” A GCM can only simulate the teperature trend of the last century if you include CO2.”
No thats completely and utterly wrong. If you hypothesise that CO2 is a warming gas you cannot fit the twentieth century data at all. Since the 30’s was warmer than the nineties, and the 40’s saw CO2 surge upwards to its highest level in the century, therefore its proven that CO2 is a bit player, and probably a net cooler. At current air pressure ranges its almost definitely a net cooler. Absolutely and in the marginal sense as well. Remembering that there is no such thing as a greenhouse gas nor any greenhouse effect, nor even any anomaly for greenhouse to fill.
And why only try and fit the century? Thats leaving 99.9% the other data out of it. There has never once been a CO2-biased model that fit the data. Never. Not one, not ever. Nor will there ever be any such beast. Working backwards in this way for a limited time period (on both rigged-up CO2 levels and rigged-up temperature readings) is not anything beyond a callibration. You need to show predictive power both farwards and backwards after the callibration is made.
I am confident if I had firing powers and running a team of computer modelers (and on the basis that no Gramscian-wankers need apply) that I could put together a model pretty quickly that fitted for long stretches, but only if I could feed in solar wind in real-time, and ocean currents in retrospect.
The two most important factors are solar wind and oceanic currents. The most important human factor is not CO2 but rather actual pollution. That is to say SO2 and particulates. The human race has almost certainly been a cooling influence since the industrial revolution, except on the local level where human activity with roads, concrete and tall buildings causes some local warming.
Solar wind is very hard to predict. Some success with predicting forbush events comes with tracking planetary alignments. Also Dr Marohasy had an excellent article about the effect of the orbit of the moon which ought to have legs. The author may not have had this particular transmission effect in mind. But if the moons orbit impedes the circulation of the ocean conveyor it will act as a delayed COOLER. If the moons orbit then stops impeding in this way, or even actually assists the ocean currents it will act as a delayed warming agent. The transmission of this effect will work through Stefan Boltzmanns law.
If the moon assists upwelling and downwelling likewise it will be expected to have a short-run cooling effect but will add to the heat budget of the oceans to produce a priori warming in the long-run. The moon also warms via more well-known “tidal warming”, which is a feature we see through-out the solar system. For example one of the moons of Jupiter (IO) has virtually constant volcanic activity due to tidal warming from Jupiter itself, as well as from the other moons.