The diminishing influence of increasing Carbon Dioxide on temperature

AlecM 2:03am: “Therefore, you cannot consider transfer of energy by EM radiation as a form of heat exchange. Anyone who does so is teaching junk physics.”
Transfer of energy by EM radiation is energy exchange. Anyone teaching it is a form of heat exchange is teaching confusing physics as heat is nonexistant in nature therefore cannot be exchanged.
The top post is correct; there is diminishing effect on Earth surface Tmean = 288K of increasing CO2 concentrations since the upcoming surface emitted photons are in essentially the same supply due balance with sun exists and eventually fewer single CO2 can find unmarried surface born photons for nuptials.

Ah, so much to reply to, so little time. Let’s start with:

And a CO2 molecule “can absorb a photon of appropriate wavelength” why? Because the wavelengths are tuned to the electron orbits? Because the wavelengths are tuned to the C-O bond lengths? I think you are correct that it is the second. Thank you for your reminder.
The bonds are electron orbitals, but the distinction you made is still important, and I wrote the matter incorrectly.

The wavelengths in question are many orders of magnitude larger than the physical dimension of the molecule. The molecules can absorb photons of the appropriate wavelength because the electron clouds have quantum levels which, when mixed, have an electrical dipole (or sometimes other multipole) moment. The dipole in question has a “natural” (quantum) frequency associated with the energy difference between the two levels, such that . If the molecule (or atom) is driven at resonance by a light wave at this frequency, it causes transitions from the lower level to the upper one and vice versa. If there are many molecules in a small volume (less than a wavelength) and the light is sufficiently intense, the response to this driving light is likely to be semi-classical — it will simply oscillate into and out of the excited state at resonance, occasionally fluorescing via spontaneous emission to scatter some of the light sideways. If the molecules are widely separated and/or the light intensity is low, the interaction will involve single (hence distinguishable) photons — the molecules will simply absorb a single photon and remain in the excited state until it either collides with something, absorbs an additional photons to get kicked into a third level, or spontaneously emits the radiation to fall back into the ground state.
In the specific case of CO_2, the spacing between the molecules is large enough and the field weak enough that the photon picture is more appropriate. The resonant photons absorbed carry both energy and momentum, so when the molecule catches the photon it recoils in the direction the photon was travelling, with some of the energy of the photon transformed into kinetic energy. The lifetime of the excited state(s) are usually much longer than the mean time between collisions of the molecule and the surrounding gas (O_2 and N_2 molecules). These collisions irreversibly transfer enough of the kinetic energy of recoil to randomize the quantum state of the CO_2 molecule two ways — by kicking it into a superposition of many states (and losing still more of the original photon energy) and in the process by “broadening” the quantum transitions involved between the participating states (very crude description, phase interruption leading to a spread Fourier transform). These collisions, which occur even when the molecule is in the ground state, make it somewhat easier to accomplish the original resonant capture by allowing the molecule to have a decent probability of absorbing photons in a band of nearby energy/wavelengths, not a single infinitely sharp wavelength. The spontaneous emission process does the same thing, giving each of the lines a “natural width” that can be broadened further by interactions and collisions. Finally, the molecules are all moving relative to the frame of reference where the photons were emitted, and this introduces a small random doppler shift of the photons relative to the molecule, further broadening the lines relative to the light.
Spontaneous emission and collisional/pressure broadening are usually referred to as “homogeneous broadening” and this is the dominant process at higher pressures near the ground. High up in the atmosphere, collisions become comparatively rare and doppler “inhomogeneous” broadening can be dominant. Even this creates a problem for model builders, because the broadening of the molecular lines causes nearby lines to overlap, creating bands of frequencies wherein absorption is very likely within some distance/time known (on an exponential average basis I won’t explain) as the “optical depth” of the gas. The gas is effectively opaque over a few optical depths — nearly 100% of any emissions in the CO_2 absorption bands will be absorbed as the intensity of the light is exponentially attenuated, transferring its energy to the surrounding gas (warming the O_2 and N_2, which otherwise would be transparent to this energy and hence unwarmed by it). The CO_2 molecules themselves remain in quasi (or rather, dynamical and local) thermal equilibrium with the surrounding (open system) gas, and radiate energy away via thermal radiation — but not a pure blackbody spectrum, rather one convoluted with the strongly coupled bands.
At ground level, the air remains “close” to ground temperatures both because of conduction/convection and radiation all three. It radiates energy back towards the ground in the GHG-linked LWIR bands at a “blackbody” temperature characteristic of the ground. You can take a look at figures from Petty reproduced here:
http://wattsupwiththat.com/2011/03/10/visualizing-the-greenhouse-effect-emission-spectra/
