Tuesday
Jul052011
by 
Bishop Hill
Bishop Hill IPCC on climate sensitivity
Jul 5, 2011 Climate: Models Nic Lewis, best known as one of the co-authors of the O'Donnell et al paper on Antarctic temperatures has a must-read post up at Judith Curry's place. The title tells you all you need to know:
The IPCC’s alteration of Forster & Gregory’s model-independent climate sensitivity results.
This is pretty shocking stuff.
Again.
Reader Comments (141)
Nic Lewis has unearthed yet another piece of corrupted science. Why is there so much evidence of hubris in the clisci world, where the same cast of dodgy characters keeps recycling onto the stage in the manner of a Greek chorus. What will be unearthed next?
@Alexander
...Global cooling.
Looks like a solid piece of 'reverse engineering' to help ascertain how a particular chart was obtained. One day, there may be a McIntyre McKitrick Award for such work - if ever sufficient sanity and respect returns to our scientific institutions. That would require their excursion into overt political activism to be subdued, and for more level-headed professors of stronger character to be in leadership positions.
There is informal appeal in Bayesian methodology for those who have a strong opinion of some effect or other, since the stronger that belief, the less impact a few conflicting observations will have on it when the handle is turned on the number-cruncher. This example, though, is of a formal analysis being used to obtain a desired effect. And once again, the idea of scientists as artists, trying to create a work of art that conveys their insight, comes to mind - as it did with the hockey-stick fiasco and the revelations about 'losing the MWP' and 'hiding the decline'.
It's just endlessly fascinating that the IPCC's good little commissars themselves believe their case is so weak that they feel a need to resort to statistical malpractice like this.
Yes, a very good piece of work by Nick. And yes, we need a McIntyre Prize and soon.
Wow. How many more straws can be placed on the back of this IPCC camel?
O/T sorry but Tory MEPs have just helped block EU plans to increase co2 reductions.
The worms are turning
Can I quibble just a little with PaulH from Barcelona:
This assumes that the argument against the IPCC is cumulative, with every straw having essentially the same weight. As propaganda goes that's effective but it doesn't get to the heart of the matter.
Nic's work is not about another little straw. It deals with something that is as central as you can possibly get, within the IPCC's own framework (written with all respect for those who doubt even the framework): climate sensitivity to a doubling of CO2. And it concerns the only estimate of sensitivity in AR4 WG1 that was based solely on observations, not on software models of one form or another. It shows that that carefully argued result, in the original Forster & Gregory paper, was completely distorted by the 'adjustment' made by the editors of WG1 (who, sadly for us and for them, included Forster and Gregory).
With glaciers, and even the hockey stick, we hear again and again that any errors are far from central to the 'real evidence' for dangerous man-made warming. But the treatment of Forster and Gregory cannot be said not to be central. Any errors here don't just count double: they make all the others look like veritable sideshows.
Yup, Richard, this rattles the foundations. There's something shaky about the IPCC and I can feel the quivers all the way over here.
===========
Is just me that is tired of the radiative inbalance at the top of the atmosphere having a direct relationship with surface temperature and giving the 1 or 1.2 degrees direct effect to doubling CO2? The sensitivity studies seem to rely on this. The best summary I can find to the arguments is here http://judithcurry.com/2010/12/11/co2-no-feedback-sensitivity/
I think the argument basically goes the radiative change at the top of the atmosphere leads to a temperature rise at the top of atmosphere/tropopause and you can follow the lapse rate down through the atmosphere to give the same rise at the surface.
The problem I have with this is it is like saying assuming heat doesn't rise this is the answer. But it does rise and convection, lapse rate and height of the tropopause completely linked together, hence why the tropopause is higher in the tropics for example. To me this is like saying if you ignore convection then the basic physics say we should all throw our central heating away as you need to stand very near to the radiator to feel the radiation effect and air is a terrible conductor so all you will do is heat near the radiator and maybe a slight skin of air on the walls which do absorb the radiation that goes through the air.
