Discussion > That CO2 thing again..
Rhoda,
I think that's the point. Because of a multitude of confounding effects, such as H2O content, convection, albedo, planetary climatic oscillations, heat sinks and sources, aerosols, soot and a dozen other things, we don't know if or how adding more CO2 affects the opacity of the atmosphere to IR radiation, or what sort of change that may make to the equilibrium temperature. If anyone tells you they do, then slap them.
It doesn't mean there is no effect, it just means it may not be calculable or measurable.
TheBigYinJames
There was discussion at Judith Curry's website some time ago. I took what Judith Curry said to be definitive and at that point I gave up trying to understand how the supposed greenhouse effect works in reality.
She had asked what there was that described the operation of the greenhouse effect in more detail than the super-simplfied black body model whose only atmosphere was a shell of greenhouse gas. (And which is so simple that its working can be analysed from 1st principles with no doubt about the validity of the result.) What she sought was something that might need the level of knowledge from, for example, the final year of a physics degree, to understand fully, but that would include the major physical effects involved - transport of heat by convection, radiation, latent heat, planetary rotation, and so on.
Judith Curry came to the conclusion that there was nothing available between the simple black body model and computer models. If someone in touch with the pulse of climate research concludes that, then I have no option but to accept that's how it is.
If you subscribe to the view that an unvalidated computer model is nothing more than an illustration of a hypothesis, you then have to conclude that the greenhouse effect is simply not understood. The absence of answers to rhoda's questions provides confirmation.
Martin A
Is it worth my suggesting you read http://scienceofdoom.com/2014/06/26/the-greenhouse-effect-explained-in-simple-terms/
Raff
Raff - Thanks for the reference to the relatively new page on SOD. SOD is always good for a clear explanation of things. Perhaps I should have made it clear that what I meant it is that the greenhouse effect is not understood in the sense that we can calculate results of known accuracy, with the contribution of different effects and their interactions being understood quantitatively.
I doubt than anybody here would dispute what SOD says in his qualitative description of how the greenhouse effect works. It's nice that he does not invoke the concept of 'back radiation', unnecessary to understanding how the GHE works, and which results in all sorts of fruitless discussion about thermodynamics and so on. Essentially the same explanation is given in Pierrehumbert's book.
Martin A
Scienceofdoom is simply a school teacher with an AGW bias and a strong copy and paste habit. I don't think I'd regard him as the guru that some seem to.
RKS
I've read the simple explanation. I've occasionally found SoD not too bad, at least he sees the distinction between the radiative balance average temp and the global average measured temp. But when you come back and ask a simple question or two, the simple explanation falls to bits. If back radiation is not part of the explanation, half of the simple treatments go away. Even the met office can be found spouting 'CO2 traps heat'. I know that is wrong. Then we are left with the radiating altitude/lapse rate idea. Which may be right, but it works the other way round too. That the surface temp is what sets the mean radiating alt via the lapse rate. Now, what made the air warm? The surface? What made THAT warm? The lapse rate made the air warm from above? How? What thermalized the air molecules? Collision with energised CO2? But we are told that happens but so does energy transfer the other way too, ie N2 and O2 collisions can energise CO2. And I hear that a photon can't get far without hitting CO2. If that's the case, does it matter how many times that happens? If not, how can the proportion of CO2 be relevant?
rhoda
"Which may be right, but it works the other way round too. That the surface temp is what sets the mean radiating alt via the lapse rate."
A body in a vacuum (which the Earth is) emits radiation to outer space at a rate proportional to the fourth power of the temperature of the surface emitting to space. It's got to balance the incoming energy. Therefore, it's the emitting surface that has to adjust its temperature to balance the energy.
"Now, what made the air warm? The surface?"
Yes.
"What made THAT warm? The lapse rate made the air warm from above? How?"
The sun.
But also, there are convection cycles where the air rises over the equator, rolls over, and descends over the deserts. (And similarly with further convection cells up to the poles.)
"What thermalized the air molecules? Collision with energised CO2? But we are told that happens but so does energy transfer the other way too, ie N2 and O2 collisions can energise CO2."
Yes, that's right. It goes both ways, until the energy flows balance. It would be a bit weird if you had a mixture of gases in intimate contact with one another where one gas was somehow a lot hotter than the other.
"And I hear that a photon can't get far without hitting CO2. If that's the case, does it matter how many times that happens? If not, how can the proportion of CO2 be relevant?"
I don't know, but I think the mean free path is quite long. The optical depth of the atmosphere is not that big.
