed objects to take on the temperature of their surroundings.

However, while it is an accounting necessity, it doesn't explain anything about why things are the temperature they are. We can see that easily enough with the example of a (convecting) pool of water, which certainly emits back-radiation, but where the cancellation is complete. Sunlight enters and is absorbed at the bottom. The warmed surface radiates, but the radiation is almost immediately absorbed, and re-radiated, and so on. If radiation was the only mechanism by which heat could be transported, even a shallow pool would soon boil. But convection is such that any excess of heat near the bottom causes the water to rise and eliminate the build-up. In a convective fluid, the convection completely short-circuits the resistance to radiative heat transport due to the fluid's opacity.

It's the same sort of principle as a boiling pot of water. Turning up the gas increases the heat input, but that doesn't change the temperature at all because all the extra energy just goes into evaporating more water which immediately escapes. The extra heating is exactly cancelled by an equal amount of evaporative cooling. But if the temperature drops below boiling point the evaporation abruptly stops and the heat starts building up again. The sharply non-linear process dominates the dynamics, and all other factors, like the IR back-radiation the pot absorbs from the surrounding kitchen, are rendered irrelevant. It exists, and is thermally absorbed in the normal way, but it has no effect on the *temperature* because the non-linear boiling process means its only effect is to make the boiling slightly more vigorous. If the walls of the kitchen were to become warmer (or if you put the pot in an oven) then it would boil faster, but not any hotter.

OK, back to the GHE. You ask where the radiation is thermalised, since the lapse rate can't heat the surface. Well, the radiation is thermalised any place it's absorbed. That's primarily at the surface, but also in the air due to the GHGs there.

However, there's no physical reason why the lapse rate can't directly warm the surface, although we have to be a little bit careful here because in the one-dimensional model we've been considering up to now it wouldn't. In the 1D model, net heat flows from the surface to the air, and from there to space, and we're ignoring any other effects. And we likewise have to be a bit careful about what is a physical mechanism and what is an explanatory 'cause' - the warming of descending air due to the lapse rate cancels the cooling that occurred when it rose, so the ultimate source of the heat is elsewhere.

However, in 3D the convection forms loops that transport the heat horizontally, and it can easily transport it to somewhere the surface is naturally cooler, and warm it. On Earth, moist air rises over the equator, rains out the water, dumping the latent heat into the air, and then rolls over and descends over the latitude bands just above and below. At altitude, even with the addition of the latent heat, the air is a chilly -50 C. The descending air warms adiabatically, and because it is dryer the lapse rate is steeper so the rise in temperature is even greater. If this descending hot dry air is warmer than the ground, then it will warm it. (Or more precisely, reduce convective losses since there will only be convection when the ground is warmer than the air.) The desert bands are hotter than the equator, not only because the dry descending air tends to be cloudless, but also from the more intense equatorial sunlight and the latent heat left behind by the tropical rain at the equator.

The same applies to the transport of heat polewards generally. Convection transports it horizontally, and the poles are warmer than they would be on pure radiative grounds alone. Indeed, on pure radiative grounds, the 6 months of darkness through the polar winters ought to cool things there to near absolute zero. It's only thanks to the atmospheric circulation that they don't.

The situation is even more extreme on Venus. There, almost all sunlight is absorbed at the level of the cloud tops around 50 km above the surface, and the surface itself is very dim and dark. Less than 10% of the incoming sunlight penetrates. It's like the oceans of Earth a few tens of metres down, in that you're getting relatively little input, and although the fluid is opaque to IR so there's little radiative transport, it's also strongly convective and so there's no real barrier to heat escaping. On Earth, it's pretty cold down there. On Venus, the only barrier - and the only fundamental difference from Earth's oceans - is the adiabatic lapse rate, that stops convective escape when the gradient falls below about 8 C/km, and that's why the weather forecast on Venus is always 'Hot'.

The physics is a little bit indirect, so it's not easy to ascribe 'causes' so straightforwardly, but in a sense yes the lapse rate does warm the surface. But I will admit, understanding that 'sense' is difficult.

Martin A,

"That's my rapid explanation prior to getting back to things that await my attention. Does it make any sense?"

It makes sense, and is the right explanation for the situation when there is no convection.

You can extend the same principle to the case when there are many thin concentric shells, all stacked on top of one another. In a pool of water, you can slice it into thin horizontal layers 20 microns thick, each layer being just thick enough to be totally opaque to thermal IR. The topmost layer emits the same energy upwards as the pond absorbs, and downwards as well. The layer below that must emit twice that energy both upwards and downwards for the top shell to maintain equilibrium. The third layer three times the energy, and so on. Thus, the radiation emitted by each layer increases linearly with depth, proportional to the number of layers, and thus the temperature increases with the fourth root of depth. A metre down we have 50,000 layers above us, and so the temperature in Kelvin must be the fourth root of 50,000 times the temperature at the top, or about 15 times the temperature. If the surface is at 300 K, the bottom of the pond would be at 4500 K!

