The magnitude of the guesses, the provenance and the error bars we use for these fag-packet calculations should be of concern to everybody. It's a fair point ssat, but it's a bit of a goalpost shift.

Mmm. First, it does actually have a pressure-driven lapse rate, although because of the low compressibility it's a lot smaller. (About 0.1 C/km, IIRC.)

Second, it's not intended as a model of the atmosphere, it's intended as a separate physical 'greenhouse' situation that your theory needs to be able to explain. The backradiation argument gives one answer, the emission altitude argument gives another. It's therefore an ideal experiment to distinguish the two cases, and illustrates how they're not equivalent.

And thirdly, it *is* heated from below - that's why I said a shallow pool in my earlier description.

Consider a pool of water 1 metre deep with a perfectly black bottom liner. Water is transparent to sunlight (near enough) so most of the sunlight passes through and is absorbed by the bottom. This radiates thermally upwards into the water, where it is absorbed within the first 20 micron layer. This layer emits up and down, back to the bottom surface and into the 20 micron layer above it. And so on.

Divide the entire pool into minimally-opaque 20 micron horizontal slices and start from the top. The top slice radiates p W/m^2 of power upwards and out of the pool, which for energy balance is required to equal the energy the pool absorbs from the sun. It radiates the same p W/m^2 downwards. It is losing a total of 2p W/m^2, which for local energy balance it can only get from the layer below. The next layer down must therefore radiate 2p W/m^2 upwards to supply this, and 2p W/m^2 downwards. It's losing 4p W/m^2, and only getting p W/m^2 from the layer above. So it must be getting another 3p W/m^2 from the third layer down, radiated up into it. And so on. The nth layer down from the top radiates np W/m^2 both up and down, power radiated increasing linearly with optical depth, and must approach a temperature high enough to make this possible. Since black body radiation goes as the fourth root of radiated power, the temperature of the water must rise in proportion to the fourth root of the optical depth.

At 20 microns a layer, there are 50,000 layers, and so the temperature at the bottom (in Kelvin) has to be about 15 times the temperature at the top. Assuming that the sunlight is sufficient to warm a surface to 250 K (ignoring the atmosphere above the water), without the water, then adding the water should heat the bottom to 15*250 = 3750 K. If we instead require the top surface to radiate at a more familiar 280 K, taking the atmosphere into account, the bottom will be at 4200 K.

The emission altitude argument, on the other hand, predicts that the surface temperature equals the effective radiative temperature plus the lapse rate times the emission altitude. That's 250 K + 0.1 K/km * 0.001 km = 250.0001 K. Or 280.0001 K if you prefer to allow for an atmosphere.

Which theory's prediction fits our observations better?

If each layer of water emits np W/m^2 of backradiation downwards and p is around 400 W/m^2, then the total backradiation emitted by 50,000 layers of 1m^2 area each would be 500 GW in every cubic metre. Even if you consider a real cube of water emitting only 400 W/m^2 from each layer, that's still 20 MW in every cubic metre. How, in the name of all seriousness, can you have 20 megawatts of backradiation have *no* effect on temperature, if backradiation is the real reason for the warming?!

Backradiation exists, and is generally a very large number, but in a convective fluid its effects cancel out. But convection only works to do so up to the limit of the adiabatic lapse rate, so if the fluid is compressible then the cancellation will not be complete. It's this non-cancellation, controlled by the properties of convection, that controls the temperature and provides the better explanation.

"At the equator a maximum of around 100C is possible yet because of cosine dilution this drops to around -270C at the poles. What about the night facing hemisphere?"

This is why I said "*uniform* temperature surroundings". This would be the situation where the Earth was surrounded entirely by a surface at the same constant temperature. (Imagine the Earth was located inside a hot gas nebula.) As soon as you posit temperature differences, you'll get convection again.

"My problem with that is that trapped energy has to not reveal itself in the atmosphere because to do so would increase temperatures at ground level immediately."