But it doesn't matter in the slightest. The only thing that matters is the altitude of the last collision, before it escapes to outer space. The only reason the CO2 matters is that more of it raises this altitude.
Nullius in Verba
The surface? What made THAT warm?
Er, visible wavelength (mostly) radiation from the Sun ?
Even the met office can be found spouting 'CO2 traps heat'.. Yes, inexcusable. Like a lot of what comes out of the Met Office. In the case of that particular MO scientist, the explanation is perhaps that they were trained at UEA.
Obviously back radiation (and front radiation and all other directions of radiation) comes into it as soon as you ask about the details of how energy gets from the warm surface to the cooler layers, where IR photons finally escape for good, and what sets the lapse rate. A minimally understood soup of heat being transferred up, down, left, right, by radiation in multiple hops, by convection, by evaporation, condensation, precipitation, expansion, advection.
I'll stick with Judith Curry's conclusion that a nontrivial model does not exist, outside of computer models.
Martin A
If the surface warmed the air (yes, of course it did, I know that) then why does a little more CO2 warm it more? There has to be a mechanism. Can it be anything other than the dreaded back-radiation? Not set by the height of radiative equality then?
rhoda
"then why does a little more CO2 warm it more? There has to be a mechanism."
It doesn't warm it more, it changes the way it escapes.
You pour water from a tap into a bucket. Eventually it fills up and starts pouring over the sides. The level of water in the bucket increases until the amount pouring over the sides matches the water flowing in. If you raise the height of the bucket sides higher, more water accumulates.
Raising the height of the sides does not itself put any more water in the bucket. The water comes from the tap, not the bucket sides. But the sides of the bucket control the water level. The source of the water in the bucket and what controls the amount of water in the bucket are clearly distinct.
Similarly with heat. Heat pours in from the sun. It sloshes around with convection, evaporation, and internal radiation and finds its level with the adiabatic lapse rate. It pours out of the Earth's system by radiating to space. The level of the heat at the point it pours out (the TOA) determines how fast it flows, and this adjusts until the outflow balances the inflow. It's the flow at point at which it pours out that controls the level.
Nullius in Verba
I think I'll have to dig out my unfinished essay "why back radiation, although it is there, does not actually warm the surface which is done solely by the incoming visible light". I was advised it was not easy enough to follow to convince anyone who did not specially want to be convinced (although perfectly correct).
The argument involved starting with the world at zero temp and letting it warm up until it reached equilibrium. All the heating of the surface is easily seen to be done by incoming radiation. The surface rises in temperature until equilibrium is reached. The surface has to emit extra IR photons because of back radiaiton so it has to be heated to a higher temperature by the incoming visible radiation than it would be without the back radiation. The 'back radiation photons' are, *in effect*, photons that the surface did not manage to get rid of the first time of emitting them. So they have to be re-emitted (in effect).
According the gospel, with more CO2, the radiation that finally escapes does so at a higher mean altitude. The mean rate of energy flowing out has to be equal to the rate of incoming short wave energy irrespectiuve of how much CO2 there is around - so that is fixed (assuming the world is in equilibrium - not heating up, not cooling down).
To get the same rate of energy to flow from the surface to a higher altitude than before means the surface has to be warmer, as the energy it has to travel a longer distance through the atmosphere before finally getting out for good..
[longer distance for the energy to flow, same rate of flow, so temperature difference has to be greater. I can see there are some implicit assumptions in there.]
Did that make any sense? I'll look tomorrow and see if it does.
Whilst typing that, I see that NIV has answered it more succinclty and clearer than my stuff. I'll hit create post anyway.
Martin A
Martin,
I personally find the easiest argument is to ask how much backradiation a shallow pond of water emits downwards.
There are a number of ways heat is transported: conduction, convection, evaporation, radiation. In any IR-opaque material, radiation goes both ways. It's emitted both up and down, and if the temperature is uniform, the two cancel out. All this is saying is that in a sufficiently opaque material radiation is not a significant contributor to heat transport. The radiative resistance to heat transport is high.
If radiation was the only means by which heat was transported, then this would indeed mean that heat getting in wouldn't be able to get out, and the temperature would rocket up. But it's not. There are other means by which heat is transported that act in parallel, with much lower resistance. And as any electronic engineer will tell you, when connecting widely differing resistances in parallel, the lowest resistances tend to 'short circuit' the high resistances. The heat all flows through the big pipe - the fact that there's also a small pipe there beside it that is blocked makes no difference.
