I’ve just encountered a new global warming site called Global warming solved (archived). The site appears to be run by a family who also owns a journal called The Open Peer-Reviewed Journal, the first 8 papers in which are all authored by them. You can find the journal quite easily if you wish, but these papers include a review of millenial temperature reconstructions, a review of temperature homogenisation, and a major correction to the physics of the greenhouse effect (which, according to them, is negligible). These people are either incredible polymaths or ….
On their brand new website, they summarise their papers. In one of their summaries (archived here) they say,
This means that we should be finding a very complex temperature profile, which is strongly dependent on the infrared-active gases. Instead, we found the temperature profile was completely independent of the infrared-active gases.
This is quite a shocking result. The man-made global warming theory assumes that increasing carbon dioxide (CO2) concentrations will cause global warming by increasing the greenhouse effect. So, if there is no greenhouse effect, this also disproves the man-made global warming theory.
To be fair, this is quite a tricky concept and one that I’ve had to think about for a while before understanding myself. However, it is – I believe – roughly correct that the greenhouse effect does not influence the temperature gradient (typically called the lapse rate). What it does, though, is make the lower atmosphere opaque to outgoing long-wavelength radiation and hence makes the lower atmosphere adiabatic (As per various comments below, it’s not the atmosphere as a whole that’s adiabatic, it’s that we can treat this as a reversible adiabatic process. As parcels of air rise and fall they don’t lose any energy, but either expand and cool, or contract and heat up. This is, also, very much a simplifying assumption and the actual processes involved in setting the lapse rate are much more complicated than this simple, illustrative model). One can therefore estimate, very simplistically, the lapse rate using fairly basic physics.
For a reversible adiabatic process (isentropic),
The can be differentiated to give
The term PdV is the work done which, for an adiabatic process, is also given by -ncvdT, where n is the number of moles, and cv is the constant volume heat capacity. This allows us to write,
where I’ve used that γcv = cp (the heat capacity at constant pressure) and that if this is written in terms of mass, rather than moles, n/V is the density.
In hydrostatic equilibrium, the downward force of gravity on a gas parcel, must be balanced by an upward pressure force. Therfore,
We can combine this with the relationship between dP and dT that we derived above to get
The term g in the above is the acceleration due to gravity and is g = 10 m s-2. The specific heat capacity at constant volume is cp = 1000 J kg-1 K-1. Therefore we get that the adiabatic lapse rate is 0.01 K m-1, or 10 K km-1.
Therefore, in a fully adiabatic atmosphere (or, more correctly, one which is isentropic), the temperature should drop at 10oC for every 1 km increase in altitude. This is actually a little too high as the atmosphere will typically contain water vapour that can condense and heat the air. In this case we get a saturated adiabatic lapse rate, which is closer 5 K km-1.
So, where does the greenhouse effect come into play? Firstly, greenhouse gases in the atmosphere means that the lower atmosphere can be regarded as, approximately, adiabatic (i.e., it is opaque to outgoing radiation and hence doesn’t lose energy – directly to space – through radiative processes. It is isentropic.). Secondly, the greenhouse gases determine an effective radiative surface at which energy can escape into space via radiation (i.e., some level at which it effectively becomes optically thin). In equilibrium, the temperature at this altitude will be the temperature the planet would have if there were no greenhouse gases present. The more greenhouse gases there are in the atmosphere, the higher this radiative surface, and the higher the surface temperature will be (because of the lapse rate). So, just because the lapse rate does not appear to depend on the existence of greenhouse gases in the atmosphere, does not invalidate the greenhouse effect, although this does appear to be a common misconception.
I should say that I just worked this out this morning and may not have explained this quite as clearly as I could have. There are also almost certainly subtleties that I’ve missed that others (Tom Curtis, Eli Rabett) could probably highlight. However, I do believe that the discovery that the lapse rate does not depend on the greenhouse gas concentration is not new and does not mean that the greenhouse effect is negligible. As usual, corrections welcome through the comments.
...and Then There's Physics 
I should add that if you want a nice explanation for the greenhouse effect, Stoat has a good post. ReaclClimate has a similar one that goes into a little more detail and is a little more complex.
You’ve gone to a lot of trouble explaining something that doesn’t really need explaining. That mob with their dodgy journal are claiming that 150 years of established Laws of Physics are incorrect and they offer no alternative. I stopped reading after about 150 words because facepalming hurts my face.
uknowiss,
I’m fairly sure you’re right. To be honest, one reason I wrote this was because I hadn’t worked through this myself for a long time and I can see how it might not be immediately obvious to some. I learned something, even if noone else did 🙂
I should add that a tweet that I got
was pointing out that the lapse rate can change due to feedbacks. If feedbacks produce larger changes in the upper troposphere than the lower, than the temperature gradient get shallower (i.e., changes more slowly). Alternatively, feedbacks could produce more warming in the lower than upper troposphere, producing a steeper temperature gradient. This is explained in more detail here.
FWIW
The atmosphere isn’t adiabatic. That would mean that no energy enters or leaves. Greenhouse gasses don’t affect this one way or the other.
The lapse rate is because of the isentropic (adiabiatic AND reversible) expansion of a parcel of air as it rises (or falls) and expands (or contracts). Adiabatic is not a sufficient condition and the conditions applies to the air parcel, not the atmosphere.
The atmosphere is not opaque. Radiation captured by GHG at any level in the atmosphere is re-radiated in all directions, including to space.
Increasing GHG increases, in a sense, the “recycle rate” of radiation. But this can be (and should be to some extent) offset by convective heat transfer.
Increasing GHG increases surface temperature by raising the effective emission altitude. At the effective emissions altitude the temperature should be effective emissions temperature, which is fixed for a fixed incoming radiation budget. If it is raised, then because of the ALR, the surface temperature increases.
kdk33,
Yes, you’re right. The term “adiabatic” refers to the parcels of air, not the atmosphere as a whole. I thought I’d kind of got that right, but maybe I didn’t express myself clearly enough.
Yes, that is the fundamental point, which I thought I’d stressed but, again, maybe not as clearly as I could have 🙂
You introduce this post with “new global warming site.” A more accurate way would be a “new global warming denier site.”
BG, true and given that they appear to deny the existence of the greenhouse effect, that would appear to be a fair characterisation. Admittedly, I have started using archive.is, rather than linking to these sites. A step in the right direction 🙂
andthentheresphysics,
Thank you for your interest, and taking the time to read some of our summaries. 🙂
Your derivation of the dry adiabatic lapse rate is indeed a standard result, and as you correctly point out is independent of greenhouse gas concentrations.
However, in our “The physics of the Earth’s atmosphere” Paper 1 and Paper 2 (which you are referring to here), we were looking at a different phenomenon.
We were looking at the absolute temperature profile. That is, we weren’t just trying to understand how temperature changes with height (“lapse rate”), but trying to understand the absolute temperatures at each height (in Kelvin).
Surprisingly, when we analysed a large sample of weather balloons (several thousand for Paper 1, and more than 13 million for Paper 2), we found that the temperature profile for each balloon could be described entirely in terms of the thermodynamic properties of the bulk gases (i.e., nitrogen, oxygen & argon) and water vapour concentration.
This was an unexpected result (at least to us!). According to the greenhouse effect theory, the presence of greenhouse gases should be substantially altering the temperatures at different altitudes. Roughly speaking, they should be:
(1) Increasing the mean temperature (and height) of the troposphere, relative to a pure nitrogen/oxygen atmosphere.
(2) Decreasing the mean temperature of the stratosphere, relative to a pure nitrogen/oxygen atmosphere.
What we found is that the mean temperature profile throughout the troposphere and stratosphere is essentially the same as that of a nitrogen/oxygen atmosphere (with some variation near ground level due to water vapour).
Ronan,
I’ll be quite blunt. I think you’re making a fundamental and potentially very embarrassing mistake.
I’m not sure that I quite understand what you’re getting at here, but I think you’re horribly mistaken.
Let me see if I can explain. Consider a scenario where an atmosphere has no greenhouse gases. In that scenario, there is no troposphere because the energy can simply be radiated directly from the surface and the surface temperature will be set by the non-greenhouse temperature.
Now consider an atmosphere with greenhouse gases. The energy cannot simply radiate from the surface because it is trapped by the greenhouse gases. In this case, the lower troposphere temperature profile can be approximated by the adiabatic lapse rate. If there’s water vapour, it will be the saturated lapse rate, but that doesn’t really matter for this illustration. The temperature will therefore drop with altitude until the atmosphere becomes sufficiently optically thin to simply radiate the energy into space. Because, on average, the energy radiated into space has to match the energy received, the temperature at this altitude will be the non-greenhouse temperature. Since the lapse rate is negative (i.e., temperature decreases with height) the surface temperature has to be higher than the non-greenhouse temperature.
If one continues to add greenhouse gases then, in a simple sense, the lapse rate doesn’t change. What does change is that effective emission height of the atmosphere. The higher it is, the higher the surface temperature has to be since the effective emission temperature remains unchanged, but the lapse rate means that the higher the altitude of the emission the higher the surface temperature has to be.
This is a little simple because, as illustrated in some comments above, feedbacks can change the lapse rate. But, I do not that what you’ve found in any way invalidates the greenhouse effect. I think it may well be precisely what’s expected.
andthentheresphysics,
You say:
Why do you believe there can be “no troposphere” without greenhouse gases? One definition of the “troposphere” would be the lower part of the atmosphere with a mean lapse rate of about -6.5K/km, as opposed to the “tropopause” (lapse rate = about 0 K/km) or the “stratosphere” (positive lapse rate).
As you point out in your blogpost, this lapse rate is independent of greenhouse gas concentrations. Are you using a different definition of the troposphere?
ATTP,
You are 100% right that the expected outcome is that the lapse rate changes only a little. The dry adiabatic lapse rate changes so little from the change in Cp that this is not observable in practice. The moist adiabatic lapse rate changes much more as the influence of latent heat release from condensation increases with moisture.
Otherwise your post is not a particularly good presentation of the mechanisms. As kdk33 wrote, saying that atmosphere is adiabatic does not make sense, nether does your comment that GHGs make the lower atmosphere more adiabatic. It’s not possible to make your description correct by minor changes, it would be necessary to discuss, how convection takes place, and how the role of GHGs is essential for the presence of such convection in the troposphere. I wrote just two sentences on that, but much more is needed to explain the issues correctly.
Ronan,
Because without greenhouse gases, how can it be adiabatic? The energy can simply radiate straight through the atmosphere. Any energy radiated by a gas parcel can also simply radiate through the atmosphere. Therefore the adiabatic assumption fails and the temperature profile would be very different. It would depend on what energy can be absorbed by the different molecules and since you’re arguing against a greenhouse effect, they can absorb nothing and hence I’m not even sure how one would estimate the temperature profile for such an atmosphere. An an example though, Mercury has a thin atmosphere, but no troposphere.
One way to look at the problem is that heating or cooling by radiative flux divergence is a “slow” process, with rates on the order of a degree or two per day. Atmospheric dynamics, in which I will include convective mixing, is a “fast” process — think of how fast a thunderstorm grows and mixes up the air. So any tendencies for concentrations of greenhouse gases to make a complex temperature profile don’t have a long enough time to accumulate and directly affect the lapse rate.
There are places where tendencies due to radiative flux divergence are larger, such as cloud tops, but they don’t have a direct effect on the large-scale lapse rate.
Pekka,
I don’t doubt that you’re correct that much more would be needed to explain things clearly and I agree that my statement that the atmosphere is adiabatic isn’t strictly correct, although I do think you’re being a little pedantic. Even kdk33’s comment was a bit pedantic as I wasn’t trying to suggest that the entire atmosphere was adiabatic, simply that one can use an adiabatic approximation – although I didn’t phrase that very well. It is just a short blog post.
I also think that saying GHGs makes it more adiabatic is a reasonable thing to say (unless you can convince me otherwise). Without GHGs the adiabatic approximation fails and one cannot represent that temperature profile in the lower troposphere using an adiabatic lapse rate (saturated or not). In a sense, that the temperature profile in the lower troposphere is well represented by an adiabatic lapse rate essentially proves the greenhouse effect and this post was intended to counter claims that the greenhouse effect is negligible.
ATTP,
You have misunderstood the meaning of the word adiabatic. The atmosphere without GHGs may well be adiabatic, but it cannot maintain the adiabatic lapse rate.
Contrary to what Ronan Connolly writes some GHE is needed for adiabatic lapse rate. It turns out, however, that an extremely small amount of GHGs is enough for that. The lapse rate would be close to the present one (or actually closer to the dry adiabatic lapse rate). The temperature drop from the surface to the tropopause would also be close to the present one, but both the surface and the whole troposphere would be tens of degrees colder that they are now.
ATTP,
Actually the atmosphere is the more adiabatic the less we have GHGs, with zero GHE it’s really close to fully adiabatic, but without an adiabatic lapse rate.
Pekka,
Okay, if an optically thin gas does not radiate it could still be adiabatic (by the way, I do know what the word adiabatic means – I’m not quite convinced that you do at this stage). Would an atmosphere without greenhouse gases be adiabatic? It would be optically thin and I can’t see why it wouldn’t radiate, so my guess is no. Maybe you can convince me otherwise.
Pekka,
I think you’re going to have to explain this very carefully. A non radiating atmosphere, sure. Is that a reasonable assumption for an atmosphere with no GHGs? I’m yet to be convinced.
Pekka,
I will say that it seems ironic that you said
followed by
ATTP,
Do you understand the theory of an optically thin atmosphere as presented in the book of Pierrehumbert?
That’s a good starting point. It tells that the temperature at tropopause is the surface temperature divided by the fourth root of 2. Such an atmosphere does, however, not protect the surface from cooling. thus the surface would be essentially as cold as without any atmosphere, and the troposphere correspondingly colder than the present one.
Pekka,
That doesn’t answer my question. Can you please do so. Would an atmosphere without GHGs be truly adiabatic?
ATTP, I think what these guys confused is climatic variability (which determines the local temperature profile) with slow changes in the trend. Therefore they fail to identify the extremely weak GHG signal. I didn’t bother to check (and I won’t do so), hence I might be wrong. No doubt though that they are terribly wrong.
In any case, Chris Colose elaborated on the issue in an easily understandable fashion a while ago (it’s also in the spirit of kdk33’s earlier comment). Always worth a read: Greenhouse-effect-revisited
Karsten,
Yes, I’m sure that’s true. I think their observations were over a very short time interval, so identifying any GHG signal is very difficult.
Adiabatic means “without transfer of heat”. Without GHGs there’s no transfer of heat in the atmosphere, and with very little GHGs only very little transfer of heat.
With less GHGs the convective flows get weaker and have less turbulence. Therefore mixing of air between neighboring volumes is decreased, and the flow more adiabatic.
Pekka,
Exactly. Maybe we’re getting somewhere. So, if you have an optically thin atmosphere with low emissivity then it would still be adiabatic. So, there’s a conundrum here. If you have no GHGs then you have no absorption and no emission, so the surface would have the non-greenhouse temperature and the atmosphere would have a temperature profile set by the adiabatic lapse rate with that temperature at the base.
So, one could have an idealised non-GHG atmosphere which is till adiabatic. The point of the post, however, was to point out that the temperature profile isn’t strongly set by the GHGs (although, I acknowledge that this is a simplification). What they set is the height at which we effectively radiate into space and the surface temperature is then set by the lapse rate from this height to the ground. That’s one way of looking at the greenhouse effect, there are certainly others that are more detailed and better, as pointed out by Karsten.
Karsten,
No, in the two papers that Andthentheresphysics is discussing here, we are looking at each individual weather balloon, as a snapshot of the atmospheric profile for a given region over a few hours.
In Paper 1, we carried out a detailed study of all the weather balloons for North America on several particular days.
However, for Paper 2, we wrote scripts to analyse all of the available weather balloons from the Integrated Global Radiosonde Archive. Some of the balloons we weren’t able to analyse (e.g., some of the balloons burst very early), but we were able to analyse about 70% of them – roughly 13 million balloons (mostly post-1960s).
Ronan,
But that’s kind of the issue. The effective emission height in the atmosphere has changed by a few hundred metres over a few decades. On the timescale of hours you could not possibly identify some kind of greenhouse effect signal in your data.
ATTP,
To me it’s totally obvious that any paper that claims to present a proof that GHE is bogus, has nothing to do with correct physics. In most cases the only reasonable thing to do is to simply ignore those sites. in some exceptional cases it makes sense to debunk them.
When you choose to start debunking, it’s, however, better that your own arguments are correct. it’s not enough that you are correct in observing that the site is full of crap, you must in addition make your own arguments right. In this case the weaknesses of your post are so essential that pointing that out is not nitpicking.
Ronan,
No, in the two papers that Andthentheresphysics is discussing here, we are looking at each individual weather balloon, as a snapshot of the atmospheric profile for a given region over a few hours.
Which tells you exactly nothing about the long-term effect of change in GHG concentration. Zilch. You wouldn’t be the first who confused different timescales. Try to submit to a reputable journal, then you’ll get at least two ppl (the referees) to read your stuff.
Pekka,
Why don’t you read my post again and tell me where specifically I’m definitively wrong. You yourself used “atmosphere is more adiabatic”. I also acknowledge that it was simple. I really don’t mind people correcting my mistakes. Pedantry, however, tends to rather tick me off, especially when it’s accompanied by an element of irony.
I think we actually agree and apart from maybe me including that an optically thin atmosphere with low emissivity could still have an adiabatic lapse rate, I think the point I was trying to make in my post is valid. You were simply highlighting a highly idealised scenario which is illustrative but probably not realistic. If it was, why do the Moon and Mercury have exospheres and no tropospheres?
Pingback: The lapse rate | Garry Rogers Nature Conservati...
