OK, now on to the “X 2” in Gavin’s post,
realclimate.org/index.php/archives/2007/04/learning-from-a-simple-model/
The model assumes a single layer of greenhouse gases above the surface that allow visible radiation in from the sun but traps a certain proportion of the outgoing longwave radiation from the earth.
Now your own link says that the “radiation is isotropic, meaning the intensity is equal in all directions”. This is true. When a greenhouse gas molecule intercepts a photon, it doesn’t “know” which direction it should re-emit it, so it re-emits it in a random direction.
Half of those directions are towards the surface, and
half are in the direction of space. (Actually, very slightly less than half are reflected back to towards the surface, because the surface is round, not flat, but that’s immaterial – i.e. the difference is too small to matter – so we’ll ignore it from now on.)
If a photon is emitted at an angle rather than straight up or straight down, does it matter? No – either it will go far enough that it either intercepts the surface
or escapes into space, or it will hit another greenhouse gas molecule and be re-radiated in a different direction which, again, will either eventually hit the surface, escape into space, or hit another molecule that re-radiates it in a different direction, and so on.
Eventually
all photons end up either hitting the surface or escaping into space, and the
proportion of photons that go each way is 50:50.
Therefore we can simply say that the amount of radiation going back to the surface is R, say, and the amount of radiation escaping into space is
also R. They
must be the same because the CO2 molecules re-radiate the photons in
any direction with equal probability and half of those directions are back towards the surface.
So that’s why Gavin has “lambda x A” radiation going in both directions from the greenhouse gas layer, which is equal to my R. He could have called them “half-lambda x A” if he wanted to and avoided the factor 2 that appears in the upcoming calculations, but he wanted to define a lambda that would have a useful physical meaning (emissivity).
Now let’s look at the three equations he derives from the diagram. (We’re assuming radiative balance, i.e., after the system has reached equilibrium.)
Surface: S + lambda x A = G
This means that the radiation that is absorbed by the surface coming from the sun (S) plus the radiation that is absorbed by the surface coming from the greenhouse gas layer above (lambda x A) must be the same as the radiation leaving the surface (G), since we’re at equilibrium.
Atmosphere: lambda x G = 2 x lambda x A
This means that the total energy absorbed by the greenhouse gases in the atmosphere (lambda x G) must be equal to the total energy re-emitted by the atmosphere (half of which eventually goes up, half of which eventually goes down).
Planet: S = lambda x A + (1 - lambda) x G
This means that the energy being absorbed by the planet as a whole (S) must be the same as the energy being
released by the planet into space (lambda x A = the half of the radiation emitted by the atmosphere that managed to escape into space and (1 - lambda) x G is the proportion of the radiation from the ground, G, that managed to slip straight through the atmosphere without being absorbed.)
Now we can do some basic algebra. First, take the second equation and rewrite it in terms of A:
lambda x G = 2 x lambda x A
therefore
lambda x A = lambda x G / 2
Now substitute that into the third equation:
S = lambda x A + (1 - lambda) x G
= lambda x G / 2 + (1 - lambda) x G
= lambda x G / 2 + G - lambda x G
= (1 - lambda/2) x G
Re-arrange for G:
G = S / (1 - lambda/2)
G is
also equal to sigma x Ts^4 from the Stefan-Boltzmann equation, so now we can calculate the expected surface temperature Ts if we know S and lambda, both of which can be directly measured.
Now this is
not the model used to estimate the
actual temperature of any body – you have to make it more complex by taking into account conduction of heat from the surface on the side of the sun through the rest of the body as well as how quickly it is rotating, and you also need to use many “layers” rather than just a single layer for the greenhouse gases, and incorporate convection and clouds. It is not
meant to be used for that. It’s exactly what Gavin said it was: a simple model to help explain the basics of the greenhouse effect. To claim that
Schmidt wrote that he and his colleagues took the Stefan-Boltzmann blackbody numbers and multiplied them by an additional factor of two to devise NASA’s official Earth energy budget.
means the author of that website is either doesn’t understand what Gavin wrote (in which case you shouldn’t pay him much attention) or he’s deliberately misleading people about what Gaven wrote (in which case you shouldn’t pay him much attention). I’ve shown you every step of the working above, and as you can see,
it doesn’t even multiply the “Stefan-Boltzmann blackbody numbers” by “an additional factor of two”! That sentence doesn’t even make sense! What is multiplied by two is lambda, the emissivity, and the factor of two comes in
precisely because the CO2 molecules re-radiate in all different directions with equal probability which he himself mentioned on the very same page!