The approach SoD is taking comes to the same answer from a different perspective. When you push an object across the table, unbalanced forces distort and accelerate it until it arrives at a new position. One perspective is to imagine the situation when you have started pushing but the motion in response hasn't started yet. You push against the stationary object with 10 Newtons of force directed to the left. The forces are unbalanced, and you can estimate what the effect is by calculating the size and direction of those forces. Another perspective is to assume that equilibrium between the object and your hand has already been achieved, and figure out where that new equilibrium is. Move your hand 10 cm and the object will have moved 10 cm, and there are no unbalanced forces.

Personally, I find the 'moving equilibrium' perspective easier to grasp than the 'unbalanced forces' perspective, but I'm sure different people differ.

All your pool analogy is describing is how convection is much more dominant than IR transfer in denser materials. This is something we all agree on. If IR was the only mechanism for liquid water to cool, then the bottom of the pool would indeed boil away, because for the surface to radiate the sun's heat, the bottom would have be really hot.

But since convection is much more dominant, this situation never arises. This does not mean the phenomenon of IR transfer isn't real. It doesn't mean you can extrapolate the fallacy from liquids to gases and say it never happens in atmospheres, because you don't ever see pools of water boiling. What you absolutely can't say is that because the two-layer optical analogy works better for the case of the pool, then that makes the optical theory superior to the back-radiation theory when it comes to gases in the atmosphere. That's called fallacy by association.

The pool is not a gas. The pool does not contain a large mass of non-participating diatomics. The pool is made of a substance that is incredibly opaque to IR and incredibly more dense which permits conductivity/convection much more easily. You cannot say that because a ludicrous effect is not seen in the pool, then it can never be seen in the air, where the materials, densities and behaviours are completely different.

"But since convection is much more dominant, this situation never arises."

Exactly. In convective fluids one mechanism applies, in non-convective materials a different mechanism does. Applying the non-convective mechanism to a convective situation and describing the effects of convection as a mere 'correction' is misleading.

"This does not mean the phenomenon of IR transfer isn't real."

I didn't say it was.

"It doesn't mean you can extrapolate the fallacy from liquids to gases and say it never happens in atmospheres, because you don't ever see pools of water boiling."

The point of having universal laws of physics is that they *can* be extrapolated from one situation to another. The water is simply a more extreme example of the same physical principles.

"What you absolutely can't say is that because the two-layer optical analogy works better for the case of the pool, then that makes the optical theory superior to the back-radiation theory when it comes to gases in the atmosphere."

No, but when the ridiculousness of the predictions of the backradiation argument in water force you to reconsider one's reasoning, the new understanding gained can be applied to the situation in the atmosphere.

"The pool is not a gas. The pool does not contain a large mass of non-participating diatomics."

Why do you think these are relevant?

"The pool is made of a substance that is incredibly opaque to IR and incredibly more dense which permits conductivity/convection much more easily."

Opacity has nothing to do with the ease of conductivity and convection. And convection in the air is pretty efficient.

"You cannot say that because a ludicrous effect is not seen in the pool, then it can never be seen in the air, where the materials, densities and behaviours are completely different."

What's ludicrous about it? The materials are different so the magnitude of the effect is different, but it's the same effect.

It is a standard approach in physics when trying to understand some interesting new mechanism to push it to the extremes and see what happens. Greenhouse gases are such because they are transparent to visible light and opaque to infrared. What would happen if you pushed that property to the extremes? It seems like an obvious enough thought-experiment, and conveniently, liquid water gives us a real-world exemplar to try it out on. The answer is that as the defining 'greenhouse' property gets stronger, we approach a situation we're familiar with - that of internal radiation inside opaque materials - and we know that in that situation it cancels out. To the point where we rarely ever even mention or consider it.

And if pushing what you thought was the relevant property to an extreme gives you an odd/unexpected result, it's a signal to the careful scientist that he may be missing something and needs a deeper understanding of what really causes the effect.

