We have been expecting too much from the IPCC about its confidence level increase: the explanation may actually be simple… and surprising.
Guest essay by Stephane Rogeau
Image: From IPCC FAQs
Many people are wondering what actually made the IPCC raise its confidence level about the fact “human influence has been the dominant cause of the observed warming since the mid-20th century” from 90% to 95%, since its last report in 2007.
Most of the time, it is argued that there has been no warming since 2007 whatsoever, which makes the increase of confidence level in the IPCC’s statement very dubious. But it may be the other way around: because there has been no warming, the IPCC raised its confidence level! And it actually makes sense… at least inside the thought paradigm of the UN organization.
The reason may actually be simple, for one reason: the theoretical warming due to anthropogenic CO2 is, for the IPCC, a given.
It is what makes the climate models “work” (i.e. match more or less historical records). It is actually the basis of the IPCC’s line of reasoning: we cannot find any other way to match our models with the data than by entering feedback assumptions that give climate sensitivity to CO2 the value x… therefore its value has to be around x.
Based on this given assumption that cannot be disproved by facts anymore, and knowing the quantity of CO2 released by human activity, one can easily calculate the theoretical human-induced contribution to global warming since 1951 (let’s call it HIC). Discrepancy with observed global warming (OGW) is, of course, due to natural variability. Therefore, the proportion of human influence in observed warming between 1951 and year “n” is simply p(n) = HIC(n) / OGW(n). If, in year n, the theoretical human-induced contribution since 1951 is for example 0.4°C, and the observed global warming is 0.5°C, then the calculated proportion of human influence is 80%.
Obviously, as we release more and more CO2 into the atmosphere, the human-induced contribution is an increasing function of time: HIC(2012) >HIC(2007). On the other hand, the so-called “hiatus” means OGW(2012) = OGW(2007), as no warming has been observed since 2007. Then it’s just basic arithmetic: p(2012)>p(2007).
Long story short: the proportion of human influence in observed global warming has increased since the last IPCC report because temperatures have leveled off. Translated in terms of confidence level: if the IPCC was 90% certain that human activity was responsible for more than half of the observed warming in 2007, it is not surprising that the confidence level for this same proportion has now risen to 95%.
To conclude: the less warming, the more confident the IPCC about its claims to policy-makers.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Jquip says:
November 21, 2013 at 9:22 am
Tim: “If you play around with these numbers on a spreadsheet so that the probability of a pause in the AGW scenario was only 30%, then the decrease in our belief is 6%.”
Not that I don’t understand your point. But it’s worth noting that experimental results are not contingent on belief
=======================================================================
Do you believe that is true about IPCC Climate Science? I thought that the confidence level should be empirically determined, but I don’t see how the IPCC could derive this without data, unless climate models are data. I even wandered off into Bayseian statistics to see if that could be the source. Maybe, but I bet the increase in confidence is a gut feeling wrapped up in mathematical hocus pocus.
The worse their computer models perform the more confident they are. What if temperatures decline up to AR6? What then? 99.99% confidence. Only in Climastrology.
@Tim: your calculation makes sense… but you have to change one thing: for the IPCC, p(AGW)=1
The % of confidence, for the IPCC, is not about reality of AGW: it is about whether it is responsible for more or less than half of observed warming. Their answer used to be 90%, it is now 95%.
Bob Greene: “Maybe, but I bet the increase in confidence is a gut feeling wrapped up in mathematical hocus pocus.”
Nah, sophistry works too. Just redefine your hypothesis. eg. If it used to be ‘The planet will warm lots, because people.’ then you need only restate it as ‘The planet will warm less than lots, because people.’ The only trick is to pretend it’s the same hypothesis. This can be done explicitly, or via differing bounds and estimates in the models; which are a concrete implementation of a hypothesis.
You can take it as a confession that the previous hypotheses in play were overstated; a case of hot air from the researchers.
Stephane:
What is the event for which the IPCC’s probability is 95% and by what algorithm do they get the 95% figure?