That doesn't happen, because convection short-circuits the resistance to the transport of heat. But if you can somehow suppress convection, the effect comes back. That's how solar ponds work. But it's a fun calculation, and a good test to see if somebody understands how the greenhouse effect really works. Liquid water makes a very good greenhouse material - so can they explain why the oceans don't boil? I've known quite a few who couldn't.

The earth radiates as a BB and the IR is absorbed by GHG. The molecule of gas becomes excited, in other words, an electron moves to a higher energy level, increasing its potential energy. If the electron then drops back down again, the molecule will re-emit the photon in a random direction. This is the so called DWIR. However, in reality, collision between the excited molecule and other molecules in the atmosphere is much more likely. The collision transfers energy as kinetic energy resulting in higher molecular velocities. In other words, the molecules in that part of the atmosphere travel faster, collide more and result in a higher air temperature. This will result in expansion of the a and convection.

Down Welling IR must therefore be relatively rare because collisions take place almost continuously.

We were doing so well without the back radiation argument rearing its head. Fortunately there is a simple physics/engineering formula which explains things very well for our spheroid in a uniform environment;

q = εσ [T1^4 - T2^4] A

If you know the temperature of both, put them in the equation. If the answer is +ve then the q is flowing from T1 to T2 if it is negative the the q is flowing from T2 to T1. The energy available to do work is q. As back radiation by its very definition is from the lower of the two temperatures then it is doing no work (heating). The rate of work being done is proportional to temperature difference. Therefore with the small temperature difference between surface and atmosphere compared with that between atmosphere and space it can be seen that q will be far greater for the latter over the former. However, if you inspect a climatology diagram invoking back radiation you will find that to be reversed (typically 240W out to space and 333W atmosphere to surface).

But climatology doesn't stop there, oh no: using this they move on to declare that there is an imbalance (back to S-B but they started it) of 0.9W, recently revised to ~0.6W, that is the declared 'trapped heat' from anthropogenic CO2. If you believe that then you have just passed *1st year climatology.

*Including field exercises measuring back radiation locally.

"1. There is continued accumulation of heat in the global ocean despite the supposed "pause" in surface temperatures, and the ocean is where the overwhelming majority of the increased heat from GHGs goes;..."

This is pure speculation. The record of decent measurements at depth is too short to tell, there are several different reconstructions that are inconsistent with one another, and the error bars are huge (and bigger than they report). And if Bob Tisdale is right, it seems peculiarly localised in the Indian Ocean - not at all what you would expect from global warming.

It's not clear to me that there even is definite ocean warming (although the evidence inclines that way), and it's certainly not clear to me that it can be distinguished from ordinary random weather. It's the same fight we went through with the surface temperatures - correlation doesn't imply causation. Even if the ocean temperature is going up, that doesn't mean it's due to global warming. The oceans are probably as complicated, but far less easily observable, than the atmosphere. And far less well understood.

"... therefore claims that "global warming has stopped" are mistaken."

I would advise the most extreme caution about such statements - it's potentially making the exact same mistake the alarmists did about the 1990s rise in surface temperature. If weather is truly random, then when it next moves it's as likely to move up as it is down. If you stake all your credibility on it moving down, you can lose it all if nature doesn't follow your agenda. Even if it moves up, which it might well do, that doesn't mean AGW is starting again; if it moves down, it doesn't imply that AGW is entirely wrong. It might only mean that the natural component of the variation is bigger than either we or they think.

If you only say "global warming has stopped" in the technical sense of changes in surface temperature anomaly on a 1-2 decade timescale, it's pretty near unarguable. But if you start trying to extend that to underlying mechanisms, you're staking everything on an unknown. The alarmists did that, for short-term polemic advantage - and we've seen now how their gamble turned out for them. Do we really want to follow their example?

We don't know. There's no solid evidence that it is as they suggest, but there's no solid evidence that it isn't, either. And the difference of opinion about whether the mere possibility is sufficient to act on has nothing to do with the science, but is about values and politics.

Its worse than that.

We know that they don't know. They don't.

NiV: The lapse rate is fairly constant up to the tropopause, but you get emission to space from all altitudes from the ground up to about 10 km, and the *average* emission altitude is about 5 km. (Manabe and Strickler 1964 give some graphs - it varies quite a bit with altitude.)

