Simon wrote:Because when about half of them are alarmists it biases the sample.
Your definition of alarmist is no doubt somone who projects alarming consequences from the status quo. If half the GCMs do this (as they do) it is not surprising that the scientists running them are alarmists. You argue:
they are alarmists => they are prejudiced => their science must be bad
Alternatively:
Their science is alarming => they publish alarming projections
is simpler.
But either way, your assessment of prejudice (based on judgement) is fallacious. You can detect prejudice by inconsistency, bad arguments, lack of logical support, etc. Not by disagreement with your own prejudice.
You & a few others on this thread attribute all these characteristics to AGW argumnets, I know. When I have investigated the deatils they have not added up. I see prejudice in the simplistic and often (though not always) internally inconsistent anti-AGW arguments.
The case that AGW is reading trends into data that has so much internal variability this is unwarranted is much more difficult to shoot down. I am not inclined to shoot it down. It must remain a hypothesis for any scientist working in the field.
Were I such it is likely that after having reviewed properly all the evidence, following the consensus, I would give that hypothesis a low probability. But how I rated it could not be determined by the simplistic arguments (pro or anti) that get trotted out by those who do not have the inclination or time to read the source material carefully and with an open mind.
The case against the consensus can only be made if you agree with TallDave:
This is known as argument by authority. It is a logical fallacy to rely on authority in an area with so much uncertainty. Expert predictions have poor track records under such circumstances (they actually do worse than non-experts).
This argument was I believe advanced by Armstrong & Green. Armstrong is well-known as an academic studying economic forecasting. His views are therefore understandable. Economic arguments tend to involve assumptions about the behaviour of human agents which are clearly (in the case of classical economics) wrong. And there is no accepted theory of how to include human complexity into economics that allows prediction.
If these statements are applied to any forecasting based on physical systems they will underestimate the skill of proper quantitative techniques. Of course, you can posit physical questions which cannot be answered because chaos predominates. Simon argues (on no evidence) that global temperature evolution is such a question. The "consensus" argues that climate is predictable:
At very short timescales - well (accurate local weather forecasts)
At short timescales - not so well (regional long-term forecasts)
At medium timescales - not at all
At long timescales - only when averaged over time and space.
The GCMs model a parameter - climate sensitivity - that has the max possible averaging, all the globe, and a timescale of 10s of years.
Given the known underlying physical basis for this it requires some creativity to imagine that when spatial and temporal variations are averaged out it would not be predictable.
Simon has this creativity. He supposes that some parameter of the climate behaves chaotically with strange attractors having a periodicty of more than 150 years. Random dynamic evolution from one strange attractor to another can then result in the observed 20th Century temperature trend.
While this is theoretically possible, it stretches credibility. What is the set of dynamic parameters that has such a long time constant? What is the chance of this random process just happening to duplicate 20th C CO2 forcing?
So I can't dismiss, this, it just seems less likely than that CO2 forcing causes this increase.
All this is said from a non-expert viewpoint - since TallDave believes non-experts do better than experts. Sounds to me rather like the school boards in Kansas who tried to prevent teaching of the "consensus" view that species evolved through natural selection. They no doubt had also read Scott Armstrong.