Pingback: The lapse rate | GarryRogers Conservation and Science Fiction: #EcoSciFi
ATTP, Peka,
Don’t forget conduction. No GHG required.
Convection creates the ALR, which is an isentropic result (not adiabatic, which is an insufficient condition to derive the usual ALR, as was done by ATTP, an many others). Any mechanism that drives convection is sufficient for ALR.
which should actually be the ILR.
Pekka,
Okay, let’s see your nitpicks and how essential they are. I guess this
isn’t quite right. The atmosphere does lose some energy directly to space, and one could have an idealised atmosphere that’s optically thin and with low emissivity and that still has an adiabatic lapse rate. However, I was simply trying to point out that one could treat the lower atmopshere as adiabatic (or the gas parcels) and that the GHGs essentially prevent energy from being lost directly to space. Admittedly simple, but not completely wrong, I think.
I then said,
Again, a simplification but I’m failing to see how this is fundamentally wrong.
Am I highlighting the bits you were thinking of, or are there more?
kdk33,
Just to be clear, this post wasn’t meant to be a definitive explanation for the atmospheric temperature profile. You may need to explain this
a little more.
Sure,
You start your derivation with PV^gam = const. This is only true for an isentropic system; not generally true for adiabatic systems. To be isentropic is to be adiabatic AND reversible. The lapse rate comes from the interconversion of heat and work (h/t to the second law) which can occur any number of ways, but in the atmosphere it occurs in a very specific way, and we describe that way as reversible..
When the atmosphere convects (transfers heat by bulk movement of stuff, in this case “air”), it does so (must so) along the lapse rate. If the atmosphere is heated from below (the surface), it will convect. Any mechanism that drives convection will work.
My explanation is probably not perfect. It is not my intention to nit-pick. .
kdk33,
I agree that it’s isentropic, I disagree that it’s not generally true for an adiabatic system, although I agree that in an adiabatic system with shocks, it is not true, but then that’s not isentropic either.
I was trying to avoid discussing, in the post at least, how the energy was transported, but I don’t think what you’ve said here is inconsistent with what I wrote in the post. I agree though, with what you’ve said here.
When conduction is important, the atmosphere approaches isothermal. The theory of the optically thin atmosphere assumes that radiative heat transfer within atmosphere is weak, but still so much stronger that convection can be disregarded.
Explaining the GHE correctly requires that both the origins of the lapse rate and the influence of the lapse rate on the outgoing radiation are considered. GHGs are essential for both, but the extra effect from more GHGs is mainly due to the radiative part. I have never seen that anyone would link that to the concept of adiabatic.
Pekka,
I think you’ve misunderstood the motivation behind my post. I wasn’t really trying to explain the GHE. I was trying to point out that looking at temperatures and temperature profiles in isolation cannot really be used to disprove the GHE. I also don’t quite agree with your final sentence. In a simplistic sense, the GHGs set the effective emission altitude. The lapse rate (which is negative) then sets the surface temperature.
As you’ve already said yourself, if the atmosphere was optically thin and had a low emissivity, then the surface temperature would be the greenhouse temperature and the atmospheric temperature would decrease with altitude. As one increases the greenhouse gas concentration in the atmosphere, the effective emission altitude rises as does the surface temperature because of the negative lapse rate.
I accept, however, that that is simplistic and ignores many processes but I still think that it’s not an unreasonable explanation for why the surface temperature is higher then the greenhouse temperature (especially as it was motivated by others arguing that they could look at atmospheric temperatures and use that to rule out the GHE).
ATTP & K.a.r.S.t.e.N.,
Do you agree that, according the the greenhouse effect theory, there should be a greenhouse effect present at all times (whether 2014 AD or 1714 AD)? Even during the pre-industrial era, there should have been a greenhouse effect (according to the theory).
It is true that, according to the greenhouse effect theory, the greenhouse effect should have been increasing since the pre-industrial era (since CO2 concentrations have been rising significantly from fossil fuel emissions). As you & K.a.r.S.t.e.N point out, this increase in the greenhouse effect signal should be a gradual process (decades or longer). But, the greenhouse effect should at all stages be having a substantial affect on the entire atmospheric temperature profile. That is, it should be detectable in any snapshot of the atmosphere (regardless of timescale).
I think the “GHG signal” you are referring to is the predicted increase in the greenhouse effect from increasing GHG concentrations. Is that correct?
That’s not we were looking for. We were attempting to calculate how much the greenhouse effect alters the underlying temperature profile of the atmosphere for a given weather balloon. Do you know what I mean?
This GHG signal should be detectable over any time interval (short or long), but we were unable to detect it.
Now, I know that “absence of evidence is not the same as evidence of absence”! But, in our Paper 1, we directly compared our results to the theoretical “infrared cooling models” used by the current Global Climate Models.
[By the way, in case any of you are unfamiliar with the term “infrared cooling model”, these are just the implementations of the greenhouse effect which are used in climate models – we give a brief summary in our paper]
P.S. Sorry if I haven’t had to chance to reply individually to everyone else who has commented here. Thanks for your interest in our work – it is appreciated (even if the comments on this blog have been mostly negative! 😉 ). But, it’s hard to respond to everyone – while I was writing this reply, I see there have been several new ones!!!
So, I’ll try & get back to you later…
ATTP,
I understood your motivation, I have no complaints on that, but as I wrote above, it’s not enough that the one being debunked is wrong, the arguments must also be correct. Your description is very different from every correct description I have seen. In particular you use the word adiabatic in a very different (and wrong) way. That’s at best confusing.
Ronan,
What do you mean by a greenhouse effect? What I was trying to illustrate in this post is that, to a first approximation, the temperature profile in the troposphere is not affected by the GHG concentration. What the GHGs do is set the effective emission height. The surface temperature is then set by working from that height, back down to the ground along the lapse.
If one increases the GHG concentration, then that increases the emission height but doesn’t change the temperature at that height. If you work back down from this new height along the lapse – to a first approximation – the temperatures at all heights in the troposphere should increase. Similarly the temperatures in the stratosphere at given heights should decrease. However, this can only be measured by considering data over a very long time interval. Decades.
So, what precisely are you measuring that you’re claiming shows no GH signal. I don’t think there is any signal one can measure by considering measurements over a few hours.
Pekka,
I’m still rather confused as to why. I don’t think that I’m using it in a different way to the way you mean. Maybe I haven’t made that clear, but I’m as yet to see you provide a different term. From everything I’ve read, it’s called the adiabatic lapse rate. I know it’s idealised, but I didn’t say otherwise.
Ronan,
But, the greenhouse effect should at all stages be having a substantial affect on the entire atmospheric temperature profile. That is, it should be detectable in any snapshot of the atmosphere (regardless of timescale).
No, it shouldn’t! As ATTP is discussing here, the lapse rate is not a function of GHG concentration (to a first approximation).
This GHG signal should be detectable over any time interval (short or long), but we were unable to detect it.
If that’s what you did, I’m not surprised you weren’t able to detect anything. That’s exactly what anyone with good textbook knowledge would expect.
One more go and I am done.
ATTP,
I guess we disagree. Adiabatic refers to heat in or out of the system. Reversible refers to how work and heat are interchanged. To derive the ALR, you have to constrain both – adiabatic AND reversible, or isentropic. PV^gam = const is only true for isentropic systems. Not true for adiabatic systems. I think you can confirm this in most undergrad thermodynamics books.
Pekka,
Important is a relative term – relative to what. It is convection that creates the lapse rate, without convection there would be no lapse rate. So, I think convection is important to understanding the lapse rate.
I agree that If there was conduction with no convection the atmosphere would be isothermal, but I don’t’ think that’s an interesting model for here. If the atmosphere was heated from the top, there would be no convection, no lapse rate, but maybe an inverse temperature profile.
We could go one deeper and acknowledge that the ILR is an asymptotic condition. The atmosphere convects when it is perturbed away from it. If heated from the bottom the atmosphere convects in an attempt to restore the lapse rate. So, it is actually never quite at the lapse rate. So what would happen if conduction dominated convection: Convection would work very hard to restore the lapse rate, but if conduction were strong enough it could keep the atmosphere perturbed, a lot even, but convection would never give up. Providing it is heated from the bottom.
Might also be worth noting that all of the energy lost to space is lost by radiative transfer. So the radiative versus convective question is only interesting when talking about how heat moves about in gasses of the atmosphere.
Have a nice Sunday.
Ronan Connolly,
The GHE has always been and is presently very visible in the atmosphere. Without a strong GHE it’s not possible to have an atmosphere with a strong circulation. The strong circulation is one way of obtaining the atmospheric temperature profile, but a similar profile can be obtained also with the weak circulation of a weak GHE.
The main change from more GHGs is an increase first in the net energy flux and then in the temperature. The temperature profile changes very little.
kdk33,
I feel like I’m in the middle of a discussion with a bunch of people with whom I agree, but who insist on disagreeing with me 🙂 I agree that it is for an isentropic system, but an adiabatic system without shocks is isentropic. I believe that the form I used is – even in textbooks – referred to as a reversible adiabatic process.
kdk33,
I haven’t noticed any significant disagreement between our thinking.
The lapse rate is controlled by convection, which sets an upper limit for the lapse rate (i.e. the temperature cannot decrease more rapidly with altitude, but all smaller values and all values of the opposite sign are possible.) Maintaining convection consumes free energy. The source of free energy is heating of warm parts of the atmosphere and heat loss from cooler parts. In addition the loss should occur at higher altitude, heat loss at the surface of high latitudes is not efficient in driving convection.
We have the adiabatic lapse rate in all locations where the lapse rate would be higher without convection. Where that’s not the case convection is either absent or driven by circulation in neighboring areas.
Tropopause is the altitude above which convection is not need to make lapse rate equal or less than the adiabatic lapse rate. At those altitudes convection is mainly absent, but may occasionally be driven by lower atmospheric circulation.
Pekkka @ 4:19pm – I think it clear that ATTP is engaging in an internet based mix of thinking out loud, discussing things down the pub with their friends and trying to learn from others mistakes.
Thus your comment comes across as a bit bitchy and irrelevant.
Of course you can say there are better ways to educate yourself, but different methods work with different people, and I myself sometimes find it gives an extra spur to learning when working out why some denialists are wrong. We all know that they are wrong, the fun and educating bit is working out why.
ATTP,
A reversible process is always isentropic, thus a reversible adiabatic process is the same as isentropic. Shock waves are not the only source of irreversibility, i.e., of deviations from isentropic. All turbulence has such an effect.
Anders & Pekka Pirilä, I did understand what you were saying, but I would prefer not to use the term “adiabatic air” or “adiabatic atmosphere”. The air or atmosphere has an adiabatic lapse rate is a clearer formulation. Adiabatic refers to a process not to an object.
In general, do I understand it right that the Connolly family rejects the greenhouse effect in general. They write: “Increases in atmospheric CO2 do not and cannot cause global warming.” Nothing in their story seems to be specific for CO2, thus I guess they would make this claim about any greenhouse gas. And if increases in greenhouse gasses cannot warming, I guess that also decreases cannot cool.
In that case we have no need to look at subtle questions around the atmospheric lapse rate, but we should ask the Connolly family, why it in not freaking cold on Earth? How do they explain the warm surface temperature on Earth?
P.S. Can I have your bets on whether Anthony Watts will report on this family of Dragon Slayers?
ATTP,
We tried to provide a reasonable summary on our blog summary. I appreciate it is a very long post for a “summary”, but we were trying to summarise the results of three quite technical papers, and we didn’t want to use too much unexplained jargon. 😦 Have a look at the subsection titled “Greenhouse effect theory: The version used by climate models” (it’s in section 3).
Pekka Pirilä,
When you say The GHE has always been and is presently very visible in the atmosphere., what do you mean by “very visible”?
I agree that within the GHE model, the GHE influences atmospheric circulation, and I agree that there is atmospheric circulation. But that doesn’t in itself prove there is a GHE.
It’s true that this shows that the GHE is consistent with the existence of an atmospheric circulation. But that doesn’t mean that the atmospheric circulation proves there is a GHE, does it? So, why do you say “very visible”?
Ronan – what would it take to convince you that there is a green house effect from certain gases, vapours etc in the atmosphere?
Hi Victor,
To be honest, I’m not entirely sure where this discussion about “lapse rates” arose from. I think it was because we were referring to the temperature profiles of the atmosphere, and ATTP assumed we were referring solely to the lapse rates. Yes, lapse rates are an important component of the temperature profile. But, there is a lot more to the temperature profile than just the rate of change of temperature with altitude (which is essentially all the “lapse rate” is).
We provide a more detailed discussion in our papers, but we give a summary on the Global Warming Solved post. Have a look at Figures 12 and 13. They give examples of what we mean by the “atmospheric temperature profiles”.
You asked “why is it not freaking cold on Earth?” Because there is an atmosphere! If we were on an asteroid orbiting the Sun at the same distance as the Earth, it would be a lot colder, but that’s cause it has no atmosphere!
guthrie,
Have you read our papers or the summary on our website (here) yet?
Victor,
I agree. Adiabatic is used with a process, and my dissatisfaction with the post is in part due to the way the word is used there. When I used it in some comments I had always in mind the totality of processes going on in the atmosphere, but didn’t say that explicitly.
Pekka,
Okay, I finally get it. I should have said “reversible adiabatic process” rather than the way I used. You’re correct. I do, however, refer you back to my earlier comments about pedantry and simply pointing out that “reversible adiabatic process” would have been a better way to phrase it would have seemed a good deal more constructive. If I seem irritated, I am.
Pekka,
Okay, yes any dissipative process is a source of irreversibility. I refer you, however, once again back to my earlier comments about pedantry.
Ronan,
And how does our atmosphere ensure that it “is not freaking cold on Earth”?
I’m having a read just now. It’s amusing; I too have a chemistry degree, but find the current scientific explanation covers everything, whereas you clearly don’t.
ATTP,
If my comments seemed aggressive, that was not the purpose. My purpose has been to get the physics presented correctly and in a non-confusing manner. As the mechanisms of the atmosphere are essential for the GHE, and as they surelyare a bit difficult to understand for non-physicist and in addition understood purely by most physicists who have not spent some time in learning atmospheric physics. The ingredients are familiar to every physicist, but how they operate in the atmosphere is not.
That’s true also in my case. I’m a physicist, and I have worked also with thermodynamics and fluid dynamics, but I didn’t understand the atmospheric physics before reading texts related specifically to it.
Can CO2 cause global warming?
Simplistic models which assume energy transfer in the atmosphere is dominated by radiative transfer predict that “greenhouse gases” such as H2O and CO2 slow the rate at which energy is lost to space. So, these models predict that increasing CO2 concentrations should theoretically cause “global warming”.
However, when we actually looked at the experimental data, we discovered that atmospheric temperatures are independent of the concentration of “greenhouse gases”. In other words, the models were wrong. CO2 doesn’t cause global warming.
I am afraid that this is a typical idea among climate ostriches, but that is not the way science works.
If you find a problem, you shout EUREKA and are happy to have found something interesting. This is, however, only the beginning. Then you start investigating why there is a discrepancy. Before you can claim that greenhouse gases do not warm, you will first have to have found that the reason for the discrepancy is in the radiative transfer equations or in the laboratory measurements of the spectrum of CO2.
Unfortunately for the Connolly family the most likely explanation is that their empirical analysis is not able to find the relationship because it is not designed in a way that it could find the relationship. I did not read the paper yet, but above it was claimed that the analysis is based on radiosonde data, even single radiosondes. This will not reveal the relationship between temperature and greenhouse gasses. Let my explain.
Lets take a simple example with a step change in CO2, and let’s assume that the greenhouse effect is right, then this would create an energy imbalance in the atmosphere and the atmosphere (and especially the ocean) would start heating. In the beginning, however, the temperature is still almost the same and one might be tempered to conclude that CO2 does not influence temperature. This would clearly be a contradiction, the method is thus wrong.
The problem is, as Karsten indicated above, that time and spatial scales are very important, only for a CO2 change on global scales and on decadal time scales, can you see the relationship between temperature and greenhouse gas concentrations.
Pekka,
Constructive would also be nice. I’d quite like to know what you thought I meant other than a reversible adiabatic process. Given that I’d rather not carry on this discussion, I wouldn’t bother explaining.
Indeed, and I often learn a lot from those who comment here, including yourself. Some discussions I enjoy more than others though.
If anyone, like me, wants the greenhouse effect for dummies version, this <6 minute video from Ray Pierrehumbert is well worth watching: How CO2 warms the climate
Myles Allen also has a good one here
guthrie,
If you have a scientific background, you might find the papers themselves more relevant. As you’ve probably guessed by now, the summary on the Global Warming Solved website is tailored for a wider audience than the papers which are more technical, precise and discuss the literature in more detail. In particular, if you’ve a chemistry background, you might find Paper 2 interesting. But, I recognise that they are quite long, so the summary is probably a good place to start.
Victor,
I think we’re coming at this from opposite perspectives. You seem to be assuming that the greenhouse effect either is or isn’t. For our study, we were specifically trying to quantify the magnitude of the greenhouse effect.
For the Global Warming Solved summary, we don’t discuss the radiative transfer equations (which lead to the infrared cooling/solar heating models) or the laboratory measurements of the CO2 IR spectrum in much detail. We felt the summary was already quite long and technical for a general audience. However, we do discuss these issues in some detail in our papers. In particular, we give a 6 page discussion in Section 3.2 of Paper 2 (starting page 16).
P.S. Again, sorry if I’m not able to respond to everyone individually. By the time I’ve written one reply, I see there’s several new ones!
Ronan,
I’d be really keen for you to explain how our atmosphere ensures that we are warmer than an asteroid would be at the same distance from the Sun.
