rss feedCan you believe the European Commission – or the IPCC?
When I heard that a draft report of the IPCC Fifth Assessment Report had been leaked by a climate skeptic, I had a look at Chapter 9 (Evaluation of Climate Models) to see if the latest versions of their computer models were up to speed on the decline of Arctic sea ice. I wanted to see if the experimental climate scientist was correct when he told me
The trouble with climate modellers is that when there is conflict between their models and the real world, they believe their models.
At the beginning of the leaked draft, it says
There is very high confidence that CMIP5 models realistically simulate the annual cycle of Arctic sea-ice extent, and there is high confidence that they realistically simulate the trend in Arctic sea-ice extent over the past decades.
But when I looked at part of their figure 9.24 I saw this
In this diagram the latest climate models (the CMIP5 models) are represented by the blue line and blue shading. The blue line is the modelled average Arctic sea ice extent (in millions of square kilometres ) of all the models for the month of September. This average declines from about 8m sq.km in 1900 to just under 6m sq.km in 2011. Estimates of the year when the Arctic becomes sea ice free in summer are not available from Chapter 9 but we know from other sources that some climate modellers expect that to be around 2040. – This is supported if the shape of the graphs are similar the earlier (CMIP3) models. See “Models are improving, but can they catch up?” on Neven’s sea ice blog)
The red line in the diagram represents the actual Arctic sea ice extent from 1952 to 2011. The brown line in the diagram represents the modeled average Arctic sea ice extent for the earlier CMIP3 models.
I have simplified this diagram to show only the results for the latest (CMIP5) models and have also added in the actual Arctic sea ice extent for 2012. The blue shading represents the error range for the models.
Should we believe the IPCC when they say “There is very high confidence that CMIP5 models realistically simulate the annual cycle of Arctic sea-ice extent”?
Should we believe the European Commission when they say “The European Commission bases its climate policies on the best available science and on the scientific consensus of experts in the field of climate change.”
I believe “the best available science and on the scientific consensus of experts in the field of climate change” and be found on the amateur blogs of Neven, Tamino, DOSBAT and others.
OK, these amateurs do rely on the publications of hard working professional scientists, who are the backbone of the IPCC.
So why are the amateurs doing a better job in warning us of the dangers we face??
Or why do the professionals of Chapter 9 publish self contradicting nonsense and why does the European Commission tell such lies?
comment
Neven has a comment on his Sea Ice Blog,,,,
I have been informed that the 2012 melting season will not feature in the final draft of AR5, for the simple fact that no peer-reviewed papers have dealt with it before the cut-off date, and that’s because the cut-off date for submitted papers was 31 July 2012. A déjà-vu of 2007.
It’s logical and insane at the same time.
Also there are some papers that have been ignored, such as Rampal 2011: IPCC climate models do not capture Arctic sea ice drift acceleration:
Consequences in terms of projected sea ice thinning and decline.
Neven’s Sea Ice Blog has now covered this issue. http://neven1.typepad.com/blog/2012/12/the-real-ar5-bombshell.html
One contributor argued that ” very high confidence re annual cycle is right”. My reply…
Do you argue that if we apply some unspecified trend function to the results of the CMIP5 models, it is possible to say
“There is very high confidence that CMIP5 models realistically simulate the annual cycle of Arctic sea-ice extent.”
This means the models should get realistic results for the phase of the annual cycle and the amplitude of the annual cycle. Do they catch the slippage of maxima and minima? Is the asymmetric change of maximum and minimum values simulated correctly?
But in these arguments over semantics are we missing a more damaging possibility – that the draft of Chapter 9 is designed to confuse and paper over the failings of the CMIP5 models?
I hope not.
That would be very dangerous for the whole world.