Convection creates all the weather on Earth so you can't just ignore/parameterise it's effects as it drives any changes you might get if you heat up the air at the surface. Another way to look at this is if you remove convection/wind to stop heat being transferred then that is like standing in a real greenhouse. I'm perfectly willing to concede that that would be much hotter on a sunny day but I don't accept the 1 or 1.2 degree number that everyone is supposed to agree on and outlined in Judiths post as it doesn't make sense to me.
Well, Forster & Gregory haven't replied to it yet so the jury is still out on this one, although it doesn't look good from the initial analysis.
I'm shocked.
http://www.youtube.com/watch?v=SjbPi00k_ME
I don't accept any of the numbers. The earth radiates to space directly from the surface (window) and from high up in the atmosphere in a pattern that is changing throughout the day and seasons and years. As any variable changes to maintain some long term equilibrium, so the whole system readjusts in a way that is too complex to calculate. For instance, if one were to postulate a bit of extra warming (due to some mythical greenhouse effect and mythical gases which "trap" heat"), then the radiation from the surface would increase and convection and advection and evaporation and condensation would all change to carry the extra energy up through the artmosphere to enable it to be radiated away. Trying to represent all this complex temporal and spatial behaviour with a single number (forcing or sensitivity) is pure unadulterated BS (IMHO).
A valid and accepted quibble Richard.
Perhaps it's more likely that the straw will belong to the men that are undoubtably being constructed to misdirect from the impact of this discovery.
Who would have thought that statistics would become such an important battleground?
Philip Bratby
Do you know what happened to Mike or Michael, the guy you tried to encourage to write pieces for various bloggs. Much of his writing is on Jonova,s debate on the missing heat?
@PaulH
'Who would have thought that statistics would become such an important battleground?'
When I first stared reading about climatology, one of the more knuckle-headed thugs who inhabit CiF at the grauniad chastised me soundly for referring to McIntyre. 'What can a statistician possibly know about climate science?' was the edited (and politer) version of his remarks.
Seemingly a great deal more than the average climatologist knows about statistics. And as there is very little (anything??) about it that isn't just torturing data until it confesses, that is a very worrying place for climatology to be.
The trouble with Bayesian uniform priors is that you have to decide what variable they are uniformly distributed over. As shown here this apparently innocuous choice can be hugely significant when the data is not overwhelmingly strong, so that the form of the prior matters.
See the Bertrand paradox for an example of this kind of thing (not strictly a Bayesian prior problem but closely related).
Richard Drake:
Phillip Bratby:
My 'written with all respect' had you in mind first and foremost, PB. But the self-consistency of the IPCC is not small matter, even if you are right. Nor is the gradual enlightenment of those steeped in the current paradigm, like Curry and Lindzen. I don't presume to correct at that level. But respect I can do, as long as some self-consistency aka integrity is present. With this piece of work Nic shows that the editors of AR4 WG1 had none. That's not a small matter.
pesadia: Would that be Michael Hammer? He still contributes at JoNova's.
Richard Drake: I agree entirely with you about the centrality within the IPCC's own framework and the lack of integrity of the authors and editors of AR4 WG1.
Noble cause corruption, rent seeking or mathematical blunder? Forster & Gregory need to be asked.
Additionally, all those who use(d) IPCC climate sensitivity from and in AR4 may want to re-think their conclusions.
Can I suggest this be cleared up before the end of the week before we break-out another trillion.
It's bleedin obvious innit. When the earth gets too hot, lots of people go on their holidays to the beach. And when they get there, if it's windy so you've got lots of convection, they put up a wind break.
The local reduction of wind speed means convection is reduced and to balance heat input with heat output, the sand temperature rises and so does outward going long radiation.
That bit going through the atmospheric IR window goes directly to space, therefore the combination of convective plus radiation being trapped in the atmosphere greenhouse gases falls a bit. That means back radiation falls and less global warming heat is added to the solar energy coming down to the beach.
This is how people on the beach control the Earth's climate. Can I have my Nobel Prize .
