Bishop Hill Slingo at the IoP
Sep 4, 2014 Climate: MetOffice This report of Julia Slingo's recent lecture at the Institute of Physics was originally posted in the discussion forum by a reader. I thought it worthy of elevation to the main blog. My thanks to "Colonel Shotover" for his efforts.
Frances Saunders, president of the Institute of Physics, introduced the lecture, telling us she was particularly delighted to welcome JS, first because her work on climate models showed the importance of physics to everyday life, and second because she was a woman, and so critical in supporting the objective of getting more women into physics. A cynic might suggest that these represented the twin pillars of government science: obtaining funding by demonstrating ‘relevance’ and supporting government policy objectives in return.
JS opened by telling us her lecture wasn’t really about climate change at all, but then showed a couple of slides showing how deadly serious our situation was. The rise in CO2 levels in the atmosphere was unprecedented during the last 800,000 years. Citing John Beddington, she told us that climate change was just one part of a dangerous future, playing into difficulties with water and other resources, population growth, food and energy supplies, politics and economics, health and migration. The planet ought to be ‘pretty much in balance’, but it wasn’t. The Thames barrier had been raised more in 2013/14 than ever before.
So far, so predictable, and I began to regret coming to the lecture. But at that point the tone and subject matter changed entirely. JS put fears about thermageddon to one side and launched into an eloquent disposition on climate models. Her immense enthusiasm for using computerised mathematical models to mimic weather systems was immediately apparent. She described the physics that went into the models, showed us the relevant equations, and talked about her drive towards better resolution, showing us how that increased resolution improved the ability of the models to reflect what happened in the real world. Here was a real scientist, making predictions from models, examining them against reality, investigating discrepancies, striving for an ever better understanding of the way the weather and climate work, immensely proud of how these incredibly complex models could simulate such things as global evaporation and precipitation, and the development of hurricanes. Frequently there were animated graphical comparisons between the models and reality, and we were shown how simulations had improved over the years. There was no alarmism here, although we were shown diagrams showing the absorbtion properties of CO2 and an explanation was given as to why these were important. She was open about problems with current models and that there was an immense amount still to be learned, reflecting this in her final slide: the words of Sir Isaac Newton: “I was like a boy playing on the sea-shore, and diverting myself now and then finding a smoother pebble or a prettier shell than ordinary, whilst the great ocean of truth lay all undiscovered before me.”
JS needs a certain amount of fear of climate change to keep the funding coming for ever-larger supercomputers, so it’s hardly surprising she’s part of the “consensus”. Nevertheless I came away with a much better opinion of her. She hadn’t shown us much about what climate models tell us about the future, and certainly hadn’t blunted my scepticism about the dire predictions we hear so often. But she had given us a better understanding of how weather and climate models work, accepted that there was much still to be learned, and shown us her passion for continuously improving the models to better reflect the real world.
Reader Comments (40)
This is amazing. She is helping the "consensus" to drive us into fuel powerty and impending shut-downs of the Grid because a couple of nuclear power stations have been shut down through her public utterances about impending doom, and yet she seems to know that the foundation for the 95% certainty is built on clay even if it is "model clay".
Thanks for this Colonel S - Is there a publicly available copy of her presentation?
hmmmm... Slingo on predictions from the Met Office.
Julia Sllingo,on BBC feedback when new decadal forecast (Dec2012) showed flat temps to 2017.
Roger Bolton:
"The Met Office, on whose figures the report was based, also had a lot of people getting in touch. Professor Julia Slingo is their chief scientist. I asked her if she thought that the Today headline "The Met Office says it does not believe that global warming will be as severe as it had previously predicted" was accurate.