Also,
As others have pointed, I don’t think you can do that by considering radiosonde measurements over a time period of a few hours.
Ronan Connolly
Our host asks, twice, himself repeating Victor Venema’s original question:
You can add me to the list of those eager to hear your response.
Ronan,
I’ve just quickly read those pages and they seem to be describing the greenhouse effect, which – on those pages – you don’t dispute. I’m now confused by what you’re claiming in your papers.
You say,
This seems like an approximate description of the GHE. How are you then concluding that it’s negligible?
ATTP – it’s simple – they created out of whole cloth a mechanism whereby molecules in the stratosphere start clumping together into groups, which releases energy, warming the stratosphere.
The actual errors start, as has been said above, with assumptions about the state of the atmosphere, equilibrium, and mistaking the lack of precise data on temperature changes due to increased CO2 with the greenhouse effect itself.
SO far it reads exactly like the usual stuff, with a lack of attempts at following things to the logical conclusion, i.e. if there really isn’t a greenhouse effect, as you have asked, why is the earth the temperature it is? The consideration of only one particular area without relation to the wider world is a common error.
guthrie,
Thanks, I think my eyes may have glossed over somewhat when I got to “pervection”.
ATTP,
The problem is often not in what we mean, but in how we say that. That applies surely to all of us. Using concepts of physics differently from their normal use is one problem, but the more difficult one is that you skipped so much essential that the explanation was not really be useful. You noted correctly the error in the paper discussed, but you presented the short derivation without telling, what the physical context is that makes that derivation meaningful. Mathematics without proper context is not a good physical argument.
The most essential physical factor needed is that radiative processes alone would lead to a stronger temperature gradient than is stable in the atmosphere, and that convection enters automatically to reduce the lapse rate, when that’s the case. This is the missing piece in the arguments of many skeptics. They present your formula and tell that the lapse rate is there, we don’t need any GHGs. It’s important to counter that kind of thinking every time. Convection is not the source of the lapse rate, it’s a limitation for the lapse rate.
Victor, I think we’re coming at this from opposite perspectives. You seem to be assuming that the greenhouse effect either is or isn’t. For our study, we were specifically trying to quantify the magnitude of the greenhouse effect.
I am responding to your claim: “In other words, the models were wrong. CO2 doesn’t cause global warming.”
That sounds rather to be an either-or claim. I am happy to hear that you do not deny that there is a greenhouse effect, but only that it is smaller. The next question would be: how much smaller? Science lives by clear and thus quantitative claims. And I hope you will update your homepage and articles on this point, people may get the wrong impression.
If you are interested in quantifying the strength of the greenhouse effect, then your next step should be to compute how sensitive your method is. Only once you can show your method should have seen a clear effect, would it be informative that you did not find a significant one.
Anders:
Strictly speaking, the air parcels are not adiabatic. Indeed, strictly speaking, there are no air parcels. Just individual molecules cannoning of each other in a gas. For convenience, physicists treat the atmosphere as being divided into air parcels (ie, as not mixing with adjacent portions of the atmosphere), and those air parcels being adiabatic (ie, as neither gaining nor loosing energy) as a simplified model. As such, it is a lot less crude than a spherical cow, and on a par with the assumption of frictionless surfaces.
Pekka is correct, also, in claiming that a non IR active atmosphere (ie, one with no greenhouse gases) would be more adiabatic than the actual atmosphere. A completely transparent atmosphere at all wavelengths of light would approach closer still. It would fail to be perfectly adiabatic all the same because it would still lose energy at the top of the atmosphere from the escape of light particles from the Earth’s orbit, and from the gain of energy from meteorites and cosmic rays. The energies involved in these cases relative to the energy exchanges from IR or UV active molecules is, of course negligible. The atmosphere (and “air parcels” within it) would still not be perfectly adiabatic ignoring even these due to differential rates of heating at the surface due to either the day/night cycle, differences in lattitude, or differential rates of heating between the ocean and land. These later, are sufficient to divert real observations from calculated lapse rates sufficiently that calculating any small divergence from the convection only lapse rate resulting from the presence of IR active gases would not result in a testable distinction.
You did, however, get one thing very right. The cause of the lapse rate is a tricky subject. Far more so than, for example, explaining the basic methods of radiation models.
Pervection looks fun. They consider the addition of more air to a body of it contained in an upturned tube in water, to be the addition of energy, and are surprised when the levels of water oscillate up and down, as you would expect with changes in pressure with such a small connector as 100m of thin tubing. If you repeated it with 6 inches of 1inch diameter tubing any oscillation would be much quicker, I think.
In fact at that point I started wondering if they were trying to troll us, in the tradition of Steorn.
Ronan Connolly,
In that you ignore the fact that atmosphere would be essentially isothermal without GHE. The whole profile is totally due to GHE. GHE does not determine the adiabatic value of the lapse rate, that’s determined by thermodynamics of gases, but without GHE we would not have anything near that value, but essentially isothermal conditions.
As I have already written, with a weak GHE we would have a much colder atmosphere with a large lapse rate, but with a very weak circulation. Again a totally different atmosphere from what we have.
Pekka,
You seriously aren’t helping. Continually telling me that my post isn’t very good without specifically explaining why is really starting to tick me off. I agree that there is much more to this than what I put in this post. FFS, I wrote it on a Sunday morning.
You’re clearly missing the point I was trying to make and maybe you could actually say whether or not the point I was illustrating was wrong or not without being pedantic. The basic point is that the temperature gradient is to a first approximation independent of the existence of GHGs. One can derive (as I tried to do) an approximation of the lapse rate by assuming that the process is adiabatic and reversible (okay with that). As we’ve already discussed, if there were no GHGs the surface would have the non-GHG temperature and if the atmosphere were optically thin and non-emitting, would have a lapse rate that meant that the atmospheric temperature dropped with height from the non-GHG surface temperature (okay, so far).
If one adds GHGs, what they do is increase the height at which the planet appears to be emitting its energy. The more GHGs, the higher that level. If one then works from that height back to the surface, given that the lapse rate is negative means that the surface temperature will be higher than the non-GH temperature. That’s really all I was trying to illustrate. Yes, I know that the actual temperature gradient will depend on other factors, but that doesn’t change what I was trying to illustrate. I also know that there are other ways to illustrate the GHE, I just happened to choose to do it this way because it seemed relevant to the topic. If you think that this is fundamentally wrong, feel free to tell me why. Bear in mind that I wasn’t trying to develop the ultimate blog post for describing atmospheric temperature profiles.
On the other hand, if you’re just trying to tell me that you think my post isn’t very good. Fine, you may well be right, but you’ve made that abundantly clear already so unless you really want me to tell you what I think of your comments, maybe you could stop pointing it out. I’ve got the message. I don’t claim that every blog post is going to be the ultimate in blog posts. It is just a blog for goodness sake.
Pekka,
Either I’m really bad at explaining what I’m trying to express, or you’re extremely bad at understanding what I write, because that seems remarkably similar to what I was trying to get across in this post. And you did it without expressing all the other factors that you seem to think are crucial and without which my post is – according to you – largely useless.
Having a brief look at paper 1 from Connolly and Connolly, they state in the abstract:
First, this is transparently unscientific argument. Suppose that it is true that the temperature profiles can be accounted for by “… changes in water content and the existence of a previously overlooked phase change”. Then that means you now have two theories which account for the same phenomenon. Proving that your theory accounts for a phenomenon does not prove that alternate theories do not. Rather than assuming they have won the battle simply because they turned up for it, Connolly and Connolly need to now expand the theory by showing independent tests.
Further, this “previously overlooked phase change”, later described as a multimerization of oxygen and nitrogen is, because previously overlooked and not independently verified, entirely ad hoc.
At this stage it should be easy to test Connolly and Connolly’s theory. A combination of O2 and N2 into groupings of 3 or four molecules should effect the spectral signature of those molecules. They need only determine that spectral signature and demonstrate, independently of their lapse rate observations, the existence of this extraordinary (and extraordinarily unlikely) “multimerization” of O2 and N2. Until they do, they have nothing worthy of publication – except of course in an open access journal, published by Connolly and Connolly and peer reviewed by Connolly and Connolly.
Would it be a good idea to have a 24 hour timeout on the topic whether this is a good post? Preferably also on the topic how an atmosphere would look like where the air does not interact with radiation in any way. It gives me an headache.
Tom – I was hoping for experiments in a large vacuum chamber at the least.
Tom,
Indeed, I didn’t ever doubt that Pekka was right. I just didn’t think I needed to do an entire lesson on thermodynamics to illustrate how measuring temperatures and temperature profiles in the atmosphere are not really a good way to determine the existence of the GHE as, to a first approximation, the lapse rate doesn’t depend strongly on the existence of GHGs.
Victor,
I’d certainly second that. I try to think about what aspects of my posts might be controversial and lead to lengthy and somewhat frustrating discussions. Sometimes, I get it horribly wrong.
Given their discovery of “pervection” would it be fair to call them “pervs”?
(I’m sorry, but somebody had to say it…)
Pekka Pirilä:
No.
Consider a single molecule in the atmosphere with a positive velocity in the z axis. As it gains altitude it also gains gravitational potential energy. By conservation of energy, it therefore loses energy from other forms at the same rates. Initially that loss of energy will be a loss of kinetic energy as it slows in the z axis. On average, however, collisions will restore some of that energy (and hence velocity in the z axis) at the cost of energy stored in other forms. Assuming no radiative transfer of energy, and no energy from phase transitions, the available energy comes from other two axis of motion, plus whatever forms of vibrational (and rotational) energy it has. Consequently, it will lose kinetic energy at a rate of equal to the gain in gravitational energy divided by the heat capacity.
The algebra comes out as dT/Dz = -g/cm, where cm is the specific heat at constant mass, rather than the specific heat at constant pressure (as in the standard formula). At least, it does if I got my algebra right 😉 What difference the difference in heat capacity used makes to the lapse rate, I am unsure.
However, regardless of the details on which I am unsure, I am certain that a non-radiating atmosphere would not be isothermal, for if it were it would violate conservation of energy once the gravitational potential energy of the molecules is taken into account.
I leave the computer for an hour and there’s another dozen comments to reply to!?! 😮
Ok, I’ll have to leave you all for the rest of the evening – I’ve been at the computer too long already on a Sunday afternoon (now evening). But, I’ll check back in tomorrow evening. And, just to reiterate, while your comments on this post have so far been mostly negative, I do appreciate the discussion, and the fact that several of you have made the effort to read some of our work! 🙂
Several people (e.g., guthrie, Victor, ATTP, BBD) have asked me above to justify our conclusions about the greenhouse effect… or for that matter, what our conclusions actually are!!!
We are acutely aware of the fact that our conclusions differ from the conventional “textbook” understanding of atmospheric temperature profiles. We didn’t come to our conclusions lightly. However, when we looked at the results of our studies, we found that the conventional understanding seems to be incomplete and inadequate.
Notice that I said “incomplete”, not wrong! We believe that much of the current understanding is correct. E.g., CO2 is infrared active, incoming solar radiation = outgoing terrestrial radiation (+ albedo), etc. We are just saying that there are additional factors which need to be considered. We find that when we account for these additional factors, we are able to explain all the results that the greenhouse effect theory was able to explain and several previously-unexplained results.
I just mentioned to ATTP in reply to a comment on our website, that so far all of the discussion on this post seems to be about our conclusions and how we came to them.
The simple answer is that we came to our conclusions on the basis of our results! We try to approach our scientific investigations with an open mind. The results of our experiments and analysis seem to us to identify several key flaws with the greenhouse effect theory.
I would be very interested to hear if any of you guys are able to explain our results in terms of the greenhouse effect theory. So far, you have been just talking about how our conclusions must be wrong… 😉
guthrie, Victor, ATTP, BBD, and anyone else, I’m sorry I haven’t had a chance to respond in detail to each of you individually. But, we have tried to summarise what our conclusions on the greenhouse effect are (and how we came to those conclusions) in the Global Warming Solved essay.
Maybe you can find answers to your questions there.
If not, we do go into more detail in our papers (particularly Paper 2).
Failing that, I might have more time to answer you tomorrow evening…
Until then, take care… 🙂
Anders, perhaps you can put my comment down to my agreeing with him in this case that you needed a more accurate presentation. I understand that you generally don’t sweat details and rely on commentors to pick up on them if they are important. Normally I have nothing against that approach, so long as it is explicit (which you are). I just feel that in this case you were a little too loose in your description, and also that correction referring kdk33’s comment is not a big improvement. The essential points I think are missing are that it is the parcels of air that are adiabatic, and that that is just a simplifying approximation.
I’m sorry to pick you up on this, but you did ask me to pick up on important details 😛
I try to put the main pieces in order without any reference to what others have said similarly or differently.
1) The overall energy balance tells that the effective radiative temperature of the Earth must be around 254 K with an albedo of 0.3. That’s a weighted average of temperatures of the locations where IR escaping at TOA is emitted.
2) In an atmosphere without convection (i.e. air somehow locked where it is) the temperature drops rapidly with altitude, more rapidly than the adiabatic lapse rate. The surface would be something like 30 degrees warmer than it is presently.
3) Lapse rates higher than adiabatic lead to instability against convection. Thus convection sets in and lowers the lapse rate to the adiabatic value (or some “adiabatic” value as we have many of them).
GHGs are essential for the points 1) and 2), not for the 3). 1) and 2) are fundamental for GHE, 3) is just on additional consideration. It’s highly misleading to emphasize 3) over 2). The formula derived in the post is an upper limit for the real lapse rate, not a formula for the actual lapse rate, unless we have additional information that convection is present.
The fundamental error of Ronan Connolly is that he ignores the importance of 2).
I come back to 1). In an optically thin atmosphere almost all radiation to space originates from the surface. Therefore the surface must be effectively 254 K for that albedo. With more and more GHGs more and more radiation originates much higher. Thus the high altitudes affect the effective average temperature. When we combine that with the lapse rate, we see that the surface gets the warmer the more we have GHGs.
This is my short version of the GHE.
Tom,
I agree and kind of thought I had indeed done that. Clearly not well enough.
Tom Curtis,
In thermodynamic equilibrium the atmosphere is isothermal. Every system in thermodynamic equilibrium is isothermal.
The loss of energy of the molecule going up is compensated by the effect that energetic molecules are more likely to go up and less energetic fall down. These effects cancel exactly – they must. That’s a fundamental result of thermodynamics.
If you wish to see one mathematical answer to your complaint you may look at this note I wrote perhaps two years ago.
Click to access barometric_derivation.pdf
Pekka,
Most of what you’ve said seems largely consistent with what I was saying in the post.
I wasn’t trying to emphasize one thing over another. IMO, it’s misleading to suggest that I was. It was just a post motivated by something I read and in which I chose to illustrate something the way I did.
However, maybe you can explain how you get the following. What are you assuming here?
Pekka,
I’m unconvinced that this is correct for an atmosphere. Certainly if one were to consider a gas placed on the surface of a planet and then allowed that gas to respond, through a reversible, adiabatic process, to its environment it should expand until it were in hydrostatic equilibrium and, as far as I can tell, the vertical temperature profile would be as I describe in this post (dT/dz = -g/cp). You seem to be then arguing that it would continue exchanging energy until it were vertically isothermal.
ATTP,
You should not ask (and I don’t assume anything on), whether you tried to do something or did something by purpose. Readers see only the actual text, and the question is, how they are likely to understand it.
Furthermore, if you wish to discuss the origins of GHE you must emphasize 2) over 3), because 2) makes the GHE, while 3) limits its strength to a lower level.
Pekka,
Really, I must? Maybe you could then explain how you got 2, as I asked you in an earlier comment.
ATTP,
As I wrote above, GHGs cause the temperature gradient, without them that would not exist in the present way. The formula is true, when convection exists, but convection leads always to some dissipation and some mechanism must continuously add free energy to keep convection going (that means in practice that atmosphere must emit IR at high altitude). When convection stops, some stratification develops. First it’s very weak, but even that prevents permanently convection. If we don’t have any radiative heat transfer, we have only conduction left. That will very slowly bring the atmosphere towards isothermal state.
As Earth is rotating and as we have latitudes, there will always be a hottest spot on the surface. At that spot we have convection as long as atmosphere is not as hot. On the other hand we have no effective means to cool the atmosphere. At higher latitudes the surface is colder. That leads to a layer of temperature inversion. How thick that layer would be, I don’t know, but surely much thinner than the present troposphere. Above that level the atmosphere would be ultimately essentially isothermal. The isothermal atmosphere is warm, but does not warm the surface significantly.
All that is purely theoretical as no atmosphere is totally without radiative heat transfer, but that’s the nearest thing the atmosphere may be to thermodynamical equilibrium.
When we start from exactly zero radiative heat transfer within atmosphere we have the above result. Adding a little GHGs changes that to the case of optically thin atmosphere, which has a simple solution as soon as radiative heat transfer is much stronger than conduction. When that’s not yet the case, the troposphere is thinner as conduction has always an effect in that direction, but very little is needed to get over that transition. Then adding GHGs has the effect of increasing circulation without any change in the temperature profile or the surface temperature as long as the atmosphere remains optically thin. When the atmosphere is not thin we start to have warming and a situation basically like that we have now.
2) is the standard radiative heat transfer calculation for a atmosphere with GHGs when radiative heat transfer is the only heat transfer mechanism. Pierrehumbert has surely that somewhere as have many others.
Yes you must, because without that the skeptics would be right.
Pekka,
I have a feeling that we’re starting to disagree. I don’t think it is correct that the GHGs cause the temperature gradient and that was the point I was trying to get across in the post. To a first approximation, the temperature gradient can be determined by simply considering the adiabatic lapse rate. Maybe that’s why you think it’s a bad post – because you disagree with the premise. Maybe I am wrong though. It’s, however, late and I need to think about this a little more. I’ll respond in the morning.
