Bishop Hill Ducking, diving, dodging, weaving
Jul 3, 2013 Climate: Models There was an interesting written answer in the House of Lords yesterday, with Baroness Verma singularly failing to answer a question about falsifying climate models.
Lord Donoughue (Labour): To ask Her Majesty’s Government whether the Met Office has set a date by which, in the event of no further increase in global temperatures, it would reassess the validity of its general circulation models.
Baroness Verma (Whip, House of Lords; Conservative): General circulation models developed by the Met Office are continually reassessed against observations and compared against international climate models through workshops and peer reviewed publications. The validity of general circulation modelling has been established for over four decades, as evident in the peer-reviewed literature. Such models are further developed in light of improvements in scientific understanding of the climate system and technical advances in computing capability.
Short term fluctuations in global temperature do not invalidate general circulation models, or determine timelines for model development. The long term projection remains that the underlying warming trend will continue in response to continuing increases in atmospheric concentrations of greenhouse gases.
It was Baroness Verma who was on the receiving end of Lord Donoughue's questions on surface temperature trends, and her ability to duck, dive, dodge and weave, often at the same time, was something to behold. It was many, many weeks before Lord D was able to pin her down to an answer.
Here we go again.
Reader Comments (109)
"Such models are further developed in light of improvements in scientific understanding of the climate system and technical advances in computing capability."
In which case the "improvements" have been ineffective since the models continue to diverge from reality by factors of 2-3.
The whole response is gobsmackingly rhetorical, but the icing on the cake is reaffirmation that its all caused by CO2 despite this being increasingly shown to be unlikely if not actually impossible.
These zealots need to be challenged at every opportunity. Roll on the revolution.
Judging by Entropic Man's recent change in writing style (plus the added insults) I would hazard a guess that he has morphed into Bit Bucket. Or is that the other way round?
It has?
"The validity of general circulation modelling has been established for over four decades ..."
This is a version of the argument Mosher ran recently at Judith Curry's, along the lines that the models aren't perfect, but they are the best we have, so we should go with them.
When I pointed out that not very long ago dirty, rusty barber's tools were the best we had for performing surgery, but that was hardly a reason to go with them, he went quiet.
It's a pathetic argument.
Given Michael Gove's justified drive to introduce clear and concise responses from civil servants it appears the good Baroness needs to spend a considerable time on the naughty step before being consigned to the dustbin of politics unless she mends her ways.
What I would give to see Michael Gove in charge of the DECC.
Damn, I hate it when I agree with Entropic Man, but in this case I think it was a stupid question, and it received a reasonable answer.
Of course the Met Office will continually re-assess the model performance, and it will improve. I don't think there will ever be a "day" on which it is decided that the models are wrong. Each minor improvement over the next 20 years will make the models perform closer to reality, and eventually they will provide much more accurate forecasts, and no doubt when the models in 2050 are compared with those of today, it will be obvious what was wrong with them, but it isn't currently known.
Entropic Man: "A model is as good as the information available at the time."
That statement is ambiguous. Are you referring to "information" in the form of data that would inform initial conditions for the model or are you describing the understanding of the physical processes that the model tries to describe?
If the former, that has not changed with time. Our knowledge of the initial conditions in, say, 1970, have not changed in any substantive way.
If the latter, well just 12 years ago when the IPCC TAR was published in 2001 we were confidently told by HMG, led by our Chief scientist and informed by the Hadely Centre and Met Office (and others) that the GCM's incorporated all natural effects and they could not explain the short warming trend during the 1980's and 1990's without incorporating anthropogenic greenhouse gases. They also said they could only explain the cooling from the 1940's to the 1970's by incorporating the effect of anthropogenic sulphates and aerosols.
It seems the modellers lacked imagination, because the hiatus in upward temperature trend of the last 17 years was not predicted by any models (as late as 2007 Trenberth and NOAA were quoted as saying a ten year plateau in the warming trend was not found in the model runs), but now they are informing us that natural causes have temporarily overcome the anthropogenic greenhouse gas effects. The Met Office is on record as saying that global warming will return with a vengeance from 2014. Good, not long to wait then.