In particular, it's the non-linearity of one of the other methods - convection - that eliminates the effect of even this small contribution. Convection turns on and off. The resistance to heat transport therefore jumps up and down massively, and it's this switch that dominates the result. In water, any reduction of temperature with altitude triggers it. In air, because it is compressible, only temperature gradients exceeding the adiabatic lapse rate trigger it. Once the gradient is achieved, the convection shuts off and the resistance jumps massively. This 'thermostat' effect totally dominates the result.
When the temperature 5 km up settles out at -18 C, the surface is warmer than that because of this ALR gradient. The further apart they are, the bigger the difference, because the gradient is constant. The surface temperature is roughly the effective radiative temperature (at which in/out balance occurs) plus the adiabatic lapse rate times the difference in height. It's not warmer because a thicker slice of atmosphere has more resistance. It's warmer because a thicker slice spans more of that fixed gradient. If the gradient was somehow negative, the surface temperature would be cooler.
Nullius in Verba
"The only reason
the CO2 matters is that more of it raises this altitude."
Is this the case though? What does the data say? I don't really see why the lapse rate has to be constant when you are convecting lots of moisture/latent heat.
Rob Burton
It's not warmer because a thicker slice of atmosphere has more resistance. It's warmer because a thicker slice spans more of that fixed gradient.
Are those two things different? I'll have to think about that. I assmed they are the same thing.
( more resistance= greater temp difference for given heat flow = more fixed gradient spanned)
Martin A
"Is this the case though? What does the data say?"
This is the simplified theory. What data would you look at to test it?
"I don't really see why the lapse rate has to be constant when you are convecting lots of moisture/latent heat."
The basic adiabatic lapse rate is a thermodynamic property of gases and fixed by their physical properties. Latent heat from humidity is affected by altitude in a similar way, and is usually handled by simply adapting the constant. Changing the humidity *does* change the constant, and hence the lapse rate, and it's this change in lapse rate that leads to the predictions of the tropical upper troposphere hot spot.
"Are those two things different? I'll have to think about that. I assmed they are the same thing."
You're right. I was being imprecise. The usual 'thicker blanket' approach treats the resistance as fixed, proportional to thickness. That would mean that for a *fixed* thickness, the heat passing through is proportional to the gradient. The gradient varies with the heat flow from nearly flat at zero heat flow smoothly up to whatever the level is. In the electronic analogy, V = IR: the temperature difference is equal to the heat flow times the thermal resistance, which is constant. But in the convectively-limited adiabatic case, it doesn't work like that. For a given thickness of atmosphere, the temperature difference V is fixed, and so as the heat flow I increases, the resistance R drops.
There is, actually, an electronic gizmo that behaves like that. A reverse-biased Zener diode has a breakdown voltage at which the resistance suddenly drops in such a way as to hold the voltage constant, whatever current you put through it.
For a given heat flow, the slices of atmosphere all have a particular resistance, and stacking more of them up in series increases the resistance in the expected way. In that sense, yes they're the same thing. But it's best not to think about it that way, because it's likely to make you think they act like fixed resistances in other ways, and they don't. It's like a stack of Zener diodes, not resistors, and the lay intuition isn't generally built to cope with that.
I probably should have said something like "It's best not to think of it as being warmer because..." but it was late, and I wasn't analysing what I was writing properly. Apologies.
Nullius in Verba
What is this resistance to convection you talk about? IMHO most of the heat redistribution is happening in the tropics. There are massive vertical flows there every day. Transporting H20 in vapour form up to where it can dump latent heat way above the tropopause.
Why does the height of the radiative balance layer vary with co2 concentration? I though it was 5km because that's half way up the atmosphere, by mass. Or halfway up the troposphere, by coincidence or not. Why does more CO2 mean it isn't halfway any more? Or was the 50% pressure alt just a coincidence too.
And yes, if I had a hypothesis, I'd be looking to show how to measure it in real life. BEFORE I started putting up analogies to entirely unrelated physical processes.
rhoda
Oh dear, we're back into another radiative physics thread :) I think I've been in at least a dozen, with Martin, rhoda, RKS, geronimo, shub, Nullius, ssat, Dung, Roger Longstaff, Sandy, Paul Dennis, splitpin, Jonathan Jones, and many more...
The danger is that certain helpful newbies may think we poor fools are coming to this concept for the first time, and we really don't understand what it is about and deign to explain it to us by posting helpful links.
