Chris, you're the one that's being obtuse. Peiser doesn't withdraw his criticism of the 928. He only withdraws his criticism of the 1247 article survey. It's immaterial what Bolt might say, the issue is with Lambert. I'm not trying to defend Bolt, I'm saying Lambert's critique is flawed.

David Latimer, you're making a straw case about doubt in scientific language. For what it's worth, in the 1994 abstracts "characterize" turns up twice (in Abstract 15), and neither have anything to do with doubt, so hoist with your own petard again.

If science thinks something exists it says so. You have Newton's Laws of motion, Newton's Law of Gravity, Boyle's law, Heisenberg's Uncertainty Principle, Planck's constant. I could go on and on. None of these say this "may" be the right answer. Anymore than scientists say there "may" be atoms, muons, quarks, electrons, protons etc. etc.

Thanks for importing Popperian concepts into this argument. The basis of Popper's theory of science is that for something to be science, the proposition must be falsifiable. This of course excludes climate modelling from the realm of science, because you can't falsify a climate model. There's no experiment you can do to test them, you just have to wait to see whether their predictions occur.

Warwick Hughes is running a score card on climate models to date, and assuming that he has done his figures correctly etc. we can say that earlier climate models have been falsified, therefore can't anymore be regarded as scientifically correct. To check his conclusions go to http://www.warwickhughes.com/hoyt/scorecard.htm. Of course, climate modellers will say that they have learnt from their past errors, and that they are getting better results with current models.

cont...

Now, all the scientists in the IPCC framework rely on these models for their forecasts of climate change. The 90% figure isn't some sort of poll of scientists, as you seem to think - it is an opinion expressed by the authors of the summary which has been released. Most of the scientists who have contributed to the 4th review wouldn't be competent to express an opinion - they didn't do the climate models and aren't modellers.

And, applying your argument about joint papers, the only person we should hold responsible for the certainty is the person who wrote that particular sentence! At least that should be the reading if Reveille can't be held responsible for something that Singer wrote in a jointly authored paper.

Who knows how they come to the 90% figure. It can't be scientific, because you can't assign probability to an event which doesn't recur regularly. It's just a guess.

Now, I don't dispute the usefulness of projections, but I do get annoyed when people claim a scientific label for something which is just opinion, especially when those people are scientists. It devalues science, and if they are shown to be incorrect (and I'd be willing to put a bet on the temperature staying below the IPCC projection for say 2020 - I want to be alive to collect) it brings it into disrepute.

Chris O’Neill,

The Tuvalu 2005 report Lambert links to says this about the 5 and 4.3 mm sea level rise figures: “A fundamental goal of the Project is to establish the rate of sea level change. It has been recognised since the beginning that this would require several decades of continuous, high quality data. However, in response to increasing requests from the region for information regarding the trends as they gradually emerge from the background “noise”, combined with concern that less experienced users might attempt to fit a trend line to the data without properly accounting for processes such as seasonality that can bias the result, the preliminary findings are now being provided. We caution against drawing conclusions prematurely, as illustrated in Figure 5 which shows how the trend develops as more data becomes available.”

There is no accurate figure for sea level at Tuvalu. Scientist Lambert had to know this when he posted the figures. Why do you continue to make excuses for him?

Alex the drummer,

So, you reckon Bolt, Blair et al get things wrong out of ignorance. Lambert’s errors, however, are often by design: he aims to mislead. Ignorance can be excused; deception cannot. In Bolt’s latest thread he challenges Lambert’s points on Tuvalu sea level rise and the frequency and strength of hurricanes. Lambert chooses not to discuss these because he knows he’s wrong.

C J Morgan,

Your “pedantic comments posted by denialists” jibe seems to be directed at me so I’ll field it. At no point have I denied the existence of AGW. My goal is simply to point out that Lambert’s entry in Best Blog Posts 2006 is misleading throughout and therefore undeserving of being honoured. Had Lambert taken a less cocky tone and given Bolt credit for his points that are valid and been more careful with his evidence I wouldn’t have bothered. That said, any reasonably knowledgeable layman with a bit of time on his hands could easily dismantle Lambert’s post. Nonsense like Lambert’s reflects rather badly on science and scientists.