What are these natural causes that they now so blithely accept, when just 12 years ago they claimed that all natural causes were accounted for? Perhaps Lord Donaghue could ask Baroness Verma. And when are they going to model clouds, or remove the unproven positive feedback mechanism of water vapour in the models, when all the current evidence suggests water vapour feedback may be negative? Perhaps those questions should be asked: explicit questions about what is in the physical model and what is not?
We have sufficent data to show the models are running consistently too hot compared to the temperature record and that there are clearly natural physical processes we do not understand that are not accounted for in the models. I have recently started to read a few of the backgournd papers published describing some of the limitations of the models. Its interesting, because the models are always promoted as being based on known physics. To some extent that is true - but my cursory reading suggests that is also a very misleading statement. The "known physics" the models are based on seems to refer primarily to the large scale circulation processes, based on coriolis and so forth, so they get the general pattern of ocean and atmospheric circulation patterns correct. Whoopee Doo! That physics was included in the undergraduate Oceanography degree I studied back in the early 1980's and it is not really relevent to climate evolution over the next 100 years. Then we have spectral absorption, basic physics of greenhouse gases in the atmosphere - again pretty uncontroversial - this stuff has been known and measured for a long time.
But then we come to the problematic physics, the physics which actually affects our ability to model climate change. Water vapour feed back - is it positive or negative? Clouds? Phase changes between precipitation as rain or snow (albedo sensitivity). And the natural processes: volcanos are probably ok, but solar effects are not as well understood. Nor are feedbacks with the biosphere and some of the ocean feedbacks. Remember the CO2 atmospheric flux with the biosphere is substantially larger than any other atmosphere exchange flux system. The biosphere flux is estimated at 120 Gt /yr (compare with human emissions from fossilf fuels of 6.4 Gt/yr, all figures quoted from IPCC AR4 2007). The models may have an exchange with the biosphere built in (at this point I am unsure) but do they model the temperature/CO2/water vapour response of the biosphere? Not as far as I am aware, because I don't think anyone understands it well enough to model it.
All the evidence I have seen currently says the GCM's model the ocean atmosphere circulation (coriolis plus land masses) pretty well, plus some spectral physics. Volcanos are introduced as spike events. The rest of the model fit is obtained by tweaking two curves - CO2/water vapour positive feedback and, to balance it all out, a negative feedback sulphate/aerosol curve. So the models fit the trends by tweaking two opposite effect feedback responses, both of which are largely arbitrary and empirical, not physical (in the case of aerosol repsonse and water vapour feedback) and both of which are assumed to be anthropogenic.
So, its man wot did it, simples. Except I think the temperature data of the last 17 years shows the models to have no predictive power currently and I suspect its going to get much, much worse unless by some stroke of luck temperature starts to climb again. Then it will taken as proof the models were right. That of course is not science - a model cannot be "proved" correct. However, if the temperature falls, the models will be shown to have been wrong, as I think has already been demonstrated by their failure to predict real world temperatures over as short a period as 17 years without significant divergence.
Well said that man!
I knew that was the reply Lord Donoughue was going to get because it was pretty much the same answer I got back in 2006 when I asked a similar question. Version numbers - a software developer's way of never having to admit they got it wrong.
"Mr Smith, in your statement you said that on the 3rd of January you were with your girl friend all day.'
"Ah, no, that was just the output from miAlibi which works out where I was from my usual patterns. After additional data and an update of the software I can assure you I was watching TV with my mate John."
EM
[1] How much alcohol did you consume last night? The difference in tone from your normal postings plus your lack of logic and the insults suggest it was well above the legal limit for driving.
[2] In the above, you come across as someone who is lecturing a class of 13 year olds, not people who know much more about the subject (mathematical/simulation modelling) than you do yourself.. As I have said previously, the things you say suggest strongly that you have never been involved in programming and validating simulation models of any significance.
[3] The Met Office has denied it does the tuning that you describe above. Yet it is clear (from 'paramaterisation' if nothing else) that they do it. You can't have it both ways. If such tuning is done using historical data, then testing that the model can reproduce the same historical data does not validate the model.
[4] Neither I nor Lord Donoughue, nor anyone else so far as I have noticed, has claimed that "no further increase in climate" (sic) will occur. Where did you get that idea? Did it just come into your head?