If adding CO2 caused more heat to transfer from the upper atmosphere to the surface, it would change the lapse rate. The upper atmosphere would cool by radiation and the surface would warm and the gradient between the two would increase. But the lapse rate in the real atmosphere is fixed. Forces trying to exceed it are countered by increased convection and no change in gradient. Thus, to the extent that the adiabatic lapse rate provides a fixed maximum gradient (which is what is observed), the effects of changing any radiation transfers internal to the atmosphere are precisely cancelled out by convection. Only the radiation exchanges external to the atmosphere count.

I'm grateful for the replies. I'm still considering them. Let me ask this: The atmosphere provides a gaseous gradient from the land/sea surface to the edge of space. There is also a temperature gradient. This is due to convection and the lapse rate.

If our atmosphere contained no greenhouse gases these two features would still exist and the blackbody radiation would still be emitted from somewhere in the atmosphere. Do you agree?

Now I agree that the presence of GHG gives an extra twist on the passage of heat through the atmosphere. This may enhance the convection in the lower atmosphere by converting IR radiation to kinetic energy via collisions and more convection. (I know it is said to slow things down, not enhance convection and heat loss) In the upper atmosphere there is a lower probability of collisions but there is a lower concentration of GHG molecules too. Now I know you are going to say that the GHG slows the passage of radiation to space. I'm unable to decide but question it for the reasons given.

These different effects may or may not be finely balanced and there are probably others. This is where my faith in the theory gets a bit weak.

Now I see clouds very differently. Just as they constitute albedo for visible light, I can see clouds as a very definite barrier for outgoing IR. Clouds contain huge quantities of finely dispersed water. Could it be that clouds are actually responsible for much of the GHE and we wrongly attribute equal power to GH gases?

Another point that I ponder is that our planet consisting of earth, sea and gases radiates IR energy until it is in balance with the incoming energy. The internal processes such as conduction, convection, IR absorbance and emission, ocean currents, latent heat, humidity and so on may distribute or modify that energy within the system but the total energy is what matters and is what drives the heat loss. The GHG may slow down the passage of heat in the atmosphere today, but that which was slowed down yesterday or last week is now getting to the TOA. It is a dynamic system, continuously adjusting to stabilise itself. I'm therefore not convinced that slowing down is a factor in heat loss. When you consider the differences between the tropics and the poles, the temperature range of the diurnal and seasonal cycles and the effects of cloud cover, there are lots of major changes happening all the time, yet in the bigger picture we have a very stable climate.

What I am edging towards is that the GHE is real. We can witness it at night for example. It is particularly effective with clouds. But whether it influences the total energy available for emission to space, I very much doubt. I think that this view is probably consistent with observations concerning the current lack of correlation with CO2 if we attribute some of last century warming to natural variability.

Finally, much of this comment should be regarded a distillation of thoughts/questions to give you some idea of where I am coming from and the questions I am grappling with and may explain some of my scepticism. I don't claim to be right, I'm just identifying my ignorance. I also realise that I'm questioning conventional wisdom and may appear stupid, but some of these things need to be asked.

If we switched on the GHE tomorrow we may have an energy imbalance for a while but then things would catch up and the warmed up earth would start radiating more heat. We are in a dynamic situation. I don't believe that internal processes will affect the heat loss from the system in a permanent way. The very act of increasing warming will result in more heat loss.

I can accept that these internal processes can change global or regional climate for a period of time, but the GHE is not unique in that respect. Oceans do this all the time in multidecadal timescales. I question whether these processes can bring about permanent or ever increasing change.

"If our atmosphere contained no greenhouse gases these two features would still exist and the blackbody radiation would still be emitted from somewhere in the atmosphere. Do you agree?"

Yes, these two features would still exist, but no, it would come directly from the surface - the sea or land - not the atmosphere.

"Now I know you are going to say that the GHG slows the passage of radiation to space."

Not significantly, and it doesn't matter whether it does or not. Photons move very fast, and space is pretty close, as places go.

