Vindicarre wrote:
And yet, everyone cites the models when they speak of the coming catastrophe. If "the models" that are used to prop up the proclamations of doom are wrong, then the science used to attain those inputs is "bad".
What else to we have to show us how terrible it's going to be in 10 20 50 100 years but "the models"?
If I'm told that my car is projected to get 50mpg and in reality I get 30, yeah, I'm going to discount their claims.
First, "everyone" is typically political-motivated non-scientists. Typically. You should not hold "science" responsible for people taking results and running amok with them.
Now, understand - I have never created a predictive model that is correct. I'll go out on a limb and say
nobody's created a predictive model that's correct. In fact, that's not even the goal. The goal of modeling is to prepare a set of calculations that explain the present and/or past (i.e. calibration) and use this basis to,
based on a set of assumptions for future inputs, predict the future for
a specific scenario within a certain range of accuracy.
The fact that a model is wrong does not make it bad science. Declaring it is correct makes you a bad scientist.
If you ask a good modeler a question like "do your results forecast death for all?", and they do, you won't get a "yes". You'll get a long explanation of the circumstances under which it appears to indicate high probability of, based on what we know, within this margin of error, with these assumptions, yada yada yada.