
GCMs and public policy



In the thread beneath the posting about the Chen and Tung paper, Richard Betts left a comment that I thought was interesting and worthy of further thought.
Bish, as always I am slightly bemused over why you think GCMs are so central to climate policy.
Everyone* agrees that the greenhouse effect is real, and that CO2 is a greenhouse gas.
Everyone* agrees that CO2 rise is anthropogenic
Everyone** agrees that we can't predict the long-term response of the climate to ongoing CO2 rise with great accuracy. It could be large, it could be small. We don't know. The old-style energy balance models got us this far. We can't be certain of large changes in future, but can't rule them out either.
So climate mitigation policy is a political judgement based on what policymakers think carries the greater risk in the future - decarbonising or not decarbonising.A primary aim of developing GCMs these days is to improve forecasts of regional climate on nearer-term timescales (seasons, year and a couple of decades) in order to inform contingency planning and adaptation (and also simply to increase understanding of the climate system by seeing how well forecasts based on current understanding stack up against observations, and then futher refining the models). Clearly, contingency planning and adaptation need to be done in the face of large uncertainty.
*OK so not quite everyone, but everyone who has thought about it to any reasonable extent
**Apart from a few who think that observations of a decade or three of small forcing can be extrapolated to indicate the response to long-term larger forcing with confidence.
So, let me try to explain why I think GCMs are so important to the policy debate.
Let us start by considering climate sensitivity. As readers here know, the official IPCC position on climate sensitivity is largely based on the GCMs. This time round we have had some minor concessions to observational estimates, but a significant proportion of the probability density of the observational studies remains outwith the IPCC's likely range of 1.5-4.5°C. Proponents of GCMs might counter that the upper end of the GCMs are ignored too, but I would suggest that one should conclude that an ECS of 5-6°C in the light of temperature history.
Estimates of climate sensitivity - and therefore in practice GCM estimates of climate sensitivity - directly inform estimates of the social cost of carbon. So when people like Chris Hope are arguing for a carbon tax of $100/tCO2, this is a function of GCMs. I recall, I hope correctly, that Chris suggested a figure of $18/tCO2 if one used an ECS of 1.6, in line with observational estimates. This matters of course, because the policy response, if any, to an $18 problem is significantly different to that for a $100 problem.
Wherever we look in the interactions between scientists and politicians on climate questions, we see an emphasis on catastrophe. We see no confessions of ignorance, but only occasional reference to uncertainties. Here's some notes of Tim Palmer addressing the All-Party Climate Change Group:
With the amount of carbon dioxide already in the atmosphere, future emissions will need to be reduced by half to that of historical emissions to limit global average temperature rise to 2°C. However, if emissions are not curbed (under the business as usual scenario), the amount of carbon dioxide in the atmosphere will be three times the historical emissions and the temperatures might rise up to 4°C.
And on the other hand they might not. This idea does not, however, seem to have been put forward for consideration.
Readers might also wonder what explanations were given to our political masters on the credibility of the GCMs. Here's what Palmer said:
Climate models are only flawed only if the basic principles of physics are, but they can be improved. Many components of the climate system could be better quantified and therefore allow for greater parameterisation in the models to make the models more accurate. Additionally increasing the resolution of models to allow them to model processes at a finer scale, again increasing the accuracy of the results. However, advances in computing technologies would be needed to perform all the necessary calculations. However, although the accuracy of predictions could be improved, the underlying processes of the models are accurate.
Apart from the transport of heat to the deep ocean, if Friday's paper from Chen and Tung is to be believed.
You can see that policymakers are getting a thoroughly biased picture of what GCMs can do and whether they are reliable or not. They are also getting a thoroughly biased picture of the cost of climate change based on the output of those GCMs. They are simply not being asked to consider the possibility that warming might be negligible or non-existent or that the models could be complete and utter junk. They are not told about the aerosol fudging or the GCMs' ongoing failures.
And this is just scratching the surface.
[BTW: Could commenters who like to amuse themselves by baiting Richard please refrain from so doing!]
Reader Comments (306)
"Jorgekafkazar has it right: GCMs are voodoo masks that the stupid hide behind." --Radical Rodent
Surely, RR-san, one could quibble about whether the person behind the mask is more or less stupid than the "patient" on the other side, particularly when the latter has let the witch doctor make repeated house calls. That's why I deliberately applied the word "ignorant" to the former. Climate scientists' knowledge is to the Earth as the dot over the 'i' in ignorant is to the entire OED.
