I follow several climate change blogs and two of them recently published some interesting articles on what we knew about global warming back in the early 80's and how the models of that time predicted the changing climate of the past 30 years.
From Climate Denial Crock of the Week, by Peter Sinclair, comes the following video:
In that video we get to see how the scientific predictions from 1982 have been validated. What I found really interesting in that video was the explanation of the "fingerprint approach", which is the method used to distinguish warming caused by carbon dioxide from warming caused by other factors.
On the Real Climate blog, two contributors wrote a post about a paper by James Hansen and co-workers (here's the pdf) that was published in 1981. The real eye-opening part of that post is a graph (click here for a direct link as I'm not sure about permissions for reproducing images) where they superimpose the temperature data of the last 30 years over the model predictions by Hansen et al from 1982. The correlation is astounding! It's not perfect, but I can tell you from experience that in science, such an agreement between two curves is considered a home run. If I could get experimental data that fitted just as well with models, then publishing science papers would be a breeze!
- J. Hansen, D.
Johnson, A. Lacis, S. Lebedeff, P. Lee, D. Rind, and G. Russell,
"Climate Impact of Increasing Atmospheric Carbon Dioxide",
Science, vol. 213, 1981, pp. 957-966.