Replies

Both papers are interesting and I'll have to read them more. But the first one admits their analysis was for clear skies only. The second tries to estimate the effect of clouds as a single parameter. Both are radiative models, not atmospheric simulations. While a simulation of the atmosphere including smaller scale weather features could provide a single parameter for the effect of clouds I think it would be faulty because climate is not determined by radiative energy with a tweak for clouds, but by both radiation and real weather conditions. Perhaps a dozen or two dozen parameters might do it, but certainly not just one.

It is not a fair statement to say "since we can't predict the weather, we can't predict climate". I wish people wouldn't say it so you would stop jumping on it. The fact is, we predict the weather quite well although timing and placement can be off particularly the timing of vortex intensification and decay. But as you have pointed out many times, we don't need to model the weather accurately in any area to use weather models for a broader purpose such as climate prediction. It simply doesn't matter if the timing and placements are off, only that they are modeled.

Unfortunately, smaller scale features are not modeled. Convective systems are not modeled and their effects of throwing uneven moisture into the top of the troposphere is not at all adequately described by a single parameter as they have done in the first paper. The question that I have not seen the answer for so far (but I'll read some more) is what effect these smaller scale phenomena have on climate and more importantly, what effect the hypothesized climate (i.e. increased water vapor) has on weather.