And so finally we have reached the stage where we will explain why the atmospheric insulating effect is inherently a ‘massive’ one and not a ‘radiative’ one. The answer is quite intriguing, maybe even a bit surprising to some, the solution rather subtle in many respects. I have settled for two posts, but could probably have written several, considering the bewildering amount of different aspects in some way or other pertaining to this whole issue.
I hope you can bear with me on what might seem like a rather repetitive style of writing in this first post. I have only done so in a humble attempt to punch through the basic idea presented, which might at first come off as a novel or unfamiliar one to most people.
The second post is more lengthy, gradually winding its way towards the final resolution. When reading it, always bear this first one in mind.
I will most likely at some point publish a (strongly) condensed version of these posts. However, their content and interconnected nature might take time to digest.
OK. Let’s begin …
TO NECESSITATE > TO ENABLE > TO CAUSE
In his ‘Physics Today’ feature article of January 2011, “Infrared radiation and planetary temperature”, Raymond T. Pierrehumbert stated the following about the proposed rGHE surface warming mechanism:
“An atmospheric greenhouse gas enables a planet to radiate at a temperature lower than the ground’s, if there is cold air aloft. It therefore causes the surface temperature in balance with a given amount of absorbed solar radiation to be higher than would be the case if the atmosphere were transparent to IR. Adding more greenhouse gas to the atmosphere makes higher, more tenuous, formerly transparent portions of the atmosphere opaque to IR and thus increases the difference between the ground temperature and the radiating temperature. The result, once the system comes into equilibrium, is surface warming.”
This is a most interesting quote, one that reveals a central misconception lying at the heart of the rGHE and AGW hypotheses. In order to get his message across, Pierrehumbert employs two quite specific terms – “enable” and “cause” – as if they were almost interchangeable. They are not. Read the two highlighted sentences once more. “An atmospheric ‘GHG’ enables a planet to radiate at a temperature lower than the ground’s, if there is cold air aloft. It therefore causes the surface temperature to be higher than would be the case if the atmosphere were transparent to IR.”
How did he get from “enables” to “therefore causes”?
He seems to forget that there’s crucially a third term that needs to be included before this chain is complete and one is able to see the whole picture, and that term is “necessitate”.
Something necessitates an effect, but cannot cause the effect before it is enabled to do so.
I will explain …
Let this be your initial guide:
- An atmosphere’s mass NECESSITATES a rise in a planet’s surface temperature.
- The atmosphere’s radiative properties ENABLE such a rise in surface temperature to take place.
- The atmosphere’s mass then CAUSES the necessary rise in surface temperature.
Figure 1. A highly schematic rendition of what happens when putting a massive atmosphere on top of a solar-heated planetary surface like ours. a) Earth without atmosphere: 240 W/m2 of radiant heat IN to surface from Sun (SW; thin, yellow arrow), 240 W/m2 of radiant heat OUT from surface to space (LW; thick, orange arrow); b) Earth with massive atmosphere: 240 W/m2 of radiant heat IN to surface from Sun (SW; thin, yellow arrow), 240 W/m2 of radiant and non-radiant heat from surface to atmosphere > 240 W/m2 of heat moved non-radiatively THROUGH the massive atmosphere, from surface to top of atmosphere (thick, light-blue arrow), finally 240 W/m2 of radiant heat OUT from top of atmosphere to space (LW; thick, orange arrow). (The thick, horizontal, white arrow does not represent any energy/heat being transferred, just the act of us ‘forcibly inserting’ the massive atmosphere in between the solid surface and space, thus expanding the Earth system, adding an extra reservoir for the storage (and movement) of internal energy.)
The simplest way of expressing the ‘massive’ effect of an atmosphere on a planetary surface is probably like this:
The solid, solar-heated planetary surface goes from disposing its heat directly into space, close to absolute zero, to disposing it rather into a massive atmosphere, a new, added, intermediate energy reservoir that will – by absorbing the heat transferred to it from the surface – naturally grow much warmer than space. Since the surface still has to shed as much heat per unit time as before, only now into a much warmer reservoir than earlier, it will now necessarily have to be considerably warmer itself also, in order to still be able to ‘drive’ the required heat out.
IOW: A solar-heated planetary surface cannot warm space, but it can (and does) warm a massive atmosphere, and in doing so, will – by necessity – be forced to become warmer itself as well.