Previous discussions worth reading before repeating again:
Detailed discussion of the GHE
The original GHE thread
Lukewarmer thread
And another
Modelling insolation
TheBigYinJames
TBYJ, I'm asking people to repeat it. To go over it again. Now, I may be resistant to ideas that don't suit me. We all are. But my problem here is that people who think they understand the mechanism have different stories, or can't answer simple questions, or can't propose ways to measure what they say is so obvious.
I'm inclined to go along with the Curry version, that it's pretty complex. But there are still folks who thinks it's all so obvious, it comes from the physics, if CO2 'absorbs' more heat it just MUST get warmer, without describing a mechanism. That attitude is not scientific, to an oxfordshire housewife. So how do so many accept it without apparent thought? Because if the story fits your prejudice, you don't need to examine it too much.
rhoda
"This is the simplified theory. What data would you look at to test it?"
The test that whoever came up with the hypothesis came up with to try and verify it.
"The basic adiabatic lapse rate is a thermodynamic property of gases and fixed by their physical properties. Latent heat from humidity is affected by altitude in a similar way, and is usually handled by simply adapting the constant. Changing the humidity *does* change the constant, and hence the lapse rate, and it's this change in lapse rate that leads to the predictions of the tropical upper troposphere hot spot."
I assume we are talking about the environmental(ie real) lapse rate. It gets a bit confusing here when you start changing the values of 'constants'. I still think that a boiling pan of water (ie IR opaque) is a good analogy to the Earth's atmosphere. Obviously doubling the heat on the hob here doesn't have much effect on the temperature of the 'atmosphere' in the pan. You will just get more convection and evaporation in this case.
Rob Burton
Jul 31, 2014 at 10:47 AM | Unregistered CommenterTheBigYinJames
Perhaps an 'expert' will answer rhoda's question if she keeps asking. I don't remember reading any good realistic explanations in any other friend. It seems a pretty fundamental question to just handwave away.
Rob Burton
I don't think there is an answer. Anyone who thinks they know is lying or mistaken.
We understand a bit about a few of the vectors, such as GHE, evaporation, etc., but we have very little idea how they predictably interact with one another in the wild. "The best model we have" (an ensemble of climate models) hasn't been very good so far, they have breached all reasonable measures of verification.
Politicians don't like "I don't know" as an answer, so will keep asking different scientists until one of them gives an answer, right or wrong.
TheBigYinJames
"What is this resistance to convection you talk about?"
I'm talking about thermal resistance, a general concept in the physics of heat transport. It's the ratio of the temperature difference to the heat flow that results.
http://en.wikipedia.org/wiki/Thermal_resistance
"Why does the height of the radiative balance layer vary with co2 concentration?"
Because if you look at the atmosphere with "infrared eyes" it would appear to be a cloudy opaque mist with the visible surface being well above the solid surface. (Think about how Venus looks from space.) This 'visible' surface is what emits radiation to space. Adding more opaque gases makes the fuzz more opaque and the 'visible' surface appear higher.
Mix milk and water in a glass so you can see into the liquid but not all the way to the bottom. Now add more milk. Does the visible surface move higher?
"I though it was 5km because that's half way up the atmosphere, by mass."
The bulk of the opacity is from water vapour, which is predominantly found in the troposphere. You get emission to space from the ground up to around 11 km, the average being somewhere around half way. (Stuff like MODTRAN and HITRAN are used for this sort of calculation.) But it doesn't have to be that way. If there were no GHGs, it would be at ground level. If the atmosphere was IR-opaque, like Venus, it would be at the top.
Manabe and Strickler calculate some graphs. See Figures 8b and 8c for the profiles here: http://idwebhost-202-127.ethz.ch/ese101/Papers/manabe64a.pdf
"But my problem here is that people who think they understand the mechanism have different stories, or can't answer simple questions, or can't propose ways to measure what they say is so obvious."
Well, I at least *try* to answer questions.
"I'm inclined to go along with the Curry version, that it's pretty complex."
It is. So what do we do? Answer simple questions with simple and simplified analogies, or say "It's too complicated for you to understand. Trust me, I'm a scientist."?
Even the simplified analogies are difficult to understand, as this discussion illustrates. There doesn't seem to be much point in going into the complexities until we've got a general understanding of and agreement on the simplified version (understood *as* a simplified version). Otherwise the discussion becomes defocussed, diverging into dozen of irresolvable arguments and disputes and demands for evidence.
