Dr. Judith Curry has an interesting post on Non Equilibrium Thermodynamics. This is right up my alley since I have been trying to explain why the potential warming cause be CO2 is half of estimated. Maximum/minimum Entropy is the controlling range of the atmospheric effect. Perfect insulation, would be minimum entropy and maximum cooling would be maximum entropy. So this should at least get a few people on the same page.
The main reason I am considered a whack job is the conductivity impact of CO2 on the atmosphere. As I have mentioned before, the thermal coefficient of conductivity for CO2 is non linear peaking at - 20 C degrees. The potential impact that has on climate is obvious to me, but no so obvious to others for some reason.
The paper does address conductivity somewhat but only as mass transfer and mixing. CO2 has some impact there, but the main conductive impact is at thermal boundary layers, mainly the ocean/atmosphere but also at the latent/radiant layer as well.
CO2 enhances conductivity at the ocean/atmosphere boundary layer really two ways. First actual conductivity or collision with warmer molecules. CO2 has twice the sensible heat capacity of standard air at -20C and about 5 times at 20C degrees. Also by photon absorption and collisional transfer to other gas molecules.
Of course, CO2 is just another molecule between boundary layers, all though it can transport more heat than the other molecules save water vapor due to its latent heat.
The tougher part to explain is the latent/radiant boundary layer enhancement. Here CO2 can only direct half of its absorbed energy generally down via emission, but all of its absorbed energy via collision just as at the surface boundary layer. Both transfers would heat the surrounding gases increasing convection. Should that heat involve a water phase change, there is the major enhancement. This is where I need better Maximum Local Emissivity Variance information.
With the Non Equilibrium Thermodynamics principals in mind, perhaps some may notice that CO2 does have a significant impact on both maximum and minimum entropy.
I am still at a loss on how explain the 65Wm-2 sink, which I have not found a paper yet that addresses. Should that get resolved, then I may be able to get the modified Kimoto equation somewhat accepted since it explains pretty much all this crap.