Sunday
Jul212013
by
Bishop Hill
![Author Author](/universal/images/transparent.png)
Low-sensitivity model outperforms
![Date Date](/universal/images/transparent.png)
![Category Category](/universal/images/transparent.png)
![Category Category](/universal/images/transparent.png)
![Category Category](/universal/images/transparent.png)
Steve McIntyre has a must-read post about a low-sensitivity climate model which, when loaded up with actual greenhouse gas levels, completely outperforms the Met Office's HADGEM2.
Reader Comments (68)
Sometimes I think SM's mischief makes his point a bit obscure but then he does it to wind up alarmists as was evident through the whole Hockey Stick saga.
However my (maybe dumb) points are:
1 Why do skeptics get excited when new work shows less and less "sensitivity" when observation shows no relationship between CO2 and temps except for the 800-2,000 year lag from ice cores? Obviously the latter does not support the "sensitivity" theory.
2 Since we know that we don't know much about the climate - admitted frequently by the IPCC but rarely mentioned- and therefore the models are all parametesised guesswork and therefore total junk, why does anyone give them credence at all?
Just asking.
Paul
Paul, I don't think SM is trying to wind up alarmists but to unwind one of their weakest arguments: that GCMs prove high sensitivity. You already think that argument is weak, as I do. But Steve is about to take it step by step from some hilarious first principles. At least I hope. That's why I was excited about this last night and remain so today.
Paul
Thanks for adding a bit of common sense to this thread.
note: This sentence from Steve's original post is striking for what it implies about where the discussion may go (yes we read it on Climate Audit and Bishop Hill and WUWT first):
[emphasis added]
Yep, too high sensitivity.
Because (as I posted somewhere else recently) of all the unverifiable concepts of "climate science" whose only existence is in the interior of climate models, climate sensitivity is unique in that 97% of sceptics also believe it is meaningful.
Settle boys, this speculation is no good for your Pacemakers(tm).
McIntyre has merely picked up an old model, which was probably state of the art in its time, sans bells and whistles that represent the 'advancement' of the 'modern' models, and observed that this old model appears to regurgitate output that more closely resembles reality than what the 'modern' models do. Further, he has also observed that, unlike the 'modern models', this old model is programmed with 'low climate sensitivity' parameters.
The irony of this is awesome ... and I await his further revelations <|:0
It doesn't bother me that 97% of sceptics accept that climate sensitivity may be meaningful, if that stat is even roughly right. Note I changed the verb there. It's only a maybe for me and I don't much care about it beyond that. I remember seeing Lindzen once being quizzed on this stuff and finally being asked if he thought humanity could cope with an increase of more than 2 deg C in globally averaged temperature anomaly - from 1850 or from when I can't even remember. He said "Of course we could cope, no problem." Another sacred cow slaughtered. I like that kind of thing. So why not sensitivity itself?
Because even if the relationship is far more complex than the simple logarithmic one it's a first cut at saying what we expect, all other things being equal, given that man's emissions are a new factor in the system, compared to the previous four billion years. I know all other things aren't equal. But baby steps help in many areas. Not that they've helped much in climate science, I admit, for reasons on which we'd no doubt agree.
The sensitivity question becomes a kind of proxy for the key policy-driving question: is the expected temperature rise in the next n years going to be dangerous?
So one defining question from me you Martin is, if you drop all mention of CS, do you have any handle on this key question from a policymaker: if humanity burns all the fossil fuel it finds is there any possible danger?
I understand that you think CS is a worthless link in the chain. I take a more pragmatic view that low sensitivity is associated by chaps like Andrew Neil and Ed Davey with safety and thus the chance to delay very costly policies. I'm happy to go with that.
So tell me how the interview goes without mentioning CS.
did John Daly ever create a climate model ?
OK, JJ got it and I'm trying to figure out who said what about glaciers.
========
Heh, Stephen Schneider defended the IPCC's basic point that glaciers were disappearing. I've seen an early video of Stephen Schneider, perhaps from the '70s, in which he confessed that climate science doesn't know the direction the earth's temperature is taking. He fooled a lot of people, but first he fooled himself.
============
er, H/t AT @ CA.
============
So tell me how the interview goes without mentioning CS.
Jul 22, 2013 at 11:53 PM | Registered CommenterRichard Drake
I would not expect Neill to start his interview with "Well, obviously climate science based on unvalidated models is rubbish" and take it from there. I've got no problem with pointing out the inconsistencies in so-called 'climate science'.
But that's not the same as fighting bullshit with bullshit.
Thanks Martin. Climate science based on invalidated models is rubbish though. If Andrew Neil is the boy who just took his finger out of the dyke it may not be as long as any of us think. It's the way I like to think.
CS I see as separable from the above. But no worries. Respect.
kim, the glaciers clue is deadly (yes, thats another clue).
And there's another good clue from Ian Blanchard.
Richard, what we have in the armoury is the 15/17 year temperature hiatus and that hiatus disproves the CS figures in just the same way that it disproves the MO models of CAGW.
Oh, Boy, I went from Hansen to Arrhenius to da Vinci to Schneider only to find it is Guy Callender. Watch Steve make a bigger fool out of me tomorrow.
==================
Is this a good time to revisit this thread? I reread it just now and realised I'd missed this from Jeremy Harvey:
I apologise that I launched into my own limited defence of the concept without referring to this. Will the simplicity Guy Callendar's model brings back to the table underline the points made?