What part do you want measurements of? The thermodynamic properties of gases? You can test the effect of compression/expansion on temperature with a bicycle pump, and you can test the adiabatic change in temperature by climbing up a mountain, or flying an aeroplane. That certain gases are opaque to IR is easily tested in the lab. There are IR-camera pictures of the Earth, that show it. That a heated body in a vacuum changes temperature until it radiates as much heat as is absorbed is again easily tested, and that it is the bit of the object that emits that so adjusts should also be testable, if it isn't obvious. The temperature and emitted radiation throughout the lower atmosphere is obviously measurable, and can be checked against predictions. (And has been. MODTRAN emission/absorption coefficients in general have been found to be accurate to around one part in a thousand.)
If you understand the basic mechanism, you should be able to come up with such tests for yourself, and have a think about how you might even be able to do some of the experiments yourself. Believe me, thinking about how to design experiments yourself is one of the best ways to understand a phenomenon, and is what scientific sceptics should be doing anyway. It's not enough to say "I'm not convinced." You have to specify what *would* convince you, or the best you can really say about it is "I don't know." For some purposes, that's sufficient to win the argument, but we should be capable of more.
The story doesn't fit my prejudices. From a political point of view, I'd like nothing better than to be able to show that the GHE isn't real, and CO2 has no effect. When I first heard the backradiation argument I thought there was something funny about it, that didn't quite add up, and so I chased it around and tried to understand. The lapse rate argument on the other hand immediately rang true to me. And I've had a lot of arguments about it since with people on *both* sides of the debate, and nobody has yet shown me any critical flaw in the reasoning. Heat balance requires that the surface emitting to space settle at the equilibrium temperature. The lapse rate fixes the surface temperature to be warmer than that in proportion to its height. The gross non-linearity of convection turning on and off dominates and cancels out all other internal effects, like water boiling in a pot. My mind is still open on the matter, which is a large part of the reason I still participate in these discussions, but so far it's been tested and has survived. That gives me some confidence in it.
But I can't magically transfer that confidence across to you. Everybody has to do their own thinking, and arguing, and testing. All I can do is try to answer questions about how the theory works and what it predicts.
---
"I assume we are talking about the environmental(ie real) lapse rate."
No, I'm talking about the 'moist adiabatic lapse rate'.
http://en.wikipedia.org/wiki/Lapse_rate#Saturated_adiabatic_lapse_rate
Nullius in Verba
TheBigYinJames
Rhoda, if at height h in the stratosphere the concentration of CO2 molecules is some nominal X, do you accept that after doubling the % CO2 in the atmosphere h will increase for the same X? It all hangs in this.
TBYJ, "helpful newbies may think we poor fools...": if you have discussed it so often why are there still doubts? Maybe you haven't explained it very well. That is generally the fault of the teacher, not the student. Being "simply a school teacher", as RKS describes SoD, is actually not simple. Imagine the difference between having Feynman as your teacher and RKS. And RKS might consider that the AGW bias he detects might be in the science not the teacher.
Raff
I still don't quite get it. I am perturbed by the lack of an agreement, a consensus, so to speak, as to how an increase in CO2 from historic to current levels ACTUALLY changes the climate.
Here's a comment I put up on the last go-round, just before the thread dried up.
Quote:
I've stayed out. I've tried to understand the various illustrations/explanations. Some seem sounder than others, but all seem to lack in the area of observations and predicted checkable occurrences.
I don't accept 33 degrees still, and I can't be convinced based on conditions on some other planet of which we don't know much. Tatooine and Hoth were both filmed on the same planet, after all.
Now, this may be my failure to understand, but there seems to me to be something circular about that TOA/lapse rate thing. Too many things being averaged at too early a stage in the explanation. Arithmetic averages count for nowt when t^4 is involved.. Purely radiative explanations will not do when massive heat pipes are dumping thousands of tons of water vapour at heights way above the tropopause every day in the tropics.
And then there's the photons. Apparently they don't just get absorbed and re-emitted, they thermalize O2 and N2, they in turn energise other CO2 or H2O molecules, they get absorbed and re-emitted all the way up until they make their inevitable way out. Now, why does it matter if they do that a gazillion times or 1.5 gazillion. They leave the planet pretty quick either way. What gets warmed up more because of the extra density of CO2? All the energy leaves. It is an open system, there is nothing to stop it.
Now, equilibrium. Many of these arguments hinge on equilibrium, but it does not exist in the actual system. What we have is an infinite number of bits none of which are in actual balance but all are seeking it. That really is not the same thing. You can (by a stretched example) change the amount of energy input into a pendulum. The average position will stay the same.
Enough. Rant over.
end quote.