to get a feel for the spectrographs at different locations and conditions. Note that on the ground looking up (e.g. figure 8.1 comparing Nauru and Barrow) the CO_2 band and Ozone band are particularly clearly displayed, and the contrast between the two figures illustrates nicely the role that water vapor (a factor at Nauru, not so much at Barrow) plays. The measured back radiation spectra has resonant peaks/bands at the characteristic temperature of the ground, plus a much smaller general background at the temperature of the height (in the case of water vapor at Nauru) where the radiation is emitted. The radiation observed at the top of the atmosphere looking down has a complementary structure. There, radiation in the absorptive bands is observed at the temperature of the top of the troposphere where the atmosphere finally becomes transparent to in-band LWIR, where elsewhere the radiation is “direct” from the ground at ground temperatures.
These figures are as direct evidence as one could possibly ask for of the atmospheric radiative effect, badly misnamed the greenhouse effect. They clearly show (what one can verify at will with simple handheld IR detectors) that there is a substantial radiative energy flux, even at night, downward from the LWIR GHG-coupled bands, and during the day this flux is added to direct solar flux in the visible. Whether one wishes to count this energy as “heat”, or “radiative energy” or anything else one might want to call it, it is part of the energy budget of the dry-land surface and has to be accounted for in the energy flow equations that ultimately determine the dynamic equilibrium temperature of the surface and surface layer of the atmosphere. But ultimately it is the lapse rate as well as this process at all heights from the ground to where the thinning atmosphere becomes transparent in the bands that determines the effective reduction in outgoing flux in the resonant bands.
The lapse rate itself is a complicated thing established by a mix of vertical “slow” convection in both directions and radiative transfer. Because the Earth is unevenly heated and cooled, the atmosphere is highly unstable, nearly always moving from one place to another and one height to another. In the rare cases where it becomes stagnant, or when a warm front overlays cooler air underneath, an inversion can occur where the lapse rate is partially negative over the troposphere. Finally, wet air and dry air behave very differently and establish different lapse rates, and wet air can transfer far more energy via direct transport of latent heat. Convection, especially when laden with moisture, represent an additional “channel” for heat delivered to the surface to leave the surface (cooling it) and be carried aloft, where it can more effectively and rapidly be radiated a way. For all practical purposes, though, only incoming solar radiation warms the Earth (with tiny contributions from other sources of free energy) and only outgoing thermal radiation cools it.
Note well that nobody should care in the slightest what one calls this process, but there is absolutely overwhelming evidence that the process exists almost exactly as I’ve described it above, and because we can see how the process varies with the concentration of at least water vapor (as a GHG) at various heights, humidities, and from surfaces at different temperatures, we can affirm that the atmospheric radiative effect results in a surface that is warmer than it would be, all things being equal, if the atmosphere were composed only of N_2 and O_2 with no H_2O or CO_2 or CH_4 or O_3 or… Furthermore, there is really little doubt that the linearized response in temperature to increased concentrations of these gases in the atmosphere is at least initially positive — more water vapor in a water vapor free atmosphere is going to (comparatively, remember) warm the ground, more CO_2 in a CO_2 free atmosphere is going to warm the ground.
It is less clear what will happen when the gases are saturated to where the atmosphere is optically thick in their absorption bands, in part because some of the bands overlap — note well the overlap between CO_2 and H_2O at Nauru or the TOA looking down tropical Pacific — and in part because nonlinear phenomena and latent heat modify this process substantially by directly modifying the albedo at different heights looking down. That is, in addition to water vapor carrying heat up through much of the troposphere, once there it tends to condense into clouds that have a very high albedo. Albedo reduces ground insolation before absorption and any need to participate in the greenhouse cycle, and does so with an extremely negative derivative, with a much stronger cooling effect than the combined greenhouse effect during the day and as a potential active and passive transport “short circuit” in the greenhouse process at night.
Once again, the atmosphere proves to be highly unstable over oceans, where a truly complex process of absorption, re-radiation, evaporation, and condensation all occur, and all occur nonlinearly and differently as things like surface wind speed, time of day, overhead humidity, overhead cloud cover, height of overhead cloud cover, temperature profile of the atmosphere, temperature profile of the ocean all vary, with clouds forming and disappearing, spreading and raining out, with water vapor losing the latent heat it carries at much warmer temperatures than the “bare” DALR plus GHE computation would have it. Here it is by no means clear that the partial derivative of average temperature in response to changes in saturated CO_2 is strongly positive, or even positive at all. Overall, we know that the atmosphere has powerful negative feedbacks because it does not exhibit runaway warming from water vapor — to put it bluntly, the slope of the partial derivative of the greenhouse effect from water vapor alone manifestly has a zero point, preventing exponential or sub-exponential growth of temperatures. If it were not so, given 70% of the Earth’s surface made up of oceans, we would be Venus.
