A study recently released by the UK Met Office with much fanfare has been generating considerable excitement in the media. The study led by Peter Stott claims that clear fingerprints of human-induced global warming are evident. As noted in the New Scientist:
it’s no surprise that when new papers confirm the IPCC’s conclusions, climate scientists are not shy about advertising them………..It’s hard to take the promotion that Stott’s review received – it was press released and presented at a press conference – as anything other than a response to the unremitting onslaught of climategate-related accusations being hurled at climate scientists at the moment.
The Stott study was referenced in a Guardian article by climate researcher Chris Huntingford called How public trust in climate scientists can be restored. Huntingford comments:
Second there is the question of whether major policy decisions should really be made on the basis of simulations of the climate system, as performed on a few specialised computers dotted around the world? There are compelling reasons to trust these computer models, but at the same time, more direct evidence underpinning the claim that climate is changing is needed. That is why the work by Peter Stott and colleagues is important.
So Stott’s study provides “direct evidence” rather than relying on computer models which were the basis for the catastrophic climate predictions in IPCC AR4. But how did Stott come up with this direct evidence so free of computer contamination?
The answer is that he didn’t. The basic methodology in the Stott study was to run climate models with both anthropogenic and natural forcings (Case A) and then with natural forcings only (Case B). The results from the two cases are then compared. So contrary to the claims of new direct evidence we are back to relying on the results of climate models. But even then, how good is this model evidence?
It appears that the only natural forcings considered are volcanoes and solar energy. As ever, model calibration and verification are not discussed but I would have expected more of an attempt to justify the Case B model by comparison to known millennial fluctuations in the global climate. Given that the ice ages and inter-glacial periods are poorly understood and are thought to be caused by a multiplicity of factors, why would anyone have any confidence in the Case B model? Or the Case A model for that matter.
Peter Stott is keen on using the term fingerprints when comparing the model results for the two cases. Fingerprints become evident when the model results are different. However if you develop a Case A model with a number of different climate forcing mechanisms and then remove some of those forcings in the Case B model, it is self-evident that the model results will be different and fingerprints will be exposed. The fingerprints are clearly an artifact of the modelling procedure; you will always detect fingerprints if the models are configured this way regardless of whether the model algorithms are valid. The results could only be used as evidence of anthropogenic changes if you believe that the climate models adequately represent all the global processes affecting climate. And even the IPCC consider that they do not.
Stott addresses the concern that model predictions may be incorrect. The solution advocated in the Stott paper is to add more models to the analysis so that more results are obtained and the suite of results can be statistically analyzed. More models provide more precision.
A further enhancement is to include several climate models in a single analysis thereby making it possible to estimate the uncertainty in response patterns and a more comprehensive estimate of attributable changes.
The Stott study then goes on to look at observations, and surprise, surprise, fingerprints are discovered there too. The conclusion is that the fingerprints are from the same source, anthropogenic climate change, but in fact they are two different sets of fingerprints. The first set of fingerprints is an artifact of the modelling procedure. The second set of fingerprints merely represents climate changes that have been observed over the period of record that could have resulted from anthropogenic influences, or not. They could be entirely natural and reflect global changes since the end of the Little Ice Age. The Stott paper is in general agreement with this:
While the observational record leaves little room for doubt that the earth is warming, the evidence does not by itself tell us what caused those changes. We could be experiencing natural fluctuations of climate operating on multidecadal timescales. Alternatively, drivers of climate change, such as volcanic eruptions or human induced emissions of greenhouse gases, could be forcing sustained changes in climate. Detection and attribution seeks to determine whether climate is changing significantly and if so what has caused such changes.
The section of the Stott paper on extreme events is even more inconclusive. The study reveals that the models are incapable of simulating floods, droughts and hurricanes so no fingerprints can be detected for extreme events though he prefers the terminology many challenges remain. So contrary to the many media reports, there is still no direct evidence that catastrophic climate change is happening or will happen.
So the Stott study admits nothing much can be said on extreme events except more work needs to be done and we are back to being dependent on model results even for attribution of changes in global mean temperatures. Stott’s findings presuppose that the models properly characterize both the natural and anthropogenic influences. Given that the models are so poor at replicating even known and frequently occurring climate phenomena such as the PDO and ENSO, then Stott’s study appears to be little more than playing yet another round of computer games.
Despite the total reliance on models, the conclusions of the Stott study are quite definitive:
The wealth of attribution studies reviewed in this article shows that there is an increasingly remote possibility that climate change is dominated by natural rather than anthropogenic factors.
The Stott study results were released to the media with a summary before the detailed paper was generally available. The media accepted the summary and dutifully published the findings with headlines such as Humans must be to blame for climate change, say scientists. The media were led to believe that this was new direct evidence, not recycled results of climate models. Even climate researchers believed that this was the case. If there are any fingerprints to be detected in this process it is the fingerprints of media manipulation and the fingerprints of promotion of climate alarmism.
Asked if the new research would help to silence the doubters who question man-made climate change, Dr Stott said: “I just hope people will make up their minds informed by the scientific evidence.” With fingerprints so revealing no more evidence is required.
This looks like it is simply following established procedure to me.