Julia Slingo:
"Absolutely not. I mean, just to put the record straight, we had not put out, ourselves, a report. We have, over several years, on an annual basis, placed our decadal forecasts on our research pages. They are experimental, they are research in progress, and these were picked up by the sceptic blogs and the story was taken from there.
https://sites.google.com/site/mytranscriptbox/home/20130111_fb
AND YET
ref same decadal forecasts - Met Office Vicky Pope.. (2007)
Dr Vicky Pope:
By 2014, we're predicting that we'll be 0.3 degrees warmer than 2004. Now just to put that into context, the warming over the past century and a half has only been 0.7 degrees, globally - now there have been bigger changes locally, but globally the warming is 0.7 degrees. So 0.3 degrees, over the next ten years, is pretty significant. And half the years after 2009 are predicted to be hotter than 1998, which was the previous record. So again, these are very strong statements about what will happen over the next ten years."
https://sites.google.com/site/mytranscriptbox/home/20070905_vp
--------
one thing for scientists, totally different for the media.. that is Julia.
the Met Office press release at the time made these predictions, no caveats, no mention of experimental, just how useful they were for business and policymakers, this 0.3C warming by 2014 made it into government advice.
http://web.archive.org/web/20080708230357/http://www.metoffice.gov.uk/corporate/pressoffice/2007/pr20070810.html
Met Office - News release
10 August 2007
The forecast for 2014...
Climate scientists at the Met Office Hadley Centre will unveil the first decadal climate prediction model in a paper published on 10 August 2007 in the journal Science. The paper includes the Met Office's prediction for annual global temperature to 2014.
Over the 10-year period as a whole, climate continues to warm and 2014 is likely to be 0.3 °C warmer than 2004. At least half of the years after 2009 are predicted to exceed the warmest year currently on record
These predictions are very relevant to businesses and policy-makers who will be able to respond to short-term climate change when making decisions today. The next decade is within many people's understanding and brings home the reality of a changing climate." - Met Office 2007
http://ipccreport.wordpress.com/2013/11/20/the-skillful-predictions-of-climate-science/
numerous repoirts.
Thats almost worse than her being a "true believer" she knows it is at best wrong, at worst a lie, yet she perpetuates that lie in order to secure funding. I'd say that constitutes fraud at the very least, and should warrant prosecution. Not to mention the moral and ethical bankruptcy it is indicative of.
When I hear Julia Slingo on the BBC or in the Guardian discuss the following, then I might believe she is a scientist 1st and foremost.:
At a Royal Society meeting in 2013, Julia Slingo of the Met office played devil’s advocate and posed the following question to Prof Jochen Marotzke of the German Max Planck Institute of Meteorology, see the 42:46 mark royalsociety.org/marotzke.mp3:
Julia Slingo
“…it’s a great presentation about 15 years being irrelevant, but I think, some of us might say if you look at the Pacific Decadal Oscillation and it’s timescale that it appears to work, it could be 30 years, and therefore I think, you know, we are still not out of the woods yet on this one. … If you do think it’s internal variability, and you say we do think the Pacific Decadal Oscillation is a key component of this, and it’s now in it’s particular phase, but was previously in the opposite phase, could you not therefore explain the accelerated warming of the 80s and 90s as being driven by the other phase of natural variability?”
Marotzke answers after much incoherence of his own:
Um…I guess I’m not sure.”
- See more at: http://notrickszone.com/2014/01/10/oops-trenberth-concedes-natural-ocean-cycles-contributed-to-1978-1998-warming-after-all-co2-diminishes-as-a-factor/#sthash.cEqJdf56.dpuf
Pierre Gosselin summarized
Simplifying Slingo’s incoherence: “If the current cooling is due to the negative PDO phase, then wouldn’t the warming of the 80s and 90s be a result of the positive PDO phase back then?”
I was at this lecture, and also came away with much the same impression as the Colonel.
Prof. Slingo was eloquent in her explanation of the underlying physics of climate models, and proud of her field's achievements over the years. I found it hard to disagree with her; it was obvious that things had improved significantly since the 1980s.
What also interested me was her declaration that the long term climate models run on exactly the same code as the short term weather forecasting models, leading her to the conclusion that the underlying modelling code is amongst the most tested in the world. She showed a graph of the increased forecasting accuracy over the last few decades; a 5 day forecast today is about as accurate as a one day forecast 40 years ago, she said.