ATTP,
Convection reduces the lapse rate, it cannot add to it (not globally, but perhaps locally). Something that reduces the effect cannot be the cause of it.
Pekka,
I haven’t ever said it was the cause of it.
As I said, I’ll have to give this more thought, but I’ll post this link which seems broadly consistent with what I’ve been saying. To be clear, I’m well aware that this post is a simplification and well aware that many processes transport energy in the atmosphere. I’m slightly concerned that we’re discussing details that were never the intent of this post. I’m also concerned that we’re going to end up in a discussion in which we each have different opinions about how best to express something that we may well agree about generally. I don’t see the point in that.
ATTP,
I have still the suspicion that you have not understood the very basics.
One point is that convection does not drive the lapse rate, when something else tries to make the lapse rate larger than the adiabatic one, that drives convection. That something else is the reason for the existence of the lapse rate, convection gives the upper limit for that. If you miss that something else you miss everything on the lapse rate. In the Earth atmosphere that something else is the GHE.
I wrote a short post with my first impression of the new journal and the disappearance of the greenhouse effect.
A few thought experiments might help:
1) an isolated planet – no heat or mass in or out. The planet has a mass equal to earth, but is impermeable and a perfect insulator. The planet has an ideal gas atmosphere with a molecular weight and mass equal to earths. The planet has a mass average atmospheric temperature of about 305K. (not all these are important, but they make the problem entertaining). The planet is allowed to come to equilibrium. What will be the temperature profile?
I think it will be isothermal at 305 K. The atmosphere may convect initially and settle at the ALR, but conduction will soon cause the temperature gradient to be less steep, which will stop convection. At this point the atmosphere rests and conduction will equalize the temperatures. Hence isothermal.
2) The same planet, but now let it orbit a sun that showers it with radiation. Let’s make the effective emissions temperature of the planet at thermal equilibrium be 305K. Make the atmosphere perfectly transparent to any incoming or outgoing radiation. What will be the temperature profile?
I think the surface temperature will average 305K, but will be hotter on day side and colder on night side. The atmospheric gradient will follow the ALR. Why? I think convection will occur. On the hot side of the planet the atmosphere will be heated by the surface and that air will rise and colder air will rush in to takes it place. This natural (conductively driven) convection will seek to impose the ALR.
I think the key is that heat will be transferred from the hot side gasses to the cold side gasses by a combination of convection and conduction.
3) the same planet, but instead of orbiting a sun, let it be heated uniformly by some imaginary radiation source. Again, let the effective emissions temperature be 305 K. And again the atmosphere perfectly transparent. What will be the temperature profile?
I think it will be isothermal at 305. With no energy sink, a rising parcel of air will have no place to transfer energy to. Even if initially convecting, conduction will soon make the temperature profile less steep than the ALR and convection will stop and the atmosphere will soon be isothermal.
4) the real earth. I don’t think GHG are required for convection. I think conduction would drive it as in #2. GHG provide an additional driving force for convection – by heating at the bottom and cooling at the top. Either way, heating from the surface causes the atmospheric temperature gradient to be steeper than the ALR and convection seeks to restore the atmospheric temperature profile to the ALR, which is the least steep profile a convecting atmosphere can attain.
I’m probably wrong, but it sure sounds good after a tumbler of small batch rye.
Conduction in the atmosphere as an efficacious mechanism of energy transfer. Hmm. Conduction. Gas. Tricky.
What if there’s a few IR absorber/emitters in the gas mix? What if we knew they existed and caused a radiative imbalance in the climate system? What if we knew that already?
Or maybe I need some small batch rye to improve my focus a bit.
pekka, in the stationary state, w = 0, which blocks your inference to equation (3) in your barometric derivation. Indeed, it also requires that δn/δw = 0 in the stationary state, meaning the function n(z,w,t) is independent of w which seems like a reductio to me, but my maths isn’t that good. In any event, that means at least one of your assumptions, ie, the “… influence of the change of velocity to the vertical coordinate … can be neglected” or that the “… time differential is
… so small that collisions can be left out of the consideration” is false.
Pekka,
Hmmm, I’m always happy to be proven wrong, but in this case I think it may be you who is mistaken, not me. I also start to get rather tired of dealing with others who seem unwilling to consider that they may be mistaken. Doesn’t give me much confidence that the discussion will be constructive.
Firstly, I’ve never said that convection drives the lapse rate. Maybe you could stop putting words in my mouth. It’s my understanding that the lapse rate does not depend, to a first approximation, on the GHG concentration. What the GHG concentration does is set the effective emission height in the atmosphere. If one then works from that height back to surface, the lapse rate then gives the surface temperature, which will be higher (because the lapse rate is negative) than the temperature at the effective emission surface, which would be the non-GHG temperature. I’ve said nothing in this post, or in any of the comments, about how the energy is transported through the atmosphere. That was partly intentional because I was aware that it was a complex process, as this discussion appears to be illustrating very effectively.
Maybe the confusion comes from my suggestion that, to a first approximation, the lapse rate does not depend on the GHG concentration (although, as pointed out in one of the early comments, feedbacks can influence the lapse rate). Without GHGs the surface of the planet would have the non-GHG temperature. The atmosphere above that would then have a profile that depended on whether or not it can absorb some energy from the radiating surface, whether or not it absorbs energy from the Sun, and whether or not it emits energy.
So, that we have a troposphere with an adiabatic-like lapse rate presumably is a consequence of the GHE (which was – despite your rather constant dismissals of this post – what I was trying to illustrate) but, unless I’m mistaken, the actual lapse rate does not – to a first approximation – depend on the GHGs.
So, it is possible that we agree (in principle at least) but haven’t quite recognised that. Maybe you could spend a few seconds giving that some thought. It’s possible that we disagree, in which case one of us is mistaken. It’s possible that I just haven’t explained myself as clearly as I would like. Whatever it is, it’s not clear that continuing this much further is likely to be particularly constructive.
Pekka,
I’ll add the caption to the figure in this post.
This is essentially what I was trying to illustrate in my post. If that isn’t what it seemed like I was saying, then I clearly didn’t express myself very clearly. If you disagree with this, we could keep going around in circles, or you could take it up with RealClimate who typically seem to do quite a good job of explaining the fundamentals of the greenhouse effect.
ATTP,
Your description is correct. Nothing in that would have made to react.
Having said that, that doesn’t explain why Connolly paper is totally wrong, because it does not tell clearly enough what’s the role of GHGs in having the adiabatic lapse rate in the troposphere. That’s the piece that I’m missing in all your answers.
Don’t you believe my explanation, or what’s the reason for your reluctance?
Pekka,
Well, given that – to a first approximation – GHGs don’t influence the lapse rate, the point I was making was that looking at temperature profiles in the troposphere and stratosphere over a period of a few hours is not going to tell you anything about the GHE. I’m not saying the Connolly paper is totally wrong, I’m suggesting that their claim that they can find – or not find – a GH signature is wrong (or, rather, they cannot use what they’ve done to infer anything about the GHE).
I don’t understand your explanation or, maybe more correctly, how it differs from what I’ve said. I feel that you’ve taken this discussion into areas that I didn’t mention in the post and that aren’t quite relevant to what I was trying to illustrate. If my description is correct, then what is the point you’re trying to make?
kdk33,
The case of an totally transparent atmosphere is not important enough to continue this exchange much longer, I repeat, however, one point.
The atmosphere is heated effectively when the mechanism that you describes works, but nothing cools it effectively anywhere. Therefore it heats up until the heating mechanism is also very weak. At that point there’s nothing to drive convection that extends to higher altitudes than a near surface thin layer that loses very slowly heat by conduction through the temperature inversion to the surface. That leads to stratification and towards isothermal.
ATTP,
There are two questions:
1) What’s the adiabatic lapse rate?
This you have answered.
2) Why part of the atmosphere has an lapse rate close to the adiabatic lapse rate?
This you haven’t answered, but GHE is the answer.
If we have just real convection that always involves also turbulence and related dissipative processes the lapse rate drops very rapidly to exactly the adiabatic lapse rate, or slightly below and the convection stops. To maintain convection the adiabatic lapse must be exceeded, not much but a little. In other terms some mechanism must all the time push the lapse rate up to higher than the adiabatic value. That’s GHE. Where GHE does not do that, we have stratosphere.
Tom Curtis,
w is the vertical velocity of an individual molecule, n(z, w, t) tells the distribution of that velocity at altitude z at the moment t. In equilibrium the expectation value of w is zero, but individual molecules have all their own values.
My note does not present anything new, it’s just an attempt to explain, how standard physics works in a situation that lead many to doubt standard physics.
In thermodynamics the answer is completely clear: Every system in thermal equilibrium is always isothermal, and left alone all isolated systems move towards thermal equilibrium, if all parts are in thermal contact with each other in some way like conduction or radiative heat transfer. That’s how the Second Law works.
Some people seem to think that the laws of thermodynamics should be modified for gas under gravity, but the above paragraph applies also in that case without any modification. Otherwise it would be possible to build a perpetum mobile of the second kind. A practical way for that is given by thermoelectric generator (Seebeck pair).
Pekka,
Did you read the second to last paragraph in my post? What you’ve said above is precisely what I was trying to do. I may not have done it well, but I was trying. In fact, I don’t think I did it all that badly. I fail to see how this
isn’t consistent with what you’ve just said.
Yes, again roughly what I was trying to say. Again, maybe not all that well.
Maybe this isn’t obvious, so I’ll try and explain how things typically work here. I write posts and probably too many. I consciously do them quite quickly as I have a life (some of which I think you’ve just wasted). I sometimes make mistakes, or don’t explain things as well as I should. Others then comment and suggest corrections, but typically they’re pleasant, constructive and polite (see Tom Curtis for example). Sometimes, I make my posts intentionally a bit simple and leave out some details (in fact, that’s probably always true). Others can then comment to elaborate or clarify, again normally done pleasantly or constructively.
That’s how it normally works and I normally quite enjoy that. I don’t claim to know everything, I don’t claim to be an expert, and I don’t mind being wrong (part of being a scientist, in my experience). What I dislike, are discussions that end up being virtually strawman arguments, either because I didn’t explain myself that clearly, or the other person didn’t read it all that well. If you’d simply written a comment to point out that the GHE essentially determines why part of the atmosphere had an adiabatic-like lapse rate, I would have said “exactly, thank you very much” rather than wasting hours of both your time and mine.
Can I ask that we drop this topic. I do try very hard not to get annoyed and irritated with those who comment, but even I have my limits.
It is probably best to drop the topic, if only for the tone of the discussion. Unfortunately, I studied this theory too long ago that I am able myself to explain it more clearly.
Still a last try, the problem is that the adiabatic profile described with the above equations is right for a parcel that is forced to rise (or drop) adiabatically. What is does not describe is why the parcel would want to move if it is not forced, then you have to consider the temperature profile of the surrounding and the buoyant forces on the parcel due to this surrounding.
That is why Pekka talks about the importance of convection. For many, this second step is quite an eye opener and I know several meteorologists that are evangelic about it. If you are interested, I would suggest to find a text book. I have the feeling that the discussion here does not bring anything any more.
Pekka,
To understand the role of GHG in convection and in establishing an ALR, thinking about a transparent atmosphere would seem to me important and useful.
The ALR is the asymptotic temperature profile that convection seeks to restore. It is the flattest temperature profile achievable by convection. In the absence of convection the ALR has no meaning. The atmosphere has an ALR because it convects.
I don’t think GHG aren’t necessary for convection. I think conduction will do. The intensity of convection and the height of the atmosphere that follow the ALR my be different. But the underlying physics doesn’t require them.
In the mechanism I describe at #2, the atmosphere gains no overall energy. It simply moves energy from the hot side to the cold side – the net is zero. (My imaginary planets losses energy via radiation through a transparent atmosphere.). It is heated from below. The energy is moved by convection so my planets atmosphere has ALR.
Thanks for chatting.
Victor,
I have no doubt that you’re right. It’s my understanding though that in the troposphere the exact mechanism that transport the energy is, to a first approximation, not that important if all you want to determine is the temperature profile. The temperature profile will, unless I’m mistaken, be typically reasonably well described by the adiabatic (saturated or not, depending on the specifics) lapse rate.
So, yes, if one wants to understand how the energy is actually transported in the troposphere, then the details are of course important. On the other hand, if one wants to simply estimate what the temperature profile will be (approximately) then one needs only consider the lapse rate.
So, if I had been writing a post about how energy was transported through the troposphere, I should have indeed discussed the role of convection. Since I was really just trying to illustrate that the temperature profile is reasonably well represented by an adiabatic lapse rate and that the GHE sets the effective emission level, that didn’t really seem necessary.
And to end on a positive note (hopefully), I’m sure Pekka is quite right in what he is saying.
kdk33,
Conduction tends always to reduce temperature differences. It moves always the system towards isothermal. Convection can persist only when something tries to make the temperature difference larger, i.e. dives the lapse rate to a larger value than the adiabatic lapse rate.
To everyone pondering these issues I present two questions:
Why do we have both the troposphere and the stratosphere?
What determines the altitude of the tropopause?
These questions are really fundamental. As long as they are not fully understood, the atmosphere is not understood, and neither is GHE.
ATPP,
I didn’t write my comments to show that I disagree with you. I wrote them because the importance of the point I emphasized is so commonly not understood, and because that’s the point where the Connolly papers made their most essential error. That’s the error that makes also everything else in their papers about physics worthless.
I addressed many of the comments to you, because they were responses to your comments and because I hoped to see you to indicate directly that you understood the point. You got close to that but not quite there.
The title of this site is one reason for my interest in seeing that the physics is right.
FWIW, and Eli has gone through this stuff before,
1. The adiabatic nature of parcels of air means that local thermodynamic equilibrium holds pretty much everywhere and you can measure a temperature.
2. The greenhouse effect enters the picture in your calculation by setting the boundary condition (the temperature) of the surface.
3. The lapse rate enters the picture because the level at which the Earth radiates to space in the IR is set by the greenhouse gas concentration. As both rises the effective temperature at which the earth radiates decreases. To maintain radiative balance the surface heats. See the fourth figure down in this post.
4. Amusing things happen if the level goes above the tropopause, but we may not be here to see them. Basically ozone absorption creates a permanent inversion.
5. Energy balance maintains the temperature profile and mass balance in the troposphere, e.g. energy in from convection and radiation equals energy out by convection and radiation at each point.
Eli,
Thanks, I think that is roughly what I was trying to express, but a nice – and probably clearer – summary nonetheless. BTW, did you miss a link in your comment?
Pekka,
We can probably just leave it a that, I guess.
Pekka,
I was expecting you to reply that radiative heat transfer was much more important and so conduction could be ignored. But you gave an unexpected reply, which makes me think you are not understanding my model. Which means I have not explained it well.
I suspect I have used conduction in a confusing way. What I am describing is usually referred to as natural convection. That is, convection within a gas phase that is driven by conduction at the boundary of that phase – the usual model is a hot plate.
How that applies here: if the atmosphere is transparent, then incoming radiation heats the surface. At the boundary of my imgainary planet, heat is transferred by condution to that layer of gas that is at the surface. Once heated, the gas convects. So conduction only occurs at the boundary. Within the atmosphere heat is transfered by convection.
That is why I say: The atmosphere follows ALR because it convects. It convects because it is heated from the bottom. That heating can be by radiative heat transfer or by conduction. On earth, both mechanisms occur.
In my thought experiment, I eliminate radiative heat transfer. It isn’t what happens in the real world, but it is useful for thinking about the physics. Convection will be more or less vigorous and the height of the convecting layer will be different, but either mode of heat transfer will drive convection.
In my #2, convection moves heat from the cold side to the hot side and this temperature difference allows convection to be maintained. In my #3, there is no hot and cold side, so convection cannot be maintained.
I hope that is more accurate.
kdk33,
I didn’t understand that you were thinking about a system with a hot surface. In that case conduction may be a significant form of heat transfer over small distances near the surface. (I thought that you discussed conduction within the atmosphere far from the surface.)
In absence of a mechanism of cooling somewhere else the temperature difference disappears gradually as the gas is warming towards the temperature of the surface.
To have a persistent temperature gradient both heating and cooling is needed. You have introduced the heating, but heating stops without the cooling. The ultimate state in that case is isothermal at the temperature of the surface.
I did wonder what you were on about too.
Possibly, just possibly Eli has caught a virus. Here is the link [link] fourth figure from the top
@Pekka,
what is it with the cooling? I don’t get it. The stratospheric temperature profile is determined by the absorption/emission characteristics of gases residing there. That ultimately determines the tropopause height and that’s about it. Temperature at the tropopause is cool enough to maintain the moist adiabatic lapse rate forever (given that effective emission height is well below the tropopause). Convection can’t do more than creating a vertical temperature profile which is equivalent to the moist adiabatic lapse rate. Plenty of cool air left higher up as far as I can tell. It won’t ever warm to surface conditions, because convection ceases if conditions are stable. What doesn’t cease is radiation to space, which cools things down as we certainly agree. Hence, I’ve got no idea which point you are trying to make.
Eli,
Thanks. Yes, that was how I understood it 🙂
Karsten,
The profile of the atmosphere is simple in the simplest (one-dimensional) case of optically thin atmosphere that has weak absorption and emission of IR, and is totally transparent to SW:
A constant adiabatic lapse rate up to the tropopause and an isothermal stratosphere. A discontinuity of the lapse rate at tropopause. Everywhere in the troposphere emission of IR exceeds slightly absorption of IR coming from the surface (IR from the rest of the atmosphere is negligible in the limit of optically thin atmosphere) the difference is covered by convection. At the tropopause the difference goes to zero, and convection stops. Above the tropopause absorption and emission balance at the constant temperature of the surface temperature divided by the fourth root of 2.
