[byerley]

Actually, it does in this case. People seem to forget that in climate science (like cosmology, ect.), you can't use the normal scientific method because you can't really do meaningful experiments, just observations.

What you do instead, is look at available information, intuit an explanation, then systematically compare the implications to known results. If all goes well, you might then do a simulation or predict other effects that can be observed. However, the process is very dissimilar to other fields of science.

[thanatropism]

Sounds like climate scientists need to read a lot into the epistemology of economics. (It may well be that "economic science" is an oxymoron, but the work that has been done to try and make it one...)

https://en.wikipedia.org/wiki/Natural_experiment

https://en.wikipedia.org/wiki/Design_of_quasi-experiments

https://en.wikipedia.org/wiki/Regression_discontinuity

https://en.wikipedia.org/wiki/Instrumental_variable

etc. etc.

[JamesBarney]

One problem with all of these approaches is they require more then one subject. Unfortunately we only have one earth, so there is no way to have a natural control without CO2 rising, or an earth where the CO2 rose at a different time.