Long discussion misses a very grounding point which is a Bolt's presentation of a global warming.

Interesting, the UNDP, as to my understanding, preferred kicking the USA and so-called "first class" world with some assertion of linking a climate change with human activities predominantely rather than simply accepting a reality of natural process, which is the Earth's ageing.

Response to GrahamY:

If you continue to do word counts, then you are going to get nothing but tireless objection from me. It is unacceptable that you examine scientific abstracts via a word count, as already explained. Those of us who are literate can readily see that scientists understate their conclusions when writing abstracts.

Is Karl Popper a disreputable name in your opinion? (For everyone else Popper (1902-99) was a respected philosopher who contributed to the definition of the scientific method) All the theories you list are subject to revision; eg Newton's theories were superseded by Einstein’s theories. And they are all models: http://en.wikipedia.org/wiki/Model_(abstract).

Imagine suggesting that a climate model cannot be falsified! You say "no experiment you can do to test them, you just have to wait to see whether their predictions occur." Obviously testing in all experimentation and modelling depends upon predicted outcomes occurring.

Why should I take Warwick Hughes seriously? Just one of the anti-science crowd.

If you have an incorrect understanding of science and take counsel from those trying to discredit science, then it is no surprise that you appear so confused about the scientific basis of climate change.

You are so off-track about Revelle. Bolt used this joint-paper, as others have, to somehow discredit Al Gore. Fred Singer wrote those words, not Revelle. He died three months after the joint publication. In 1957, Revelle warned that Human beings are now carrying out a large scale geophysical experiment" by releasing greenhouse gases" (http://aip.org/history/climate/Revelle.htm). There is no reason to doubt Gore learnt about C02 increases initially from Revelle and the circumstances of the joint paper are irrelevant.

There is far more to probability than randomisation. Almost all experimentation and meta-analysis involves the mathematics of probability. It is not mere opinion. You deal with opinion and you are entitled to your own, but don't assume scientists think, write and work like a columnist, politician or lobbyist. They do not.

Finally, I point out that climate change is not a bet any person can afford. The wager includes coastlines, diseases, mass extinction, refugees and agricultural decline.

Well in all the preceding, the dialogue between David Latimer and Graham Y at least yields some useful insights into the misunderstanding and misapplication of scientific methods. Their dialogue at least has the potential to arrive at a better understanding of what makes a climate model a scientific model, and not just a scientist's opinion of what they reckon might happen. Graham Y, please take note.

Graham Y's statements about falsifiability shows a confusion between falsification and disproof. Falsification is a priori, disproof is a posteriori. All climate models can be falsified in principle, which was Karl Popper's minimum standard for falsifiability. All it would take to falsify a climate model is to find something in the paleontological record, or the laboratory, or the natural world, or in logic, that disproves one of the underlying assumptions on which a given climate model is based.

The paleontological record, the laboratory and the natural world are all open to experimentation that could yield data which falsifies current climate models. We could also examine the logic of the models for contradictions. So Graham Y, climate models certainly are falsifiable in principle. They're not just a scientist's opinion, they're robust models open to a priori falsification. We don't just have to "wait and see", as you believe.

This sheds further light on Graham Y's apparent self-contradiction where he said that one "can't falsify a climate model" and in the very next paragraph said that "earlier climate models have been falsified". Those earlier models haven't been falsified Graham, they've been disproved, and the distinction is crucial.

Climate models are both falsifiable a priori and open to disproof a posteriori. They meet all the relevant criteria to be valid scientific tools. I don't claim to know whether they are true models or not, but I know a scientifically valid model when I see one. As scientific models, they potentially present a more reliable and valid standard of knowledge than does my opinion, or yours, or Graham Y's, or Tim Lambert's, or David Latimer's.

Hope this clarifies things a bit.