@terry Oldberg: I’m not sure there is a right answer, and I guess even IPCC members would have difficulties to really clarify that point.
The 95% we’re talking about is more a confidence level than a probability, about the fact human-induced contribution to warming is responsible for more than half of the observed warming. The figure is high, as the IPCC thinks that “the best estimate of the human-induced contribution to warming is similar to the observed warming over this period”, which is, of course, much higher than half of it!
But how this 95% is actually calculated is one of the biggest mysteries of science 🙂
Bloke down the pub says:
November 21, 2013 at 2:39 am
Or to put it another way, the threat of agw doing something nasty to us is equal to the warming multiplied by their confidence that it’s all our fault. To maintain the threat level, and therefore their income, as the amount of warming goes down so the IPCC’s confidence has to go up. What happens after it exceeds 100% is as yet undetermined.
Well, obviously you can’t exceed 100 percent confidence. The problem is that the closer the IPCC’s ‘belief’ moves towards the figure of 100 percent, the more it becomes resigned to the realms of superstition ( I use that word to reflect an SkS post). Climate change and weather events are very diverse and random events over which we have no control. For the IPCC to suggest such high confidence levels is in the realms of religion. “I believe in this – you cannot prove otherwise, therefore it exists”. A bit like aliens really
Wait till the IPCC discovers Bayesian statistical methods! Then we’ll be hearing about probabilities from them and guys like Anthony will start criticizing the IPCC’s priors.
@Jquip
>Just redefine your hypothesis. eg. If it used to be ‘The planet will warm lots, because people.’ then you need only restate it as ‘The planet will warm less than lots, because people.’ The only trick is to pretend it’s the same hypothesis. This can be done explicitly, or via differing bounds and estimates in the models; which are a concrete implementation of a hypothesis.
Nail, meet hammer. Once you slip in the small change in terminology, point loudly at the change in the confidence number. Here is a good title for an article on the words and methods used and abused:
“AR5 – The Confidence Game”.
Tad says:
November 21, 2013 at 12:27 pm
Wait till the IPCC discovers Bayesian statistical methods! Then we’ll be hearing about probabilities from them and guys like Anthony will start criticizing the IPCC’s priors convictions.
FIFY
MikeB says:
November 21, 2013 at 2:54am
“You do realise, I hope, that when it is day on one side of the planet it is night on the other. Even if this wasn’t the case, the amount of radiation that the Earth receives and loses is always in approximate balance (The Radiation Balance)…”
——————–
Now I wasa thinking that iffen there was such a thing as the earth having a Radiation Balance ….. then we would not be engaged in all of these long winded discussions about climate and CAGW, …… would we?
Jquip says:
November 21, 2013 at 5:56 am
“Remember that if we’re not in a quantum radiative mode, then we’re colliding molecules. Or, essentially, atmospheric gasses act as if they’re in a pinball machine. The surface imparts energy to them by kicking them back up the table, where they go collide with anything and everything.”
———————-
Right you are, Jquip, the observed “action” of an ole time pinball machine is probably the best example to explain the transfer of thermal energy in earth’s atmosphere to those persons who don’t quite understand it.
Assign different “atmospheric gas” names to the 5 steel balls and “shoot” all 5 of them out at the same time …………….. and observe the utter randomness of energy transfer when collisions occur.
HA, …. then ask if they can measure or calculate how much “energy” each of the steel balls absorbed and emitted before it disappeared into the “return slot”.
Stephane says:
November 21, 2013 at 7:15 am
Let’s not mix confidence level with proportion of human-induced contribution. With this IPCC logic, if the climate cools:
– the proportion of human-induced contribution may exceed 100% (actually, it is almost the case, according to the IPCC, as they wrote that “the best estimate of the human-induced contribution to warming is similar to the observed warming over this period”)
– the confidence level about the fact the proportion of human-induced warming is more than half of observed warming will approach or reach 100% (but will of course not exceed it).
