Sigh. Trendline. Look at that. Of course something as complex as the climate can't be modeled with precise resolution. The models had broad agreement where warming was concerned, where effects from increased greenhouse gases like Co2 are concerned.
That means we're really on to something!
Do you know anything about modeling?
Of course they're not going to agree. They're not the same goddamn model, what the actual hell would you expect? A small difference in the logic on model A vs model B means a potentially large difference in predicted outcome for X.
Differences exist because we're not perfectly sure how all relevant interactions in the physical atmosphere work -- so we try different approaches, different algorithms. Over time we monitor how the models do, which are more accurate -- which bits of physics in which models are more accurate -- and adjust. Toss out the bad bits, recombine good bits, spin up new models.
How do you think weather prediction went? We started with nothing and over the last 30 to 40 years have become able to predict large weather events more than a week out; tornado threats over a limited geographical region a couple days out. And we keep getting better.
The evolution of models.