Is Peter Stott not duplicating the original IPCC procedure for identifying and blaming the ‘human’ fingerprint – the ‘A’ in AGW?
I’m pretty sure that the only ‘evidence’ for the level of human induced warming claimed by the IPCC is this absurd ‘We ran the models with and without, and this is what we got, and nothing else can explain the warming type approach.’
The trouble is that all the ‘unique’ variations that ‘prove’ AGW-such as arctic ice melt, receding glaciers, rising temperatures, changing weather patterns etc-are nothing of the sort and can be clearly traced back through historical record as having occurred before.
The Met office take the view that there was very limited climatic variability prior to the modern era which is given the lie by their own instrumental records back to 1660.
How CO2 can be a constant 280ppm throughout these climatic upheavals yet still be considered the primary climate driver still needs explaining.
In this regards Chuckles is correct, in as much in believing the temperature variations since 1850 are ‘unexplained’ so human made CO2 MUST be the cause, is yet more of the same failed computer modelling.
Perhaps they ought to model in the unknown factor ‘x’ which I suspect is 90% of the climate drivers and comprises things we currently know about but dismiss, plus things we haven’t even thought about yet
tonyb
TonyB
To the “guilty by default” verdict proclaimed by Peter Stott relative to human attribution for late 20th century warming, you observed:
Yet it is precisely this same Met Office that now attributes the 21st century cooling (despite record CO2 increase) to “climate variability” (a.k.a. “natural forcing”).
Sorry, folks. You can’t have it both ways.
There is an underlying arrogance in all this, starting with Stott’s absurd assumption that we know everything there is to know about what makes our planet’s climate change, therefore if we cannot identify the cause, it does not exist.
Ignorance itself is excusable, but when it is combined with arrogance, it is not.
Max
Max
As you say the Met office can’t have it both ways but they are certainly trying to :)
That we don’t know everything about even CO2- let alone all the other aspects of climate- is abundantly clear from the nearly 200 comments on my Historic CO2 variations thread at Air Vent.
For Stott to continue to use models whilst pretending to have found new evidence is deceitful-why do the media keep falling for it?
Tonyb
TonyB
“The media” hear what they want to hear.
Stott has claimed he has evidence to prove human cause for climate change, but he really just cites the same old climate model stuff that IPCC has been citing.
Are the media astute enough to know the difference?
Maybe.
Do they want to know the difference?
Probably not.
Then there is the question of Stott’s motivation.
If Stott would simply admit that there is much more that “we do not know” about what makes our planet’s climate behave as it does than there is that “we do know”, he could qualify as a real scentist.
But using the tired old “our computers cannot explain it any other way” saw, reveals that he is an arrogant activist selling a bill of goods and not a true scientist looking for the truth.
So I’d say Stott and the media share the blame for this silly so-called “fingerprint study” (which proves nothing as it brings absolutely nothing new to the argument).
Max
Max
I think Donald Rumsfeld was unfairly maligned for this quote as it could be tailor made to describe our shocking ignorance of what drives the climate whilst everyone tries to pretend the science is settled;
“There are known knowns. These are things we know that we know. There are known unknowns. That is to say, there are things that we now know we don’t know. But there are also unknown unknowns. These are things we do not know we don’t know”
tonyb
Tonyb:
You observe:
I think the media fall for it because they do not have the aptitude or the expertise to investigate claims like Stott’s. It is far easier to simply accept the press release and regurgitate it.
I am more bothered by the attitude of climate scientists. We know of at least one (Huntingford) who did not bother to read or understand the paper. But what about others? In this supposedly new, enlightened era post Climategate, why are other scientists not speaking out and challenging Stott’s conclusions? It would help build trust in climatology if there was an open debate. I guess we hear nothing from them because they think it would not help their “cause”.
tonyb:
You are right. Donald Rumsfeld was unfairly maligned for the quote on unknown unknowns. The notion has been around for a long time in engineering and project management. I first heard it in the early 1990s and it was probably in use much before then.
I suspect there are alot more unknown unknowns in climate. As a recent historical example, the Pacific Decadal Oscillation was only discovered in 1997 and was only found because of an investigation into the variability of salmon stocks!
Just to say very good post, potentilla. The headline of the article in the Independent just threw all doubt to the winds – none of all that “likely” or “very likely” stuff, but “Humans must be to blame”, with the “must” underlined for good measure. And for all that, the Stott paper seems to be saying nothing that has not been said already, hardly the “powerful riposte” that Steve Connor is proclaiming.
It seems that climate scientists are still under the delusion that computer runs constitute experiments.
TonyB
And these are the ones that make ALL claims of human cause plus future predictions on our planet’s climate totally worthless.
Stott is either purposely ignoring these to get his point across (i.e. arrogant) or so silly that he really believes they cannot exist (i.e. ignorant).
(I personally believe that it is the former.)
Max
I recently asked Monbiot directly for evidence of dangerous man-made global warming in the present or in the future, and he quoted the Stott article at me. I pointed out that Stott doesn’t even begin to claim to tackle the question of future warming and its dangerosity. Of course, Monbiot didn’t reply. I plan to worry this particular bone to death every chance I get.