The question is if the CO2 effect is detectable:
http://claesjohnson.blogspot.se/2012/03/can-greenhouse-effect-be-detected.html
Assuming CO2 increase it is x% of the warming is also an assumption not proven yet.
As Claes says: “The discussion gets complicated by the fact that “the greenhouse effect” is not clearly described in the literature, but it is somehow connected to the radiative properties of the atmosphere:”
See further post at Claes site:
http://claesjohnson.blogspot.se/search/label/OLR
“When climate skeptics state that for sure they understand very well that there is a CO2 greenhouse effect, as any knowledgable scientist must do, they refer to the radiative forcing of 3.7 W/m2 with 1 C warming from doubled concentration, but then forget that this effect has very weak scientific support.”
As we have seen this effect is also called into discussion:
http://joannenova.com.au/2013/05/major-30-reduction-in-modelers-estimates-of-climate-sensitivity-skeptics-were-right/
“The hallowed forcing due to a doubling of CO2 was 3.7Wm^-2 is being lowered to 3.44Wm-2.”
What caused the rise in confidence?
“Spin”
In public relations, spin is a form of propaganda, achieved through providing an interpretation of an event or campaign to persuade public opinion in favor or against a certain organization or public figure. While traditional public relations may also rely on creative presentation of the facts, “spin” often implies disingenuous, deceptive and/or highly manipulative tactics.[1]
Politicians are often accused by their opponents of claiming to be honest and seek the truth while using spin tactics to manipulate public opinion. Because of the frequent association between spin and press conferences (especially government press conferences), the room in which these take place is sometimes described as a spin room. A group of people who develop spin may be referred to as “spin doctors” who engage in “spin doctoring” for the person or group that hired them.[2]
http://en.wikipedia.org/wiki/Spin_(public_relations)
Expected in an organisation where politics are the main aim and all the facts are going against them
What is important is not what they said in the Science Report but what was put in the Summary for Policymakers (SPM) and then which part of that got media attention.
The whole game was and is to get headlines with carefully worded phrases and selected, usually exaggerated data. You only have blue boots on the 1995 person, but provide the phrase about “discernible human influence.” In fact the original portion of the SPM written by the authors of the chapter said there was no human evidence discernible.
It was changed by the lead author of the Chapter (8) Benjamin Santer, (a CRU) graduate, without his colleague’s approval, to include the phrase human influence. This phrase became the headline and the complete summary for the media and the public of the 1995 Report. I explain the entire corruptive process here;
http://drtimball.com/2011/early-signs-of-cruipcc-corruption-and-cover-up/
It was the first major manipulation of results for the political agenda and alone should give the 1995 Report a totally blue person.
Very clever explanation, Stephane. I enjoyed it greatly and will chuckle about it for some time. Your explanation could very well be the truth. Many a time I have seen academic committees produce exactly this kind of brazen stupidity. It seems that academics have to feel themselves to be ensconced in the very lap of truth before they can reason at all.
“…no warming since 2007…”
?!
It may be that this article is for entertainment only. Apologies if I take it more seriously than it was meant.
Confidence on human contribution is dependent on improved monitoring and analysis of various contributions (forcings/fluctuations) to global temps over the last 60 years or so. This is a completely different kettle of haddock to the slow down in surface warming of the last decade or so (or 5 years!). The attempted conflation of these two issues in the article is putting apples next to turnips.
…Any chance to get explanations from IPCC ? Any contact ? A good scientist cannot feed people results without explanations.
Thanks Stephane Rogeau for picking up that non-sense.
Yeah, but this still leaves a 5% chance that HIC < 50.1% which of course includes zero.
If I had to close this topic, I would say the following:
The main message of the IPCC to policy-makers (the fact their confidence level in the proportion of human-induced warming has increased) is basically anti-scientific: as long as the IPCC does not review its basic (ideological) assumptions, their message is strengthen by lower warming rates.
How ironic: the IPCC managed to use the very sign of its failure as a deception tool in favor of its theory!