Steve,
My question wasn't rhetorical. Why would you possibly think it was?
jorge - I would be a bit more charitable, and put it as the dot plus a dash. But, in principle, you are right. We don't even understand terrestrial ecosystems yet - and our ignorance of maritime ones is almost bottomless. FGS, we haven't even mapped the bottom of the sea. Then there are the aerial systems - all those critters, pollen, bacteria etc that fly around. We know bugger all about it all, after thousands of years of observation and study. And, that's not even counting what goes on outside our planet.
But, these hubris-riddled clowns try to tell us that they not only understand climate on Earth, they can predict it for hundreds of years into the future.
Dear oh dear. I just wish that they would apply their amazing methods to telling me which horse will win the Melbourne Cup in November.
Spence_UK:
Sure isn't. Can I have that as a wall plaque, please.
Judith Curry has an excellent post wrt GCM's. I quote her:
http://judithcurry.com/2014/08/24/the-50-50-argument/
Richard, there is quite a consistent story here from many people who have considerable experience of modelling (you, Dan Hughes, johanna etc). And of course our good friend Dr Koutsoyiannis has had peer-reviewed papers published outlining the same basic arguments. Yet it falls consistently on deaf ears. Frustrating!
Disappointed that no one from the MO or Government took the opportunity to clarify the situation.
Lord Beaverbrook,
Really? I guess you can be disappointed, but it's hard to see why you'd be surprised. Given what some on this site say about the Met Office and those who work there, I'm continually amazed that people from the Met Office actually comment here. I think it's commendable, but still surprising.
Spence_UK,
If you going to appeal to authority, then I think those who actually do climate modelling win.
aTTP: argument from incredulity? Yes, you could be right, there. I find it incredulous that someone who selects a name for themselves that infers a belief in the hard facts of physics can be so suckered into the whole AGW / ACC scam.
Let me make some predictions about global climate, based on the data that is freely available, and without any need to construct huge, number-crunching computer “models”, and ignoring the possibility of catastrophic events, such as volcanic activity, earthquake or meteor strike, etc.:
The average global temperature by the year 2020 will have dropped (there might even be a growing fear of another ice age);
The average global temperature by the year 2050 will have risen (possibly with a growing fear that we are all going to burn, yet again, if lessons from today have not been learned);
The average global temperature by the year 2100 will be less than 1°C different from the present. As almost everyone alive now will be dead by then, who is going to care?
As my first prediction is not too far away in time, you might get the opportunity to gloat should I be wrong. You will note that I do not consider the possibility of warming being substantial (more than 3°C); there is no historical precedent for that, so why should we fear it, now? The alternative that is offered, and the direction that you seem to want us to follow, is one that involves the impoverishment and massive de-population of most of the “Western” world. As the social turmoil that would generate would probably lead to the summary execution of most of the (perceived) proponents of the idea, and their supporters, do you feel that is a risk worth taking?
Radical,
I'm guessing you don't particularly like being called a conspiracy theorist?
Without a significant increase in volcanic activity or an asteroid strike, this is physically implausible.
Okay.
If we can sustain an energy imbalance of greater than 1 W/m^2 for the next 86 years and only warm by 1 degree or less, I think we'll have to rewrite the laws of physics. Anything's possible I guess. Some things are just extremely unlikely.
Jorgekafkazar: GCMs are voodoo masks that are made by the ignorant, but it is the stupid (i.e. wilfully ignorant) that hide behind them, in the hope of fooling those being deliberately kept ignorant.
Love your analogy of the level on knowledge we presently have, by the way.
aTTP:
Call me what you like – I have the hide of a rhino (or hippo, if you can understand that allusion to Reggie Perrin). Oh? Please explain. Better still, let’s wait and see. Oh, purleeeze… this argument is going round in circles. Where is the evidence that there is an energy imbalance greater than 1 W/m^2? Oh, yes… it’s all going into the deep, deep oceans. How terribly convenient. Meanwhile, we all shiver in one of the coldest Augusts on record (but that is only weather). Also, please read what I wrote – I did NOT state that it would necessarily be warmer! For one who purports to be scientific, your observations can be very selective....and Then There' s Physics:
"If you going to appeal to authority, then I think those who actually do climate modelling win."
Win what?
Radical,
Oh, I see, you're suggesting that it could even be cooler. Apologies for missing that. It is, however, an even sillier idea than the idea that we might only be 1 degree warmer than today if we continue to increase our emissions (which I assume we all agree is the outcome of doing nothing).
