Nic Lewis has a new post at Climate Audit that deals with some assumptions that went into IPCC AR4’s use of a uniform prior for estimating climate sensitivity.
He has shown it to be faulty, to the point that would normally be cause for a retraction, but this is Climate Science, where being wrong is simply a shade of grey, not black, nor white.
He writes:
Frame and Allen’s original graph (Figure 1) showed that use of a uniform prior in ECS gives a very high 95% upper bound for climate sensitivity, whereas a uniform prior in Feedback strength (the reciprocal of ECS) – which declines with ECS squared – gives a low 95% bound. A uniform prior in the observable variables (AW and EHC) also gives a 95% bound under half that based on a uniform in ECS prior; using a prior that is uniform in transient climate response (TCR) rather than in AW, and is uniform in EHC, gives an almost identical PDF.
Figure 1: reproduction of Fig. (c) from Frame and Allen ‘Observational Constraints and Prior Assumptions on Climate Sensitivity’, 2004 IPCC Workshop on Climate Sensitivity. Vertical bars show 95% bounds.
However, the Frame et al 2005 claim that high sensitivity, high heat uptake cases cannot be ruled out is incorrect: such cases would give rise to excessive ocean warming relative to the observational uncertainty range. It follows that Frame and Allen’s proposal to use a uniform in ECS prior when it is ECS that is being estimated does not in fact answer the question they posed, as to what the study tells one about ECS given no prior knowledge about it. Of course, I am not the first person to point out that Frame and Allen’s proposal to use a uniform-in-ECS prior when estimating ECS makes no sense. James Annan and Julia Hargreaves did so years ago.
…
The noninformative prior used for method 2 is shown in Figure 3. The prior is very highly peaked the in low ECS, low Kv corner, and by an ECS of 5°C is, at mid-range Kv, under one-hundredth of its peak value . What climate scientist using a Subjective Bayesian approach would choose a joint prior for ECS and Kv looking like that, or even include any prior like it if exploring sensitivity to choice of priors? Most climate scientists would claim I had chosen a ridiculous prior that ruled out a priori the possibility of ECS being high. Yet, as I show in my paper, use of this prior produces identical results to those from applying the transformation of variables formula to the PDFs for AW and EHC that were derived in Frame et al 2005, and almost the same results as using the non-Bayesian profile likelihood method.
Figure 3: Noninformative Jeffreys’ prior for inferring ECS and Kv from the (AW, EHC) likelihood. (The fitted EHC distribution is parameterised differently here than in my paper, but the shape of the prior is almost identical.)
…
Whilst my paper was under review, the Frame et al 2005 authors arranged a corrigendum to Frame et al 2005 in GRL in relation to the likelihood function error and the miscalculation of the ocean heat content change. They did not take the opportunity to withdraw what they had originally written about choice of priors, or their claim about not being able to rule out high ECS values based on 20th century observations. My paper[v] is now available in Early Online Release form, here. The final submitted manuscript is available on my own webpage, here.
The complete Climate Audit post is here: http://climateaudit.org/2014/07/30/paper-justifying-ar4s-use-of-a-uniform-prior-for-estimating-climate-sensitivity-shown-to-be-faulty/
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
latecommer2014 says:
July 30, 2014 at 11:57 am (Edit)
And all this proves what?
That's the point, after all, isn't it? Circular logic. You can't calculate A using A.
To me he wrote this piece of obfuscation so that …….. ?
That’s why god made peer review (and like all such creations, made imperfect by us unhairy apes). But its what we have.
1. This is not a null. There is no quantifiable statement that can be falsified
That sword cuts both ways, of course. We have gotten far away from normal scientific method and into the post-normal science realm. I will even go so far as to say that (proper) PNS evaluation is all we have in cases where there are large unknowns, and is not an inappropriate tool.
But climatologists tend to abuse this valuable-but-dangerous tool as badly as today’s Keynesians abuse the theories of Keynes. Comfort-zone hunch becomes “probability” before you can say “four out of five dentists recommend”.
After all that, my PNS hunch puts me in with the “97%”, which exceedingly broad definition, of course, includes the lukewarmers. But Curry, Spencer, Lindzen, N-G, Christy (and many other prominent skeptics) and and even including Anthony all fall into that category. We don’t ask “how”, we ask “how much”.
2. It is not a null that is related to the core issue. to wit, c02 causes warming
It appears to. At least the Arrhenius experiment can be reproduced in the lab. And it does roughly correlate to our admittedly flawed observation.
There does appear to be a net radiation imbalance, in spite of the huge unknowns. We are seeing a “stepladder” progression of flat and warm periods (negative IPO/PDO phase followed by positive). Skeptics point at the flat periods and alarmists point at the warming periods. But one must take the average in order to arrive near the truth.
It is only lukewarming (at ~1C/ century or even less if microsite is what we expect), since 1950, but it is what one would expect from a mild, constant upward pressure from CO2 and other anthropogenic contributions (e.g., Arctic soot).
evanmjones says:
July 31, 2014 at 8:29 am
1. This is not a null. There is no quantifiable statement that can be falsified
That sword cuts both ways, of course. We have gotten far away from normal scientific method and into the post-normal science realm. I will even go so far as to say that (proper) PNS evaluation is all we have in cases where there are large unknowns, and is not an inappropriate tool.