Pulleeeze!
Sorry alistair, but you forgot that by the time the sand starts to heat up, the rising sea level cools it and the heat is transferred to the bottom of the ocean where it is a travesty that we can no longer find it.
Phil Bratby.
I think so, this subject sounds to me just the sort that he could get his teeth into. I was very impressed with his work.
Alistair, you've missed a fundamental side effect of extra beach going.
All those children (and Dads) on the pebble beaches of the world are trying to get a six-bounce skimmer, resulting in enormous volumes of rock being thrown into the oceans, resulting in sea level rise!
SimpeSeeker:
I think it is more a case of Nobel Cause Corruption.
Phillip - your post at 4:31 pm is interesting. Clearly, the IPCC's view of the world is that it is a bit like an old-fashioned calorimeter: it has a constant response to an extra input of a given amount of energy (or a fixed change in one aspect of its insulating jacket). Hence it is sensible to ask the question "how would the average surface temperature change if you doubled CO2" - you'd imagine that if you took the world, and ran 100 experiments in which you doubled CO2, the temperature would go up by about the same amount each time (especially if you started from broadly similar conditions). I guess you are saying that the world may be so complicated, with the opportunity for weird and wonderful things to happen to so many of its components if CO2 is doubled, that one might in fact get a huge range of different responses in terms of temperature, even starting from rather similar starting points?
I guess I have mostly been a prisoner of the "forcings" and "sensitivity" language, so I had not really found a way to imagine that "sensitivity" may not even even be a meaningful concept. Whether I've misunderstood you or not, thanks for pushing that thought my way!
One can only hope that Lynas, Monbiot et al (who undoubtedly visit this and other blogs which deal with the science) have enough maths to understand the enormity of the IPPC fiddles and the significance for the whole CAGW "hypothesis".
MSM interest will, as usual, be zero.
This again shows that you can obtain practically any answer you want by applying your preconceptions to the data handling, and by torturing the data to tell you what you want it to say. It is tempting to say that the results should 'speak for themselves', but, they can't, of course. Data is just a bag of numbers and has no meaning outside an interpretative framework.
This sort of analysis is good, though, because it does demonstrate a particular bias and preconception being applied. None of that is meant in a pejorative sense, but it is really important for readers to understand the bias, prejudice and preconceptions being applied to the data. AR4 and its predecessors doesn't go anywhere near pointing out to the 'layman' reader, and hence policy maker, what the doctrines of the faith of the high priests (and priestesses) of climatology are and what are the methods of their priestcraft. Readers would be forgiven for assuming that all analyses were dispassionate and unbiased, when they are not, and never can be.
Especially since The Team was rumbled somewhat in the Climategate emails, we have a much better understanding of their modus operandi - their espousal of Postnormal science, their disparagement of those climate heretics who have traditional ways of working, their desire for witch hunts and professional burning at the stake of such heretics, and their disdain for any who don't use Bayesian methods.
I have no problem with Bayesianism in relation to pure logical or mathematical puzzles, and I don't suppose anyone else has. The problem is that The Team apply Bayesianism to epistemology, and there it has some serious weaknesses as it is based on subjective beliefs and subjective probability. Point out the problems with Bayesianism and a Bayesian will retort that the problem is that one has a certain belief, but based on the evidence one OUGHT to have a different belief. When science gets mired into subjectivism and issues of OUGHT rather than IS then you know that's a place you don't want to be at. Bayesian conditionalization automatically assign probability 1 to 'evidence', but new 'evidence' can never lower the probability of prior 'evidence'. This is fatal when applied to science:
"new evidence could support defeaters for the arguments for earlier evidence, and contrary to Bayesian conditionalization, that should lower the probability of the earlier evidence... And when we substitute a more realistic kind of epistemic probability for subjective probability, it becomes clear that Bayesian conditionalization is simplistic. The general lesson to be learned is that epistemic cognition has a more complex logical structure than countenanced by Bayesian epistemology." [John Pollock]
Bayesian analysis is cocaine to a post-normal / noble cause corrupted scientist. With a sufficiently strong prior, all analyses yield the desired result...