To re-iterate the Colonel's point, it was not explained how the longer term models relate to the short term forecasting - why an increase in accuracy of short term forecasts means that the longer term projections can be "trusted". There may well be a corresponding increase in accuracy in the long term projections, but the relationship is not necessarily proportional, i.e. a 50% increase in accuracy of short term models does not lead to a 50% long-term increase. Indeed if the accuracy declines in inverse proportion to the timescale, then even a large increase in short term accuracy may not be at all significant for projecting to the year 2100.
All in all it was enjoyable, and there was an amusing "speech" by Piers Corbyn, who managed to get Prof. Slingo to reluctantly admit that she, and her fellow scientists, certainly did not rule 'sceptic theories' out, and that the 22 year solar cycle may very well turn out to be of great importance to phenomena like the Jet Stream.
On reflection I actually felt a lot of empathy toward her, for several reasons:
- She has chosen arguably the most complex field of empirical science
- She probably did not go into the field knowing how politicized it would become
- She is being attacked from all angles
- She is getting on, and probably wants to conduct herself with a degree of decorum not compatible with the field
- She knows about the uncertainties, but also understands that to admit uncertainty implies "weakness" in the eyes of the uneducated, and therefore she has decided - probably against her principles - to become political too
I think you saw the two sides of science, sides that need to be kept very much separate:
A. The pure pursuit of the science, with all its ifs and buts and uncertainties.
B. Science & Policy, much more likely to work well if A is recognised and celebrated, not swept under the carpet by those blinded or corrupted by belief.
So it seems that everyone agrees the climate models are extremely complex, and probably excellent examples of combined research and work, but have severe limitations.
I guess my question is that in light of these issues/discrepancies, are the models useful enough to warrant the severe economic/policy response based on them? Or is that unfair - perhaps they are just one of many inputs?
@Ian Shurmer
See the recent posts here re Richard Betts.
If by 'other inputs' you mean drivers to raise taxes, appease green pressure groups and implement UN Agenda 21. Yes they certainly exist. The models are just supporting 'data'
re: Slingo's statement,
What an appalling outlook from a prominent climate scientist! We should be relieved if so much alleged warming turns out to be a natural cycle, since that would imply that sensitivity to CO2 is much lower than previously believed. Alleged global catastrophes averted (the ones which are said to be caused by human generated CO2 emissions). Nature will still throw natural disasters at us but we would have far less reason to expect that CO2 will add to our woes.
Slingo's slip of the tongue there reveals her fervent desire for belief in impending catastrophes.
Why would that be??
I have long thought that a truth has been ignored widely: We are in an interglacial period during which the earth will continue to warm up ... until it doesn't, then back into an ice age.
Hmmmmm. This from the woman who said Met Office forecasts re 2012 were not wrong, nor were their calculations, because they were "probablisitic!" In engineering parlance this says it doesn't matter what the calculations say they are never wrong! Utter tosh!
If as she decently indicates, climate models are a work in progress, why are their outputs good enough to be a basis for policy?
Re-usable...re-used code??
Whats that, the OS, the functional code. Fred s spreadsheet
Was there a Q&A and was the question asked...what do you mean exactly by underlying code?
What's Vicky Pope been smoking
hadcrut4 shows a fall from 2004
woodfortrees.org/plot/hadcrut4gl/from:1995/plot/hadcrut4gl/from:2004.42/trend
the trend for rss from Nov 2000 is -0.004 +/- 0.003 C/year
so it's not a pause it's a fall
Thank you, Colonel Shotover.
I am glad to see that Dr Slingo can discourse reasonably on the limitations of models. But where were all those selfsame caveats when she ran the Met Office? With all due respect to Dr Betts, it *is* the models' predictions/projections which drive the urgency -- one might say the hysteria -- of renewables subsidies, emissions limits, carbon taxes et al. The 0.8 K increase over the last century or so, would not impel any action in and of itself.
P.S. to the Colonel: "disquisition" intended rather than "disposition" in paragraph #3?
Were there no questions at the end?
