I have some thoughts on why the GHG effect is not going to fry the planet. I don’t claim to be right, but I can’t find major flaws in my logic. I would welcome your thoughts.

I totally accept that greenhouse gases are active in the IR. I accept that the planet is warmer than our black body temperature would suggest. I even accept the principle of the GH effect.

I find it difficult to accept that a trace gas is going to dramatically change our climate. I find it difficult to believe that the positive feedback of water vapour will lead to runaway warming. Our water based planet would have gone out of control years ago if warming could trigger such an event.

I have been racking my brain for years on how to resolve these matters and after a few daft explanations I think this one might have some merit.

Central to my thinking is that the release of a photon by an excited GHG molecule can take milliseconds to a tenth of a second. This is quite slow. The electron transitions for atoms is very fast, almost instant, but with molecules where the excitation takes the form of faster vibrations and bending, the conversion of kinetic energy to a photon takes some time.

The probability of molecules colliding means that at least one collision will occur within 0.1 microsecond. Now, the probability of a collision depends on the numbers of molecules, their sizes and their speed. The speed of course is a function of their kinetic energy or temperature and the collisions with the walls of the hypothetical containing vessel is the cause of the measured pressure. All of this means that collisions are more likely near ground level and less likely at high altitude where there are fewer molecules.

Given the large difference in the time taken to release a photon and the time before molecules collide, I think it is reasonable to suggest that a high proportion of excited GHG molecules will collide with other molecules before they have a chance to emit the photon of excess energy. This means that the energy will be transferred as kinetic energy and this warmer quantity of air will rise by convection.

If the GHG molecule collides with another GHG molecule of the same type, then the potential to emit a photon is retained. This is not going to be a significant route for several reasons. A trace gas molecule like CO2 is unlikely to collide with another one and it may take many of these collisions before a photon is released, assuming the time is not reset to zero on each collision.

So, greenhouse gases are likely to create convection rather than return heat to the surface.

The problem at this point is that I have just demolished 99% of the GH effect so I started to consider the conditions that would permit downwelling IR.

Clearly, GHG molecules at high altitude will take longer to encounter random collisions. Interception of outward going photons will be less likely too, but the excited molecules will probably have time to emit their photons.

Then I thought about clouds. A cloud might contain a billion kilograms of water. Water will absorb outward going IR photons. Water has a substantial capacity for heat compared with air. This warmed water will emit IR as a blackbody, a proportion of it back down to the surface. This could be a substantial form of back warming, but not involving gases and the GH effect as defined.

Each droplet of water will be in equilibrium with saturated air around it so a cloud is full of water vapour as well as liquid water. This water vapour can perform as a greenhouse gas. This time, if collisions occur, the chance of hitting another water vapour molecule is quite high, so the chance of emitting an IR photon his good.

This is how I ended up with the thought that downwelling IR does take place, mainly from the water in clouds. The GH effect is much less, mainly from clouds, some from the upper atmosphere and a little from lone molecules in the lower atmosphere that have managed to emit a photon before losing the energy in a collision.

What are the implications of these thoughts? The dominant role of clouds doesn’t affect the argument about the earth being warmer because of back warming. In fact, back warming by clouds at night is well within our experience. The warming of a desert by GHG back warming is something that is never experienced.

The positive feedback by water vapour is never going to fry the planet. Clouds are also the sun shade, reducing solar warming, so it is a wonderful, elegant and simple thermostat designed by Nature.

So, a trace gas like CO2 will perform as a GHG but the additional warming will be negligible, apart from an increase in warming the air followed by convection.

Clouds are the main source of back warming and it is the liquid water that does most of it.

Have I got it all wrong?

A few points:

I have some thoughts on why the GHG effect is not going to fry the planet.

I don't think anyone except the foaming crazies ever thought it would 'fry' the planet. The runaway greenhouse idea was never scientific, we don't have enough GHGs and don't have enough fossil fuels in the crust to make anywhere near enough of them to cause a runaway. As long as we have water in the air, we don't get runaway. So there's no point trying to argue against a strawman, that's what they do.

I think it is reasonable to suggest that a high proportion of excited GHG molecules will collide with other molecules before they have a chance to emit the photon of excess energy.

We covered this in a previous discussion. At ground level with air under pressure, the majority of energy transfer will be kinetic (because of the relaxation time versus kinetic collision time). Almost all heat transfer at or near the surface will be convective. It's only higher up in the atmosphere where IR becomes dominant.

This means that the energy will be transferred as kinetic energy and this warmer quantity of air will rise by convection.

