Regarding the thermodynamic analysis WRT global warming.
I had the same idea and wrote about it here in the context of the CSALT model that I am working on
I guess that's what happens if one reads John's forum and you get ideas. I provided a link to John's post where he discusses [variational principles](http://johncarlosbaez.wordpress.com/2012/01/19/classical-mechanics-versus-thermodynamics-part-1/).
A more thorough evaluation is here:
I don't know if this is a correct evaluation because it is of course impossible to do a controlled experiment on the Earth's atmosphere and ocean. The fact that the ocean immediately takes up likely 90% of the energy imbalance says that this may in fact be second-order. Yet, one can still do some impressive projections based on training the climate on thermodynamic factors that can absorb the excessive heat coming into the system, with CO2 being the ringleader:
This model also has relevance to the El Nino project, since one thermodynamic factor is variations in pressure, which are taken from the ENSO SOI index. The reason that I got started on the El Nino project was because I wanted to see if I could predict the SOI, and therefore make CSALT into a predictive model.
BTW, I didn't include nad's suggestions for the electrochemical potential implicit in photosynthesis, only because there are no known numbers for this dating back far enough. In my opinion, that is pretty much guesswork. But it certainly would have some effect.