The real Earth atmosphere the structure is basically the same but in the upper stratosphere the absorption of UV by ozone leads to a temperature that increases with altitude. The troposphere and the lower stratosphere have the same nature as in the case of optically thin atmosphere, but the details are much more complex. Therefore the temperature profile is also smoother and more complex.
The point is that the stratosphere exists and is fundamentally different from the troposphere independently of the emission/absorption characteristics of the gases there.
If the planet is heated uniformly (as in my #3), I agree that isothermal is the ultimate condition. But, if the planet is heated non-uniformly (as in my #2), then that non-uniformity, will maintain convection. Convection will continually move heat from the hot places to the cold places – the atmosphere will be heated at the hot places and cooled at the cold places. That internal recycle won’t change the overall energy budget for the planet, but it will maintain convection.
Good day.
kdk33,
I discussed that case in my comment of February 23, 2014 at 9:51 pm of this thread.
@Pekka,
what we have is radiative-convective equilibrium. The lapse rate is (to a first approximation) fix and won’t change (significantly) under increased GHG conditions. I’m with kdk33 for how it works under realistic conditions here at planet Earth, with it’s strongly different diabatic heating rates (day/night) and its atmospheric circulation in general.
Karsten,
As far as I know, no one has studied really how an atmosphere that’s fully transparent to all radiation behaves on an rotating planet like the Earth. Standard atmospheric models are probably unsuitable for that as they have been built to work for a very different atmosphere.
I consider it, however, virtually certain that something similar to the stratosphere will start from a much lower altitude than in the present atmosphere. As soon as the lapse rate has become less than the adiabatic one, the convection cannot enter those altitudes, and to me it appears certain that that would be the case at a rather low altitude.
Some circulation would certainly remain as the atmosphere would keep on loosing some energy to the surface in the colder areas by conduction through the layer of temperature inversion that would develop in those areas (this is a point I’m fully certain about). All that heat would be replenished by convective heat transfer near the hottest spot of the surface. The most important quantitative details I’m not able to estimate include the height of the inversion layer at various locations and the total value of the circulating heat flux. I believe that estimating these values would require a GCM modified to work in a fully transparent atmosphere where conduction is an important factor in energy transfer.
Regular conduction might be enhanced by some turbulent mixing, but I would expect that there would not be much turbulence anywhere far from the hot spot of the surface, because the circulation would be very weak.
To add my 2 cents… on the topic of adiabatic, my understanding was an adiabatic process (no heat exchange with the rest of the world) is generally a good approximation for a system undergoing relatively fast change – i.e. on a time scale much much shorter than internal energy divided by heat transfer rate. Reversibility on the other hand is often a good approximation for slow changes, where the system remains in local equilibrium with its surroundings. For a system where heat transfer rates were fast (high GHGs or high conductivity?) and convection slow (high viscosity) the adiabatic approximation here would probably fail. But for conditions as they are in Earth’s atmosphere I’m sure it’s fine.
Pekka,
I think one more thought experiment is in order.
We seem to agree that for an isolated planet (my #1,) there will not be convection and the atmosphere will be isothermal. I think we also agree that for a uniformly heated planet with a transparent atmosphere there will not be convection and the atmosphere will be isothermal. I even think we agree that for a non-uniformly heated planet with transparent atmosphere there will be convection as air rises from hot places and sinks at cold places – the vigorousness of this convection being less, of course, than if the atmosphere were optically active.
The new thought experiment i’d like to introduce is this:
Same planet, but now make the atmosphere optically active to outgoing radiation in the same way as earth’s. Let this planet be heated UNIFORMLY from some imaginary radiation source. Will there be convection?
I think yes (and I think you will immediately agree). Air will be radiatively heated by the surface and will cool higher in the atmosphere as it gives off radiation to space.
Whether the adiabatic approximation is good depends on the question. It’s probably good within a dry rising column of air that’s not disturbed by horizontal mixing during the ascent. The real lapse rate is, however, rarely the dry adiabatic lapse rate, and that’s due to the failure of the adiabatic approximation.
Sometimes the lapse rate is the moist adiabatic lapse rate. As condensation is delayed the process is not fully reversible even in the cleanest case.
The adiabatic approximation fails also in telling that circulation consumes always free energy due to all kind of dissipative processes including condensation that leads to precipitation, turbulence, horizontal mixing, … The dissipative processes reduce the lapse rate that results from ascending flow, and increase that from a (usually forced) descending flow.
The validity of the adiabatic approximation is thus quite limited.
kdk33,
We agree at some level on the case on which we have already had some exchange, but we may not agree on, whats the dominant nature of that, as I emphasize the importance of the part that’s like a stratosphere (but not controlled by radiative heat transfer as the real stratosphere is).
On your new addition I agree. Heating like that from below combined with cooling at the top leads to instability and onset of convection. (That could be called spontaneous breakdown of translational symmetry).
This is a fascinating discussion and at the risk of going off topic I look forward to somebody modelling these imaginary atmospheres on imaginary planets and then comparing those results to actual extrasolar planets with real atmospheres via giant space telescopes. These kinds of idealized models are useful (even if wrong, which they surely are), and the one that I came up with was the rocket engine modeled as an infinitely long atomic accelerator. The thermodynamic equations for rocket engines are pretty well known, but by looking at this idealized model one can easily see that a finite length single atom accelerator will result in the highest exhaust velocity per unit energy for the lowest atomic weight atom. Sorry for going off topic. Carry on.
To be simple and blunt about it, the stratosphere is a result of the ozone and oxygen absorbing UV solar energy.
The resulting heating produces an inversion with the temperature rising until one reaches the stratopause @ ~ 50km
The inverted temperature profile kills off convection in the stratosphere.
The term stratosphere refers to the stratified nature of the beast. with little vertical flow.
Because of this heating effect the simple calculation And Then (Eli may use your first names, please?) shows for the lapse rate fails as the temperature decline near the tropopause slows and then reverses.
However, as far as the greenhouse effect is concerned this is not an issue, with the effective radiative height to space being well below the tropopause.
Yes, what Eli describes is my understanding of the temperature inversion in the stratosphere. Also, as interesting as it is, as Eli also says, it isn’t really relevant to what I was trying to highlight in the post and which seems to be consistent with Eli’s post that he linked to in an earlier comment (well, after a second attempt 🙂 )
I’m not quite sure – at this stage – how to respond to this.
Is it the case that the stratosphere does not convect because it is heated from the top?
kdk33,
The inversion, I believe, is because of the ozone and oxygen (as Eli says) absorbing solar UV, so it is certainly heated from the top. So, yes, I think that would kill off convection between the tropopause and the peak of the temperature in the stratosphere.
The stratosphere would be there even without heating of air by the solar radiation. The tropopause would move a little without that heating, but not much.
The point is that at some level the total emission up and down becomes as strong as the total warming by radiation from below and above. That’s the altitude of the tropopause, because above that altitude the atmosphere cannot take any extra heat carried by convection.
Another way of saying the same thing is that at altitudes below the tropopause the lapse rate would be higher than the adiabatic lapse rate without the extra heat from convection, while above that level radiative heat transfer alone leads to a lesser lapse rate than adiabatic. That makes the stratosphere stratified.
Heating by solar radiation contributes to that balance, but is not essential for that.
If anyone is interested I think Manabe and Strickler (1964) is the first paper where the models started to grow some teeth. This is discussed in relation to this current topical venue here and on another post there at the bottom of the page. Ramanathan and Coakley (1978) I guess would be the gold standard. I had forgotten all about the other attempt to falsify global warming and the ‘greenhouse effect’.
Some more details on the stratosphere.
Pierrehumbert discusses in his book on Planetary Climate the properties of the stratosphere.
First he discusses the case of an optically thin atmosphere that has an isothermal stratosphere. Then he continues with an atmosphere that’s not optically thin but has the IR properties of the real Earth atmosphere without solar heating of the stratosphere.
In this more realistic case the tropopause can be observed as the level where convection stops. Above that we have the stratified stratosphere. So far nothing has changed fundamentally and the tropopause is at roughly the same altitude as in the real atmosphere. Above the tropopause the lapse rate is less than the adiabatic lapse rate. Thus the potential temperature starts to increase with altitude. The spectral properties of CO2 mean, however, that the atmosphere is not optically thin at all wavelengths even in the stratosphere. That leads to an temperature profile where the real temperature keeps on decreasing while the potential temperature increases.
The absorption of solar radiation in stratosphere is not essential for the existence of the tropopause and the stratosphere, but it’s essential for the actual temperature profile.
Pekka, care to explain why Venus and Mars DON’T have stratospheres?
I know I’m way in over my head here, but as far as Then’s toy asteroid atmosphere models, there may be some opportunities to devise, calibrate and compare imaginary planetary atmospheres during the Pluto Flyby, and I hear Ceres may even have some water vapor, although it’s hard to imagine that it’s dense enough and persistent enough to affect thermal equilibrium. More on that next year.
Seems to me that the troposphere convects because it is heated from below and the stratosphere does not because it is heated from above. The boundary is where these forces balance.
Absent heating from above, why would convection stop?
A related and interesting topic: what are the fluid properties that determine if a fluid will convect or simply conduct. Perhaps the fluid properties in the stratosphere meet these conditions and so will not convect regardless of the heating from above – but that’s hard for me to imagine.
Eli,
Are you sure that they don’t?
I don’t really know much about the other planets beyond what I have read from Pierrehumbert’s book.
According to Pierrehumbert Venus has a thin isothermal layer just above the troposphere. I would imagine that such a layer must be part of a stratosphere.
On Mars he writes: even a layer like that of Mars’ upper atmosphere, whose temperature decreases gently with height, can be stably stratified.
He defines stratosphere as the stratified part of the atmosphere where potential temperature increases with height. That’s also my definition.
kdk33,
The stratosphere can also be heated from below, but only by radiation. The important thing is that it’s heated as much by radiation as it emits, i.e. the temperature is determined fully by the radiative energy balance, and that does not lead locally to a lapse rate that exceeds the adiabatic lapse rate.
Peka, neither Mars nor Venus have a stratosphere. They do have tropospheres, mesospheres and thermospheres. The demarcation between troposphere and thermosphere is the point where the lapse rate becomes determined by radiative transfer rather than convective transfer of heat.
As I am citing myself, and indirectly a graph prepared for a course at the University of Colorado, here is a more authoritative source on Mars, at least.
I may have mistaken on the terminology, as that definition is exactly what I used for the lower limit of stratosphere. In both cases it’s the upper limit of troposphere, and that’s the significant factor here. Up to the question of Eli only the Earth atmosphere was discussed, and the main subject was the troposphere.
Prekka, go look up the definition of the tropopause.
A rapid net search brought up a variety of definitions, some in full agreement with my thinking, some not. All agree on the present Earth atmosphere, but diverge when the structure of the atmosphere is different.
The section in Pierrehumbert’s book that has influenced me is somewhat vague. He starts by saying that the definition that deviates most of my thinking is too restrictive, and that the boundary between the convective troposphere and the next stratified layer is the essential one, but he does not really tell, what’s the name of the next layer on Venus or Mars.
When another planet has a different series of layers above the troposphere, all names used for the Earth atmosphere don’t fit, nor do they for a different assumed atmosphere on Earth. I took naively for granted that the first stratified layer is always stratosphere, but evidently that’s not true or at least not a universal practice.
I propose now a new answer to your earlier question. Mars and Venus don’t have a stratosphere because the layer that might be called stratosphere has another name.
Nothing in this changes the physics, but using different names for concepts is certainly confusing in the discussion. You may interpret all my earlier comments by replacing the word stratosphere by the descriptive expression the first stratified part of atmosphere above the troposphere. It would be nice to have a name for that, on which we could agree.
Could these be from the same person 🙂
It would certainly seem so. I’ll leave it at that in the hope that a little reflection may take place.
ATTP,
I admit that I used a word in a way that may be confusing. I apologize for any confusion that this lapse may have caused in those few latest comments that are affected by this.
That changes otherwise nothing in my argument which was about troposphere, and nothing in most of my comments, because this issue entered late in this thread. Stratosphere entered earlier only as the name for the next layer. I do still think that tropopause is the upper limit of troposphere, whatever is the name of the next layer, and that troposphere ends, where stratification starts.
Troposphere is the lowest part of the atmosphere where persistent convection limits the lapse rate, tropopause is the level where this convection stops. That level exists independently of the absorption of solar radiation in the upper layer be it stratosphere or not.
It’s really fundamental that GHGs are central in maintaining the convection and that this happens trough the mechanism that they try to force the lapse rate to a higher value than the adiabatic lapse rate. With little GHGs we have a weak circulation, with increasing GHGs the circulation gets stronger. That’s certain up to a point. it’s more difficult to tell, how added GHGs affect the complex circulation of the present atmosphere, but in some sense it’s likely to get stronger even now with more GHGs.
Pekka,
If you thought my last comment was implying you should apologise, in particular about getting some terminology wrong, then you really do have trouble understanding what I write (for balance, maybe I’m not sufficiently clear though).
As for your last paragraph. Hmmm, yes I agree. As I tried to point out (I think) and others as well, if we have an atmosphere in which greenhouse gases set an effective radiative surface, then we have an atmosphere that gets heated from the bottom, and cools from the top. In between, the gas is optically thick (because of the GHGs) and hence we can treat it (approximately) as adiabatic (or the processes as reversible adiabatic processes). Hence the most stable configuration for that atmosphere is that it is in hydrostatic equilibrium in which the temperature profile settles to the adiabatic-like lapse rate. Whatever process acts to transport energy through this atmosphere, it will always tend back towards the most stable state. Yes, convection will act to do this. However, since the simple point is that the GHGs set the radiative “surface” and the lapse rates sets the temperature profile, it is my view that this specific is not really necessary in order to make the point that the GHE exists. You may, of course, disagree. You may, of course, also be wrong. It’s certainly my experience that one way to start an argument is to discuss how best to discuss a complex topic. I find that tedious.
ATTP,
We still do not agree about physics.
I don’t agree that optical thickness is a reason for the sufficient validity of adiabatic approximation. The processes are not approximately reversible because of the optical thickness. Furthermore all important processes within the troposphere are not approximately reversible at all. In this connection we have some important subprocesses that are approximately reversible. Most importantly the collective motion of air and it’s expansion when it rises or compression when it subsides. Close enough to reversible to serve as an approximation is also the condensation of moisture (which leads to the moist adiabat). None of these processes involves directly radiative heat transfer, and a major part of the radiative heat transfer is not reversible (the atmosphere is everywhere too transparent for a major part of the wavelengths, therefore the temperatures differ at the point of origin and point of absorption, which makes the transfer irreversible).
The adiabatic lapse rate is not everywhere the most stable state, it’s the limit of stable states, not the most stable state. Temperature inversions are stable under many conditions and the whole temperature profile is stable far from the adiabatic lapse rate in high latitude winter. It’s not a general rule that the atmosphere moves towards the adiabatic lapse rate. That’s true when and only when something forces vertical air flows (uplift or descending flows). The absence of such flows is stable in absence of external forcing.
Yes, you may restrict your argument to some parts of full description. The problem was that your original post was not only limited in coverage, but did contain some erroneous formulations. All this discussion is the result from noting that. Then we have gone on, because some similar errors have popped out at every step.
I have been ready to stop several times, but then something new has been brought up that have made me think that I must respond again. For most of this exchange I have written responses to comments that have been directed to me, or interpreted my earlier comments in a way that differs from what I tried to say.
Pekka,
I apologise for bringing this up again. I agree that what I’ve described is too simple. I agree that the actual processes are much more complicated than how I’ve described them. I think I agree with you, although you’ve made things so complicated, I really don’t know. The simple point, though, is that the adiabatic lapse rate is a good description – in general – of the temperature profile in the lower troposphere and that, to a first approximation, one can determine the lapse rate without having to consider the details of the processes that transfer the energy, or consider that the actual approximations are not, strictly speaking, completely valid. When I say “approximately”, I use that term as I mean it – i.e., it’s not perfect, it’s approximate.
As I said in my previous comment, it appears that one way to start an argument is to discuss how best to explain a complex topic, as we’ve aptly illustrated here. I don’t know about you, but I find it incredibly tedious and annoying and would really rather not do to too often. If you are unable to comment here in a manner that is constructive and clear, and if you are unable to understand the meaning of the words “approximate”, “almost”, “about”, “roughly” and why I use them, then maybe give some thought to not commenting at all. You do appear to have a reasonably good understanding of the basic physics, but this blog takes up enough of my time that I really don’t need to spend it arguing with someone who seems unwilling to recognise an approximation and seems intent on being pedantic about details.
I apologise if that seems unduly unpleasant, but this blog is not a public service and there are enough frustrations and annoyances without adding more.
ATTP,
My comments may have been too difficult. They may have required some prior knowledge that was unreasonable to expect. I have done my best to explain the issues, and from earlier experience I know that I do often succeed in that, but not always.
In one pair of comments you wrote a description on the issues that I accepted as one that I would not react to. That presented the main argument without adding something that’s not quite correct any more.
From my previous comment you should see, where the problems start to build up. Getting around them requires an overall picture of the full process that creates an atmosphere with the main features of the real one, heating by sun, cooling to space, radiative heat transfer within the atmosphere, convection and full circulation at least at the level on one cell. Understanding stratification is also essential. Evaporation, condensation and precipitation would be good additions, but not as essential to start with.
I continue to believe that the most fundamental errors of the Connolly papers and of many other similar claims by skeptics are closely linked to the issues I have tried to explain. Therefore I consider my comments relevant for this thread.