They start with Stefan-Boltzmann, they calculate how much the ground radiates at 15 C, and then they point out that this is more energy than it receives from the sun.

*average* emission altitude is about 5km 'Greenhouse Effect' 33K. MALR 6.4C/km = 5.16km.

Hmmm.

First I would like to say that I have really enjoyed the comments on this thread. I read BH regularly, but don't normally comment. The following are my thoughts on some of the discussion questions, and are intended to view from a different angle. I hope they are not perceived as negative to what others have already posted, nor overly simplistic for those that are well versed in these discussions. I only found this thread yesterday, and hope there are still folks here to read and comment on my descriptions.

Where does the absorption take place?
The concept of 'path length' is useful here.
Early testing was done projecting different wavelengths through tubes with different gasses. They looked for what, if anything, came out the other end of the tube, hence transmittance (or absorbance which is 1-transmittance). With a vacuum or a non-GHG, they observed 100% transmittance (no absorbance). For the 'greenhouse' gasses, they observed absorbance at particular wavelenths, and that the amount of absorbance was dependant on both the concentration of the gas and the LENGTH OF THE TUBE. For a given GHG concentration there can be 100% absorbance if the tube is long enough, hence the 'path length'. With increased concentration, the path length is shorter. (An aside - this may be where the idea of 'doubling' came from, since they could 'double' the tube length using mirrors.)

For CO2 in our current atmosphere, I have seen the nominal path length reported in the range of 10 to 100 m. Increase in CO2 concentration makes the path length shorter, perhaps at say 90 m instead of 100m, but even if we rolled back to something like 350 ppm, all of the CO2-ready photons would be absorbed relatively near the surface.
Then the next important question is ..... What happens after the photons are absorbed?

What happens after the photons are absorbed?
Now we get to 'thermalization'.
Temperature is measuring the Kinetic Energy (KE) of the gas molecules banging into one another and onto the thermometer. Warmer air, higher Kinetic Energy, more energetic collisions on the thermometer, the reading goes up.
When a GHG absorbs a photon, the energy is INTERNAL to that molecule. If that energy is re-emitted as a photon, the thermometer does not change, because at this point the KE of the bulk gas has not changed. One could picture that if photons hopped from GHG to GHG and eventually escaped to space there would be no influence on the thermometer.
However, if a GHG molecule becomes energized by absorbing a photon, and then collides with a non-GHG air molecule, the internal energy is converted to Kinetic Energy in the bulk gas, and now we have an event that can change the temperature. This 'thermalization' is what warms the air.
How likely is that? My understanding is that near the surface, due to the rapid collisions, it is 10,000-to-one more likely to result in thermalization (collision) versus the re-admittance of the photon. Is that it then? Game over? Well apparently not because the reverse also happens - collisions can also re-energize a GHG molecule, which could then emit that energy as a photon. The physics folks describe how collision energy is partitioned into various forms, but that is outside my ability to describe, suffice it to say there is another 'layer' in that discussion.

But think of the irony here. If the photons are not 'thermalized' there is no temperature affect, and if they are thermalized, they are not photons any longer (no back-radiation from them). Even for photons that are re-emitted, the path length tells us that only low level photons have a chance to reach the ground. From the mid troposphere, the re-emitted photons are not directly reaching back to the surface, since they are multiple 'path lengths' away. It seems that, near the surface, the photons help the energy jump from surface into the low level atmosphere, and then help spread the energy within the affected 'path length' of nearby bulk air.

It is Convection and Latent Heat effects that do the 'heavy lifting' to move the energy up through the atmosphere, in my opinion. I tend to think of surface to air radient heat transfer as an enhancement to the convective heat transfer process. Latent Heat effects are significant where there is liquid water and minimal in the absence of water or water vapor, as discussed up thread in the ocean and desert comments. Note that convective heat transfer is proportional to the difference in temperature between the surface and the air and can go either direction. Natural Convective heat transfer can slow down as the air warms, without needing a back-radiation postulate.

DiD, a very interesting addition to the thread. You mentioned 'doubling' by which I take it you mean it as it appears in the "for a doubling of CO2 average temperatures will increase ~1F" sense which we are told is incontrovertible. Does anyone know what led to that conclusion? Out of interest, 1F would raise the height of the average emission temperature (-18C) by 86m, all other things being equal. What I don't see is why this suggests that there would be an energy imbalance at TOA as predicted by climate models and searched for by CERES satellites.

reply to SSAT
There is an AGW thought model that goes something like this - The radiative Top of Atmosphere can be estimated using computational tools such as Modtran. If one then adds extra CO2 and recalculates, the Top of Atmosphere moves higher (harder for radient flux to escape due to more GHG). The CO2 photons escape from the Tropopause (part of the TOA estimation) and at higher CO2 concentration, the Tropopause shifts higher. Because the Tropopause is higher, a back calculation using the Lapse Rate necessarily projects a higher surface temperature as a result.