Figure 1 should depict quite neatly what’s going on. We start with a situation without an atmosphere, Scenario a). The Earth system in this case basically constitutes the surface (+ the near subsurface). There is no ‘climate’ in the normal sense of the word, only direct radiative heating and cooling of the surface (+ conductive heating and cooling of the subsurface). This is essentially the lunar situation. The heat exchange between planet and surroundings (Sun/space) happens directly at the solid surface.
Then we ‘forcibly insert’ a massive atmosphere to lie sandwiched in between the planetary surface and space (Scenario b)), and by that we have all of a sudden extended the Earth system – the effective planet/space boundary is moved from the surface up to the top of the atmosphere (ToA).
Note of course that Figure 1 presents a simplified version of reality. It ignores the slightly complicating matters of 1) atmospheric reflection/scattering and absorption also of the incoming radiation from the Sun, and of 2) some outgoing radiation from the surface escaping directly into space through so-called ‘atmospheric windows’. It is assumed that all the solar radiant heat through the ToA also reaches the surface. In reality, it doesn’t. It is also assumed that all the radiant heat from the solid surface is ultimately transferred to the atmosphere. That would be the case on Venus, not on Earth and Mars.
With this in mind, what is crucial to point out at this stage is the following: In Scenario b) the Earth system’s heat input (Qin, from the Sun) still penetrates all the way down to the surface, but the Earth system’s heat output (Qout) – which in the steady state needs to equal the input – cannot possibly still be delivered in full straight from that same surface and back out to space. The inherent reason for this is however NOT, as you might instinctively think, that the atmosphere intercepts some or all of the outgoing radiation. The solid surface could not dump all its heat directly to space with a massive atmosphere on top, even if this atmosphere were completely transparent to EMR.
And this fact is what ultimately explains why it’s not radiation …
The reason is simple: By placing an atmosphere around a planet, you have effectively wrapped the solar-heated surface of that planet in a “conductive” insulating layer, that is, you have placed it in direct thermal contact with a massive medium. And as soon as you surround a heated object with a massive medium of any kind, there will be ‘massive’ heat transfer mechanisms available to that object, meaning, the object is no longer able to shed all of its heat through radiation alone. There will be non-radiant heat losses eating into the radiant one.
For a solar-heated planetary surface overlain by a gaseous atmosphere subjected to gravity, much or most of the heat loss will be convective, that is, the surface heat will be drawn away from the surface and up into the massive atmosphere by way of bulk air movement. Conduction (and – importantly, on a water planet like ours – evaporation) is what does the actual transfer of heat from the very surface to the air above, but natural convective uplift is what makes this heat loss in any way effective, by pulling the transferred heat swiftly away from the surface air layer, making room for more. I refer you to the previous post discussing the balsa wood wall analogy to see how convection makes the ‘massive’ heat transfer through a fluid medium immensely more effective than pure conduction.
So, what have we? In Figure 1, Scenario b), even as the total heat input to the surface is still radiative, its total heat output could no longer be. And since the heat output to space from the Earth system as a whole still has to be by way of radiation, the (substantial) part of the surface heat drawn into the massive atmosphere (rather than going straight to space) via non-radiative processes will have to be radiated to space rather from the atmosphere.
In other words, the ToA has effectively become the Earth system’s new emitting “surface” to space, replacing the actual, solid surface now at the bottom of the massive atmosphere, no longer at the radiative boundary between Earth and space.
We go through it again:
As soon as you put a massive atmosphere on top of a solar-heated planetary surface, some of the heat from that surface meant for space (its ultimate ‘heat sink’) is automatically drawn rather into the atmosphere (functioning as an intermediate energy reservoir), warming the atmosphere. This heat is not easily returned to the surface (convection vs. conduction), so will, for all intents and purposes, have to be shed to space from the atmosphere.
And this can only happen via radiation.
Which is why a massive atmosphere has to be IR-active. If the atmosphere weren’t IR-active (able to absorb and emit EMR), the necessary transport of heat from the surface to space, back out of the Earth system, would partly stop short inside the atmosphere – the orange arrow from the ToA to space in Figure 1 b) simply wouldn’t be there to release to space the heat continuously transferred non-radiatively from the solid surface to and up into the massive atmosphere (the light blue arrow). The surplus energy (the heat input) would essentially be ‘trapped’ in the atmosphere. Which is an unviable situation.
- The natural non-radiative heat losses from a solar-heated planetary surface to a massive atmosphere lying on top of it NECESSITATE that massive atmosphere to be IR-active.
