Wharfie, Can you come up with a model that will be this close to observations in 35 years? I can lead a project that uses an enormous amount of compute power that models every molecule in the atmosphere. I can run that model for weeks, if not months at a time, and come up with a set of results that may or may not hit expectations.
If they do hit expectations, great, I'll publish the data and brag about the amount of compute power it took to arrive at those conclusions.
If they don't hit expectations, I'll take a look at the variables and see how I can tweak them in order to arrive at the desired conclusion. Then I'll repeat the process until I get the desired results.
Remember, there is absolutely no way to verify that my results are accurate, especially if they're meant to predict the future 35 years out. By that time, it won't matter whether I was right or wrong, because I will have already secured even more funding to continue my research.
Why am I so confident that I could do this? Because my whole job is to run models. However, in my line of work, the accuracy of my simulations can be verified in less than a year. If they aren't accurate, I'm out of a job, period.
Not so with climate models, where the accuracy can never be measured until it's too late for it to affect my career.
Tenchusatsu |
