[Symmetry]

So far the spherical cow model of climate change that Arrhenius used, with better numbers for absorption spectra and so forth, have historically out performed computer models without disagreeing with them by a huge amount. We can hope that the most recent computer models are more accurate but we won't know quickly. You do need to model that CO2, while weak, blocks a spectrum of infrared that isn't blocked by the much more influential water vapor. But that still doesn't require looking at feedback loops, much less cell modeling.

To be honest that uncertainty about exactly how things will turn out is what makes me most concerned about global warming. Better to play Russian roulette with 6 blanks that'll ruin the hearing in one ear than with 4 blanks, one entirely empty chamber, and one actual bullet. IPCC forecasts are bad but they're not unbearably bad. The real danger is that thing'll go far worse even if we can also plausibly hope that they'll be far better.