I'll repeat, "where are the "sound, repeatable, demonstrable, falsifiable, testable" *experiments* to check that yes, our theory about how CO2 behaves in the atmosphere (or rather "toy atmosphere", for testing), is brilliantly confirmed by observations?" Or, to be perfectly precise, where is the experiment with (1) an insulated tube with various mixes of gases you wish to test (and various lengths), (2) a radiation/light source of variable wavelengths, and (3) measurements of radiation at both ends.
If spectrometry and the radiative forcing equation apply, they accurately predict the experiment, then congrats, you can use those equations. Physics equations aren't magic, and it's dishonest to misrepresent your argument with them without experiment or derivation for confirmation, just as it's dishonest to claim it's settled science, and make ad homiem attacks, when you haven't actually done the fundamental experiment, and it's fraud when you attempt to use intimidation by consensus.
Fortunately, nature has been kind enough to provide us with Venus. The adiabatic lapse rate (convective heat transport in the atmosphere) = dT/dz = -Mg/R*(y-1)/y = ~7.82K/km (I was lazy and used 100% CO2 for this, also y = gamma) which isn't too far off from the ALR calculated from measurements using least squares = ~7.74K/km. Surprise, we have experimental evidence that Venus, a planet with ~20,000X (92atm, ~96% CO2) the concentration of CO2 is quite explainable without resort to vague computer models.
As for newscientist:
Chaos does not disappear by averaging, no matter how many time you repeat it, the averaged N-S equations are just as hard to solve as the instantaneous, and iterating a crude weather model far past the point where it is valid does not magically make climate pop out. The only way you can predict a chaotic system is if it's periodic (or quasi-periodic), and then you have to determine the boundary conditions.
Still waiting on the natural variability of climate calculated from observations, a quantitative reason when weather turns to climate, and (correct) model predictions, say 3 standard deviations (or some equivalent probability) outside natural variability.
All in all a mess of non sequiturs, fuzzy sciencey analogies, and basic lack of understanding of math and physics, as well as the complexities of numerics. You probably should work on learning something about science, instead of just parroting terms and equations if you don't want to look foolish/dishonest. But then again, cargo cult science rarely figures out why planes don't land.
% APL is a natural extension of assembler language programming; ...and is best for educational purposes. -- A. Perlis