Guest Post by Willis Eschenbach
I got to thinking about the classical way to measure the very poorly-named “greenhouse effect”, which has nothing to do with greenhouses. To my knowledge, this method of measuring the greenhouse effect was first proposed by Raval and Ramanathan in a 1989 paper yclept “Observational determination of the greenhouse effect“.
Their method, followed up to the present by most everyone including me, is to subtract the upwelling (space-bound) longwave (LW) radiation measured by satellites at the top of the atmosphere (TOA), from the upwelling surface longwave radiation. Or as they describe it in the paper, which was only about the ocean:
” We obtain G by subtracting longwave radiation escaping to space from estimates of the radiation emitted by the ocean surface.”
This measurement is said to represent the amount of upwelling surface radiation absorbed by the atmosphere. This can be expressed either as watts per square meter, or as a percentage or a fraction of the surface emission.
Figure 1 shows this measurement of the all-sky “greenhouse effect” around the world. It shows the amount of energy absorbed by the atmosphere expressed as a fraction of the underlying surface emission.
Figure 1. Atmospheric upwelling longwave (LW) absorption as a fraction of surface longwave emission.
Figure 1a. As in Figure 1. Changes over time of atmospheric longwave (LW) absorption as a fraction of surface longwave emission.
So … what’s not to like?
Today, while pondering a totally different question, I realized that the Ramanathan measurement, while not useless, is also not accurate. There are two issues I see with the measurement.
Other Energy Inputs To The Atmosphere
About 40 W/m2 of upwelling surface longwave goes directly to space. The rest of the ~240 W/m2 of upwelling LW comes from the atmosphere, not the surface.
The first issue with the Ramanathan method is that the atmosphere only gets about two-thirds of its energy flux from absorbed upwelling surface longwave radiation. The other third of its energy flux comes from two totally different sources— 1) solar energy absorbed by the atmosphere, aerosols, and clouds, and 2) latent (evaporative) and sensible (conductive) heat loss from the surface to the atmosphere.
As a result of these other energy fluxes entering and leaving the atmosphere, changes in the top-of-atmosphere (TOA) longwave measured by satellites using the Ramanathan method may merely reflect changes in solar absorption or changes in latent/sensible heat loss. Here’s the total of the other energy going into the atmosphere.
Figure 2. The sum of two other sources of energy fluxes absorbed by the atmosphere.
As you can see, these other sources of atmospheric energy flux vary over time. Part of this additional energy flux is radiated to space, messing with the Ramanathan estimate of the greenhouse effect.
Up Versus Down
The second issue is that the atmosphere radiates in two directions, up and down. However, the ratio between upwelling and downwelling longwave (LW) radiation is not constant. Here is the variation in TOA upwelling longwave due solely to the changing upwelling/downwelling ratio.
Figure 3. Variations in top-of-atmosphere longwave (TOA LW) radiation due solely to the variations in the ratio of atmospheric energy going upwards and downwards.
The variations in these two other energy fluxes, variations that will appear in the amount of energy heading out to space, will cause spurious variations in the Ramanathan greenhouse measurement.
A Better Metric??
Seems like if we considered the TOA LW as a fraction of the total energy entering the atmosphere, rather than as a fraction of upwelling surface LW, it might be more instructive … hang on, never done this … well, dang, this is interesting.
Figure 4. As in Fig. 1a, except comparing the upwelling TOA longwave radiation going to space to total atmospheric energy flux, rather than comparing it just to upwelling surface longwave.
Hmmm … not sure what to say about that. It does seem that the fraction of atmospheric energy flux going out to space hasn’t changed much over the 22-year period of record. And it certainly has not increased by the amount we would expect from the increase in CO2 forcing …
All ideas welcome.
My best wishes to all,
w.
The Usual: Please quote the exact words you are discussing. It avoids a host of misunderstandings.
Iowa Climate Science Education 
