"my complaint of the climatologists modelling of the planet rests at several levels.
- Their models appear to be partial, incomplete if you will, and do not accurately reflect the actual planet-wide system. This is not good news given that weather and sea systems are not only non-linear, they are chaotic too. Any minor omissions from a non-linear model are likely to have a very big impact - e.g. the 'butterfly wings' effect. But the models have very big omissions - and so we might expect some big errors!
- As far as I can see there has been no effective raw material or model sharing or corroboration at the base level of model design, construction and testing. In fact, this has been resisted and there has not been a process of mutually 'looking under the hood'. As far as I am aware, and I could be wrong, nothing meaningful has been published in this regard.
- Most communication between the various teams and groups has been focused on achieving corroboration and agreement in the conclusions. This is a dangerous ethos that can often lead to a collective myopia that produces the results you wanted in the first place.
- The long-term source data comes from ice cores, tree rings and sedimentary layers - all of which are notoriously difficult to interpret. Add to this a very short period (250 years or so) of instrumented, and more accurate data, plus the lack of comprehensive global coverage, and it is clear that the opportunity for error is very high!
- Modelling to date assumes some linear, or linear-ish, relationship between temperature and greenhouse gases. So, reduce the gases and the temperature will also reduce is an obvious conclusion. However, the reality is that the system is non-linear and it could well be bi-stable. That is, it could have flipped into a new state, or at least exhibit hysteresis. In which case greenhouse gas production could continue no matter what - after all the planet generates far more greenhouse gas than we do, and warming runaway could be real."