What has "improved significantly since the 1980s" is the satellite coverage. It's a lot easier to predict the track of the residue of Hurricane Bertha if you can follow what she's doing from minute to minute. There was a time in my lifetime when an autumn storm wa an autumn storm. Nobody ever suggested to us that it was actually the tail end of a hurricane.
To imply that this improvement in 5-day forecasting is somehow suggesting that this can tell us anything about the climate in 50 years or even 50 months or (as we keep seeing year in year out) 50 days really is the pea under the thimble.
'Here was a real scientist, making predictions from models, examining them against reality, investigating discrepancies, striving for an ever better understanding of the way the weather and climate work,'
And then throwing all that our of the window in line with rule one of climate 'science' when the models and reality differ in value its reality which is in error . There is no chance of the MET given up its unquestioning support of AGW of which its a very willing advocate , until the person that has lead the MET down that path has gone . And that is Slingo, pimping for cash is only part of the deal , the other is their personal commitment to the ideas behind 'the cause '
"What also interested me was her declaration that the long term climate models run on exactly the same code as the short term weather forecasting models, leading her to the conclusion that the underlying modelling code is amongst the most tested in the world."
This does not prove anything. When the models are used for weather forecasting, they do not have to predict slow changes like ocean currents, they only have to use current data on ocean temps as input for atmospheric models. Similarly they don't have to predict the jet stream, just note where it is currently.
What frustrates me, as a programmer with some familiarity with chaos theory, the way that errors propagate over time, sensitivity to initial conditions and all that, is the way they keep repeating that 'the models are based on the laws of physics' as if that had some relevance to the problem. It doesn't matter how closely you model the physics, you can't get away from the unpredictability of complex dynamic systems - why are they so vocal about the laws of physics but silent on the chaos? If there is some body of evidence demonstrating that the climate models have either solved, or can safely ignore, the problem of chaos, we certainly need to hear more about it.
One suspects that Julia Slingo is with Laplace's demon in believing that, with a big enough computer, she could predict the future of the whole world and everyone in it.
The modellers' working theory seems to be that only weather is chaotic while climate is predictable, on the a priori assumption that increased CO2 will inevitably and continuously cause heat to accumulate in the atmosphere, QED. If it were really that simple, we wouldn't need the models in the first place; the fact that the models bear out the assumption, while reality does not, seems rather significant.
Colonel and Rob - thank you for the feedback on this presentation.
Regarding the relations between Climate Models and NWP please can I point you to this post by Judith Curry some little while ago - (Note that Sandrine Bony was a lead author on the AR4 Chapter 8 re: Climate Model Evaluation):
http://judithcurry.com/2013/06/16/what-are-climate-models-missing/
What Are Climate Models Missing?
"Fifty years ago, Joseph Smagorinsky published a landmark paper (1) describing numerical experiments using the primitive equations (a set of fluid equations that describe global atmospheric flows). In so doing, he introduced what later became known as a General Circulation Model (GCM). GCMs have come to provide a compelling framework for coupling the atmospheric circulation to a great variety of processes. Although early GCMs could only consider a small subset of these processes, it was widely appreciated that a more comprehensive treatment was necessary toadequately represent the drivers of the circulation. But how comprehensive this treatment must be was unclear and, as Smagorinsky realized (2), could only be determined through numerical experimentation. These types of experiments have since shown that an adequate description of basic processes like cloud formation, moist convection, and mixing is what climate models miss most. "
Paper here:
http://www.euclipse.eu/Publications/Stevens,%20Bony_What%20are%20climate%20models%20missing.pdf
As far as the evaluation of models I'd hihlight this work by Demetris Koutsoyiannis et al:
http://climateaudit.org/2008/07/29/koutsoyiannis-et-al-2008-on-the-credibility-of-climate-predictions/
"On the credibility of climate predictions
Koutsoyiannis, D., A. Efstratiadis, N. Mamassis, and A. Christofides, On the credibility of climate predictions, Hydrological Sciences Journal, 53 (4), 671–684, 2008.