I really wanted only to help, I apologize, if I couldn’t make that clear at every step.
Pekka,
They may well be relevant, and as I’ve tried to explain (from quite early on) there’s no doubt that I’ve been rather simplistic in how I’ve described things. This is a blog, and I’m aware that I’m not teaching a graduate level atmospheric physics course. I’m trying (and sometimes failing) to explain something that illustrates an aspect of the GHE in a manner that could be understood by someone with a basic understanding of physics. That’s partly why I found some of what you said confusing. Not that I don’t understand (I would hope I do, given my career and where I am in it), but I wasn’t always clear what you were criticising or why. There always be more detail one can put into something like this. Choosing what to include, what to ignore, and what approximations to make are important (and sometimes contentious) parts of communcating science.
There are probably many reasons why there are fundamental errors with the Connolly (and related) papers and I have no doubt that what you’ve said is relevant. However, I would still maintain that what I’ve described here is a reasonably good (if clearly not perfect) illustration of why what they suggest is almost certainly wrong and why they’ve likely misunderstood why they can’t find a GH signal in their radiosonde data.
And I apologise for my tone, but I believe I’m very open to criticism and correction. However, my response to criticism is somewhat influence by the tone of the criticism. Possibly a failing, but one that’s hard to avoid.
Pekka –
Some unsolicited advice. If you want to help, don’t tell people that they don’t the subject that they’re talking about (I’m paraphrasing). Just tell them specifically what you disagree with and why. You’re not really in a position to know if they know what they’re talking about. It could be that they were imprecise in describing what they do know. It could be that you misunderstood something that they have said. And anyway, your judgement of what their knowledge is, is basically irrelevant.
I only add that my English is obviously not perfect. Therefore I cannot always reach the tone I would like to present. Even with better knowledge of language blog discussions pose problems for getting the reader to see the intended tone.
I try to remember the above, when I read comments directed to myself. I don’t always succeed in that either. I consider that a price to be paid for being active in blogosphere.
Pekka,
It seems that the troposphere is heated from below. It seems heat can transfer to the stratosphere from below or above, but the heating from above is enough to invert the temperature profile.
What is the feature of the stratosphere that allows it to be heated from above, and what feature is missing from the troposphere that prevents this.
It’s a little hard to keep a straight face while assigning a temperature to a thin vacuum where O atom densities approach that of O2 molecules like the thermo and ionosphere and the free ion/electron concentrations are going through the roof. There are not exactly conditions of thermodynamic equilibrium. For the same reason the upper atmospheres of planets like Venus and Mars where the temperature trend has turned around and is increasing with altitude is a very different deal than the Earth’s stratosphere and it is more than a stretch.
kdk33,
The upper stratosphere is heated by solar UV absorbed by ozone. The cooling mechanism is IR radiation by CO2 and other GHGs. As the pressure is low, the lines are narrow. The radiation up from the upper stratosphere is visible in the outgoing IR spectrum as a sharp peak (or actually a few very sharp peaks at nearby wavelengths) at the center of the broad minimum caused by the absorption of IR by CO2 at lower altitudes both in the troposphere and in the lower stratosphere.
The downwards radiation from that peak cannot proceed far as it’s at the wavelength of strongest absorption. Therefore the heat is not transferred effectively down. Furthermore the overall power of that IR is not large as the peaks are very narrow. The lower part of the stratosphere does also absorb some solar radiation, but it receives much of it’s heat from below. The local heating and radiation from above are the reasons for the turnaround of the temperature gradient, without this heating the temperature would keep on decreasing with altitude, but at a lesser rate than the adiabatic lapse rate. Therefore that region would not be part of troposphere. (I have learned to avoid telling what would be the name of a stratified part of the atmosphere where temperature drops with altitude.)
(Eli hints in his most recent message to the reason, why the upper Venus or Mars atmosphere should not be called stratosphere. That makes sense. I still don’t know whether it’s correct to call the part of the counterfactual Earth atmosphere stratosphere that’s next above the troposphere, but with a falling temperature due to lacking solar heating.)
Interesting discussion in light of what many call settled science.
Some confusion may involve different conceptions of the greenhouse effect or whatever causes Earth’s surface to be warmer than it would otherwise be without an atmosphere. As everyone should agree, the greenhouse terminology is misleading. I recommend avoiding the term and will refrain from doing so further. Instead of GHGs I will refer to IR absorbing gases or IRAGs. Instead of GHE, I will use Temperate Climate Effect or TCE.
By definition, the TCE occurs due to the presence of an atmosphere. The discussion here is about what effect changes in the atmosphere will have on the lapse rate. ATTP and Ronan discussed one case, the hypothetical atmosphere without IRAGs compared to the real atmosphere. Pekka mentioned another hypothetical atmosphere with “less” IRAGs. Most of the discussion focuses on the difference between the real atmospheres–past, present, and future–which are varying due to increasing levels of IRAGs, mainly CO2.
ATTP proposes that increasing CO2 won’t affect the lapse rate, but will raise the effective emission height and, consequently, the effective surface temperature. Has anyone measured the effective emission height of the atmosphere? For this hypothesis to be accepted, some measurements of lapse rates, emission temperatures, and emission heights would be needed. The Connollys have drawn their conclusions from actual data.
Chic,
This isn’t quite what I’m suggesting. I’m suggesting (as I believe is accepted) that adding more GHGs, to a first approximation, doesn’t affect the lapse rate. If you see one of my earlier comments however, that isn’t strictly correct, as there can be feedbacks that do affect the lapse rate. This was only meant to be approximate. Not exact. What I’m suggesting is that there isn’t an easily identifiable global warming signal in the lapse rate. In some sense one could probably argue that in the absence of a GHE one wouldn’t expect our atmosphere to be as it is, but that implies accepting the GHE in the first place. So, what I’m suggesting is more that there isn’t a special, easily identifiable, greenhouse effect signal in the temperature profiles in the atmosphere that would convince someone who is skeptical of the GHE that it does indeed exist.
I believe, however, that there are measurements indicating that the troposphere height in the atmosphere has increased by a few hundred metres over the last 30 years or so. There is some discussion and some links to papers in this Skeptical Science post.
As I said above, there appear to be measurements showing an increase in the troposphere height over the last 3 decades. There’s also some basic physics that one can apply. In the absence of a greenhouse effect, the average temperature of the surface of the Earth would be 255 K. It’s not, it’s more like 288 K. That there is a layer in the atmosphere that is 255 K, tells is that the Earth is effectively emitting from this layer. Why? Because if you measure the outgoing spectrum, as has been done, to a first approximation, the Earth appears to be a body with a temperature of 255 K. Since the surface is clearly not this temperature, one could conclude that the Earth is emitting to space from a layer in the atmosphere that has this temperature. This occurs at around 5 – 6 km in the atmosphere.
What I think the Connollys have done is make various measurements in the atmosphere and somehow concluded that there is no evidence for a greenhouse effect. I’m still slightly confused as to precisely how they’re drawing that conclusion. However, until they can explain how – in the absence of a greenhouse effect – the surface doesn’t cool down to an average temperature of 255 K, I think what they’ve concluded is not correct. Just because some measurements have been taken does not immediately mean that the conclusions drawn from those measurements make sense.
Has anyone measured the effective emission height of the atmosphere?
Yes, it is trivial. You look at the effective temperature of emission to space from greenhouse gases. This walks you through it, but there are lots of low earth orbit and satellite observations
Eli, thanks, that’s a really good post. I hadn’t seen that before.
Ah, but there is an easily identifiable signal in the lapse rate because amount of water vapor in the atmosphere increases (observed)
Eli, yes, but to identify such a change, one would still need, presumably, take measurements over many decades(?) not many hours.
Chic Bowdrie:
“Interesting discussion in light of what many call settled science.”
Well, there are also “interesting discussions” to be found in “academic publications” “demonstrating” that the earth is only 6,000 years old, despite the fact that the earth is rather older than that being “settled science”.
Not to mention those “interesting discussions” of perpetual motion, 2nd-law-overturning machines that university physics professors occassionally run across in their in boxes …
Such work is always based on, as you put it, “actual data”.
“interesting discussions” which claim to overturn wide swaths of settled science are almost always just a waste of time.
Connolly’s effort falls into this category.
Fortunately for us, Chic Bowdrie demonstrates his knowledge of physics here:
http://scienceblogs.com/stoat/2014/02/16/the-idealised-greenhouse-effect-model-and-its-enemies/
@Chic,
did you see anyone here disputing the fundamental basics? Just wondering … cauze I certainly didn’t.
Chic, surely what’s interesting is that this discussion (of “settled science”), has rather tunnelled down towards a common perspective with much of the apparent “disagreement” relating to (i) confusion over terminology, (ii) a desire from some quarters that the subject should not be described with any degree of approximation, and (iii) various degrees of pedantry (see (ii)).
One can pull out the essentials of the “settled science” since they shine through the discussion, namely that increasing levels of greenhouse gas result in a higher altitude for radiation to space of LWIR required to restore the Earth system towards radiative equilibrium, with the resulting warming of atmospheric layers below and right down to the surface according to the lapse rate. The empirical observations of enhanced water vapour and raised tropospheric altitude are entirely consistent with the physical understanding of what one might choose to call “settled science”.
So your term “settled science” is an appropriate one here 🙂
ATTP,
I don’t know which GHE you refer to and I don’t want to waste my words by parsing yours. However, I would like to be convinced of some kind of signal due to increasing CO2, etc. I will check out your SKS reference and the one from Eli as well.
“In the absence of a greenhouse effect, the average temperature of the surface of the Earth would be 255 K.”
I assume you mean in the absence of an atmosphere, the idealized grey body model of Earth would be 255 K. This doesn’t necessarily mean that emission to space occurs in the layer 5-6 km up in the real atmosphere. More likely there is some sort of skewed bell-shape distribution of energy emission from altitudes centered around 5-6 km. I think how increasing CO2 will affect that distribution is speculation.
No, I mean in the absence of the greenhouse effect. In the absence of the greenhouse effect all radiation passes straight through the atmosphere. Hence all the light from the Sun hits the ground and (apart from that reflected) is absorbed and heats the surface. The surface then emits at longer wavelengths which, in the absence of the greenhouse effect, would all pass back out through the atmosphere into space. The surface would then have the non-greenhouse equilibrium temperature. This is basic physics and I don’t think that there is any alternative to this.
I get the impression that some people think that the atmosphere can both set the surface temperature (magically) and can somehow absorb radiation without producing any kind of greenhouse effect. This is incorrect. The greenhouse effect is, in its simplest form, a consequence of the fact that the atmosphere is not transparent at infrared wavelengths and hence absorbs the energy being radiated by the surface. If you think that the atmosphere absorbs radiation and re-emits it at 5-6km and that the atmosphere also sets the surface temperature, then you agree with the basic greenhouse effect; you maybe just don’t know this yet.
Eli
“Ah, but there is an easily identifiable signal in the lapse rate because amount of water vapor in the atmosphere increases (observed)”
I don’t follow. The lapse rate decreases with increasing water vapor.
Chic,
Eli didn’t specify whether or not the lapse rate would increase or decrease. He was simply pointing out that increasing water vapour in the atmosphere (as occurs when the atmosphere warms) would change the lapse rate.
ATTP,
You are implying that an atmosphere without IRAGs will result in a surface temperature the same as an Earth with no atmosphere. Isn’t it possible that an Earth without IRAGs, but O2/N2 the same, would still warm up enough to radiate away whatever it needed to maintain whatever average temperature resulted?
Concerning the post of Eli on the MODTRAN calculations, it’s worthwhile to check the first link to an earlier post and the comment of Raypierre there. That’s actually related to the some of the discussion I have had in this thread on the properties of the stratosphere.
I think the simple answer to your question is that if such an atmosphere did not absorb any radiation, then there’d be no greenhouse effect and the surface temperature would be set entirely by the balance between the energy received from the Sun and the energy radiated back into space. An atmosphere, by itself, cannot influence the surface temperature. It does so through the greenhouse effect.
Chic,
The atmosphere does not affect the radiation from surface, it has two other effects:
– how much of that radiation gets absorbed in the atmosphere
– how much downwelling IR radiation the surface receives from the atmosphere.
Without GHGs there’s virtually no absorption of IR and no DWIR. Therefore the surface cools as efficiently as without any atmosphere.
“Eli didn’t specify whether or not the lapse rate would increase or decrease. He was simply pointing out that increasing water vapour in the atmosphere (as occurs when the atmosphere warms) would change the lapse rate.”
This is the feedback caveat you referred to earlier. However, it is a transient effect with respect to varying humidity and limited by the saturated lapse rate. This still leaves me searching for the signal due to rising CO2.
It’s not transient. As CO2 concentrations increase in the atmosphere it warms the atmosphere. This produces an increase in water vapour that can then change the lapse rate. I don’t know why you think it’s transient.
It is transient for two reasons. If Earth cools, the previous increase in water vapour would become a decrease. Humidity varies. The “average” lapse rate constantly fluctuates depending on what the humidity is at the moment. I don’t like discussion averages, because what does it really represent anyway?
Chic,
You asked for a signature of warming. Yes, if the Earth cools, but if it doesn’t …. Averages have their place, especially on blogs 🙂
ATTP and Pekka,
As long as there is an atmosphere, there will be heat conducted into it. The surface will warm and warm the air above it. Without IR absorbing gases to cool it, the surface should be warmer than the case with IRAGs.
Chic,
And what happens after this? As I understand it, air is a very poor conductor.
I think you’re assuming here that IR gases only act to cool. That would be, I think, wrong.
My apologies if this has already been suggested upstream. I’m coming to this thread late in the game and do not have time to read through all of the comments that have already been posted.
Anyone interested in delving into more detail on the “Lapse Rate” than Anders did in his OP would do well to read, Temperature Profile in the Atmosphere – The Lapse Rate posted on the Science of Doom website.
Badgersouth,
That is a good post, thanks.
Chic,
The atmosphere cannot warm the surface significantly in any other way than by downwelling IR, because that’s the only effective way of moving heat downwards.
With an almost fully transparent atmosphere (very little GHGs) the surface is essentially as cold as it would be without any atmosphere. The atmosphere near the surface has approximately the same low temperature and the upper troposphere is really cold, tens of degrees colder than in the present atmosphere.
Pingback: It’s not fair! I can prove I’m better than BOD… | ThinkOrSwim (the Climatechange.ie Blog)
ATTP and Pekka,
I’m still working on my tropopause height homework, but don’t want to let the non-GHG discussion get cold (smiley).
Since the atmosphere has never been IRAG-free, no one has measured what the temperatures would be in that case. Conduction from the surface would be slow, but not negligible. Convection would still occur. Eventually, a lapse rate would develop and be larger than an IRAG-free atmosphere. Where else would the heat go?
Pekka, heat doesn’t need to move downward from the atmosphere. Some does at night, fortunately. Heat from the sun does need somewhere to go to keep the surface from becoming too hot as it probably would be without IRAGs.
ATTP,
I studied Eli’s MODTRAN simulations and failed to see how one would measure any change in emission temperature and/or levels. The CO2 shoulder remains just below the 220 K curve for both 375 and 37500 ppm CO2 at 70 km. I would predict that a series of simulations between 10 and 20 km would show the CO2 shoulder gradually move from 240 K to 220 K and level out to basically what it is at 70 km. Furthermore I see no way to tell what happens to the surface temperature as a result of this simulation.
The SKS reference leads to papers behind pay walls. The abstract on one indicates a 50m increase in tropopause height and 1.25 K drop in temp between 1978 and 1997. Using a 6 deg/km lapse rate, the surface temperature would have to be between 1 K cooler or 0.3 K warmer depending on whether or not the 1.25 K tropopause temperature difference is factored in.
I agree that there isn’t an easily identifiable global warming signal in the lapse rate.
Considering the paucity of data on tropopause height, I also think it difficult to justify a greater surface temperature due to increasing CO2.
Chic,
Well, we have measurements from the Moon that are – I believe – entirely consistent with what basic physics would tell us a non-GH temperature should be.
I don’t see the relevance of this. If there are no GHGs, all energy radiated from the surface passes through the atmosphere into space. If the surface temperature were higher than the non-GH temperature it would radiate more energy than it receives and it would cool. If it were lower, it would radiate less energy than it received and it would heat it. It would, therefore, typically settle at the non-GH temperature and the atmosphere would be irrelevant.
Radiation! If there are no GHGs, the atmosphere is irrelevant.
I think that’s kind of the point. Whatever the CO2 concentration, the Earth will appear – from space – to be a body with an average temperature of 255 K. The effective height at which it emits its energy to space, though, increases with atmosphere GHG concentration. To determine the surface temperature you need to know this height (set by the GHGs) and the lapse rate. That, at a fundamental level, is what the GHE is doing. It’s raising the height at which we emit our energy into space into the atmosphere and, consequently, causing the surface temperature to be higher than the non-GH temperature.
How else would you get it then?
Chic,
There’s no doubt that energy can be transferred from surface to atmosphere without GHGs, but it’s very easy to verify that the heat transfer in the opposite direction is very weak. That’s seen all the time in weather patterns that lead to temperature inversion. Under clear sky and with little moisture the surface cools rapidly radiating to space. It’s not protected by the warmer air above, because it,s radiates too little without sufficient water vapor. The radiation from CO2 can also be measured and has been measured under those conditions. This radiation slows down the cooling, but cannot prevent significant an fast cooling. Without GHGs temperature inversion would be present everywhere during the night and when the sun is near horizon.
Without any GHGs that same effect would be present everywhere. The warmest area with sun high in the sky would warm rapidly, but during night even those places would cool to rather low temperatures. Less sunny latitudes would not get close to the freezing point during the day. The North pole would be warm for a couple of months assuming that without GHGs we would not have snow either, but extremely cold during most of the year. The atmosphere would have a very weak effect on the surface.