[5] "Making unsubstantiated claims that the Civil Service deliberately lies to its government is not usually a sign of a fully sane individual" The evidence of the MO's bullshit presented to the govt as scientific fact is there for all to see. As I said before, saying that someone you are discussing things with its mad is not a convincing argument.
Judging by Entropic Man's recent change in writing style (plus the added insults) I would hazard a guess that he has morphed into Bit Bucket. Or is that the other way round?
Jul 4, 2013 at 8:32 AM David Porter
Sudden apparent changes of character are often explainable by the consumption of ethanol. Clearly he was pissed last night. See my post above. Nothing to do with Bitbucket.
The question is how short is "short term". In the peer reviewed literature Ben Santer says 17 years of flatlining is enough. NOAA said 15 years:
Source: http://www1.ncdc.noaa.gov/pub/data/cmb/bams-sotc/climate-assessment-2008-lo-rez.pdf
Of course the Met office did not expect any flatlining in temperature at all. They were expecting a parabolic increase. Natural variation was supposed to be in decline and manmade warming was now dominant we were told. Observations prove the exact opposite. All the rest is excuses.
steveta_uk - you assert that models will improve in future. Do you have any evidence that the models have improved in the past? Did you see this paper at Judith Curry's? :
//
CLIMATE CHANGE
What Are Climate Models Missing?
Bjorn Stevens1, Sandrine Bony2
Extract-
The increase in complexity has greatly expanded the scope of questions to which General Circulation Models (GCMs) can be applied. Yet, it has had relatively little impact on key uncertainties that emerged in early studies with less comprehensive models. These uncertainties include the equilibrium climate sensitivity (that is, the global warming associated with a doubling of atmospheric carbon dioxide), arctic amplification of temperature changes, and regional precipitation responses. Rather than reducing biases stemming from an inadequate representation of basic processes, additional complexity has multiplied the ways in which these biases introduce uncertainties in climate simulations.
http://judithcurry.com/2013/06/16/what-are-climate-models-missing/
//
Re: long term hindcasting this one looks relevant:
//
Climate of the last millennium:ensemble consistency of simulations and reconstructions
O. Bothe, J. H. Jungclaus, D. Zanchettin, and E. Zorita
From the abstract-
The lack of consistency found in our analyses implies that, on the basis of the studied data sets, no status of truth can be assumed for climate evolutions on the considered spatial and temporal scales and, thus, assessing the accuracy of reconstructions and simulations is so far of limited feasibility
http://www.clim-past-discuss.net/8/2409/2012/cpd-8-2409-2012.pdf
//
Couple of bits of trivia to lighten the mood. Wasn't Lord Donoughue the adviser on Yes Minister, (along with Lady Falkender). And is our host a fan of Bach? The title reminds me of the chorale "Weinen Klagen Sorgen Zagen", which was translated for me as Weeping, wailing, sobbing, sighing.
Here's how one experienced climate scientist might have answered Lord Donoughue's question:
//
SPIEGEL: How long will it still be possible to reconcile such a pause in global warming with established climate forecasts?
Storch: If things continue as they have been, in five years, at the latest, we will need to acknowledge that something is fundamentally wrong with our climate models. A 20-year pause in global warming does not occur in a single modeled scenario. But even today, we are finding it very difficult to reconcile actual temperature trends with our expectations.
//
http://www.spiegel.de/international/world/interview-hans-von-storch-on-problems-with-climate-change-models-a-906721.html
http://www.bishop-hill.net/blog/2013/6/21/von-storch-on-the-pause.html
> General circulation models developed by the Met Office are continually
> reassessed against observations and compared against international
> climate models through workshops and peer reviewed publications
Are the 'peer reviewers' the people who are creating the 'international climate models' that have shown to be all wrong....
http://www.drroyspencer.com/2013/06/still-epic-fail-73-climate-models-vs-measurements-running-5-year-means/
Not much of a verification method.
So the specific answer is No.
In which case if there are no circumstances under which the Met Office will consider there theories falsified the next question is "Is it possible for the met Office to consider itself a scientific organisation"
To which, shorn of woffle, the inevitable answer is No.