Well.. sort of. Remember that a the surface, some of the kinetic energy is transferred to the ground, which cannot rise up. But yes, in general hot air will rise, and cooler air will fall to replace it. This hot air gets high enough until it can release IR to space (and half of it back downwards towards the planet, or at least the layer of air directly under it). It is definitely a point worth making that the greenhouse effect is likely not dominant at lower altitudes.

This could be a substantial form of back warming, but not involving gases and the GH effect as defined.

Well H2O vapour is a greenhouse gas, so it still involves gases, and it is still the GH effect. What you're talking about here I think is droplet vapour (rather than molecules) but I'm not sure you're not talking about the same thing. A molecule's ability to absorb heat is the same mechanism that gives a bulk of water a good heat capacity. It's exactly the same thing. H2O is good at absorbing heat.

The warming of a desert by GHG back warming is something that is never experienced.

Well, that's not quite true, otherwise deserts would rapidly cool to -273 degrees C. Deserts are kept warm (or at least prevented cooling to absolute zero) in part by the atmosphere preventing this heat loss by the usual means, part of which is GH effect. Clouds just make this effect more marked.

Thanks for your comments.

Your fourth reply: I raised a point that does not seem to appear in any energy budget diagrams. The liquid water in clouds will absorb black body IR from the surface. This will warm the water, which itself can act as a black body radiating IR in all directions, including back to the surface. Given that an average cloud can hold a huge amount of water, this can be a significant amount of heat returned. This black body IR radiation follows S-B law and is not confined to H2O absorption bands, so it is not a GHG emission.

I then go on to discuss the water vapour associated with clouds, which is a GH gas.

This black body IR radiation follows S-B law and is not confined to H2O absorption bands, so it is not a GHG emission.

Well, this is not right. Liquid water does not radiate a Planck curve - liquid water IS confined to the H2O absorption bands, I'm afraid, I'm not sure where you got the idea that a liquid behaves differently.

Sure, the bands are slightly different for the phases of water, but not amazingly so. See this graph

This represents the difference between the absorption peak moving between approx 2.5 and 3.0 micrometers, But's it's still a peak, it overlaps with the gaseous absorption peak. It is in no way a SB Planck curve.

Ok but it still not a GHG if it is liquid water.

I guess the main points I am trying to make are that collisions greatly reduce GHG photon emissions at lower altitudes and must surely still degrade them to some degree in the upper atmosphere. Clouds are the main source of IR emissions from the atmosphere.

Martin A : I think your last sentence defines the problem.

It's not a Greenhouse GAS, admittedly, but it's a vapour acting like one, semantics only, it's not a new unfound physical phenomenon, it's just water (the same stuff that was gaseous before the cloud formed) cooling to below the dew point and turning into tiny blobs of liquid. The molecules are still governed by physics.

Clouds are the main source of IR emissions from the atmosphere.

You haven't proved that, only stated it, and your theory has a few problems:

1. IR photons are measured in quantity on cloudless nights. See here

2. A molecule of water in a droplet is almost as likely to take part in a kinetic collision as a gaseous one.

3. Any photons from clouds in the atmosphere will be quickly turned into convection anyway.

I see where you are going with this:

1. IR effects are minimal at low altitudes.
2. Liquid water has a greater heat capacity than the gaseous stuff.
3. Therefore clouds absorb all the IR, not 'greenhouse gas'

The flaws with this are:

1. Clouds are made up of condensed GHG anyway.
2. A liquid water droplet has no more heat capacity than the sum of the gas that created the droplet.
3. Clouds ARE made from greenhouse gases. See point 1.
4. Nobody is arguing that H2O isn't the majority of GHG warming. It is.
5. GH theory is only arguing about the 20-25% GH effect contributed by other gases, the ones that are going up.

I accept that I was nit picking to differentiate between water phases and GHG.

I did say at the outset that I'm an amateur! I'm trying to get to grips with some of the issues that don't often get mentioned never mind explained. I appreciate your comments.

The things I am still keen to get my head round are these:

(It is probably best to limit this to CO2 because of your point 5 and also because it avoids the special case of clouds)

Collisions of excited CO2 reduce photon emission particularly in the lower atmosphere. This suggests an efficiency vs altitude curve for the gas. Has anyone worked the numbers? Collisions warm the local atmosphere and create convection. I guess overall this is a failure of the GHG although the process takes longer than the loss of the SB radiation directly to space.

Photons emitted by an excited CO2 molecule towards the surface represent a win for the warming effect though the photon could be absorbed by more CO2 (or water where the absorption bands overlap) in which case we are back to the probability of a collision.

