And that has been a huge issue with me in this area for decades. If the results can not be reproduced, then it is not science.
I still remember the furor over "Cold Fusion" over 35 years ago, and how many jumped on the bandwagon. Claiming it was the energy of the future, that that was how mankind was going to change everything. End the Energy Crisis, make cheap affordable power for people globally.
Then people actually tried to reproduce the results. And failed. And even more tried to reproduce the results, and also failed. Tens of millions of dollars were wasted to reproduce the results, and all of them failed. And now it is completely debunked, but for some reason there are still people that believe in it, and even that it is real but was covered up by governments, corporations, or multiple other shadow groups.
And the thing is, their research is really not all that hard to reproduce. Simple give them double blind data from a time in the past without telling them when, and see if their forecasts then match what was in the historical record. It's rather simple, but that is something they seem completely unable or unwilling to do.
And as long as they are not able to reproduce their results, I have no choice but to consider their "forecasts" to be as scientifically real as Cold Fusion. Without replication, it's nothing but smoke and mirrors.
And especially as a hell of a lot of scientists that actually work in fields like climate modeling are simply not real.
Check out this article from The Atlantic highlighting Justin Mankin and Alex Gottlieb's research on rising temperatures.
geography.dartmouth.edu
The relationship between CO2 and temperature is more complicated than the polemics suggest.
www.hoover.org