JamesG: "Of course the Met office did not expect any flatlining in temperature at all. They were expecting a parabolic increase."
The first sentence is true, but the second is not. They are assuming an exponential increase in CO2 emissions, but a logarithmic forcing response curve, so the models effectively have linear temperature increase trend coded in.
It was a vague question based on loose language.
The Baroness had every reason to avoid a yes/no answer that could be misinterpreted. Let's face it, this is a partisan and heated topic; it would be misinterpreted.
Perhaps the question that should have been asked was one that was intended to ascertain if the models could be falsified and if so what that means. A model not being right may just need tuning, after all.
How about, "After what period of time and by what amount of real-world deviation from the mean of runs of a model does the MET Office deem that the model's projections should not be used in policy development?”
If the MET Office thinks no model should be used in policy development then I would like to know that.
If the MET Office has a limit beyond which the model is only useful as a negative understanding of the climate then I would like to know that too.
But the validity, or otherwise, of a model is too vague to be meaningful. ‘Valid’ to do what?
I think it would be better to ask which physical effects are modelled using physics, which are modelled on empirical assumptions and which are not modelled at all.
Eg are clouds modelled using empirical assumptions?
In the met office model, is water vapour modelled as a positive feedback or a negative feedback?
M Courtney Jul 4, 2013 at 11:22 AM | - fair comment, but all those points you make are now available as follow up questions. Note the reply did not start with "No, because..."
IMO, now that the claim has been made over peer reviewed validation, there needs to be a request for the specific references and the reasons for their selection. A question asking for the Gov. response to peer reviewed literature which falsifies the models is also required.
Further to those, I think specific questions need to be asked over the v and v framework and methodology in use at the MO with reference to specific individual models and their evaluation criteria.
Etc etc
In 2001 government policies were based on predictions from untested models (GCM's). 12 years later the models are clearly diverging from reality, running too hot. So now we are going to base continuing those policy decisions on revised models. So what is it that convinces us the models now have predictive ability? Nothing. Instead, we have to wait again. How long do we have to wait? 17 years? And while we are trashing our economy tilting at windmills, what we know absolutely for sure is that even if AGW theory is correct, our "noble gesture" won't make any measurable difference whatsoever to global temperatures in the future because of the growth in output from china and india.
For those who think tuning of models is science, not art, try the opening sentences of Mauritsen et al (2012) on the topic of model tuning:
"During a development stage global climate models have their properties adjusted or tuned in various ways to best match the known state of the Earth's climate system. These desired properties are observables, such as the radiation balance at the top of the atmosphere, the global mean temperature, sea ice, clouds and wind fields. The tuning is typically performed by adjusting uncertain, or even non-observable, parameters related to processes not explicitly represented at the model grid resolution. The practice of climate model tuning has seen an increasing level of attention because key model properties, such as climate sensitivity, have been shown to depend on frequently used tuning parameters"
Short term fluctuations in global temperature do not invalidate general circulation models.
Here it appears to be admitted that GCMs cannot follow short term fluctuations, although ‘short term’ is not defined. What are the implications of this?
I do not know what timelines for model development are. Does anyone?
[…] the […] warming trend will continue in response to continuing increases in atmospheric concentrations of greenhouse gases.
The Met Office leaves no room for doubt. I fear there is a rude awakening on the horizon.
thinkingscientist
To further expand the point - this is the problem. They aren't using an empircally characterised formula for effective forcing - i.e. how much of the energy gets turned into heat for frequencies of that wavelength when the absorbing surface is in contact with an atmosphere. IR heating is different that UV heating due to the surface depth penetration reduction (essentially a skin effect) and surface conduction with the air is going to be different than with a vacuum. The effective temperature relationship may be less (the percentage will be shown by test)
Also I don't see many model results where a simple CO2-on CO2 off comparison is clearly shown. I trust they are performed but you don't see many of them.
The fundamental issue to me is that modellers are theorists not scientists. A scientist has to use real data to match against theory and if it doesn't match you throw out the theory. You also make sure you have as much tested fundamental data and understanding before you make your simulations more complicated.
It seems that modellers try this for everything but CO2 forcing. That stinks of confirmation bias, whether intended or not.