Geographically distributed predictions of future climate, obtained through climate models, are widely used in hydrology and many other disciplines, typically without assessing their reliability. Here we compare the output of various models to temperature and precipitation observations from eight stations with long (over 100 years) records from around the globe. The results show that models perform poorly, even at a climatic (30-year) scale. Thus local model projections cannot be credible, whereas a common argument that models can perform better at larger spatial scales is unsupported."
Open access paper here;
https://www.itia.ntua.gr/en/docinfo/864/
Were either of these papers or similar works mentioned? Were the issues they raise discussed?
A PS on the cloud formation issue - Judith has co authored a new text with Vitaly I. Khvorostyanov:
http://judithcurry.com/2014/09/04/thermodynamics-kinetics-and-microphysics-of-clouds/
http://www.amazon.com/Thermodynamics-Kinetics-Microphysics-Clouds-Khvorostyanov/dp/1107016037/ref=sr_1_1?s=books&ie=UTF8&qid=1409861421&sr=1-1
Julia Slingo's enthusiasm for numerical climate models came across strongly when I met with her last autumn, and I'm sure is genuine. She is undoubtedly proud of the Met Office's model, and it does indeed come out as one of the best climate models on a number of measures. There is no doubt that the Met Office has an impressive modelling capability. And JS is IMO right to say that these models are very impressive achievements. But, as has been pointed out, the fact that a model is good at simulating short term weather (and the current climate) does not mean that it is good at simulating the long term response to increasing greenhouse gas etc. forcing.
It is, incidentally, ironic that the Met Office HadGEM2-ES model ranked poorly in the Sherwood et al (2014) paper - which claimed that high sensitivity models simulated atmospheric convective mixing better than low sensitivity models - despite being almost the most sensitive of the current generation CMIP5 models to increasing forcing. But that is arguably more a reflection of that study's measures of model fidelity being weak than on the relative quality of the various models it analysed.
It was great to meet a number of other BH contributors or sympathisers again at this excellent event. I made fairly copious notes on Slingo's talk that I can't convert to digital form until the weekend at least - and I want to let various ideas raised settle down in my mind in any case - but talking afterwards to Shotover it seemed we had a very similar view. And I am very glad Rob Long has added this:
If anything Slingo's claim for how tested this code is against the real world was even stronger than this. But tested for what? Weather over five days or climate (whatever that may be) over a century? As a programmer I know that if my unit tests are strong in one area they may be completely non-existent in another - with grave consequences for false positives for myself and others in regression testing down the line. Don't we need thirty to a hundred years to establish whether today's models are a good guide for policy makers? (The reverse not being true of course - a 55-year span in the past may already almost be enough to show they're not.)
As Martin A and Jonathan Jones were saying on a recent thread (partly in response to Richard Betts) we are among other things surely seeing genuine climate scientists, among whom Nic Lewis numbers Dame Julia, beginning to hedge their bets between tacit support for the alarm that has been responsible for so much of their funding and a future when that alarm is likely to look stupid indeed. Slingo was passionate about the models she's made her name by programming since the 70s and she was I think genuine. But as Shotover captured so well afterwards there are limits to candour when one's legacy and livelihood are at stake.
IMVHO I have no doubt that JS is an excellent scientist, more than worth her salt in the research world and mankind will eventually benefit from her passionate endeavors.
I am concerned that a difficult to rectify mistake has been made. The combination of a new young embryonic research unit - the Hadley Centre and Met Office might well have seemed logical and potentially productive, especially when the research unit delivered a mega USP - CAGW.
From then on the tail was wagging the dog. The "non-productive" PR machine started to earn more than the "productive" scientific product.
JS may well become known as the "Mother" of climate modelling and have no doubt I appreciate her ability, commitment and contribution to our understanding of how our world operates.
But all the appreciation, the understanding, the admiration dissolves into questions on the sight of the Met Office's Chief Scientist standing out in the pouring rain in order to deliver a TV "moment". Any true scientist would have told the PR mob where to get off and in spades.#
Split the MO and Hadley, they are by definition incompatible!