###################
huh, you are not making much sense here.
We have a very simple example; Arrhenius.
in the 1890s he made a quantifiable statement about the effect of C02
namely doubling c02 would INCREASE temperatures, not decrease them.
The Null was quantifiable.
if c02 doubles, temperatures will not change.
Since he made his prediction which has NOTHING to do with ‘unprecedented” c02 has increased and temperature has increased. We can say that over that time frame ( a century) that the null looks pretty bad. Of course that doesnt mean we know with certainty HOW MUCH warming, but we have evidence, good evidence, that a null of NO change in temperature hasnt
done very well.
The point remains that the “thesis” of natural variability as the “cause” is busted.
Its busted for two reasons.
A) natural variability is NOT a cause.. it is the effect. “something” causes the climate to vary
Saying that natural variability explains or causes natural variability is not a falsifiable
statement IN PRINCIPLE and in practice.
B) the “natural variability” “null” is not quantifiable. You cant test it.
We could say that Solar, Volcano, GHGs, Ocean cycles, GCR, land use, all drive the variability in the climate. but saying that “natural variability” explains the climate is a no op.
we could say we understand x% of the climate and y% is unexplained. we could call these
unexplained bits “natural variability” but even that would not put natural variability as an explanatory variability.
As for unprecedented. It was probably warming in the MWP than today. That does not entail that C02 doesnt warm the climate. The most it tells you that warming can be caused by the combination of many things.. the job of figuring that out is tough. But any explanation that rules out ANY role for c02 is wrong. Any explanation that assigns c02 as the one and only cause is wrong.
Seems that we are debating the properties of a quantity (ECS) for which it is not clear even exists – despite the thousands of papers, blog posts/comments and billions of dollars expended. If it was renamed as EBJS (Equilibrium Bell Jar Sensitivity (in the Mosher sense)) then there would not be much of a debate would there? Not many people except specialist in bell jar chemistry would be particularly concerned.
Seems most have accepted the premise of there is such a thing as ECS and it is positive. I don’t think that case has been made conclusively.
The current evidence indicates that the ECS (for the actual climate) is very close to zero and/or not distinguishable from noise and/or natural phenomena variability that actually drives changes in the climate (see – not a climate change denier).
Of course the best argument against the existence of an ECS seems to be an empirical one, i.e., co2 levels have been much higher in the geologic past and yet the temps seem to be non-responsive in a causal way, e.g., no tipping points, no runaway temp rise. Temps always recover despite high(er) co2 concentrations. In addition there is also evidence that co2 is responsive to temps not the other way ’round as makes much more sense physically. Or perhaps any increase (temps and/or co2) sets in motion emergent phenomenon that counteracts any increase, perhaps in some fashion similar to Eschenbach’s “climate governor” speculations.
Even assuming such a thing exists, why do the ECS probability models exclude negative numbers along the ECS axis? If using “uninformed, unbiased” priors – it seems reasonable that we might allow for an increase in co2 (at some level(s)) that would directly or indirectly cause emergent phenomena that would work to decrease temps – there is empirical evidence for this in the geologic record.
It seems that the remarkable fact concerning the climate is its remarkable stability over long periods of time. I seem to recall reading that the temps have varied much less than 1% over the recorded history.
Steve Mosher:
The data do not support the hypothesis of Arrhenius very well. The atmosphere is not the same as a laboratory vessel – it is quite different, in fact.
Bruce A says it well in his post: Climate Sensitivity is an iffy proposition and the data indicate that it is indistinguishable from zero.
“mpainter says:
July 31, 2014 at 2:13 pm
Steve Mosher:
The data do not support the hypothesis of Arrhenius very well. The atmosphere is not the same as a laboratory vessel – it is quite different, in fact.
Yes they do.
Bruce A says it well in his post: Climate Sensitivity is an iffy proposition and the data indicate that it is indistinguishable from zero.
No it doesnt.
Friends:
The “Yes they do”, “No it doesnt”, ” ‘Tis too”, and ” ‘Tisn’t” style of debate is thought to be cogent in primary school playgrounds. But in WUWT threads not so much.
Richard
To Steve Mosh….
The 15uM IR band has a nice hole dug out of it from the *averaged* 255K retransmitted light but it only exists in dry and bright conditions. The scoop goes down to 220K.
So these temperatures are -20C to -50C.
Has anyone managed to heat their house to an average of 13.8C when the central heating only supplies its warming fluid to -35C? Some cry it increases air temperature by latency. That has to be empirically wrong due to the laws of thermodynamics. Any latency in the air will heat the CO2. It would be reversed!
Actually being able to measure this band requires super dry and clean air. Making it purely a theory Earth wide.
The absorption of CO2 is highly non linear as to be logarithmic. The 15uM bucket of water is taken and the horses have drank from it. Now there is nothing left. You cannot drink more from the hole. So no new heating. It’s max’ed out as proven empirically. You don’t argue with reality do you?
Then the other issue with Earth’s heating by CO2 and that is the absorption of heat by the gas includes the greater effect of cooling effect at night times.
Reblogged this on Centinel2012 and commented:
Well we know the numbers they use 1.5 C to 4.5 C with a mean of 3.0 C that they got for the 1979 Charney Report is way off. From what I can determine its under 1.0C