So in what way, then, does such a massive atmosphere also and at the same time necessitate a rise in the average temperature of the solar-heated planetary surface?
The reason was already provided above, but let me elaborate a bit:
A 3D volume of gaseous atmosphere could always – all in all and in an optimised internal arrangement – emit as much to space as a 2D solid/liquid surface could, at the same mean temperature. The problem is that the atmosphere will never be at the same mean temperature as the surface; it would always be (considerably) cooler on average than the surface beneath, because it is heated (mainly) by the surface, the surface being the atmosphere’s “hot reservoir”, the atmosphere the surface’s “cold reservoir”.
In the schematic of Figure 1, this situation is rendered in a simplified, conceptual way by stipulating the premise that the Earth system’s effective or apparent ’emitting surface’ to space must correspond to a certain blackbody emission temperature. This temperature for Earth happens to be 255K.
The gist is this:
- The atmosphere (in Figure 1 represented by the ToA, the atmosphere’s ideal effective/apparent emitting “surface”) can and does emit as much to space in Scenario b) – at 255K – as the solid surface can and does in Scenario a) – also at 255K. But when the ToA in Scenario b) is at 255K, the solid surface underneath no longer can be. It has to be warmer. Because it has to be able to continue to ‘drive’ out as much total heat as before. It has to effectively get its own heat from itself, the Earth system’s ‘old’ emitting surface, up to its ‘new’ emitting “surface”, the top of the atmosphere, to have it radiated to space from there.
So by putting a massive atmosphere on top of a solar-heated planetary surface, you essentially move Earth’s ‘effective/apparent emitting surface’ at 255K (really determined by the incoming flux from the Sun) from the actual, solid surface and up into the cooler massive atmosphere. (Yes, it’s Pierrehumbert’s ‘raised ERL’ argument, but it is now a ‘massive’ one, no longer a ‘radiative’ one …)
Observe, in reality there is no single air layer in the atmosphere at 255K radiating Earth’s total flux to space. Earth’s final, accumulated 240 W/m2 of radiant heat loss is not tied to one single temperature “surface” anywhere in the Earth system. This is merely a mental construct. But the principle remains: The massive atmosphere is Earth’s new emitting “surface” to space, and it’s cooler than the old one, the actual, solid surface now down below.
We have already described how a massive atmosphere on top of a solar-heated planetary surface must necessarily be IR-active. If it weren’t, the new, extended planetary system could never achieve a steady state, no internal/external dynamic equilibrium (see the Addendum at the end).
However, you could also turn things somewhat around. If you look at it from the opposite perspective, the massive atmosphere needs to be IR-active for the necessitated extra surface warming process to get going at all.
The thing is, this process – upon emplacing the massive atmosphere – towards a new steady state, a new dynamic equilibrium, moving Earth’s ’emitting surface’ from the actual, solid surface to the top of the atmosphere, can’t grab hold and find traction, can’t move forward in a stable manner, unless the massive atmosphere is IR-active. In order for the atmosphere to be able to really ‘supplant’ the actual, solid/liquid surface and become the new emitting “surface” of the Earth system to space, it needs to be able to radiate away its absorbed heat. An emitting surface needs to have the ability to … emit. And with the ability to emit EM radiation naturally comes also the parallel ability to absorb EM radiation. For the old emitting surface (the actual, solid/liquid one at the bottom of the massive atmosphere) to be able to become and stay warmer than the new emitting surface, the massive atmosphere between it and space needs to absorb some (or all) of its emitted radiant heat. The absorption itself doesn’t cause the surface to warm. It enables the surface to warm via other processes.
In other words, the IR-activity of the atmosphere, its ability to both absorb and emit EM radiation, ENABLES the extended Earth system to reach its new steady state, its new dynamic equilibrium, with the atmosphere as its new emitting “surface” to space, and with the old, retreated one naturally warmer than the new, advanced one, a process – a redistribution of internal energy – necessitated simply by the insertion of the massive atmosphere between the ‘heated object’ (the solid/liquid surface) and its ultimate ‘heat sink’ (space).
As soon as the atmosphere’s IR-activity is ‘switched on’, the reworking process can commence. The atmospheric mass forces (CAUSES) the surface temperature to rise (it always starts with the surface temperature, after all, propagating upwards) until it can balance its heat gain from the Sun with its heat loss to its two ‘heat sinks’, the warm air masses of the atmosphere, and the cold vacuum of space. This will happen at the point where the atmosphere/surface in combination – the extended Earth system – finds its balance between its Qin and its Qout. (More on this in Part 2.)
