Global Warming is a measurable (measured) phenomenon. Models of global warming are models. Models that don't "back fit" existing data are not considered sufficiently good and so the models have to be adjusted to fit the data - else they don't get to be models. Famous models that don't "back fit" are things like "God made everything by speaking it into existence 8000 years ago". Has no predictive value, but worse, has no postdictive value. If your model can be run to match the past, it has a much better chance of predicting the future - not a guarantee, but one that doesn't "back fit" has almost no chance. And a model that doesn't exist at all (the model that the anti-global warming crowd unanimously prefer) has no hope of being right - or wrong. That's why they like it. The anti-global warming crowd have NO model that shows their result just as the flat earthers have NO map that works - and they like it that way.
Some models are better than others, obviously, and it's been known for over a hundred years that carbon dioxide traps heat and we now have a CO2 load bigger than anything in the last half million years and it isn't natural. Here is a
graph from NASA, who are admittedly lying about space existing and the earth being a globe but lets put that aside for a minute.
We are coming out of a glacial period, we are now in the Holocene, there is a natural warming cycle occurring, but it is currently not as warm as the Medieval Warm period or the Roman Warm period,
So why not only use the correct model and discard the other 100 models the IPCC tracks? ( they are not the same 101 models being tracked every year, models are constantly being dropped and "refined" because they are so inaccurate)
The models are all curve fitted to historical data, "
Hindcasting". Most of the worlds data sets only start in the 1870's, In Europe in the 1800's, any thing before that is a proxy estimated temperature. Accurate GISS satellite measurements only start in the 1980's
The data set is too short and the understanding of climate variables, and feedback loops is too limited for any predictive use, they are nothing but academic exercises that may have some predictive value after much refinement in the distant future.
The biggest greenhouse gas is water vapor, which has a very complex feed back loop.
The CO2 greenhouse effect is logarithmic, not linear, and CO2 is not uniformly distributed in the atmosphere, the highest concentrations of CO2 is at ground level, increased CO2 levels are beneficial to plant growth. Plants evolved at 6000 ppm, trees stop growing below 200 ppm and all C3 plants die below 140 ppm. One of the causes of the Pleistocene megafauna extinctions is because CO2 level dropped too low ( 180 ppm ) for C3 plant growth
In geological history, CO2 below 400 ppm is at the bottom of its historical range, CO2 has ranged from a low of 180 ppm to 7000 ppm in earth's history