I was skeptical of that thought model and have tried to educate myself about the Tropopause and the TOA. There is a famous quote from H. L. Mencken "For every complex problem there is an answer that is clear, simple, and wrong." While I don't claim to have all the answers, here are a couple things that cause me concern with the above model.

1 - The radiative TOA spectra as observed from satellites shows each radiative 'player' doing its thing from its temperature and 'height', which varies all over the map (literally). On a clear day the 'atmospheric window' looks like the surface, which can be 330K in the Sahara, 290K over the ocean, and 180K from an Antarctic ice sheet. Water wavelength emisission look like the top of clouds and on a 'cloudy' day they block the 'atmospheric window' ( the iris effect). The CO2 emissions are pretty much always at Tropopause temperature. That TOA is hard to pin down.
Check out the 'Satellite looking down' spectra at this link
Atmospheric Radiation Some IR Spectra Lecture for Spring 2009 Prof. Brian H. Fiedler
http://mensch.org/5223/IRspectra.pdf

2 - The satellite CO2 spectra comes from the Tropopause, but the Troposause is not a static sheet at one height as presumed in the thought model. Here is a description from The Height of the Tropopause, Geerts and Linacer, (link below)
"Thus, it is about 16 km high over Australia at year-end, and between 12 - 16 km at midyear, being lower at the higher latitudes. At latitudes above 60 , the tropopause is less than 9 -10 km above sea level; the lowest is less than 8 km high, above Antarctica and above Siberia and northern Canada in winter. The highest average tropopause is over the oceanic warm pool of the western equatorial Pacific, about 17.5 km high, and over Southeast Asia, during the summer monsoon, the tropopause occasionally peaks above 18 km."
They then describe variations with jet streams and macro climate circulations Hadley Cell, Indirect Ferrel Cell, and Direct Polar Cell (there is a nice graphic at the link). Weather fronts and tropical thunderstorms push the tropopause around, and I recall reading (not in this article) that there is a 'bump up' in the Tropopause due to the mid day sun, that then relaxes at night. With all of that going on, I have to think that while the Modtran estimate may be real, it is trivially lost as no more than noise in what is actually happening to the Tropopause.
Here is the link, good graphics and well worth a read
The height of the tropopause, B. Geerts and E. Linacre 11/'97
http://www-das.uwyo.edu/~geerts/cwx/notes/chap01/tropo.html

Martin A
"... is itself a very clear indication that, in reality, they have only the vaguest idea."
No argument from me on that.

Had seen the comment by R G Brown earlier, but good to re read it. His observations are always well organized and well expressed.

My posts above were an attempt to talk through some of the mechanisms as I understand them. If I can't at least describe them, then I won't understand well enough to critically examine comments from others. And exposing my comments to knowledgeable folks here may help me find flaws and fill in gaps in my understanding. I read with interest the earlier posts from you and others. This has been a good discussion thread.

DiD: Thank you for the references which are interesting as is the thought experiment. However, much of that is concerned with what happens within the atmosphere where energy is being transported by a multitude of effects. I think it best ignored as overly-complex for what, viewed from space, is a simple phenomenon. The sink of space will see an average temperature for Earth. That average will be driven by incident solar radiation which Earth red-shifts but does not change in quantity. Greenhouse effect can be easily explained by the lapse rate, itself driven by radiation as are the multitude of effects within the atmosphere, and which complies with the gas laws.

So, basically I accept the greenhouse effect and the role of both radiation and CO2 in it. However, I see no need for AGW to be explained by a postulated radiative imbalance computed by a model that attempts to chase radiation through a multitude of effects when a simple explanation exists. The fact that it is leads me to think that popular climate science panders to some pessimistic outlook that wants industrial CO2 to be a significant risk and follows a route which leads to maximum effect for little change.

You will have detected that I am no physicist or mathematician, perhaps you are, perhaps even a climate scientist, but I suspect that you, like me, have some major problem with what is actually known, what is claimed to be known and how we are being urged to react.

For me, the precautionary principle says 'do nothing'.

RKS recently posted this link in unthreaded;

http://greenhouse.geologist-1011.net/

I re-post it here, not only because it is very informative on the history of GE but because I see it as apposite to that which I have been 'bangin on about' on this thread.