I'm guessing you mean it slows the flow of energy to space. Radiation warms the air, and the air radiates thermally (if it has GHGs in it). But it doesn't matter how fast it happens, whet matters is what temperature it approaches in the longer term, and that temperature is the temperature that overall emits as much energy as the whole Earth absorbs. You can pump water through a narrow pipe fast or a wide pipe slow, and have the same total amount of water pass per unit time. What matters is the total per unit time, not the speed of motion.

"Now I see clouds very differently. Just as they constitute albedo for visible light, I can see clouds as a very definite barrier for outgoing IR."

Yes, agreed. I was ignoring clouds as an added complication until we got the basic GHE sorted out. Emission from clouds does indeed cause a greenhouse effect, and its magnitude depends on the altitude of the clouds, the droplet size and density, the thickness, whether there are other clouds overlapping above and below, where they are, the incidence angle of the sun to the cloud plane, and lots of other stuff. Very complicated.

And no, we don't have any answers. The models do them badly, and nobody knows exactly how much they contribute.

But just because clouds also contribute doesn't mean that GHGs don't. Don't get so stuck in trying to find reasons to ignore GHEs that you'll grasp onto anything that gives you an excuse - that's not being a scientist. That's not being a sceptic. That's being as partisan and biased as any warmist - just on the other side of the debate.

(My sincere apologies if that doesn't apply to you - it's just the impression I'm getting, and often get in these discussions.)

"I think that this view is probably consistent with observations concerning the current lack of correlation with CO2 if we attribute some of last century warming to natural variability."

We can't tell how much of last century's warming was down to natural variability - we don't have a validated statistical model of the natural variation. What we can say is that if the current 'pause' is ascribed to that natural variability cancelling the warming out, then the magnitude of the natural variability must be at least as large on a 20-year time-scale. It might be bigger, it might not. (And the answer might be different on longer time-scales.) It's therefore entirely plausible that most of last century's warming was natural, but still possible that it wasn't. We don't know.

And that should be our position. We don't know. We can't say that it was natural, for the same reasons they can't say it wasn't. We can't say it's peaked and is about to go down, because we don't know that it will. If it's random, it could just as well go up again. If you stake your position and your reputation on making statements you've got no scientific support for, you take a huge gamble on the weather doing what you hope. The warmists did that, and look at where it's got them.

We need to learn from their mistakes, make sure we don't make claims we can't support or offer incorrect explanations of the science just because it works politically, and always be ready to say "we don't know", because it always catches up with you eventually.

"Finally, much of this comment should be regarded a distillation of thoughts/questions to give you some idea of where I am coming from and the questions I am grappling with and may explain some of my scepticism. I don't claim to be right, I'm just identifying my ignorance. I also realise that I'm questioning conventional wisdom and may appear stupid, but some of these things need to be asked."

That's good, and definitely not stupid. If you don't understand, ask. Don't take anybody's word for it. You can always ask for a scientist's evidence, or for his/her reasons for thinking what they do. (They don't have to answer, but any unsupported claims pay the penalty in credibility.) Asking questions and expressing doubts are good ways to learn, and are what keeps science on track.

But in doing so, we should all be aware that everybody has biases, and everybody makes mistakes, and what separates the true scientist from the partisan is that scientists are aware of it, and try to learn from them. You always need a diversity of viewpoints, because while everyone has blindspots, different people have different blindspots and we can tell each other what is in them. You get a more complete picture of the world that way - even from listening to warmists. We need them, and they need us.

We rely on our biases to motivate our critical examination of the evidence, and that's a good thing and very much necessary for science to work, but we need to be aware not to get too hung up on them.

It's too long to excerpt, but Mill's essay "Of the liberty of thought and discussion" is well worth reading on this topic. The bit about "If the cultivation of the understanding consists in one thing more than in another, it is surely in learning the grounds of one's own opinions" in particular.
http://www.bartleby.com/130/2.html

"I don't believe that internal processes will affect the heat loss from the system in a permanent way. The very act of increasing warming will result in more heat loss."