Without clouds the height distribution of the point of emission of radiation that exits the atmosphere is determined almost fully by the GHG concentrations, the temperature profile has an almost negligible influence on that in any other way than by affecting water vapor concentration. The more GHGs the more is emitted at highest levels and the less originates at or near the surface.
Temperature profile enters in the intensity of radiation as the intensity is the stronger the temperature is higher at the point of emission to space. The height of the tropopause is significant as one factor that affects the temperature profile.
The calculations of Eli show dramatically some phenomena, but only one case has a consistent pair of GHG concentration and temperature profile (that of the present atmosphere). The other cases are severely inconsistent, and should not be thought to represent any alternative possible case on a planet like Earth, but with very different amounts of CO2. In particular the upper stratosphere would not be warm with very high CO2 concentration, but colder than tropopause. Thus the curve would be very different. (Would the stratosphere be stratosphere with that profile? For me, yes. For others, I don’t know after the nit-picking on this word by Eli.)
ATTP,
You are doing of good job of it, but you are simply reiterating the standard hypothesis. If you want to convince someone who is skeptical of the GHE that it does indeed exist, there has to be substantive evidence. There is no evidence in Eli’s calculations of a change in emission height or temperature with increasing CO2. The only radiosonde evidence I saw (albeit limited) indicates the tropopause height increases while its temperature cools. If the lapse rate is consistent there would be no net change in surface temperature. Definitive evidence will be hard to get. The tropopause height varies with season and latitude, which introduces the problem of averaging.
Chic,
Well, all I have is the standard hypothesis. I can’t really do any better than that. I’m also not really trying to convince anyone of anything. I’m simply willing to spend some of my time explaining, to the best of my abilities, the science/physics associated with global warming and climate change. If you choose not to accept it because you think it’s not strong enough, that’s entirely up to you. However, I would argue that suggesting that the surface of the planet can be 33K warmer than the non-greenhouse equilibrium temperature without a greenhouse effect is fairly close to appealing to magic 🙂
This is absurd. The radiative properties of GHGs are well established. The relationship between paleoclimate behaviour and GHGs is well established (eg early Cenozoic hyperthermals). Arguing against basic, stone-foundation physics and solid paleoclimate evidence (CIEs and hyperthermals) isn’t big or clever.
CIE = carbon isotope excursion.
ATTP
IIRC, Brandon did not answer the key question: why is the Earth’s surface as warm as it is? I’m glad to see you reiterate it now.
Sorry – Ronan, not Brandon.
Pingback: Pielke Senior ci spera ancora – Ocasapiens - Blog - Repubblica.it
For genuine skeptics, here is 134,862 distinct pieces of “no evidence” of the greenhouse effect (see discussion point (2) in the post). Briefly, the evidence is 134,862 distinct comparisons between predictions of the TOA OLR by the Fu-Liou radiation model, and observations with the Ceres satellite. The picture that emerges is one of remarkable accuracy across a wide range of latitudes, longitudes (not separately graphed), and temperatures. From the accuracy of the model as shown by these comparisons, it follows that with very high probability, that the model has got the physics of radiation correct. And the model – in fact all radiation models of even moderate accuracy – predicts both the GreenHouse Effect (GHE) and the Enhanced GHE.
Indeed, the accuracy of these models have been known from observation since at least 1970. Therefore it has been known with very high probability since 1970 that the GHE, and the Enhanced GHE exist, and are as predicted by theory (see last section of the original blog post from which the comment linked above comes).
So anytime you read somebody (Chic Bowdrie) saying, “If you want to convince someone who is skeptical of the GHE that it does indeed exist, there has to be substantive evidence” and suggesting that no such evidence exists; they are at best demonstrating their complete ignorance on the topic. They can be taken no more seriously than somebody who disputes that the axis around which the planets rotate lies (to a first approximation) within the Sun’s surface.
We now have the opportunity to see whether Chic Bowdrie is merely woefully, and stubbornly, misinformed – or whether his is actively irrational (or worse) on this topic. If the latter I recommend you stop wasting time on him by removing his posting privileges.
Tom, is there supposed to be a link in your comment?
I think this is the link Tom was intending to include.
Sorry, this is the link I intended to include; although I also like the one found by Anders 🙂
Actually, doing a google search for “134,862”, “curtis” and “skeptical science” will give you about half a dozen good comments (IMO 😉 ) on the same theme. One in particular adds something to the others in that its first point contains a discussion of the relationship between scientific models and scientific theories, which is relevant.
Hi guys,
Apologies for the delay in returning to this thread. Apparently, there’s been 148 new comments since I left you on Sunday evening! 😮 I can’t reply to each comment individually, but some arguments seem to have been mentioned several times, so I’ll try and focus on those ones for now.
In ATTP’s response to Chic on Feb 25th, 3.50pm, you said
This is one of the main issues we discuss in Section 3.2 of Paper 2. What we point out is that, yes, the greenhouse effect theory predicts what you are saying should occur. But, we point out that there are also other ways to explain these observations. If you look at page 17 of our Paper 2, we quite explicitly pointed out that our explanation is also consistent with many of the observations used as “evidence of the GHE”:
I know you were saying on my blog that you think the assumption of Local Thermodynamic Equilibrium (LTE) isn’t important for the GHE theory. But, as Eli has being pointing out, it is indeed a key assumption.
In Section 3.2, we show that, if the LTE assumption is invalid, then you still can explain all of the experimental observations typically attributed to the GHE. Moreover, we think it provides explanations for several observations which the GHE theory does not explain.
Further, our analysis of the weather balloon measurements in Paper 1 indicates to us that the atmosphere is in thermodynamic equilibrium (at least over the distances from 0-35 km covered by the balloons). This contradicts the LTE assumption.
In Paper 3, we put forward our explanation as to why the atmosphere would be in thermodynamic equilibrium over these distances.
P.S. I’m surprised that, while there has been a lot of negative criticism of our conclusions in this thread, I haven’t found any discussion of our experimental results!
A theory is only of use, if it is able to explain the available data. We started from the assumption that the GHE theory was correct. However, when we carried out our measurements, our results contradicted the predictions of the GHE theory. This is why we have put forward alternative explanations for the atmospheric temperature profiles.
Up until now, this thread seems to be focussing on the fact that in our analysis we’re not using the textbook theories. Some people have even implied that we’re somehow unaware of the textbook theories! Nothing could be further from the truth!
Instead of ranting about how our conclusions are allegedly “wrong”, why don’t you start looking at our results, and explain them!!!
I look forward to these papers being submitted for publication somewhere reputable. There’s too much blogscience about for anyone to invest the time in reviewing them.
Ronan,
Thanks for the comment.
I don’t think Eli is quite saying what you think he’s saying. Also, I didn’t say it wasn’t important. I said your measurements don’t contradict it. I also think that you’re mis-interpreting what LTE means. All it really means is that the gas is able to retain a Maxwell-Boltzmann distribution and hence one can define a temperature. Your measurements, I’m pretty sure, show that the gas is in local thermodynamic equilibrium. You seem to be suggesting that LTE means/implies that there can be large changes from one locality to the next and I don’t think that this is necessarily correct.
I don’t think that this is correct. I think that the atmosphere can both be in thermodynamic equilibrium and in LTE. I don’t believe that your measurements are showing that it is not in LTE. If it was not in LTE you would not be able, strictly speaking, to define a temperature.
I think what you’ve done is chosen something that you believe to be a fundamental aspect of the GHE and which you think your measurements don’t show. Firstly, the GHE is simply that the lower troposphere is largely opaque to outgoing longwavelength radiation. That’s really all it is. The details of how the energy is transported to the emission height is a consequence of the GHE, but not defined by it.
Firstly, extraordinary claims need extra-ordinary proof. Secondly, nothing you’ve said so far has convinced me that the reason the surface of the planet is 33K warmer than the non-GH temperature is not because the lower troposphere is largely opaque to long-wavelength radiation. As far as I can tell, you’re essentially describing the greenhouse effect. In a sense, I don’t necessarily think there is anything wrong with the measurements you’ve made. What’s wrong is how you’ve interpreted them.
If the greenhouse effect is small/negligible (as you suggest) then the atmosphere is – by definition – transparent to infrared radiation. This means that the surface will cool if its temperature exceed the non-GH temperature and warm if it is below the non-GH temperature. That’s really all it is. The fact that the average temperature of the surface if 33K higher than the non-GH temperature is, to a large extent, proof of the greenhouse effect. There really is no other physically plausible explanation.
So, until you can explain how the surface temperature can be 33K warmer than the non-GH temperature despite the greenhouse effect being negligible, I’m going to continue claiming that what you’ve concluded is incorrect. I don’t need to consider your measurements, or your chemistry or your radiative transfer to do so. I just need basic physics.
Pekka
What I thought was missing on both sides of the discussion was water vapour, which I assume you included in your comment about the need for GHG(s) as a mechanism for transferring energy.
The dry adiabatic lapse rate is on page one of the the chapter. Since in real life and in a real climate we deal with moist air, there seem not much point in labouring over the dry adiabatic rate.
Chris,
I certainly put that in the post. I believe that in the Earth’s atmosphere the average/typical lapse rate is somewhere between the dry adiabatic lapse rate (10 K/km) and the saturated adiabatic lapse rate (5K/km). From what I’ve read, a value of around 7K/km is a reasonable estimate for the average lapse rate in the Earth’s atmosphere. Of course, it can vary quite a lot depending on the region, the time, and the seasons.
Pekka:
That is not informative. What is at dispute between us is what the standard physics in fact implies.
As it happens, in 1930, Richard Tolman wrote a paper, “On the weight of heat, and thermal equilibrium in general relativity”. In it, he derives first a classical approximation and then the relativistic solution to the differences of temperature with distance from the center of a self gravitating gas in thermal equilibrium. In the conclusion he writes:
He then goes on to note the necessity of a more accurate formulation to account for relativistic effects.
He goes on, in his final paragraph to write:
I freely admit that I am unable to follow the maths in that paper. Never-the-less, the qualitative expression is exactly my physical intuition in this case, and the “first approximation” of the effect was exactly the effect I was trying to capture (whether successfully or not) with my equation. Given Tolman’s prominence, his bold assertion that this was indeed an entirely new result, and the fact that the paper continued to get cited into the 2000s, I am strongly inclined to believe that no refutation was published, and that Tolman’s treatment was sound. Of course, my formulation may in fact be inconsistent with his first approximation – but I remain very confident that it is my, rather than your intuitions on this matter that comply with what “standard physics” implies.
I start with the last comment of Tom. The result of Tolman concerns the difference between General Relativity and earlier theories of gravity. Furthermore he writes that effect is extremely small. Based on these comments his result is totally irrelevant for our present discussion. To say more about that would require looking carefully, what he and others have written on that later.
Back to Ronan Connelly. Thermodynamic equilibrium means that there’s first of all a local thermodynamic equilibrium everywhere, and further that all these local equilibriums have the same temperature and also many other local properties relate to each other in a specific way determined by the thermodynamic equilibrium. Ronan Connolly’s “theory” is so far from all textbook theories that he must first define all concepts again, and show that he has an consistent alternative theory that agrees with the whole range of phenomena explained by our present physics. I would be dishonest if I would wish him good luck, as I’m totally convinced that he would only waste more time by continuing to develop his theories. We have a very successful theory of physics, there’s no need for an alternative.
A little on the lapse rates.
The dry adiabatic lapse rate is locally valid in adiabatic ascending or descending movement of air in all cases where neither condensation nor evaporation takes place. The moist adiabat applies to a rising flow, when the air is saturated by moisture and condensation takes therefore place continuously as required by the falling temperature and Clausius-Clapeyron law. The difference between the two lapse rates is very small at very low temperatures of the highest part of the troposphere, because the condensation of the little moisture left has a small influence on the temperature. The difference increases gradually when the temperature and the saturation moisture level increase. In warm tropical lowest troposphere it’s as low as 4 C/km.
A typical value for the actual lapse rate is about 6.5 C/km (often called environmental lapse rate). That seems to be true under quite wide range of conditions. The value is surely affected by many deviations from the assumptions that lead to the pure dry and moist lapse rates including horizontal mixing of air.
Ronan,
ATTP asked at some point: “Explain how our atmosphere ensures that we are warmer than an asteroid would be at the same distance from the Sun” and just recently “explain how the surface temperature can be 33K warmer than the non-GH temperature despite the greenhouse effect being negligible ….”
You and ATTP have been discussing existence of the it-shall-not-be-named effect without agreeing on what each of you mean by it. Everyone should agree that the planet is to some degree warmer because it has an atmosphere. There are at least three possible scenarios resulting in a warmer planet:
1) An Earth-like atmosphere without IR absorbing gases compared to no atmosphere.
2) An atmosphere with IR absorbing gases compared to 1).
3) A future atmosphere with greater amounts of IR absorbing gases.
Why not define which effect your papers describe as negligible or non-existent?
Pekka, calculating the lapse rate based on the idea that the mean kinetic energy plus the gravitational potential energy is a constant, using the Newtonian formula for gravitational potential energy, and assuming a constant gravitational acceleration with altitude of g, the lapse rate is -22.8 K per Km altitude. That is a far greater absolute value of the lapse rate than the dry adiabat because it does not allow for the drawing of exchange of internal energy for gravitational potential energy. If you allow for that exchange, and the equipartition theorem, then the lapse rate is simply the dry adiabat. Given the equipartition theorem, I think the gravitational potential energy must be drawn equally from all modes of motion of the particles of the gas (ie, equally from motion in all three axis, plus from any vibrational or rotational modes), so the later result is far more probable.
I do not know if that calculation is in agreement with Tolman’s results, but if it is not, then either gravitational potential energy calculated under Newtonian principles is not a near approximation of gravitational potential energy calculated using General Relativity under Earth conditions (which would be a major surprise); or I have seriously misunderstood what is meant by Tolman when he says “a first approximation to the magnitude of this temperature gradient was obtained by modifying the classical thermodynamics merely by ascribing to each given intrinsic quantity of energy the right additional quantity of potential gravitational energy”. That would not be so surprising, but the english is fairly plain. A third alternative is that Tolman was calculating the situation for some fairly odd situations. His initial results are for a gravitationaly bound “cloud” of radiation.
In any event, I do not think you can simply brush this of. Particularly so given that I have seen the result that the lapse rate in equilibrium equals the dry adiabat derived from the virial theorem as well (but have lost the link, and the paper was on Arxiv). Regardless, your initial physical intuition that temperature must be constant is shown to be wrong, and until you in fact calculate the lapse rate under thermal equilibrium, you cannot argue that the existence of an environmental lapse rate governed by the dry and moist adiabats is proof of the existence of a greenhouse effect.
Tom,
That temperature is constant everywhere in thermal equilibrium is a very basic feature of thermodynamics. It’s not my intuition only, but follows directly from the second law. The whole thermodynamics breaks, if that breaks. Even perpetum mobiles of the second kind could be constructed.
General relativity might require some change for thermodynamics as well in situations, where its
influence is noticeable, but that’s not the case for the Earth atmosphere.
andthentheresphysics on February 27, 2014 at 1:55 pm:
You can see in figure 9 and 10 of the article Temperature Profile in the Atmosphere – The Lapse Rate that the lapse rate has significant variation outside of the tropics.
If you check out the (northern) polar regions you can see negative lapse rates in winter and low lapse rates in summer.
How can this be?
I haven’t read all the comments on this blog but from a few comments – from people who understand the subject – some people have tried to explain (correctly) that the adiabatic lapse rate is what happens to air temperature when that air is displaced.
If the air is not displaced, no convection and no reason for the atmosphere to settle into the “adiabatic lapse rate”. It’s only when you move air vertically and “rapidly” that it cools at the appropriate amount.
So if we take polar winter, the surface loses heat more rapidly than the atmosphere by radiation to space and so it cools faster. The atmosphere above becomes warmer than the surface below. Any rising air – for whatever reason – cools and is cooler than the air above. So it falls.
The atmosphere is stable. This also (often) happens at night.
And in high latitudes, even in local summer, the “environmental lapse rate” can be quite small. The “environmental lapse rate” is the temperature profile we measure, not the temperature change if we took a parcel of air and displaced it rapidly in the vertical direction. Meteorology people talk about baroclinic instability outside of the tropics/subtropics but I don’t really understand it, not enough to explain why the lapse rate is what it is outside of the tropics/subtropics.
The most important point to take away – a parcel of air rising rapidly cools. This is the dynamic effect, due to the first law of thermodynamics. Let’s call this the “adiabatic lapse rate” and not consider whether the parcel contains water vapor or not.
A parcel of air not rising at all does not follow this rule. The “adiabatic lapse rate” does not – by itself constrain the atmosphere – the “environmental lapse rate” to be the same as the “adiabatic lapse rate”. It’s only when other factors force air to rise. Suppress convection and you decouple the environmental lapse rate from the easy to calculate adiabatic lapse rate.
Surface heating by the sun (prime factor) and forced convection by circulation (secondary factor) are the “initiators” of atmospheric convection, thereby ensuring – in some parts of the climate – that the “environmental lapse rate” turns out to be the same as the “adiabatic lapse rate”.
SoD,
Thanks. Someone linked to your post earlier. I found it very useful. I’ve been trying to use it to convince Ronan Connolly that what he’s observed is consistent with the standard greenhouse effect. Not having much success yet 🙂
I agree that there can be quite large variations in the actual lapse rate and that one can get negative lapse rates (by which you mean dT/dz is actually positive) in some regions. As I think Pekka was trying to illustrate, it’s convection (or the vertical motion of gas parcels) that acts to drive the atmospheric temperature profile towards the adiabatic lapse rate.
