A couple of flaws in your back of the envelope calculations, but it doesn't change the major conclusions. > Daily human caloric intake is 2500 calories... 1250

A couple of flaws in your back of the envelope calculations, but it doesn't change the major conclusions. > Daily human caloric intake is 2500 calories... 12500 calories per human daily. 12500 calories is about 50k joules. kilocalories not calories, so 50 MJ not 50 kJ. > 20% efficiency (...) recover 99% of energy that we put into humans back out, that's another factor of 100 right there Not sure how you can recover 99% of the energy while 20% was actually used efficiently, pretty much meaning with non-heat output. And you're not making new energy here, you're just reusing the 80% that was inefficient. If you want to do that, just consider the initial process was efficient. > If we cover the planet in solar panels, why couldn't we recover most of that human-generated heat ? Surely you mean cover the clouds with bottom-looking "solar" panels sensitive to infrared. Anyway, this just changes things by a factor of 100000 or so, down to about 10^14 with some inevitable inefficiency before considering other energy sources, which is 100000 billion. Finally, the current absorption by photosynthesis, that then goes into biomass is only 3000 EJ/year. If we consider 2000kcal per individual (there are about 50% women after all...), a gross estimation of what is currently extracted by the biomass would be enough for 10^12 humans, so 1000 billion.