Bayesian logic cannot create understanding where none exists. It is best used to discover the most likely resolution of a difference of *expert* opinion. If we do not understand the climate yet, no expert in the sense required for Bayesian analysis yet exists.

Eager to hear your thoughts.

“Shouldn’t you also question the various alternatives to anthropogenic global warming?”

Absolutely, yes. And I don’t consider any alternative hypothesis to have made its case yet, either.

“I don’t know” is not an allowed answer in religion, but it is in science.

Thank you. That’s actually a response with content. I like that.

A few notes in response.

1. This is a metastudy; a form of analysis that collects a large number of not-very-good studies (in the sense of not giving tight confidence bounds or having only marginal significance) and tries to combine them into a single good study. But unless you have a good understanding of the shapes of the error distributions on those source studies, there is a major risk of grossly underestimating the resulting uncertainty. The Bayesian approach requires that the uncertainty at least be known accurately. Furthermore, the individual studies often are based on different assumptions, are subject to different caveats, and these are often incompatible. Metastudies are well-known in medical epidemiology, and don’t have a very good reputation there.

2. There is no discussion of the assumption that the error distributions in the source studies are independent, or zero-mean. There is no discussion of how they were selected, either – there is potential for all sorts of selection bias and “bottom-drawer” effects here. Especially in such a politicised topic.

3. I was a bit surprised to see the ice ages used in this way. As I understand it, the prevailing theory is that it is not changes in the total forcing that causes ice ages, but changes in its global distribution, due to changes in axial tilt and orbit ellipticity. In particular, cooling on the margins of the ice cover can cause it to spread, independently of slightly greater equatorial heating. As changes in CO2 and albedo are caused by the ice age, and are at best feedbacks that amplify it somewhat, trying to determine the proportionality between just the feedback loop and the effect, ignoring the original cause, seems to be getting things backwards somewhat. I haven’t looked at this study in detail, so I can’t say, but I’m dubious.

4. There was also Idso’s similar approach to the question here. I personally think those calculations are too simplistic, but they show that the problem is that fixing them requires modelling the contributions of other effects, and it is far too easy to pick the same unproven assumptions about aerosols, clouds, and ocean warming that lead to GCMs giving the results they do to get such calculations giving the same results. More modern approaches such as Roy Spencer’s analysis of CERES cloud data are also food for thought. I find it curious that all the source studies mentioned are ones that predict strong feedback, even though I know that studies that don’t do exist. There may be fewer of them, and you may choose to give them less confidence for reasons you can set out, but they don’t seem to be even mentioned. Curious, yes?

5. The most obvious sensitivity analysis is the warming seen so far. We have seen 0.8C warming for roughly a 40% CO2 rise. If that first number is accurate (which of course it isn’t), and if that rise was entirely due to CO2 (which not even the IPCC thinks it is) then you get a sensitivity of 1.6C/2xCO2, far short of the GCMs’ estimates. So far as I know, all the proposed mechanisms for explaining the difference are either speculative, or (so far) unmeasurable. So given that you can’t estimate it even from the best and most obvious example because of a multitude of unknowns, like natural internal variability, how are these unknowns eliminated from the alternative and more indirect approaches? Should one not suspect that the same speculative hypotheses have been used to ‘interpret’ the data in coming to the estimates?

6. Looking at the approaches themselves – Wigley’s volcanic cooling estimate uses a climate model (MAGICC), a GCM is used by Annan to model the last glacial maximum, and also Murphy, Knutti, and Piani, the Maunder minimum mentions simulations by Rind and Crowley, and at least one of the sources is another metastudy. Modelling contributes to many of the sources used. Yes, they can only fit observation to the models if they set sensitivity high, which we already knew, but that isn’t exactly a result purely of observation, is it?

In summary, this metastudy is doing more formally what the piles of existing pro-AGW studies do informally, and using a kind of ‘argument by consensus’. Since I doubted the uncertainty estimates on the original studies here (several of which I know of from the IPCC reports), the mathematical combination of them seems to me a case of “the Emperor of China’s nose”, if you know what that means.

However, it is a far more serious challenge to my views, for which I thank you. It is evidence, but very weak and still dependent on modelling. You might be able to make progress by picking one or two of the sources and showing them not to be the result of modelling in detail.

The references therein will take you to the independent estimates, most of which are not reliant on GCM’s.

*Before you get sidetracked, the difference between a denialist and a skeptic is that skeptics, by definition, don’t insist on anything being certain.

As a skeptic shouldn’t also have questioned the flatearth theory? What do you think led to the advancement and new discovery that the Earth is round? Shouldn’t you also question the various alternatives to anthropogenic global warming? The AGW skeptics don’t seem to have formed a consensus around any single alternative theory. What’s your poison? Do you think it’s just natural variability? Solar radiance? Cosmic rays from space? Have you cast your skepticism equally at the AGW skeptic’s theories?