What I am leading to is that the GHG process must have some sort of efficiency score that varies with altitude. I'm guessing that the efficiency is not high. This sort of concept doesn't seem to get talked about in the usual blogs though I accept that it may feature in papers if you know where to look.

Being an amateur I have no idea whether these considerations are all fully taken into account when GHG implications are presented.

Schrödinger's cat

In the troposphere the path lengths are short and a photon from one CO2 molecule will probably be absorbed by another.Eventually the energy will be absorbed by the ground or escape into the stratosphere.

In the stratosphere path lengths are long enough that an emitted photon is more likely to return to the troposphere or be radiated to space. This is the main heat loss mechanism at this altitude, one reason why the stratosphere is cold and why increasing CO2 is expected to make it colder

"So, greenhouse gases are likely to create convection rather than return heat to the surface. The problem at this point is that I have just demolished 99% of the GH effect so I started to consider the conditions that would permit downwelling IR."

Greenhouse gases do mostly cause convection, but convection isn't able to cancel the entire effect because of a feature called the adiabatic lapse rate. When gases are compressed their temperature rises, and when allowed to expand, the temperature falls. And atmospheric pressure varies with altitude, so air that rises expands and cools, and air that falls is compressed and warms up. If the vertical temperature gradient is less than the rate at which this compressive warming would occur, then air cannot rise because it would cool faster than the air around it. Convection stops short of equalising the temperature, and air is stable even with cold air on top of warmer air. That's why the tops of mountains are colder than their bottoms.

The adiabatic lapse rate forces a convective atmosphere to have a fixed gradient - the temperature drops by about 6.5 C for every km you climb. But it doesn't set the absolute level. If the temperature at every height rises by the same amount, the slope of the temperature-height graph is the same and the atmosphere is just as stable. So the lower atmosphere warms or cools as a whole, all together, and the surface is always 32.5 C warmer than the point 5 km up, and always 65 C warmer than the point 10 km up, and so on for all points in between.

So far, all we've need is convection and the physics of gases under compression.

So what controls the temperature? Well, we've got a certain amount of heat energy coming in from the sun, and it has to escape at the same rate if the temperature is to stay constant. A black body radiates more energy as the temperature goes up, so there will be a temperature at which the Earth radiates just the right amount of energy out into the vacuum of space. Any lower and the Earth as a whole would absorb energy until it warmed up. Any higher and it would lose energy until it cooled down. Since the atmosphere warms or cools as a whole, the part of the Earth that radiates to space will warm or cool until it reaches that temperature at which the energy balances.

And if there are greenhouse gases in the atmosphere, then the part that radiates to space is (on average) somewhere high up in the atmosphere, so it is the middle of the atmosphere (about 5 km up) that settles at the black body temperature to balance input and output. And then, as discussed above, the surface is about 32.5 C warmer than that, because of convection and the adiabatic lapse rate.

What happens if you add more greenhouse gases? The atmosphere gets more opaque, the radiation to space occurs from higher up, and there is a bigger height difference between here and the surface. So the surface temperature increases. If the average rises to 5.1 km above the surface, the surface temperature rises to 6.5 C/km * 5.1 km = 33.15 C above the black body temperature, or about 0.65 C warmer than it was.

Two other cases: Venus has opaque clouds about 50 km up that radiate to space, and an adiabatic lapse rate of about 8 C/km, so the surface is 8 C/km * 50 km = 400 C warmer than the clouds, where the temperature is such as to balance incoming and outgoing radiation. The Earth's oceans absorb sunlight 10s of metres down, but are virtually opaque to thermal IR - about 20,000 times more powerful a greenhouse agent than the greenhouse gases in the atmosphere. Because water is incompressible, the adiabatic lapse rate is near zero (about 0.1 C/km), so over 10 metres there's no greenhouse warming. Note that if you work out the effect of the downwelling radiation emitted by the water, acting purely as a "downwelling IR greenhouse" and ignore convection, the oceans would boil. The temperature would rise thousands of degrees in the topmost metre of water. In fact, the oceans as a whole do convect (the global thermohaline circulation) and do indeed warm about 0.5 C as you descend 5 km to the ocean depths. That's just as legitimate a "greenhouse effect" as on Venus, even though it's even darker down there than it is below the clouds on our sister planet.

