A Little Update for the Webster at the end:
What got me into this Climate Change nonsense to begin with was finding a glaring error in the first two Earth Energy Balance I saw. The authors truly had to force things to close their budgets. Since then I have not been very comfortable accepting any on the "given" assumptions. The original error was missing nearly 20 Wm-2 of energy absorbed in clouds. That error cause the atmospheric window, or open spectrum, to be twice as large as it actually is. Since a large part of the "fat tail" of climate sensitivity is the closing of the atmospheric window, that part is less likely than before.
The second issue I have found is the atmospheric sink temperature. Using the current best estimates of absolute temperatures, the atmospheric sink temperature has to be approximately 184K degrees which is equal to the lowest temperature recorded on Earth. That sink temperature is also required to have a climate sensitivity of 2.6C or more per doubling without considerable feedbacks.
The Meso/Turbo Pause energy is not generally considered in most energy budgets, probably because of its complexity. Atmospheric chemistry with solar ionization tends to drive this region of the atmosphere. Ozone is the most common reference used when talking about the upper 25% of the atmospheric mass, but Hydroxyl, OH* is a major player. Even CO2 can react with free Oxygen atoms to form unstable Carbon Trioxide, CO3 which can create some interesting possibilities. The net result of all this ionic chemistry laboratory is a surprisingly stable temperature region some 87 kilometers to 100 km above the surface even at night. One of the phenomena produced is OH Air Glow.
While this region is reasonably "stable" that doesn't mean what you might think. Horizontal winds can be over 200 kilometers per hour and wind shear over 30m/s-km. Minor shifts can produce enormous impacts. Solar irradiance can vary the temperature at one level by +/- 20 K degrees, but a higher or lower level can remain at the "normal" 184K degrees.
The Stratopause provides a second layer of stability by maintaining a reasonably "stable" 316 Wm-2 (0 C degrees) with less variation, about 3 K per 80 Wm-2 change in annual solar forcing.
One of the more interesting omissions in Climate Science is the actual atmospheric "shells" or near isothermal layers. The chart above focuses on the lower and middle atmosphere. These regions contain nearly 100% of the mass of the atmosphere, so the 100km altitude is normally considered the Top of the Atmosphere (TOA). Comparing the Atmospheric Boundary Layer (ABL) which is a reasonably laminar region near the surface with the Stratopause, another reasonably stable "surface", the obvious relationship ~315Wm-2 which is approximately the effective energy of water at its freezing point, the choice of a fictitious 240Wm-2 "surface" as a reference looks a bit silly with more stable shells available. In between these to shells is the Tropopause at a much lower temperature and energy.
The tropopause varies in altitude from ~8km to nearly 20 km with temperatures ranging from about -50 C to as low as -105 C in extreme deep convection cases near the tropics. Just below the Tropopause is the Troposphere with its Halley Cells and Ferrel Cells that transport energy towards the poles. This advection is the reason that the tropopause is colder while sandwiched between the ABL and Stratopause. With a difference of 185 Wm-2 you can get a rough magnitude of the amount of advected energy.
The Mesopause has an average energy of 65 Wm-2, half of the average tropopause energy. If you consider the Stratopause and Mesopause combined, the total energy is roughly 380 Wm-2. Allowing the differences between the true surface area and the area of the middle atmosphere layers, there would be about 10 to 17 Wm-2 difference, so an effective surface energy of 390 to 397 Wm-2 could be the energy source.
The MesoPause at roughly 87km and the Turbopause at roughly 100km are lower mass regions of the atmosphere. For this reason it is assumed that their combined specific heat capacity is too small to have a major impact on climate. However, the space blanket, a thin reflective plastic sheet, is often used for a Greenhouse Effect analogy. In the Turbopause where turbulent mixing is so small that molecules separate into layers based on mass, Earth has the original space blanket. Like a space blanket, the turbopause is an effect radiant barrier but not much of a general purpose thermal barrier. If it gains much energy, it would tend to advect in the 200 plus kilometer per hour winds. Both the Meso/turbopause and Tropopause have more advective potential impact than the ABL and Stratopause. Since the Meso/Turbopause have a 65Wm-2 radiant impact and the Stratopause has a 315 Wm-2 impact in the middle atmosphere, this region should have its own energy budget.
The Total Solar Irradiance average is approximately 1361 to 1366 Wm-2. Since the middle atmosphere is also spherically shaped and rotates, the normal estimate for effective energy available is TSI/4. That yields the range of 340 to 341 Wm-2 TOA. The middle atmosphere is transparent and has a larger radius than the surface by 50 to 100 kilometers, depending on the shell. This allows some degree of internal reflection and refraction that can effectively increase the surface area of the middle atmosphere relative to the lower atmosphere and surface.
With a combined 380Wm-2 outgoing and 340 Wm-2 incoming, 40 Wm-2 or about 12 percent more effective area should be required to balance the budget. With 12 to 18 degrees of twilight region that is extremely likely.
From space the apparent TOA energy emitted and absorbed is ~240 Wm-2. This is a purely up/down estimate which would depend on the "surface(s)" used to determine the effective TOA. SW anisotropic or any solar energy not in an up/down orientation would have to be huge to miss nearly 150 Wm-2. That is not likely, but with the Mesopause energy primarily down and the Stratopause energy mainly up as viewed from space, the net would be 315-65=250 Wm-2. Missing 10 Wm-2 is much more likely given the difference in the shell areas. That would infer that the majority of the Meso/Turbopause energy gained is not from up/down irradiance but through the longer path length SW advection through the upper and middle atmospheres.
If you can follow this and this is about as simple as I can state it, the majority of the GreenHouse Effect is an illusion. CO2 and other greenhouse gases still have a major impact in the lower atmosphere, but the top half not so much. That should reduce the magnitude of the sensitivity estimates by a factor of two, to ~ 0.8C per doubling or 0.195K/Wm-2. Imagine that?
This is MODTRAN screen shot for 10 ppm CO2 looking up from 6.65km which results in about 65 Wm-2.
Same setting but with 750 ppm CO2, that would be 6 plus doublings from 10. The Iout increased by 18 Wm-2. With 6 plus doublings, the altitude of the 65Wm-2 Iout would increase to 7.65km or 1000 kilometers.