It doesn't have to be there today (although the evidence suggests it's around 0.5 W/m^2 today), but if we continue to increase our emissions and somehow (magically) manage to not warm at more that 0.1 degrees per decade (which is the maximum you seem to be suggesting is possible) it will be there in the future. That's physics.
"...I think we'll have to rewrite the laws of physics."
This is a nonsensical notion for a physicist.
There are several systems where mere energy budget considerations cannot serve in a predictive role. There are several systems where knowing the 'laws of physics' of the sub-components cannot help predict behaviour of the whole system. There are situations in the same systems where knowing everything still doesn't help because behaviour is randomly variable. i.e., you predict something, the system realizes something else.
Laws of physics' type reductionism appears to be applicable to the climate system at long timescales, i.e., multi-centennial and beyond. Even at those timescale, you may not be aware of this, variability of the system is the norm. The problem in front of the GCMs therefore is not recapitulating driven behaviour ('forced') but the variability. That's is what is important at 'policy-relevant' timescales.
More importantly, applicability of GCMs depends on a full understanding of climate. This should include accommodation of questioning existing circular frameworks ('feedbacks, forcings etc method of thinking). For instance, you seemingly allow for 'small' variations in the global surface temperature to be dismissed as mere 'internal variation'. I do not know of any 'physics' that has something called 'internal variation'. What is your formalized physics-based method for excluding multidecadal trends as 'noise' and accepting certain other trends on the same timescale as signal? How did you rule out multicentennial trends being mere noise? How did you decide where to draw the line? You may think a glacier melting is a big deal but it may be noise for the earth system.
Moreover, in blog debate, bringing up 'physics' as a deus ex machina for everything is a cop-out. There are several papers describing climatic variability as emergent unceasing noise - noise sustained by the energy fed into the system. Without quoting specific examples and patiently substantiating your position, it just reduces to trolling.
Shub,
And then there's energy conservation, but I'm not convinced that you believe in that or understand it. If we follow a high emission pathway (RCP8.5 for example) then we will increase the anthropogenic forcing by around 6 W/m^2 by 2100. If we warm by 1 degree or less, we will increase the outgoing flux by 3.7W/m^2 or less. Today we have an energy imbalance of around 0.5 W/m^2. So by 2100, an RCP8.5 emission pathway with less than one degree of warming would mean an energy imbalance of around 3 W/m^2. I think this is physically implausible. We would be accruing energy 6 times faster than we are today and yet warming slower than we are today. So, yes, if this were to happen, some kind of rewrite of the laws of physics may be required. Do you understand my point now?
Yup, you'd certainly know.
aTTP - as you are around and reading and commenting, please could you answer my question @ 11:54 above? Thank you.
I'm not relying on appeal to authority. Nice strawman though.
On the previous page I gave a 10-paragraph outline of my views, which is a generic explanation of the problems of modelling complex systems (and examples of how that explicitly applies to climate modelling).
I added a footnote that others with similar experience agree with me.
Allow me to explain basic critical thinking for you, and why my argument above is NOT an "appeal to authority".
An appeal to authority takes the form: "X is correct because person Y says so, and person Y is an authority". My argument did not take that form.
My claims are grounded in the longer argument that precedes it. The fact that, in addition to my more comprehensive discussion, I note that others agree with me, is not an appeal to authority as the core of my argument is founded on other claims.
I'd recommend you familiarise yourself a little better with logical fallacies and their use before throwing them around without thinking them through.
Spence_UK,
As are the arguments of those who do climate modelling.
Why? That would be no fun. To be honest, if I was thinking things through, I wouldn't be commenting here.
not banned yet,
The internet.
Thank you. Illuminating. I thought you might have been referencing the work of, for example, Bony and Stevens. Bye bye.
ATTP says" To be honest, if I was thinking things through, I wouldn't be commenting here" - I say that is the first thiughtful think you have written. Especially given your claim the (and I paraphrase for brevity`s sake) surface temp record played no part inthe construction of the GCM's since they are all just physics and so we can ignore the current temp record.
aTTP: you still cleave to the threat of our increasing emissions. Why are you so enamoured with the idea that increasing CO2 will raise temperatures? Look at a simple fact: CO2 levels continue to rise; global temperatures do not. Can you not see the disconnect?
And which law of physics allows this energy imbalance to plummet to the deep, deep depths without having any effect upon the surface (other than, apparently, cooling)?
AndrewK,
Not quite sure how you've managed to paraphrase me saying that, as I don't believe I did or why it's relevant.
Radical,
Because increasing CO<sub>2</sub> does cause warming and because we are still warming.
No law, because this isn't what is happening.