It is in the establishment and maintenance of the stable temperature gradient up through the massive medium making up the atmosphere, the extension and added energy reservoir of the Earth system, to get the heat flow from the ‘heating end’ to the ‘cooling end’ moving at sufficient speed and momentum, that the actual warming of the surface is effectuated. This is a massive process, not a radiative one. As decribed in the previous post.
In all this, radiation is but a necessary tool, a means by which to reach an end, that end being a new steady state for the extended Earth system, with a relatively cool atmosphere as its new emitting “surface” to space and with a relatively warm actual, solid/liquid surface at the bottom, feeding the massive atmosphere above it with heat. Radiation is still only the overall system’s heat INPUT and OUTPUT mechanism, the final output merely the result of how the Earth system is organised in its steady state. The THROUGHPUT of energy/”heat”, the system’s internal transport mechanism – getting the heat from the ‘absorbing (heating) end’ down low to the ’emitting (cooling) end’ up high – is convection, the bulk movement of air. Hence, the physical processes influencing the effectiveness of this convective transport mechanism will be the ones determining the internal steady-state temperatures of the extended Earth system.
The “Raised ERL rGHE model” for explaining ‘extra’ surface warming is actually quite a nifty representation of what’s going on in terms of mechanism>effect. It is only flawed insofar as pre/misconceiving the working mechanism as a radiative one, when in fact it is simply a massive one (see Part 2). True, the radiative properties of a massive atmosphere and the mass itself are easy to mix up, for they very much go hand in hand. But the former is but an enabling tool, the latter the actual muscle force …
The original presupposed misconception at the heart of the rGHE/AGW confusion is ingrained in the fundamental premise behind the whole idea of the radiative mechanism as being somehow a progressive one (rGHE → AGW); that of the initially fully IR-transparent atmosphere:
With an atmosphere devoid of radiatively active constituents (like H2O, CO2, O3, CH4 and N2O), one completely transparent to electromagnetic radiation, there would be nothing to stop the temperature-based (thermal) radiation emitted by the surface from escaping directly into space. Hence, if the global mean surface temperature were at an isothermal 289K, the global mean radiant heat flux from Earth to space would be 398 W/m2. To balance the incoming flux from the Sun of say 240 W/m2 (global albedo kept at 0.3), the global mean surface temperature could rise no higher than 255K (assuming ε=1).
From this, the inescapable conclusion appears to be that it is not the presence of the atmosphere itself, but of the specific radiatively active components of the atmosphere, intercepting (‘trapping’) some portion of the outgoing thermal radiation from the surface on its way to space, that causes our global mean surface temperature to be 289K and not (a maximum) 255K.
The short answer to this, as strange as it might perhaps at first seem, would be:
- First paragraph: YES*
- Second paragraph: NO
(*There might be problems even with this proposition, based on things revealed in the energy budgets of Mars and Venus. I will come back to these ‘discoveries’ in a later post.)
Reason? This premise is simply intrinsically oxymoronic. A nonstarter. People might not readily see exactly why and how, because the hypothetical scenario being described here in a way feels so straightforwardly familiar, and hence so credible, convincing even, normally presented to us – quite matter-of-factly – as a sort of necessary ‘Ground State’ of things: “First we start with an atmosphere like ours, only without the ability to absorb outgoing surface IR. And only then we proceed by introducing the IR absorbers. To see what happens next …”
We hardly ever take our time to stop and think about what is actually being suggested here. We tend to just go along with it. In fact, we swallow it hook, line, and sinker. And from that point on, having accepted this basic premise of theirs, there is no escape. No matter what we say thereafter, they can always come back, simply point to it and say: “Look. Without ‘GHGs’, 255K. With ‘GHGs’, 289K. Argue the matter with gentlemen Planck, Stefan and Boltzmann.”
‘Basic physics; sound footing; settled science.’ Right?
The truth is, their whole premise, the whole thought experiment, is invalid. There can and will be no massive atmosphere without radiative properties. And none exists. For good reason. It would be an intolerable physical situation in a real universe, leading to absurd scenarios of ever-fluctuating and never-equilibrating planetary energy budgets.
- A planet with an EMR-transparent atmosphere absorbs a 300 W/m2 mean radiant heat flux from its sun at its global surface.
- Since this is a) a solid-surface, spherical, rotating planet, b) its sun shines on it at all times, and c) its atmosphere is massive, some conductive>convective heat loss from its surface into the atmosphere is physically unavoidable.
