Monday
Apr112011
by 
Bishop Hill
Bishop Hill Climate change just happens
Apr 11, 2011 Climate: Models I've always shuddered rather when people say things like "70% of the observed temperature change is due to manmade carbon dioxide emissions". Christofides and Koutsoiyannis clearly feel the same way as shown in their presentation to the EGU a few days ago.
...we should be careful when we talk about causes, and that trends and shifts do not necessarily imply non-stationarity or a change in forcings: they can just happen.
The implications for all those claims of "we can only reproduce climate history with carbon dioxide in our forcing mix" seem rather profound.
Reader Comments (30)
Retirement and climate change have given me time and a growing expertise into looking into actual figures. About a two weeks ago a ther results of a study were announced by I believe the department of Health, or whatever they're called these days. It announced that we shouldn't eat more than a 500grams, or a pound as the French call it, of meat a week because they had discovered that it causes bowel cancer. First off I wondered how they'd been able to put these figures together, i.e. how they had managed to collect the dietary habits of 16000 people dying in various parts of the UK. As this seemed an intractable problem for me to solve without the actual paper, I looked for the total deaths/annum in the UK. It hovers around 500,000. Now not all of the 16000 could have died because they ate too much meat, but I have no way of telling that without access to the report, so I assumed they all did. And bingo
Were was I? OH yes bingo! it was a maximum of 0.32% of the annual deaths that caused the department of health to recommend 63million people to reduce their red meat intake. I'm not a particular big fan of red meat myself, and probably stay within their guide line, but recommending 63 million people to reduce their red meat intake on the back of something south of 0.32% annual deaths looks more like green propaganda to me than carefully thought out government nutritional policy.
Quite so, geronimo.
And I think it's now widely accepted that the "safe limits" for alcohol arose from a conversation along the lines of:
Gov: We need to be seen to be doing something about all this excess alcohol consumption.
Sci: We can always say it's bad for your health. I mean, it obviously is.
Gov: Any studies that prove it?
Sci: I can probably find a couple that can be interpreted that way.
Gov: Can we set limits or targets or something?
Sci: Sure. What sort of figure do you have in mind?
Gov: Oh, pick a number.
Sci: How about .......... 21?
I've said before that those of us who follow (or try to) the ins and outs of "latest research shows ..." are lost without Sandy Szwarc's Junk Food Science blog. The next best authority is Brignell at Number Watch. Pity that one of them has disappeared and the other is not a well man. We need people to ferret out the facts behind a lot of the data that gets thrown at us.
Cui bono is always a good starting point, of course!
I hope this isn't too far off topic. It has always seemed to me that the sort of analysis of data that these two have engaged in would provide equally sounf arguments against the supposed certainties of climate change.
I read that stairs are responsible for 15% of accidental deaths. They must be banned!
...and allegedly have a bad reputation in police stations
I remember reading many years ago (before www) that something like 15 people a year die getting out of bed. A further 20 die putting their socks on in the morning.
I've tried not getting out of bed but got to bored so I've settled for not wearing socks.
On topic for a moment:
Thanks for the link to this subtle little argument which shows how careful one should be when dealing with complex non-linear feedback systems in assuming that an observable trend is the result of some suitable forcing.
Two of the recent departures in our family were on the loo. There may have been others in times past but delicacy must have hushed them up. Even assuming that the rest were not on the loo, this remains a very high percentage. Must find another locus.
Dad always said you wanted to stay out of hospitals, people die in them.
On a similar theme, "Nonlinearities, Feedbacks and Critical Thresholds" by Riall, Pielke Sr at al,
http://pielkeclimatesci.files.wordpress.com/2009/10/r-260.pdf.
Also several recent articles at Judith Curry's blog by Robert Ellison and Tomas Milanovic ("Decadal Variability of Clouds", "Spatio-Temporal Chaos" and "Chaos, Ergodicity and Attractors").
Basic idea seems to be that anything can happen in the next half hour. Not only is this stuff scientifically exciting, I think it undermines the usual warmist case, and demonstrates that preparedness is likely to be far more effective than windmills and taxes. I wonder when HMG will get the message?
Want another story of attributional sorcery?
http://wmbriggs.com/blog/?p=3702
Briefly, the California Air Resources Board uses statistics from a study that showed that diesel fumes 'caused' 2000 'premature deaths' in California. The study was written by a person, a CARB employee who obtained his 'PhD' from an online degree website, but claimed it to be from UC Davis. The study was revealed as flawed, and the degree as fake, by James Enstrom (of Enstrom and Kabat fame), professor of statistics at UCLA.
The result: Enstrom was fired from his job.
"anything can happen in the next half hour"
Stingray!
Not quite back OT, but I heard an interview on R4 in the last couple of days where a minister was made to squirm quite hard about future energy requirements. The interviewer seemed more clued up than usual and made some telling points about continuity of supply and the role of nuclear. Maybe the Beeb guardians are away at weekends?
There is no concern of man, either real or imagined, which cannot be manipulated for profit.
re Geronimo
But meat is murder. I'm not sure if this is part of a new anti-meat or agriculture marketing campaign, or even a start to downplay CO2's importance. This from the BBC:
http://www.bbc.co.uk/news/science-environment-13025304
I'm not sure if Mr Sutton was chewing a nice mouthful of grass while he said that, but I thought animals were good at turning indigestible plant proteins into digestible ones, and making use of land that would otherwise be less useful for growing food crops.
