Re: looking for an answer to this question for about 3 years now
Then you are an incompetent searcher or don't know much about basic science.
It took me about 3 hours with Google once to figure out how crappy the assertions of the The Holy Church of Man-Caused Global Warming are. I followed the link to a paper on the model and read some of the basic assumptions. One in particular caught my eye: the one where they assume constant solar output for their model to simplify things. Then I went and looked up measured solar radiative output. First up, the number on the charts I found for measure output was 1-3% above the assumed number for the calculation. That's a really bad start. Then I found measurable variation in the solar output. So they don't even make it out of the starting gate.
Full disclosure: I went looking for it because more than 20 years ago during the semester I was taking Astro 440 (which included calculating radiative outputs for the sun depending on certain assumptions about transfer), or as I called it Applied Mathematics, as my 3rd math course for the semester, I finally realized math and I don't get along well enough for me to have gone into the field. I did manage to understand a fair bit of the radiative transfer theory though. And the bottom line was, beyond some ideas about the nuclear reaction chain, we don't know squat about how the sun really shines. If you can't calculate that, you don't have a baseline from which to work your CO2 theory. If you don't have a baseline, you don't have anything.
You want to know why we keep working to refine the measured distance of an AU, even though we know it to more than 4 decimals in the Celestial Goofy Shit* system (CGS, or centimeter, gram, second for you non-astro types)? It's because when your baseline is long enough, those missing decimal places work out to billions of light years. The same thing applies to repetitive feedback mechanisms.
*Why the name? Because, I mean really?! When you know from the outset that you're going to wind up putting a 'x 10^Y' in your notation, why would you intentionally add more digits to the exponents than you need to?