I suspect the models have been important players in a theatrical sense in the unfolding drama of this particular climate scare. Others have come and gone, but this one has had an astonishing impact.
As pouncer alludes to above, the 'Limits to Growth' fiasco made much of computer models. Suddenly all the complexity which we poor humans could not cope with was being handled by a Big Blue, and we were to watch aghast as the dismal prospects tumbled out in the fan-folded sheets of green-lined paper. Who could, or did stand up to this early version of 'The Computer says No'? (anyone not familiar with this Little Britain reference, should look here for one of several examples on YouTube).
The lack of much progress with model-dominated climate sensitivity estimates over the past few decades has caused a naughty suspicion to arise in my mind. Recall the 'flux adjustment' phase when the carefree addition of 'external forcings' produced such arrant nonsense in the outputs that ad hoc adjustments had to be made before the outputs were ready to go on stage. Well, given that these models are primarily a playground for programmers, these programmers might well have tinkered away at them until they more or less agreed with earlier published papers such as Plass (1956) getting 3.6K, and Manabe and Weatherald (1967) telling them to keep relative humidity fixed at all times if they wanted big numbers. In this nightmare scenario, the models merely served as high-impact illustrators of previous conclusions. Not adding anything, therefore, to scientific progress, but good nevertheless for fund-raising and scaring politicians, and others, witless. So, GCMs by no means essential for guiding policy makers into de-industrialisation at breakneck speeds, but jolly handy all the same.
aTTP:
You are now saying that the observed not-warming that is taking place, and has done for over 15 years, is actually warming? Are you getting totally disconnected from reality? At least you do not agree with Trenberth, which could indicate that some progress is being made.Radical,
From GISStemp the most likely trend from 1999 - 2014 is 0.1 degrees per decade. Are you playing the "statistical significance" game? If so, maybe go and learn some statistics.
aTTP
Many thanks for allowing me to be disappointed but not surprised. Are you available tomorrow to define my mood or should I just hazard a guess.
I'm sure that you find it is within your remit to answer for those in authority, in your own world of omnipresence, but rest assured that it really is not necessary.
Your opinion counts for a lot, so please keep informing us of it, for it sets a standard that reminds us of all of the years of propaganda and misdirection that we have been fed from cultists like yourself and of which the media and scientific institutions are now withdrawing from.
Just consider the concept that when contributors to this blog don't respond to your incessant commentary it's generally not because they agree with you it's generally because you are ten years behind the curve and we have heard it all before so choose to ignore the drivel.
The only reason that you have this response from me on this occasion is that you got lucky, it's raining, it's Bank Holiday, I've finished cleaning out the fish tank and the blogs are quiet. Don't expect your luck to last.
Lord Beaverbrook,
My pleasure. Serious question. If you worked for the Met Office as a climate scientist, would you comment here? Of course, it's possible that you being a climate scientist is so far removed from reality that the question doesn't make any sense to you, but give it your best shot.
I am not aware of anywhere that the arguments I put forward are addressed by those who do climate modelling.
The reason I noted others also made the same claim is not because I think that makes them correct. If I was the only one making the claim I could fully understand the claims not being addressed - climate modellers could not possibly address criticisms from every individual in the world, no matter how cogent.
But because many people make the same point, I think it would be wise if climate modellers did address the points made here and elsewhere (including the peer-reviewed literature). But I'm not aware of those points being addressed, anywhere. They just press on with the failed deterministic model of climate, despite the repeated falsifications.
LB: “…the blogs are quiet…” not quite what I envisaged for you; perhaps you meant dogs?
Humour aside… aTTP, you are showing yourself to be rather removed from reality, so do not try to draw others into your world. You are on the ropes, now, and you know it. Time to beat a hasty retreat, perhaps? (Now, if that is not a mixing of metaphors, I have no idea what is!)
Spence_UK,
Okay, try reading the recent Annan & Hargreaves paper and maybe some of what they cite. It may answer some of your questions. The chaos issue that Dan brings up is largely a red herring. Our climate is bounded by external influences (solar flux, atmospheric composition, albedo). It is certainly chaotic within those bounds, but if we're acting to change one of the boundary conditions (atmospheric composition) then the system will evolve according to that change, even if it remains chaotic on short timescales (and constrained by the boundary conditions).