What happens if you add more greenhouse gases? The atmosphere gets more opaque, the radiation to space occurs from higher up, and there is a bigger height difference between here and the surface. So the surface temperature increases. If the average rises to 5.1 km above the surface, the surface temperature rises to 6.5 C/km * 5.1 km = 33.15 C above the black body temperature, or about 0.65 C warmer than it was.

That is the conclusion that I have reached too but it is not that of climate science as far as I understand that. Their conclusion is that during changing concentrations of radiative gasses, emission is from a greater altitude where it is colder which leads to a reduction in outgoing radiation. That energy difference is stored by thermal inertia - the trapped heat scenario. If this is correct then the question is how long does the effect of delta CO2 last: less than 1 diurnal cycle or more? Less and the theory (scare) is irrelevent, more and the theory is currently falsified by observation.

how long does the effect of delta CO2 last: less than 1 diurnal cycle or more? Less and the theory (scare) is irrelevent, more and the theory is currently falsified by observation.

Nov 10, 2014 at 9:48 PM | Unregistered Commenterssat

Sounds like a good opportunity to test the hypothesis. How might it be measured?

"That is the conclusion that I have reached too but it is not that of climate science as far as I understand that. Their conclusion is that during changing concentrations of radiative gasses, emission is from a greater altitude where it is colder which leads to a reduction in outgoing radiation."

Same idea from a different perspective. Like I said, if the temperature at the emission altitude is too low, the Earth absorbs more energy than it radiates, gains energy, and thus warms up until equilibrium is restored. That's the situation they're describing.

My approach describes the shifting location of the equilibrium point. Their approach describes the unbalanced situation pushing it towards the new equilibrium. Put an extra weight in the pan of a spring balance, and in the instant before anything moves you have the pointer still pointing to the original value, but the weight downwards exceeding the pull of the spring upwards, accelerating it. After a few moments, the extra stretch of the spring balances the extra force, and the balance is static again pointing to the new value. If you gradually increase the weight, which description is more intuitive? It's always closer to equilibrium than the former approach suggests, but never exactly on it as the latter does.

It may be best to teach both and leave it to the student to decide which they prefer.

They both describe the same thing, they just explain it from different perspectives. You need the greater IR effects of more CO2 molecules to explain the raise in the IR opacity ceiling. I love the simplicity of the IR ceiling plus lapse rate explanation, but I've found when trying to explain it, that it makes little sense to the layman, perhaps because it assumes a knowledge of how a lapse rate works, and the non-intuitive idea of opacity of an invisible type of light.

NiV / TBYJ,
I am not sure that they are the same idea or the same thing although at equilibrium they coincide. Climate science is concerned about the dynamic situation (of CO2). They introduce time into their deliberations. If the time period is short enough to be neglected then BOA temperatures will increase only by the increasing height of average emissions via the lapse rate: there is no extra energy in the system, only a redistribution of it. If the time period is significant then there is extra energy in the system: their 1 deg C per doubling. If time is effectively zero then a doubling would produce significantly less than 1 deg (my guess).

EM,
Not being a scientist I have no idea how it might be measured other than the obvious one of radiative imbalance at TOA. None has been detected and what measurements have been made show no trend. The 1 deg appears to be speculative. Why not try the calculations assuming the temperature gradient travels with the average height of emission?

S's Cat,
Radiation from the atmosphere does travel downward as well as upward but it is the net exchange that determines the rate of cooling. The net exchange is controlled by the temperature difference: downwelling radiation can't add more energy as it has already arrived into the system; you can't count it twice. When night time temperatures under cloud are higher than under clear skies is it not because the temperature difference under cloud is less?

Seat

If you are thiking in terms of local warming I think the ∆CO2 lag is considerably longer than 1 diurnal cycle., otherwise it would be obvious. For the planet as a whole the lag is generally regarded in the trade as 30 years for TCR and several centuries to equilibrium.

S's cat

Spot on. I think it is because clouds damp convection. They also reflect longwave radiation

Ssat

I found a paper describing a 25% increase in CO2 overnight compared with afternoons in residential Phoenix.(but it wont href :-( ).

The standard 5.35ln(C/Co) formula gives 5.35ln(1.25) and an increase in downwelling radiation of 1.19w. This corresponds to a nocturnal reduction in cooling of 0.3C.

Since this equation is usually used on a whole atmosphere those figures represent an upper limit considerably larger than likely in practice. Against the normal diurnal temperature variation, I doubt it would be noticeable.

Excellent comments. I've learned a lot here. I had figured out the connection between the radiation at a particular altitude/temperature, the lapse rate and the surface temperature, but your comments have helped.

I still have some bits to understand but I'll study all the comments a bit more.