As smarter people than me have pointed out to exhaustion, the fact that a model can fit past events to perfection does not give in any predictive power whatsoever.
If it did, I would be a multi-millionaire, since I have developed simple models which can fit the football results of past seasons with extraordinary precision. As predictors of the future, they simply cause me to have pockets to let.
Slingo doesn't get it ... it doesn't matter how complex the models, how precisely they churn numbers, how exquisite the handling of physics is ... the output is rubbish! That means that the models are rubbish ... the UKMet and BoM (they use the same God-awful models) cannot get medium term weather forecasts right, not even close; they have Buckley's chance of forecasting climate no matter how smart they think they deal with the physics.
Sep 5, 2014 at 1:05 AM | Green Sand
....
Phantom pregnancy!
Green Sand actual those that raise to top in such organisation are often not much better administrators then scientists, who like the 'meetings , reviews etc that the job entails rather than actually doing science.
I was there as well and I agree with the reports above. There were only three questions - from Peter Gill about whether the models included periodicities, from Piers Corby which was more of a good natured speech and sales pitch for his business, and from a young man asking the impressively original question, whether uncertainty is an excuse for inaction; maybe he works for the Grantham Institute. The last part of Julia Slingo's talk gave a nod to the impact of solar cycles on climate and weather via UV modulation of ozone, whose discovery she attributed to her ally Prof. Joanna Haigh at Imperial, although I can't help thinking of a few others not in the climate establishment who have written about this. She did talk about chaos, noting that it is not the same thing as randomness.
It would be very interesting to see a debate about GCMs between Julia Slingo and someone sceptical of them who actually understands how they work as well as she does. But I suppose that those who work on them are the people who really know how they work and nobody would become a professional climate modeller if they thought that GCMs were a waste of time. I agree with the comments above saying that improved accuracy of short term weather forecasts (measured in days), which in part she put down to increased resolution reducing the need for parameterisation, does not tell us anything about their predictive power over decades and centuries.
She did put a lot of weight on the frequently repeated argument that if you think climate models are wrong then the implication is that you think that certain laws of physics are wrong. In reflecting on this, and I'm not sure if this is a good argument or not, it seems to me that the models presuppose an equilibrium which is moved by changes in exogenous variables such as atmospheric CO2 concentration. In fact, as Roy Clark write in his book, climate is not in equilibrium at any period. It displays cyclical variation over numerous periods of changing periodicity and also secular changes. So Peter Gill's periodicity question is worth exploring further in relation to longer periodic changes than just the PDO, which is what he and she ended up addressing in their dialogue. Do the models explain glacial/interglacial cycles, by far the biggest impact periodicity in the climate of the past 2.8m years? In my view, there are two places to look for causes of cyclical climate change: feedback & damping in the earth's energy transport system, which is what the models seem to concentrate on, and the consequences of being part of a solar system with eccentric cyclical planetary orbits, which they don't seem to pay attention to. This is really another way of saying what someone said above, that if the models can't model the past, they can't predict the future and I get the impression that they simply do not reflect theories explaining actual historic climate change, not just glacial/interglacial changes but also the numerous changes since the last glacial maximum (e.g. LIA, MWP, Younger Dryas etc).
Going back to the start of the lecture, as reported by the Colonel, she predicates her alarmism on the fact that the "Uncharted territory" in the title of the lecture is the increase in atmospheric CO2 concentration above levels seen in ice cores for 800,000 years. There has been plenty written about whether the ice core records are accurate or not and no resolution in my view of whether that data is reliable.
Looking forward to more fireworks than there were at this lecture at Mann's and Cook's Bristol lectures in a few weeks!
Yes, the 'Consensus' isn't entirely made up of the the dishonest, the ideological and the gulllible. There are those whose major interest is their grants, who must be seen to agree.