-J
Jeremy at 5:52 PM. I think you may well have nicely captured the agitated activist's view of climate: simple, 'settled science', calorimeter orbiting the Sun kind of stuff. But, but, but - the complexities are many and various, covering huge ranges of time (seconds to at least many hundreds of thousands of years, perhaps millions) and space (cm to thousands of km on the surface of Earth itself, and greatly larger still when orbital variations and environments are included). Whether one can say 'there is warming' or 'there is cooling', depends on which part of these scales one is referring to. For example, we can see strong evidence for overall cooling in this past 6,000 years of the Holocene over much of the Earth. But the warming as the Sun rose over my bit of garden yesterday was clear, dramatic, and very welcome. The troposphere, where most of our climate is developed and experienced, is characterised by a daily transfer of heat from the tropical toward the polar regions. In the tropics there is an excess of incoming over outgoing radiation energy, and in the polar regions it is the other way round. Add to this convective turmoil a spinning, spherical planet with huge variations in surface properties on large space scales, and you do see something of intimidating complexity. So much so that climate studies were an academic backwater in which little more than botanical-style documenting of observations could be expected. Imagine the temptations of those who foresaw the 'CO2 controls climate' hypothesis as a vehicle for a mighty leap in status and funding, thanks to those who had already spotted it as a vehicle for enormous political influence. Away goes the complexity of 'a huge range of different responses', in comes the simplicity of a dominating factor, ambient CO2. Absolutely absurd, and all but completely unsupported by observational evidence since 'business as usual' is a fair description of the weather/ice/sea levels in recent decades.
The analysis by Nick is excellent and the article is very well written. It clearly supports the IPCC motto:
"Never let data get in the way of a good model"
I wonder what BBD thinks about this.
I posted this at CA and would appreciate any feedback the denizens of the Bishop's blog might choose to offer:
"Please correct me if I missed something, but this appears to place IPCC defenders between the proverbial rock and hard place. Either the IPCC looks really, really bad for manipulating a critical study OR they argue that the statistical treatment is simply one of several acceptable options. If they argue the latter, however, they are admitting that the science is not “settled” and, indeed, not even close."
Do y'all (Southern US term for the second person plural. Invented to overcome English's oversight in not differentiating from the second person singular.) agree? If the stats can produce such huge differences because of assumptions considered 'acceptable', then science really doesn't know enough to say anything particularly useful.
Who-dun-it????????????????
Surely that is the question!
John Shade @ 6:36PM
A minor correction. You say "The troposphere, where most of our climate is developed". I would say that as the oceans receive, store and transfer the majority of the incoming solar energy, that is where most of our climate is developed.
Statistics. The gift that appears to keep on giving...
This time to Richard Black at the BBC.
'Global warming lull down to China's coal growth'
http://www.bbc.co.uk/news/science-environment-14002264
'The lull in global warming from 1998 to 2008 was mainly caused by a sharp rise in China's coal use, a study suggests.
The absence of a temperature rise over that decade is often used by "climate sceptics" as grounds for denying the existence of man-made global warming.
But the new study, in Proceedings of the National Academy of Sciences, concludes that smog from the extra coal acted to mask greenhouse warming.
... the new study, which uses statistical models that are very different from the models traditionally used to simulate the Earth's climate, offers an alternative way of explaining the apparent halt.'
Must investigate what 'very different' means...
Interesting twist to the argument from authority. Who's more authoritative: a Cambridge maths graduate or the BBB climate science brigade?
To pose the question is to know the answer.
Appreciate your correction. I think 'unfolds' or 'is realised' would be better than 'is developed and experienced' since I was just thinking of location. Actually, the whole 9 words of that phrase could have been omitted altogether without loss. Sometimes I type faster than I can think. In fact, I think that is probably always the case...
Stan: you'ns, in the Ozarks, works equally well aand sounds even less like a word.