Are you? How do you know?

I’m sure most people who believe in a round world have never seen the calculations, would not be able to reproduce them if asked, and would struggle if anybody asked them to prove it themselves. They take it on faith, just like the people who assumed it was flat.

The difference, though, is that were I to ask an educated man any time after Aristotle, he would have been able to give me the reasons in detail, many of which are directly verifiable with a little travel.

But I think there would be many people who asked, were told only that “Aristotle said so”, and came away unconvinced. (As they should have done.) I do not think that any who had the calculations and observations explained to them would doubt for long. But in an age of Argument from Authority, calculations and observations are not a part of the argument.

And it was because of this reliance on authority that the reality of the four elements – earth, air, fire, and water – was the scientific consensus for a thousand years, attested to by thousands of experts. That, I think, might be a better analogy to CAGW.

We have to accept that there are a group of people who will never be convinced. There will always be those who deny reality. I’m sure all the flatearthers were never fully convinced by by all the observations and calculations. That didn’t keep us from pushing forward though did it? We will overcome the various challenges with regard to wind energy. Of that I have little doubt. If we can develop enough renewable energy to replace fossil fuels in time to prevent catastrophe is anyone’s guess.

I have done. I’ve spent several years doing so. And it’s precisely because I’ve done so that I’m a sceptic. I started off with the default acceptance that any scientist would have, until somebody raised some anomalies that led me to investigate further, and that’s when I found out about the state climate science was in, and concluded that there was insufficient evidence.

You appear to be reasoning backwards – CAGW is true, therefore there must be evidence for it. I prefer to do it the other way round – what is the evidence, and does it imply CAGW? And I have looked. However, evidence so far has proved a will-o-the-wisp. There’s lots of circumstantial evidence for demonstrating various vaguely related claims, but they always leap over one or more steps of logic in reaching the final conclusion.

And many of those I talk to repeat over and over that there is an “overwhelming mass of evidence”, and yet none of them seem to actually know what it is. They either come up with some mangled version of broken logic, or more often, cite a pile of papers and reports that they themselves have not read or understood. And I wonder what it is that has got them so solidly convinced, that no shred of doubt can enter their minds. “The scientific literature contains plenty of evidence.” Sure; until you actually go and look.

I asked earlier about the reasoning behind professional bodies endorsing AGW, and you gave me a link to a report that did so. (Which was a good approach. I applaud that.) It stated (correctly) that the only evidence relied upon to attribute causation is the modelling of GCMs. It stated (correctly) that the reliability of these for long term prediction had been challenged. It didn’t go into the details of their problems, but it’s not hard to do so. I’d personally be very interested to know how they get round the issue. And so what answer did this document give to that? None. It just moved on to the next topic – evidence for continued warming, which of course does not confirm or contradict the GCMs, show causality, or contribute to a solution.

In the end, the entire argument boils down to “We don’t know, there’s no actual empirical evidence, but we’re confident of the conclusion anyway.”

–

As for wind power, yes, there have been cases where people get paid for using it, because it’s so unreliable that you can’t use it on any large scale in an electricity grid. You always have to have backup conventional power available, which means that when the wind starts to blow there’s too much to handle, and you have to give it away. Solar is similarly expensive. They’re not yet practical, and only exist because of corrupt subsidies, by means of which customers are forced to pay higher prices for them via taxation.

Yes, the potential is there, with about another 30 years of research and development. And I look forward to the day when we switch over to solar power naturally – without pressure – because it saves us all money. I’m very much in favour, when it’s ready. Until then, your only large-scale alternative is nuclear. Which is of course, cheap, safe, and clean, but nevertheless politically opposed. (By the very same people, in the main, demanding viable alternatives to oil and coal.)

It seems mad to me. But it’s been going on for more than a century now, so I don’t expect it to stop any time soon. It’s all a grand application of Pascal’s Wager. I think that just about says it all.

What’s wrong with Pascal’s Wager? It’s a simple logical argument. There is more evidence for anthropogenic global warming than there is for belief in God, yet some 95% of the people on the planet believe in a higher power.

“Absolute certainty is not demanded. Just a reasonable level of evidence.”

There is plenty of evidence if you’re willing to accept it. Review the scientific literature more on the subject. Put aside any bias or belief and examine it objectively.

Your main bias from this and other threads is having to give up cheap sources of energy. I’m not sure that’s true. There have been cases where wind energy produced so much they actually paid people to use it. People who put solar panels on their roofs get paid for any excess energy they add to the grid. The potential energy needed is there, without using dirty fuels, we just need open our minds enough to see that it’s possible.