If the climate section of the Met Office was a private company selling climate advice, it would have either gone out of business by now or be entirely supported by zealous hopefuls who want there to be a CO2 crisis. As far as I can see, their advice has been seriously misleading, and therefore it is reasonable to suppose it may well have been seriously harmful to society. If a computer game had been taken seriously by governments as a basis for drastic decisions, and was subsequently shown to be misleading, would we applaud the game's designer for her enthusiasm? Or would we say, gosh we were a bit silly in taking it all so seriously? The general atmosphere of political panic has not helped with the pursuit of calm, considered assessments. Nor have the climate models, loaded as they are with the goal of illustrating a warming effect of CO2. That being where the money is. Government money, i.e. money no longer in the hands of private citizens to spend as they wish in the marketplace.
"...if you think climate models are wrong then the implication is that you think that certain laws of physics are wrong. In reflecting on this, and I'm not sure if this is a good argument or not.."
Well I know and it is entirely dishonest to suggest any such thing...
Firstly the laws actually applied in the model have to be really grossly simplified in order to get a result. Whether these simplifications are adequate is a matter of testing against real life. If you don't or cannot do that then you can have zero confdence in them.
Secondly there are a lot of physical systems that we don't yet have any law for and these are replaced in the models by abject guesswork, parameterisations based on other (much simpler) models or contradictory fiddle factors, such as adjusting highly uncertain aerosol reactions to produce cooling purely in order to match the 20th century observations while maintaing a politically expedient but unlikely high CO2 sensitivity.
Thirdly the climate establishment are quite happy to jettison Henry's law and declare a warming ocean as a net carbon sink rather than a source and also to ignore all the laws of thermodynamics in proposing that the missing heat can be stored in the ocean while not being detecting passing throught the top 700m so they needn't lecture the rest of us on following the basic physics.
Lastly, engineers, who laid the mathematical and numerical analysis foundations for most of these climate codes would never even think about presenting model results as true if the real world said the opposite: To do so is either stupid or dishonest, and for engineers also damned dangerous. (climateers get away with it only because lives don't depend on their results). There are some things we can model really well, some things reasonably well, some things well enough for a decision to be made between an upper and lower bound but many, many things cannot be modelled at all well yet. Modelling the climate of the Earth is far and away the most difficult task anyone can attempt in modelling and it is doomed to failure from the outset. All of the comparisons between real world and models have shown inadequacy of the models both spatially and temporally except for some really dishonest comparisons I've seen between highly seeded local maps with the reality that it was simulated from; ie akin to walking into the exam with the results paper in your hand and not at all useful for prediction of such a complex, non-linear, chaotic system.
The facts as they stand show that, regardless of the amount of lines of code and the amount of government lackeys tweaking them, they are inadequate for policy and every climate modeler must already know this lest they are fooling themselves.
Re JamesG, sorry, I wasn't clear: when I said "and I'm not sure if this is a good argument or not" I meant my own argument in the succeeding sentences about why the models are wrong, not Julia Slingo's about why they must be right.
She did refer briefly to simplification of some theories and to parameterisation. By saying that I am not trying to defend the GCMs whose output seems to have proved so wrong, merely to report to those who weren't there that Julia Slingo did not completely ignore some of the points made by critics, such as the points made by JamesG above.
Nic Lewis - I'd appreciate a few lines from you detailing what, precisely, you find impressive about climate models? Thanks.
JamesG:
Agreed. Thank you for the many valuable thoughts from you, Guy and others.
not banned yet: I can't speak for Nic but I remain impressed by the climate models. I will explain in what way when I write, as much as I can, my fuller account of, and reflections on, Dr Slingo's talk. But I'm far from impressed by the use to which GCMs have been put since 1988 in the arguments for radical and ruinous energy policies. They're nothing like fit for that particular purpose. On that I'm sure we agree. But we don't have to be totally negative either. More later.
Sep 5, 2014 at 10:47 AM | not banned yet
;) There's these straight lines of code, miles of it. It makes 'runs' and 'loops' back on itself ... the whirring sound is very impressive ... the flashing LEDs too. /sarc
nby (Sep 4 at 12:48 PM):
There's no mention of either video or transcript in the IoP's original page for the event or, as far as I can see, for any of its events. I'd be grateful to hear if anyone knows different.