Also I don't see many model results where a simple CO2-on CO2 off comparison is clearly shown. I trust they are performed but you don't see many of them.
Jul 4, 2013 at 12:49 PM | U Micky H Corbett
As I said to another commenter "They would show that, wouldn't they". If you have models programmed by someone who believes CO2 causes global warming, and whose job is to provide evidence via models of such warming, it would be very surprising if switching off his "CO2 = warming" module did not reduce the predicted temperature increase.
Richard Betts of the Met Office referred here some time ago to a paper reporting just that. ( FOUND IT: http://www.image.ucar.edu/idag/Papers/Stott_HadGEM.pdf )
Don't insult EM by suggesting that he was "tired and emotional" last night.
Much more likely he forgot to take his meds :-)
He deserves our sympathy, not derision.
DK - you are so right. I feel ashamed of myself.
"I do not know what timelines for model development are. Does anyone?"
Well I am pretty certain that the Met Office has never said "our climate models are crap on a timescale of anything less than 30 (say) years". They believe their models have skill at all timscales from next week's weather to next century's climate.
Murry Salby's Hamburg talk is full of gems of great significance to the field of climate science. In discussing climate models he says:
"In blue is the forecast evolution of global temperature averaged over two dozen models of the IPCC. The mean is free of decadal variability, which must therefore be incoherent between the models. (slide 58.47) It follows that, on those timescales, the models have no predictive skill. Notice this is the same timescale responsible for almost all of the twentieth century warming."
(my emphasis)
Yet the Met Office claims that agreement with climate models produced elsewhere validates the Met Office models. What Professor Salby has made clear is that there is no agreement between climate models on the decadal timescale, which falsifies the MO's claim.
thinkingscientist
You have made some brilliant posts on this thread and it seems like nitpicking to find fault with any of it but I am afraid I must ^.^
First I have to take issue with the following:
" basic physics of greenhouse gases in the atmosphere - again pretty uncontroversial - this stuff has been known and measured for a long time."
Is not the whole climate debate about the fact that what we thought we knew about greenhouse gases does not seem to be giving us the right answers? For me the problem is quantifying the basic effect and then understanding exactly the logarithmic effect, at what level does CO2 cease to have any real effect on temperature? Until these questions are answered then we can not really say we understand the greenhouse effect.
Second
"I think it would be better to ask which physical effects are modelled using physics, which are modelled on empirical assumptions and which are not modelled at all."
To be honest it does not matter what is included and what is not included, the point is that our knowledge of climate science is so poor that models should be regarded as interesting but not yet useful.
Claims to be able to predict the future based on climate models is fraudulent.
Martin, what a great line that was, to the effect that MO models are backed up by other models, it does not matter that they are not backed up by empirical observation, that is belly laughing funny!
"Claims to be able to predict the future based on climate models is fraudulent."
Never a truer word spoken.
Martin A
Thanks for the link to that modelling paper. I think I spoke too soon. Their version of modelling "natural" variations was really an exercise in back-slapping and circular logic. It seems that these models fix a value of sensitivity and then parameterise / normalise to this. Taking out CO2 is really taking out CO2 effects in "model land" not reality. I think this was your point as well.
It's funny, nowhere in the references or in the paper (and I could be wrong) do they actually show the phenomenological relationship between radiative forcing and effective temperature. It seems to be assumed in the climate sensitivity.
When that graph was first posted on WUWT, comparing the multitude (73?) of models "projections" with actual temperatures, there was a huge debate about the use of a "mean" of the models.
In all the uproar, there was a simple comment which seemed very significant.
It was to the effect that, if the science is "settled", why is there such huge divergence between the models?
How can that divergence be explained other than by the use of different assumptions, different weightings and so forth? That would show that the science is anything but settled.
Dung: "You have made some brilliant posts on this thread", Thank you :-)
"First I have to take issue with the following:
" basic physics of greenhouse gases in the atmosphere - again pretty uncontroversial - this stuff has been known and measured for a long time."
based on your comment highlighting this, and re-reading it, I would agree with you. I think the basic physics of absorption/emission of CO2 is pretty uncontroverisal, however I am not sure its behaviour in the atmosphere at large is as we expect. Fred Singer, that great (and sceptical) atmospheric physicist, has always acknowledged the greenhouse properties of CO2, but always qualified it by describing the already saturated bands and the logarithmic response. He has always stated a doubling in CO2 would only correspond to 0.6 degC, and I suspect he is probably right. I pride myself on being ok at physics, and particularly the "back of the envelope" stuff, but Singer is in another league, both experience-wise and intellectually. I defer to his position and think he is probably right.