The rest is religion meets business. The climate changes, we must be responsible. We can't burn witches any more so we'll have to tax people instead. The modern human sacrifices to mitigate climate change will be the poor, elderly and infirm, and all because of a tenuous trend between CO2 and temperature.
Atomic Hairdryer
Yes, I heard the broadcast, and immediately "nitrogen is the new co2" came to mind ..........
Whatever language you dress it up in, the answer always comes back (Rather succinctly I feel) as;
Climate changes - don't know why - Exam on Friday.
We should be wary of such studies, for in the recent past we have had;
"exaggerated the impact of eating meat on climate change"
Perhaps because if you don't eat it, it will just stay in the field and fart!
Christofides and Koutsoyiannis make an important point. The context however is equally important.
Random and occasionally persistent change emerges from deterministic chaos. This applies to climate. But so does radiative forcing by GHGs such as CO2. There is no fundamental opposition between these positions.
The problem with the aggressively pessimistic CAGW stance is that it refuses to acknowledge the inherent potential for random excursions in climate behaviour. The problem with over-focus on random excursions is that it ignores the radiative physics.
I don't think BH was suggesting that we do this, so balance and an open mind would seem to be the way forward.
Ultimately, the scientific problem is to improve the estimate for climate sensitivity to CO2. Ignoring random climate shifts and understating the importance of natural oscillations will not help us do this.
Well, they might start by looking at cultures that eat more red meat than we do. The Mongolians. The Argentinians. And cultures that eat very little red meat.
How to they fare in the bowel cancer stakes (steaks?).
Even then, they need to prove other factors are at play.
It reminds me of Gilmore and the other anti-booze medicos. Of course we need to enormously increase the price of booze to control the alcoholism deaths and costs.
Yeah. But how come Scandinavian countries where booze is much more expensive than in the UK have worse alcoholism problems? And warm mediterranean countries with cheap booze have less problems?
Funny that, isn't it?
Are all the people who die of red meat perhaps the same ones who die of smoking and of PM10 levels and of Climate Change and booze and driving fast and every other Nanny State scare?
I wonder.
Those who for ideological reasons would seek to impose vegetarianism on the rest of us are equally unbalanced.
It is interesting to see how readily the climate change lobby has embraced the involuntary vegetarianism brigade.
[No - try moderating your tone]
[Sorry]
Reminds me of Rumfeld: "Sh1t happens".
I know I shouldn't rise to this but...
This paper is a fairly simple paper, presenting a simple and obvious fact: systems which exhibit exponential error growth from initial conditions are generally less predictable, rather than more predictable, as the time scale of the prediction increases.
Climatologists claim the opposite of this, an extraordinary (but not impossible) claim; yet they fail to provide the extraordinary evidence in support of it. But nobody dares point out the elephant in the room.
Antonis and Demetris are merely pointing out the elephant in the room. And their fellow scientists appreciate it as well; Professor Koutsoyiannis received the prestigious Henry Darcy medal in 2009 for his contributions (see here). So even if blog comment trolls do not appreciate their work, the scientific community at large do.
Of course, this paper doesn't stand on its own: this pair have dozens of peer-reviewed papers that make the argument in a more comprehensive manner. This short paper is a summary suitable for a 15 minute oral session at a conference, which is exactly what it was.
BBD: you are correct in pointing out that it is plausible that the greenhouse effect could live "side by side" with long term persistence as an emergent property of a chaotic attractor. However, the latter seriously confounds attempts to estimate the former. Indeed, it is a trivial observation to argue that you cannot estimate the sensitivity without accounting for natural variability; and, given our present state of knowledge, the variability may mean the whole concept of sensitivity is an ill-formed problem and essentially meaningless.
BH
Never mind ;-)
For those interested, a more in-depth examination of the potential influence on climate of 'synchronised chaos' can be found in Tsionis et al. (2007):
https://pantherfile.uwm.edu/aatsonis/www/2007GL030288.pdf
Just for fun:
If the idea of localised chaotic attractors in climate systems prompts you to ask for some evidence, which it certainly should, there is the Great Red Spot:
http://www.wired.com/wiredscience/2010/03/jupiter-spot/
@James P
I also heard this and you've prompted me to dig it out - it was "The World This Weekend" on 10/04/2011 at around 10:00 in:
http://www.bbc.co.uk/iplayer/console/b0107wxc
and he did indeed put him right on the spot - great job.
Listened again, and it's just no good for my blood pressure. Utter weasel.
It has been known for a while that the further you look back in time the more chaotic this planet's climate appears. On a modern human time scale, looking back over many generations, climate is essentially cyclic. In one human's lifetime climate is essentialy the experience of seasonal and unseasonal weather.
When you combine physical chaos and complexity with human irrationality you invoke a simplification process that induces a societal and moral fear that the setting sun will not rise in the morning. Cause and effect is a human construct. The physical world cares nothing about human constructs - it just does what it always does.
BBD - "Ultimately, the scientific problem is to improve the estimate for climate sensitivity to CO2. Ignoring random climate shifts and understating the importance of natural oscillations will not help us do this."
FWIW I'd suggest that you are going a little too far a little too early with your "ultimate" problem. IMO there is much to be done before the search for a climate sensitivity to CO2 can be undertaken in any seriousness. The whole system of climate is in need of investigation and concepts such as "climate sensitivity to CO2" presuppose much which I think is yet to be established.