They don't make any permanent change to the heat loss. They do make a permanent change to the temperature - the internal temperature has to rise until the heat loss from the outer surface is the same again.

The point of my long post was that both the back-radiation model, and the optical thickness model both suffer from similar analogy weaknesses.. they simplify a complex situation in order to demonstrate an effect. They are not competing theories about how it works, they are different teaching tools.

Yes, and my point was that the backradiation model makes the wrong predictions. There are common physical situations with huge amounts of back radiation and virtually no greenhouse warming at all.

It may be a teaching tool, but it teaches the wrong thing.

The excellent discussion above has been very useful. You guys deserve some feedback.

I am guilty of making some assumptions without thought. One of them was that all matter above absolute zero will radiate heat according to Stefan's law. In fact gases will only do this if they are dense - I assume compressed close to being liquefied - when they will obey kirchoff's first law. Otherwise they only emit/absorb at their characteristic wavelengths. So apart from the GH gases nitrogen, for example is active in the extreme UV so forget participation in long wave radiation from the earth.

This mistake on my part will explain some of my comments which you guys corrected. I think the other major problem I had was the insistence of all the literature that the GHE is back radiation, when in fact that is probably the least part of the effect. I realised that it was wrong but that tended to make me conclude the whole thing was wrong.

So I guess the first point is that those who seek enlightenment from the internet get fed an unhelpful mechanism.

I now understand the points you make and I can live with the concept of a chunk of the atmosphere undergoing collisions, kinetic energy exchanges, photon emission and convection. I see convection as the dominant process within all of that. Of course emission of IR photons to space from the GHG molecules towards the top of this part of the atmosphere is the important loss of energy.

So I accept much of the GH theory so far. I have some more bits to think about before I comment further.

I hope this discussion thread continues. As they say, the science is not settled.

I think it is fair to say that the warming that was expected has not materialised and some of the assumptions in the models are clearly wrong. It is probably also true that natural drivers of climate have been underestimated. I agree that the feedbacks are probably the key.

Although I accept much of the discussion above, I feel that some of it is more an act of faith than a proven fact. Then again, I'm not an expert and have not studied the evidence.

If those who have studied the evidence are still visiting this thread, I have a question. What aspects, if any, of the theory discussed earlier are a bit suspect? Does some of it give you doubts? The theory is certainly plausible, but is it right?

I'm thinking of the opacity/emission altitude/lapse rate and the consequence of adding more CO2. It all fits wonderfully well without actually being terribly convincing, to me anyway. I thought it would be interesting to cover this last aspect of what we discussed before since it is healthy to challenge the consensus.

If there are any references to where all this stuff is reported, that would be interesting too.

Seasons greetings to all.

Sorry for the late reply to an earlier post, but I've been pondering.

On page 2 of this thread in his initial lengthy explanation TBYJ posted....
"You take a spectrum analyser and point it at a dark cloudless sky.
It has to be dark, so you won't measure direct solar radiation (which would drown out the smaller signal from the hot gases in the sky). It should be cloudless, because some of the effects may be from vapour clouds which are even better at bouncing IR back. We're looking to see what a normal, fairly dry warm night atmosphere is doing, not the sun, not the clouds. "

Should you not simultaneously have an identical spectrum analyser at the same altitude pointing down at the earth?

Surely it's the difference in these two measurements we should be interested in?

The problem with this subject is that it is very complicated. It seems to me that a relatively small clique has defined how it all works and this has been accepted. It is very difficult to explore the awkward questions such as saturation of the CO2 absorption bands and the consequence of so much overlap with H20.

These are just examples. Most of the stuff on the internet is superficial, misleading and wrong. It does not fire me up with confidence. When technical challenges are made in the more scientific blogs, these rarely get a direct answer or even a plausible argument. This really is a strange branch of science. Do others find this or am I being unreasonable?