On your criticism of my comment "I think it would be better to ask which physical effects are modelled using physics, which are modelled on empirical assumptions and which are not modelled at all." to which you say
"To be honest it does not matter what is included and what is not included, the point is that our knowledge of climate science is so poor that models should be regarded as interesting but not yet useful. Claims to be able to predict the future based on climate models is fraudulent."
I would say that by questioning and highlighting the important areas of physics where the models have no clue, one is shining a light onto the inadequacies and therefore undermining the absurd confidence placed on their failed future predictions.
Dung
Watch that you dont slip. You stand on a slope with reasonable doubt at the top and David Icke's green aliens at the bottom.
MikeH
Of course the science isn't settled. It never is.
Your mistake is to assume that because scientists disagree, the general conclusion of the science is wrong. Disagreement is the grindstone on which scientific ideas are honed.
Green Sand
You use mental models of the world to predict everything from train times to the behaviour of your fellow man. How does using a computer to help with the arithmetic automatically invalidate the concept?
Martin A
The question of the role of CO2 in climate models is easily resolved. Surely among all these engineers and experts is someone who can write and publish a physical model which accurately hindcast the last 130 years of climate without including the effect of changing CO2. That would be a more effective persuader than any amount of BS on BH.
Relax, I was pulling your leg. When you stuck it out and waggled it the temptation was irresistable. If I had a pound for every time a climate sceptic doubted my sanity I would be able to buy more MPs than Unison.
Speaking of which, I wonder what coin you used to buy Lord Donohue. It probably wasn't money. Would a passed-over politician be bought by the promise of increased influence?
Entropic Man
You misjudge me sir!
I do not have reasonable doubt, I have an absolute certainty that scientists do not know enough to predict future climate.
David Icke on the other hand has an absolute certainty that you are a green bug eyed monster, hmmm maybe I should give him more credit?
1. as the ipcc itself don't expect anything outside of natural variability for two or three decades, it'd be easy to build a model based on coefficients given to oceanic oscillations, with zero contribution from CO2
the plausibility of the model cannot be linked to its ability to hindcast. Anybody can fit an appropriate Taylor series to any climatic index
2. behaviour in the lab doesn't mean behaviour outside the lab, and am not referring to. white-coated nerds becoming party animals
every kitchen the world over is a little lab where every day warm air rises above cold air, and there is a positive gradient between the floor and the ceiling. This doesn't mean Mt Everest is very hot on top.
Omnologos
Respect
Dung
No scientist would ever claim absolute certainty. It is an attribute usually associated with priests, politicians and snake oil salesmen.
Omnologos.
Read my post. I am not thinking of a Taylor series, or any other mathematical function purely designed to mimic the shape of a graph. I am talking about a simulation which uses the mathematical description of physical processes to describe the evolution of a physical system from a set of starting conditions.
Given the availability of code for all the major models, and the wide availability of the IT skills to use it, I would have expected a valid physical model based on the sceptic interpretation of climate physics to have emerged long before now. If it is easy, where is it?
The lack of such a model is a major obstacle to the acceptance of any non-CO2 based climatology; an example of the way in which the sceptic case tends to be based on doubt, rather than positive evidence that it provides an alternative paradigm.
PS Let me remind you how hindcasting is done. You do not start now and work back. You start the simulation with a set of starting observations, perhaps for 1950, and let the simulation run. The output is then compared with the observations for the years concerned.
PPS What has your kitchen got to do with it? The physics of energy and heat flow in a small closed room has very limited validity as a model of the troposphere.
Entropic Man
Absolute certainty of total knowledge on a subject may well be the province of priests, politicians and snake oil salesmen but absolute certainty of ignorance just takes common sense ^.^
EM, the roll you are on takes you very close to the edge of reason and decency. I may have misunderstood you, but doubt that. Your impugning the integrity of Lord Donahue for asking questions of Baroness Verma does, at the very least, betray your own shallow understanding of morality and ethics, let alone science.