That, in turn, raises the interesting question concerning the second order terms that are generally neglected in linearized descriptions of the GHE. Changing CO_2 from (say) 300 ppm to 400 ppm in an otherwise perfectly dry atmosphere might well produce some warming. And yes, warming given an ocean might well be expected to produce an increase in humidity. But these assertions are orthogonal, or inconsistent. If you already have humidity that is in balance, such that increases in humidity cause net cooling (necessary to prevent runaway warming, as the ocean is always happy to make things more humid as it warms) and add CO_2 to the atmosphere, it could well be much like adding a strong acid to a buffer (metaphorically speaking, of course) — instead of dramatically decreasing the pH, it finds it cheaper to rearrange the much less expensive equilibrium of the buffer. To re-express this in context, one does get some warming from the CO_2, but only a fraction of what you would expect because the system finds it nonlinearly simpler to shift the H_2O equilibrium, which is already at the zero point and becomes negative with more H_2O to cancel most of the warming until a new mutual equilibrium is established.
This sort of mechanism has been suggested in the literature (Milosovic?) and is actually rather plausible, made likely to be the dominant response to more CO_2 by the simple observation of pre-existing water-vapor equilibrium, which by definition means that the partial derivative of global temperature with respect to water vapor variations must be (on average) zero, and further more must change sign as water vapor varies around the zero to push it back towards equilibrium and not away! However…
This is still only a tiny part of the story. We haven’t discussed the dual role of aerosols — to directly reflect incoming energy and to nucleate clouds, further and nonlinearly increasing the albedo and indeed enabling nonlinear feedback conditions to cause net cloud coverage growth (monkeying with the aforementioned water vapor equilibrium and keeping it from ever being simple), soot and particulates, lateral transport of heat, the effect of the Earth’s tilt, the fact that the tilted Earth is in an eccentric orbit that causes TOA insolation itself to vary by over 90 W/m^2 from perigee to apogee, in a way that is projected onto the slowly precessing axial tilt so that different continents and oceans receive maximum insolation at different times of the year, nonlinearly affecting net heat uptake on a grand scale because of the different mean albedo and albedo feedbacks at the different latitudes and seasons. Oh, and the sun itself varies by a possible significant hair over the solar cycle, which is then superimposed onto this and the possibly coupled decadal oscillations in general patterns of atmospheric circulation, which are (again necessarily) net warming and net cooling changes in order for the Earth to remain in approximate dynamic equilibrium…
Which it doesn’t! And that’s the damnedest thing, isn’t it. In even the very limited high frequency thermometric record, the Earth’s temperature and other “climate” measures vary substantially on a decadal or longer scale, with numerous visible global temperature shifts on the order of 0.2 to 0.5 C over times as little as 10 to 20 years, both with and (key!) without the help of rapidly increasing CO_2. Furthermore, the temperature has proven to be quite capable of not rising along with CO_2 on similar timescales, again suggesting that the global climate system has substantial multivariate buffering capacity that in all probability substantially reduces the expected CO_2-only Beer-Lambert warming. But no simple CO_2-is-the-only-knob system can explain the geological climate record — indeed, models that attempt to do so majorly fail to describe only the Holocene, let alone the Wisconsin, Younger Dryas, or general Pleistocene. The models of CMIP5 fail to describe HADCRUT4 outside of the reference interval singly or collectively over only the last 150 years, as illustrated by figure 9.8a of AR5.
And honestly, that’s no surprise. Nick has stated that he is “pretty sure” that the feedback from water vapor is positive and will roughly double CO_2 only warming. I’m at least equally certain that it is negative. I think my argument is better than his, because he has to explain how CO_2-driven warming will increase the water vapor warming by increasing the water vapor concentration while increases in water vapor at the current dynamical equilibrium temperature decrease the temperature — on average — or else the climate would have a runaway warming catastrophe from positive feedback in the water vapor greenhouse channel. His argument presumes a knowledge and ability to compute all of the nonlinear phenomena associated with the water cycle in the climate in models that are too coarse grained to even represent much of that dynamics. Indeed, it is generally acknowledged in many published papers that we don’t really know how to compute or account for the water cycle — what is used in the models is basically a coarse grained approximate guess, a guess that then has to be empirically balanced against other gain and loss terms as a guess in the reference period and that might not work outside of it.
My argument does not. It relies only on the observation that the climate empirically is locally stable against water vapor fluctuations because if it weren’t we’d be Venus — a positive fluctuation would grow instead of shrink if the average feedback past the set point were not negative, including all of the computable and non-computable effects of everything else.
rgb

@Trick: heat and enthalpy are different. The former is molecular vibration and translation. The latter, a thermodynamic potential, includes heat, latent heat energy and PV for the system.
Latent heat is the energy released or absorbed by a substance when it undergoes a phase transition at constant temperature.

@ur momisugly Matthew R Marler: the water cycle substantially reduces diurnal and annual surface temperature variation. it also keeps warming from well mixed GHGs near zero.