In fact, when I wrote the paragraph that you quote, I was well aware that the lapse could be outside the range I mentioned and the last sentence was meant to imply what you’ve described here, but reading it back it does sound like I’m saying it’s always between 10K/km and 5K/km, which – as you quite rightly say – is not correct.
Concerning stability, the word may refer to to the state of the atmosphere or to the value of the lapse rate.
For atmosphere it means that the it’s not subject to the possibility of changing suddenly to a significantly different state due to a small disturbance. In this sense the stable states are those with a lapse rate smaller than the adiabatic or of the opposite sign as in temperature inversion, which is stable even for a very strong inversion. The state with exactly adiabatic lapse rate is a marginal case, where a major change in the strength of convection can be caused rather easily (It’s like a metal ball on a level glass table, very easy to get moving, but stable, when not pushed). Higher lapse rates are unstable, and lead very rapidly to strong convection.
For the value of the lapse rate a state of stationary non-zero vertical movement of air leads to stability in the sense that a major change in the strength of the convection leads to a small change in the lapse rate, i.e. the lapse rate is almost independent of the strength of the convective vertical flow as long as it does not lead to breakdown of the approximate reversibility of the process through turbulence. On the other hand the lapse rate of a stratified atmosphere can change more easily as witnessed by the rapid changes in the temperature profile of an inversion layer when that develops under suitable weather conditions.
I have now worked through my issues re the lapse rate at thermal equilibrium. That has been particularly helped by an excellent historical survey of thermodynamics, including of Maxwell’s originally mistaken calculation of a change of temperature with altitude, his correction – Loschmidt’s challenge and Maxwell, Boltzmann’s and Ehrenfest’s responses. I have also worked out through more careful reading that Tolman’s result only applies the gravitational potential of the difference in mass between the observed mass and the rest mass of particles in motion at a given temperature, rather than to the gravitational potential of the entire observed mass; and as a result truly is miniscule at near Earth surface conditions.
Thank you, Pekka, for your persistence.
On the general relativistic question, I think at issue is how you define “temperature” under relativistic conditions where time is dilated and mass and length are adjusted, somewhat different from anything of concern here on Earth.
On the Connolly claims – we’re familiar with this gravitational energy story, but the multi-merization business is another extraordinary claim that goes against centuries of observations of gases at a variety of pressures in the laboratory. Gases condense at high densities, not at low densities; the lower the density (at reasonable temperatures) the closer any gas becomes to ideal. What the Connolly’s propose would have profound effects on heat capacity, pressure-density relations, and other basic observational parameters for which there is absolutely no evidence. This is pretty wild stuff.
I have another theory to throw into the mix…
The average environmental lapse of 6.5ºC/km (actual temperature profile of the air) is often said to be created by water vapour warming the air with its latent heat to reduce the dry rate of cooling at 9.8C/km.
So the air is warmer than it should be by 3.3ºC for every km increase in altitude.
This is the difference between the environmental lapse and dry of 9.8ºC/km. right up to the tropopause, well beyond where most latent heat is released as water vapour condenses to form clouds at the boundary layer 7km below.
Water vapour strongly absorbs solar near-infrared which makes up nearly half of the solar spectrum.
Although water vapour density decreases at a steady rate right up to to tropopause its molecules communicate very well radiatively to equalise any differences in pressure broadening etc.
This warms the air equally throughout the column adding kinetic energy to nitrogen and oxygen via more collisions than emissions.
This could explain the smooth reduction in lapse rate up through the troposphere from the surface.
This reduction in the cooling (lapse) rate results in potential temperature increase of 33ºC/km. at the 10km high tropopause.
This is still colder than the surface by 65ºC but not as cold as the air would be if dry at 100ºC colder.
This accounts for NASA’s infamous warming of 33ºC., but at the tropospheric height of 10km. rather than the surface and only for that highest kilometre of air.
Planetary winds and turbulent mixing can average out that difference by mixing that warmer air above with the cooler below to equalise the potential temperature increase to zero.
This mixing to the average of 16.5ºC throughout results in a warming of the surface by 16.5ºC. but also a cooling of the air at the tropopause by the same amount.
However this warming of the atmosphere has been achieved by water vapour intercepting half of the sun’s energy and acting as a parasol which prevents the surface getting as hot.
This is different to conventional greenhouse gas theory but fits with my assertion that water in all its forms is a moderator of extremes.
A planet without oceans and evaporation would be a world of huge polar to tropical differences but maybe the same meaningless average that everyone refers to constantly.
As an afterthought to explain the origins of this idea…
At dawn sunshine hits the atmosphere from the side and starts exciting water vapour molecules straight away to warm the air which then rises above the cold and convection begins. Thereafter the convection loops and eddies get bigger and bigger until they reach the boundary layer at 2 to 3km altitude on average. The boundary layer is where most condensation happens and clouds form but also where the constant gradient wind blows. When these connective loops from the warming air at the surface connect with the gradient wind it is dragged to the surface and we feel the breeze at the surface once again after the calm of night. So the stable night time potential temperature increase of 9.9ºC over 3km is averaged out by turbulent mixing of the potentially warmer above and the cooler below to arrive at an increase of 4.95ºC at the surface by turbulent mixing alone!
P.P.S.
NASA says water vapour and a few other trace gases raise the surface temperature by 33ºC.
Consider the tropics.
Would removing all the water vapour from the air result in a drop of 33ºC at the surface?
Of course not.
Would removing all the water vapour from the air above London on a hot summer’s day reduce the surface temperature by 33ºC.?
Nope.
So where would the removal of water vapour’s supposed “greenhouse gas” effect result in a cooling of 33ºC at the surface?
The north pole perhaps?
Again no.
Water in all its forms is storage heater, central heating, underfloor heating and air conditioner all in one!
A moderator of extremes.
Pablo,
I don’t really know what you’re getting at. Maybe you could summarise?
Thanks for your interest.
It is pretty clear in my head but not so easy to explain, so bear with me.
THE FIRST KEY POINT being that water vapour should be considered as a distributor of energy rather than an amplifier and that the greenhouse effect of 33ºC warming is actually defined by the mass of the atmosphere and its kinetic energy not by the radiative properties of water vapour.
NASA’s infamous 33º surface warming by mostly water vapour could be interpreted as true if it is understood that the surface they refer to is actually at the 10km high surface of the tropopause and is only a progressive warming at a steady 3.3ºC/km throughout the column by direct absorption of solar near-infrared by water vapour.
THE SECOND KEY POINT being that the heat source for that warming is not coming from FAR-infrared from a warmed ground but from sun directly in the NEAR-infrared as it passes through an ever increasing density of water vapour, (nearly half of the sun’s energy is in the NEAR-infrared).
THE THIRD KEY POINT is that this interception of solar energy results in less heating of the surface by the sun.
So the idea that there is a “Greenhouse effect” above a sun-warmed surface which reradiates that far-infrared back and forth to raise the temperature 33ºC is plainly wrong.
The other version of “Greenhouse gas theory” that a trace of CO2 could raise that effective Earth temp. of -18ºC at 5km altitude higher in altitude by any significant distance which then, via the lapse rate downwards, raises the surface temperature is also suspect.
Warming of the atmosphere by the sun directly can only be via water vapour intercepting the incoming NEAR-infrared within the solar spectrum. CO2 only gets excited by outgoing FAR-infrared.
The mass of the atmosphere defines the height of effective Earth’s temperature of -18ºC at 5km high.
FOURTH KEY POINT
Working down from that -18ºC 5km height via the reduced lapse rate to the ground means a temperature at the surface of only 32.5ºC warmer instead 50ºC warmer if dry.
So the reduction of the lapse rate by water vapour results in a lower surface air temp. of 14ºC instead of 32ºC if dry lapse. { -18 + (5×6.5=32.5) = 14.5 ) or { -18 + (5×10=50) = 32 }
So water vapour is able to both warm the air with its radiative properties alone and reduce the surface temperature via its reduced lapse rate.
This is nothing new.
But mixing of air on a planetary scale to reveal the average potential temperature increase throughout is something I have not come across. The light bulb moment for me was when I noticed this increase of potential temperature to 33ºC at the tropopause was the same as NASA’s warming which they (disingenuously?) infer to be an increase of 33ºC at the sea level surface.
Here are the workings of my (definitely unreliable and well worth checking) maths.
Mixing within the turbulent boundary layer from the surface up to 3km equalises the potential increase (3 x 3.3 = 9.9) to the average throughout to make the surface air warmer by 4.95 and the upper air cooler by 4.95.
So I reasoned that if that mixing can reveal the average throughout that 3km column the same might happen on a planetary scale over greater distances.
If the 3km of mixing that occurs daily within the boundary layer is ignored that leaves 7km of increasing potential temperature with height within the very stable and stratified but windy air above.
So assuming thorough mixing could occur eventually throughout the atmosphere, that would result in the average increase of 23.1ºC (7 x 3.3) to the tropopause being revealed as 11.55ºC. throughout. (50% of 23.1)
11.5 + 4.95 = 16.5ºC average overall.
Assuming a base temperature above freezing for the globe maintained by the storage heater of the oceans we have something very close to the surface air temperature that is observed.
Which means in conclusion that the cooling of the air at the surface by 18ºC due to a reduced lapse rate is more or less balanced by the warming of the air at the surface by 16.5ºC by mixing alone on a planetary scale.( see key point 4 )
So NASA is right that water vapour warms the surface air but only by 16.5C. and only by intercepting sunlight directly and thus preventing the surface getting as hot. And remember that water vapour is only in the air because of evaporation that has cooled the solid or liquid surface.
All in all then.
It seems to me that water vapour simply moves energy around faster than otherwise which lessens the tropical to polar differences and makes for a more equable world.
Water vapour also lessens extremes of diurnal temperature range. It slows down the warming of the surface during the day and slows down the cooling of the land at night.
Pablo,
Well, this is simply wrong. Without the radiative properties of the gases, the surface temperature would be set by simple energy balance (i.e., it would have to have the same temperature as a blackbody that emits as much energy as it receives from the Sun).
Again thanks for your interest and for having the graciousness to publish opposing views.
After all consensus is not science.
With an atmosphere of mass to store heat as kinetic energy and a mass of surface and ocean to store much more of that solar energy, the surface will always be well above the blackbody temperature. That’s without consideration of the very effective insulation for downward conduction that our nuclear core provides.
Forget about the trace of CO2 for a moment.
Water vapour, the main “Greenhouse” gas, is on average 2% of the atmospheric mass.
Therefore 98% of atmospheric mass cannot cool down significantly by radiative means.
The warmth we feel in the air around us is determined by the average speed of the molecules within it. Without radiative gases this kinetic energy is preserved.
If you were to remove water vapour molecules with their internal absorption of various bands of infrared along with the same light that it then scatters, the only thing to slow down the cooling of the surface would be the temperature of the air above it.
However there would also be no means for the atmosphere to stop the surface getting as hot by intercepting the incoming infrared. (half the sun’s energy)
The amount of slowing down of surface cooling by radiation at night is much less than the slowing down of surface warming during a sunny day.
The air would be warmed by conduction and convection but as it now contains no radiative gases could not cool by radiative means. The only way it could cool would be by warming cooler air elsewhere by diffusion. So logically all air will eventually reach a temperature above Earth’s blackbody temperature. Its kinetic energy is preserved.
My suspicion is that although diurnal and seasonal ranges of temperature range would be much greater the AVERAGE temperature would would be much the same.
The real blanket for our Earth is provided by kinetic energy within the whole of our atmospheric mass.
The icing on the cake is that gravity ensures that most of that kinetic energy is kept close to the ground to keep us snug and warm within a universal vacuum of no temperature at all !
Pablo,
No, this is wrong. This is because if the atmosphere is completely transparent (as would be the case if the gases were not radiatively active) then this extra energy would be radiated into space until the amount of energy leaving the surface (per unit time) matches the amount of energy it receives (per unit time).
The mass of the atmosphere determines the pressure at the base of the atmosphere, but doesn’t determine the temperature. This is because the equation of state is a relationship between pressure, temperature, and density. So, you can have a dense, cold atmosphere, or a less dense, warm atmosphere, both of what can be at the same pressure.
“Warming of the atmosphere by the sun directly can only be via water vapour intercepting the incoming NEAR-infrared within the solar spectrum. CO2 only gets excited by outgoing FAR-infrared.”
Nope. Twice wrong, even. Not only does CO2 have several relevant absorption bands in the near-IR, CO2 gets nicely excited in the *mid-IR* region, too (the main band we all should know is around 2300 cm-1, which is about 43 µm). The FAR-IR region is far (pun intended) further down; it is usually stated to start at a wavelength of 15 µm and longer.
If you can’t even get this basic stuff right, why should anyone look further at what you write?
Thanks for continuing.
Marco..
OK. I should have said CO2 mostly gets excited by the outgoing… and the mid range overlaps with water vapour anyhow.
For the moment let’s forget about the trace of CO2 that through positive feed back is supposed to increase the amount of water vapour that is available to fry us to death.
[That’s AT for you, Pablo. Please mind your manners. -W ]
Just because an atmosphere without radiative gases is transparent to infrared does not mean that it cannot heat up by conduction and convection from a sun warmed surface. The surface would also now be able to receive the half of solar energy in the infrared that was previously intercepted by water vapour.
This warmed dry air would be purely kinetic and has no radiative component so it cannot lose its heat to space by radiative means.
Also, I did not say pressure determines temperature but gravity does give us a density gradient.
Warmer air will always be less dense than colder air and will rise until it equalises temperature with its surroundings, which in a non radiative atmosphere would be by diffusion.
Because this is now a dry atmosphere the lapse rate would be 10ºC decrease on the up and 10ºC increase on the down.
Even with a constant outgoing of increased IR from the surface to space, the temperature of the atmosphere would eventually rise to the effective temperature of -18ºC at 5 km high, at which point the surface would radiate out any surplus. This would make for an average surface temperature of
32ºC on average but with much greater extremes.
Pablo,
Except that the surface temperature would be set by how much energy it received from the Sun. It would therefore be much lower than it is observed to be today.
So what? Currently, the surface radiates about 395W/m^2, while the amount of energy we receive from the Sun is 240W/m^2. This would not be possible if the atmosphere did not contain radiatively active gases. The reason for this is because some of the energy being radiated from the surface is intercepted in the atmosphere, rather than being radiated directly to space. This is what leads to the greenhouse effect.
> After all consensus is not science.
Yet scientists reach consensus all the time, Pablo. OTOH, I don’t recall one single scientific exchange where one of the contributors rope-a-doped from one talking point to the next that ended well. Perhaps it’s because science requires a “dryer atmosphere,” in contrast of e.g.:
Science may not require legendary metaphors.
I don’t think this is correct. If the surface is the only place radiating and absorbing, then the effective radiating altitude is at the surface, not 5km high. And so the surface is -18C; the temperature needed for the outgoing radiation to balance the incoming radiation.
The lapse rate will still be the dry rate, as you say, but the effective radiating altitude will be at ground level, where all the radiation takes place.
Yes, precisely.
The air without water vapour would also absorb heat from the surface by conduction and convection.
A blackbody temperature is in a vacuum.
We are not in a vacuum.
An Earth in a vacuum would have a surface temperature of -18C.
In an atmosphere without water vapour to intercept radiation, the sun would blast the daytime surface with an undiluted 1370 W/m2 of power. This would make the surface much hotter in the daytime from which some radiation would go straight to space but at a much slower rate than it absorbs it. i.e. it would get hot.
The air in contact with that now hot ground would get very warm indeed.
Convection would lift that warm air up and away to spread its heat around by diffusion.
As nitrogen cannot radiate at Earths temperatures the atmosphere would warm up to -18C at 5km high at which point the surface air would become 32ºC. ON AVERAGE.
The sun would still blast mercilessly down, the ground would get hotter still.
But as the average air temperature has now stabilised at 32ºC. any surplus ground heat would be radiated straight to space. Balance is restored.
To assume that the average surface air temperature without water vapour in the tropical and summertime mid latitudes would never get above -18C. may be an error.
Pablo,
This does not change that without the greenhouse effect, the surface temperature would be -18C.
No. A blackbody temperature is the temperature an opaque body would need to have to radiate
I don’t think this discussion is really going anywhere, so maybe we can draw it to a close.
AT.
I apologise for my lack of blogging protocol re. your name back a bit. I am new to this.
No problem.
@-pablo
“To assume that the average surface air temperature without water vapour in the tropical and summertime mid latitudes would never get above -18C. may be an error.”
Paleoclimate indicates that at least twice the Earth has entered a ‘snowball’ state when this happened. Surface temperatures were below freezing with very little water vapour and low CO2.
In each case the geochemical indications are that the surface only warmed again when CO2 accumulated to levels sufficient to drive greenhouse effect temperatures high enough to get water vapour feedback.
Thank you for your correction. Its been fun.
I’m sorry to report that I just googled Pekka Pirilä, and learned from his Finish Wikipedia page that he passed away Nov. 24, 2015.
I believe most people know that. Pekka was a great guy. Always patient and helpful.
Talk about having you cake and eating it. You say:
– temperature profile of a planetary troposphere is unaffected by GHG
– a tropopause only exists because of GHG.
Let’s consider a hypothetical planetary atmosphere with no GHG. With only helium and hydrogen in the atmosphere.
It presumably has a lapse rate?
Eventually the lapse rate stops decreasing with increasing altitude, at about 0.1 bar. Above that temperature the temperature change reverses and increases with altitude.
What is the inflection point (at about 0.1 bar) where the temperature change reverses direction? What do you call it? You just said it ain’t called a tropopause because planet’s without GHG don’t have them!
Mark,
You might have to provide some context for your comment.