Apologies to Blackadder III from whom I stole the title.

Climate Sensitivity. Well, the issue about clmate sensitivity is, just how much is it? Is it a dangerous 6 degrees, or a benign 1 degree. That's the question, is it not? One which has taxed the minds of the team, the IPCC and the sceptics for a couple of decades now. Scarcely a day passes without a new paper or a rebuttal or a Press Release giving a new figure, or at least a refinement of the old. The climate sensitivity has become the battlefield of the clmate wars, the salient feature of the struggle between the warmists and the lukewarmers.

My object here is to pose a question. Is there really a meaningful figure for the expected temperature rise of the Earth for a doubling of CO2 in the atmosphere? Is it even a useful concept in our understanding of the climate? (I can see how useful a concept it is for the science policy folks, and for the sky-is-falling crowd.)

What's the problem? Anybody can see that if an increase in CO2 warms the planet then there must be a way of getting a handle on that in terms of numbers. Ditto for all the other players in the forcing canon. Aerosols, soot, non-radiative effects of H2O, solar and so on. Every one of them has a forcing in degrees per..well.. something or other, and when you sum the forcings you get a result which you can use to predict what is going to happens to temperatures. conversely, if you have a rough handle on what you think are the major forcings but not CO2, you can subtract the known from the measured trend to deduce the climate sensitivity due to CO2 and its feedbacks.

That last is the stuff of much debate right now. Should AR5 acknowledge a lower figure for CO2 based on newer observations and better data on other forcings? Are better statisitical interpretations of the evidence, more enlightened bayesian approaches, giving us a CS figure which is low enough for AGW not to be a concern but possibly even a net benefit? Well, they may be, but what if the concept is nothing more than just that?

But we've measured it, haven't we? We are only arguing about the magnitude, aren't we? Not in my view. What have we done when we have subtracted the known from the measured? Well, what we have left is not the bit we knew we didn't know, it is the sum of all the things we didn't know, whether we knew about them or not. It is expressed in terms of cliimate sensitivity to CO2, but all it really is is a part of a trend which might not BE a trend less a lot of other forcings none of which are definitely known to act on a straightforward non-interacting basis. The whole idea of cumulative forcings is suspect. OK, fine as a concept, but should that concept not need to be shown to be a valid way of looking at things before it is used to inform us?

Here is my null hypothesis. Each and every influence on temperatures acts at a point in time and space. It has immediate effects in terms of temperature and other non-temperature responses. What happens at one point affects what happens at others. There is no basis for assuming any sort of linear response without confirming observations. There is no basis for assuming that any observed linear response will be the same outside the observation space. You can have your forcings, but each will act only at a local level. There is no basis for assuming you can just sum them and get anytihng other than a transient effect on any observed parameter. What they integrate into is a global effect but it is only instantaneous. It has the effect of moving your long-term measure (eg. 'global average temperature') in a very short timescale. There is no basis for assuming any sort of long timescale, much less the century-scale periods we are talking about with a doubling of CO2. What you have may well be a random walk. Going up and down as the influences take it, each day influenced by the one before and serving only to set the start point for the one following. That's what you have, in your time series. Look for trends and you will find them. Apply forcings and you will see what looks like a response. But it is all an illusion.
You can see cycles? Well, of course you can. Humans are programmed to see patterns and persist in doing so when they are not there. But perhaps even the cycles and the correlations with this that or the other variable are no more than illusion themselves. Maybe nothing actually repeats. Except possibly onions.

Is there nobody to defend the idea of a universal CS and the validity of adding forcings arithmetically? We have all been taking the idea seriously for a good decade, but where is the justification? Why debate the value of CS when it may be an entirely hypothetical concept?

Well, I'd say 'Show me' and they would have to come to my battlefield. But if we don't, if we take the position of so many on our side of accepting bad terms and trying to ameliorate them, like that economist bloke (Tim Worstall, just slipped my mind there) who accepts carbon tax and argues about the rate then we have given away most of our position first. Reminds me of Michael Foot's defence policy in the 1983 suicide note, let them invade and fight as a resistance.

Well, I can't play the dead bat if they won't bowl to me. but their entire schtick is NOT to argue with us but to tell us there is no debate or we are not qualified to be in it because we are not circularly-defined climate scientists. No tactic adopted by us has pinned the other side into backing up their assertions, they merely bypass us with their helpful media fellow travellers.

But that doesn't answer the question of the validity of CS. Someone who first mooted it must have justified it? No? Twice in the last week I'v heard people harking back to Arrhenius. Is that really the answer? Has nobody looked at it critically since?

I'm bumping this in case anybody has found a real argument to justify using past CS as a value useful in predicting the future. I've heard nothing yet, but I may be too stubborn to see it.

Can I give this one a bump? We have RB driving us down the deadend street of 97%, we have Kevin Anderson on another thread putting turbine techology into some daft time series whereby the rate of past efficiency improvements is somehow relevannt to future tech and THAT proves something or other about whether windmils will..well, I lost the will to live along with the thread of his argument.

So, while the warmists are here, would they like to read the thread above and put me right? Or the one about best evidence or experimental proof? Nothing but unanswered questions, and if you can't answer the questions of an Oxfordshire housewife how ever do you think you have a handle on GW, AGW or especially CAGW?

I'm going to review all the questions Martin helpfully listed, but on this thread rather than bump all of them. First, positive water vapour feedback. There just was no argument proposed to support it. In fact we all missed the undisputed fact that in the absence of H2O, in dry places, diurnal temperature variations are rather lager than in wet places. Hence the presence of water does not enhance the signal. Pretty much QED, unless someone can propose a mechanism which explains that and gives a positive feedback but magically does not lead to a runaway.

First finding: There is no positive water vapour feedback. There is a high probability that there is a negative feedback.

OK, not so much as an indication that there is a positive WV feedback. The next on the list is the experiment question. Can we postulate an experiment to show the claimed radiative effect is present in the lab? Can we show it happening in the outside world? You know, I didn't think this was a hard question. Anybody who hypothesises that a given phenomenon can lead to a global catastrophe can be asked to show his working, can't he? And to back that up with a demonstration that his theory works in practice, and that it can be seen to be working. It is no good at all saying simultaneously the the effect has far-reaching global effects and that you can't actually show it happening. If every proposed experiment cannot sort the signal from the noise, wouldn't you want to carry on with better and better experiments until you can? What in science gives someone the enitilement to skip that bit and go straight to the answer? Hubris, sheer hubris.

Perhaps I should rephrase the question. How about 'What experimental evidence did you use to come to the conclusion that the whole logical train of CO2 warming is as described in the hypothesis?' And is that the best evidence there could have been?

"We also know the properties of greenhouse gases such as carbon dioxide. Way back in 1859, British scientist John Tyndall showed that carbon dioxide could trap heat. So the physics is actually quite simple. CO2 is actually incredibly effective at trapping heat compared to another other gases in the air so if we increase the amount of carbon dioxide in the atmosphere, then we are trapping more heat in it. This has the effect of warming the earth."

CO2 can trap heat? That is a given, is it? Well, I thought it was not in dispute that CO2 absorbs radiation in some bands and emits it in another. Where is the trapped heat there. If it is emitted towards the surface, what happens to it then? What proportion of the heat is trapped and how much eventually escapes via radiation? Does 400ppm really trap twice as much as 200ppm? When the amount intercepted changes, does the proportion that ends up as heat change? If you think you know the answers to those questions, how? Where is the experimental backup? And of course, if the physics is oh so simple, why no experimental demo?