Mark
@PaulH from Barcelona
Roger Black on the BBC explains why the globe isnae warming: 'Global warming lull down to China's coal growth'
These zealots will explain away deviations from their faith ad infinitum.
Is Roger Black's middle name Yesbut?
I know it's a pain, BH, but what you are immensely good at is putting the significance of these things into layman language. I have visited that link, seen a load of acronyms and statistical analysis and switched off.
I'm sure it's very significant, but I would rather that it was explained—even roughly—what it was about; then I could comment on it without having to traverse a whole understanding of the central thesis.
But then, come to think of it, I might just be asking you to dumb it down...
DK
P.S. Any chance of a quick summary though...?
I assume that all the concern over forcing is because it's some sort of overriding effect that will cause irresistible warming. If this is the case, why does it appear to be so affected by supposedly minor effects such as chinese emissions, the odd volcano, contrails, purturbations in solar output etc. etc. Surely if the sensitivity to CO2 was so high, and so all important, there would be no impact from relatively minor fluctuations in these other variables? Unless of course the sensitivity/forcings are not that important.
DK - here's my one sentence stab at simplification:
The IPCC took an empirical observation based study which showed a relatively low climate sensitivity and stretched it via a statistical method (which implicitly assumes a relatively high climate sensitivity range) to shift its results upwards.
Anyone else, feel free to pile in!
So, one country in the world burning some extra coal (which according to the Green zealots is about the worst thing you can burn because it evolves so much 'greenhouse gas') can offset the global temperature rise due to its own extra coal burning CO2 emissions AND the extra CO2 emissions of the rest of the world.
Get this: coal burning is a work of supererogation: not only does it counteract the effect of its own CO2 emissions, but it has surplus negative forcing to counteract the rise in carbon emissions from other sources the world over. Every tonne of coal burned more than offsets its CO2 forcing, so its effective carbon footprint is negative, and it should be attracting huge feed-in tariffs and subsidies. It pumps vast amounts of CO2 into the atmosphere which helps crops grow to feed the world whilst actually cooling the world.
So the answer to global warming and greenhouse gas forcing is simple: burn more coal. If we all convert to coal burning, who knows, the negative forcing will be so great that we will be able to get temperatures down to where they were 60 years ago. Think about it, Richard Black, the logic is inescapable.
The Green zealots can't have their cake and eat it: if one country's coal burning really is suppressing the GLOBAL temperature rise, then coal burning has to be good, doesn't it? Good ol' King Coal!
No doubt they have anticipated this answer and knew that this paper was coming out, which explains the flurry of activity about 'ocean acidification' last month. Coal is off the hook with regard to global warming, but it's the villain of the piece for the next scam. You can just see it can't you? - CO2 dissolves more readily in cold water than warm water, so as coal burning reduces the global temperature rise it will contribute to greater dissolution of CO2 in the oceans. Believe me - that's the next nonsense you'll hear!
Who-dun-it????????????????
Surely that is the question!
Jul 5, 2011 at 8:14 PM | Mac
Exactly. Which IPCC person(s) are responsible for this?
All in all it's just another brick in the wall.
(We don't need no education
We dont need no thought control
No dark sarcasm in the classroom
Teachers leave them kids alone
Hey! Teachers! Leave them kids alone!)
Brent
"Is Roger Black's middle name Yesbut?"
I thought it was "Whiteis"
Alistair
Don't forget the increase in albedo of all those white bodies lying down... (the effect lessens as the season progresses, but is balanced by the arrival of summer in the opposite hemisphere).
Woodentop
That is good enough for me!
Also see parallel thread at Climate Audit for some extra insights by Nic Lewis et al
James P
All those white bodies lying down can't be carbon neutral due to the truck loads of "after sun" that are subsequently required.
The redbodies then do not want to go out in the sun, and prefer to stay in their air conditioned room, powered by CO2 emitting power stations
Those bodies that make it to the brown stage, are balancing out the albedo of those in the white stage
Conclusion? The sun drives climate change. But we knew that anyway!