- Let’s say that, on average, 100 W/m2 of surface heat is drawn convectively into the radiatively inactive atmosphere. This leaves 200 W/m2 to be radiated straight into space from the surface. The atmosphere will smooth out some of the spatial and temporal temperature amplitudes, but definitely not completely, so radiatively, the global mean surface temperature in this state would not be able to rise permanently above the ideal BB emission temp of 244K.
- But is the state described above – 300 IN, 100 OUT to the atmosphere, 200 OUT to space – a steady one? No it isn’t. The planet absorbs 300 W/m2 from its sun, but is only able to put 200 W/m2 back out. That’s a significantly positive imbalance between the planet and its surroundings. A thermodynamic system with a continuous 100 W/m2 positive imbalance will warm.
- So what if the surface is then forced to warm up, responding to a naturally and progressively warming atmosphere, to an average global temperature high enough for its mean outgoing radiant flux to space to match the incoming solar flux: 300 W/m2? What then? Would that constitute a steady state? Well, at this point the planet as a whole would appear to balance its incoming heat with its outgoing. Problem is, this is an unsustainable situation, for the simple reasons listed in the second bullet point above. There is simply no way you can ever eliminate convective loss under circumstances like this. It can never be zero on average.
- And so the atmosphere will keep on drawing heat from the surface, at this stage an excess, cooling it back down. And the cycle starts anew. Only, the atmosphere is getting ever more filled up with energy transferred convectively as heat from the surface, inflating it to the point where its mass starts eroding from the top, blown off into space.
This will be a planetary system incapable of reaching a dynamic equilibrium, either internally or externally. Until the atmosphere itself is gone. An absurd, inherently and eternally unstable scenario. The universe won’t have it. Such an atmosphere will have to go.
You simply cannot have a planet with an atmosphere unable to shed its heat. That would defy the whole point of having an atmosphere in the first place. Either you have a planet without an atmosphere altogether (resulting in a relatively cold surface on average), or you have a planet with a massive and (ergo!) radiatively active atmosphere (resulting in a relatively warm surface on average). If there’s a massive atmosphere, it – by definition – contains radiatively active substances, enabling it to absorb and emit electromagnetic radiation. It’s a requisite part of its physical makeup. If not, its purpose is lost on the universe. Why? Because the ability of a massive atmosphere to absorb AND pass on the heat from the underlying surface to space is what makes the atmosphere run, what makes it operate and function as an atmosphere, according to its ‘instructions’, so to speak. It is what would keep the overall surface/atmosphere system stable, what would ensure a planetary steady state, a dynamic equilibrium to be achieved. Since the atmosphere could readily absorb surface heat through non-radiative processes, but not likewise emit it to space, it needs to be radiatively active to find its balance. And so naturally any and every atmosphere that we have observed is …
And this is why the fundamental premise of the ‘atmospheric insulating effect’ should go rather like this:
Without an atmosphere, the planetary surface would achieve a dynamic radiative equilibrium directly with its sun. The only relevant constraining factors to the mean steady-state temperature would be the distance from the sun, the planetary rotation period, the radiative properties (topography, albedo and emissivity) of the actual surface and the thermophysical properties (thermal mass and conductivity) of the bedrock and/or bulk regolith from the surface down. With an absorbed solar flux of say 240 W/m2, the highest possible global average would be 255K. In reality it would be much lower, due to large temporal swings in local/regional temperatures.
With an atmosphere, the planetary surface would no longer itself need – or be able – to achieve a radiative equilibrium directly with its sun. Now it would rather dispose some (in fact, most) of its heat to the atmosphere – its new primary ‘cold reservoir’, warmer than space, but cooler than the surface – and let this pass it on to space in its stead, so to say. This makes a difference, because the extra (intermediate) stage of heat transfer from the old emitting surface (the solid one) to the new (the insulating layer on top of it, the atmosphere) needs to work its way through a massive medium up along a particular temperature gradient, hence it requires a certain ‘driving force’ to get the heat out. (Basically, the balsa wood wall insulating mechanism.)
The gist is that the inherent radiative properties of a massive atmosphere indeed ENABLE the – all radiatively heated, but not all radiatively cooled – planetary surface to reach a steady-state temperature beyond that of a pure solar radiative equilibrium, but they neither NECESSITATE nor CAUSE this warming to actually happen. That’s the atmospheric mass …