The current models cannot hindcast based on CO2 alone; they need to turn up the aerosol knob in order to keep the CO2 sensitivity to 3 while maintaining the alarmists preferred runaway warming scenario. Easy to do because of the massive error bars in the inputs. However it doesn't matter what they do they cannot replicate the hump in the middle of the 20th century with CO2 and aerosols as you'll see from the outputs. As it happens Hadley already proved that you can replicate the 20th century (including the hump) a lot easier with a higher solar sensitivity.
Now the trouble is that the current CO2 theory needs an amplification of 3+ to cause any alarm. This is achieved via positive feedbacks. The solar amplification required is around 13. Hence neither is that plausible and neither can actually replicate the current flattening; solar amplification makes the model run too cool and CO2 amplification makes the model run hot. Hence to attempt to say there is no alternative to CO2 is lieing by omission; the reality is there is no credible simulation anywhere.
This is not that surprising because climate scientists do not actually know where natural variation comes form. Clearly though it is potentially huge because it caused ice ages, hothouse conditions and abrupt climatic changes well before man became dominant. Hence the idea that there is a sort of natural stasis that is only interrupted by mans CO2 is mere philosphy that is not supported by the facts. Given the history of the planet, 0.5K per 100 years is nothing to even bother about. In a more enlightened age we'd marvel at how stable the planet currently is.
As for the ability of models to simulate. I can get displacement and stress results near 1% compared to theory. If i add frictional contact; I need to use a fudge factor that gives me something between a stick and slip state. A tribologist would say my guesswork bears little or no relation to reality and it is dangerous to rely on it. Alas there are a lot of things we just cannot model and a wise man admits it. For climate, much more work needs to be done to isolate the causes of natural variation before we can progress. As Trenberth says, you cannot just keep calling it 'noise'. Alas as Spencer tells us, no money is available for the research into natural variation. Is my guess as good as yours? If you are paid big amounts of money to produce an alarmist model and I need to do my more sane model in my spare time is that fair? Interestingly Tsonis and swanson may have the most cogent mathematical simulation of the 20th century but it depends on a big dollop of cyclical natural variation and a very steady background rate of warming that may be natural or manmade but is certainly not going parabolic.
Hi James,
Very interesting post. Do you have any further information, a reference or a published example relating to:
"As it happens Hadley already proved that you can replicate the 20th century (including the hump) a lot easier with a higher solar sensitivity."
Regarding the CO2 amplification, that is the multiplier 3-4 that I was aware of, I was unaware of the 13+ amplification factor to make solar fit, and I would be interested in finding out how good the fit is with just solar and an amplification factor only. Do you have any referneces or published examples you can link me to?
As an interesting parallel, in reservoir modelling for hydrocarbon reservoirs and modelling oil production, generally there are a few direct measurements of rock permeability available. We think we understand quite well how to upscale these measurements to the coarser grid cell size used for simulation and we have reasonable models to stochastically simulate the grid cells away from the data points (accepting there is non-uniqeness in this approach, so we need many realisations). Yet despite all this, in reservoir modelling it is invariably the case that a permeability multiplier is required (anything from 1+ to 5+ or more) in order to obtain a match between the model predictions and actual well production measurements.
JamesG
Brilliant post mate, right on the money, all the research needs to go on for a long time before we can understand and maybe predict climate. However until that time, scientists should (as you said) be honest and admit that they can not model climate yet thus taking away the politicians' justification for robbing us blind.
Entropic Man
"Speaking of which, I wonder what coin you used to buy Lord Donohue. It probably wasn't money. Would a passed-over politician be bought by the promise of increased influence?"
EM, your contributions on this post are valuable, when they make sensible arguments. It requires a certain spirit to persevere with a discussion in which most of the contributors are out of sympathy with you. So I hope you will carry on. I assume, however, that you are more interested in changing minds than in relieving your feelings. If I'm right, you will do better to refrain from attributing improper motives to those who disagree with you. It makes us think (on that point at least) that you have nothing better to urge.
This applies equally to those who abuse you, of course.