Radical,
Hmm, you think I should beat a hasty retreat because you think I'm on the ropes. Why would you think that :-)
aTTP:
Up, or down? Many graphs I have seen seem to be downward (which kinda backs up my earlier prediction, yes?). “…most likely…” so, not definite, then?Radical,
Up (or else I would have put a minus sign in front). In fact, all the temperature datasets have positive mean trends for the period 1999 - 2014. You can play the statistical significance game if you wish, but all that tells you is that you can't rule out (at the 95% level) that we haven't warmed, it doesn't allow you to claim that we haven't warming.
aTTP: So why do the IPCC, and others, say that there has been no statistically significant warming? Perhaps they ought to be getting you on their panels… or perhaps they would prefer you to stay off their panels… Who knows?
Radical,
I don't think the IPCC do say that (if they have, feel free to point it out). What I find - in the SPM - is
I'm aware that some do claim that there has been no statistically significant warming, but that - as I pointed out above - does not mean that there has been no warming. It means that we can't rule out (at the 95% level) that there has been no warming. Of course, people who do say this, typically don't acknowledge this distinction.
aTTP: "From GISStemp the most likely trend from 1999 - 2014 is 0.1 degrees per decade."
Hmm...running a linear least squares fit starting from an extensive La Niña period. I call cherry-picking. Really, I think you can do (and have done) better.
How about starting from 2001, which avoids most of the La Niña? I get 0.01 K/dec for GISS, -0.01 K/dec for HadCRUT4, and 0.00 K/dec for NCDC.
HaroldW,
Ahhh, I was simply responding to the claim that there's been no warming for the last 15 years. Here's a better way :
The HadCRUT4 trend from 1880 to 1999 is 0.052 +- 0.009 degrees per decade. The trend from 1880 to 2014 is 0.063 +- 0.008 degrees per decade. How is that possible if there's been no warming for the last 15 years? I would argue that it isn't really possible (or, maybe more correctly, it's very unlikely that there's been no warming for the last 15 years).
aTTP: alright, I’ll accept the 0.1°C per decade figure, which (quick mental calculations) puts the rise by 2100 at 0.86°C – i.e. less than 1°C. Now, what was my third prediction (which you gleefully ridiculed)?
I have read it a while ago, and I'm fairly sure they don't address any of the points raised here. But later I will go and re-read just to confirm.
Quite wrong.
1. Chaotic systems are predictable on short timescales, and become less predictable as timescales increase.
2. Boundary conditions must be shown to apply, and these can only be shown to apply if we have some reasonable understanding of the invariant properties of the attractor.
3. I am only aware of one credible model of the attractor of the earth's climate - it may not be right but it is the most credible presented to date - the fractal model. This has been verified by several researchers in the peer-reviewed literature (not all of them sceptics!) If the attractor is fractal then the first moment of the time series does not converge at all with increasing timescale.
Put simply: the average (anything) in climate most certainly does not move from an initial condition problem to a boundary condition problem with increasing timescale, by the only credible model of natural variability we have.
Far from being a red herring, the issues that Dan highlights are absolutely pivotal to understanding the timescales at which climate is predictable. These issues are not addressed by Annan and Hargreaves as I recall, neither have any climate modellers, to my knowledge, addressed this issues head on.
Note: in order to answer my question, you have to understand my question. You do not yet understand the points I am making, so you do not even realise you are not answering them. In order to be a useful discussion, you need to understand the basics behind complex, non-linear systems and their predictability, because the "wisdom" of climate scientists that you present is the exactly the reverse of the behaviour and predictability of complex systems.
What happens when the PDO flips and temperatures actually start falling soon... ATTP brain will possibly explode.......
aTTP -
"The HadCRUT4 trend from 1880 to 1999 is 0.052 +- 0.009 degrees per decade. The trend from 1880 to 2014 is 0.063 +- 0.008 degrees per decade. How is that possible if there's been no warming for the last 15 years?"
Mathematically, this is just flat-out incorrect and shows a misapprehension of the limits of applicability of OLS trends. Here's an example: Let the temperature anomaly increase quadratically from 1880 to 1999, then stay flat at the 1999 level thereafter. [I used a formula T(y) = 0.8 K * ( (y-1880)/(1999-1880) )^2 to give a 0.8 K rise over that interval, but the results are independent of scale.] Clearly, there is no warming for the last 15 years of such a record, I think you'll agree. The OLS trend from 1880-1999 is 0.067 K/decade; that for 1880-2014 is 0.071 K/decade. It's not until 2036 that the trend from 1880 is as low as it was during the period of increase.
For the avoidance of doubt, I'm not contending that temperatures resemble that quadratic curve. Merely that the comparison of two OLS trends does not support the conclusion you offer.
Spence_UK,
Indeed, but we're not trying to predict if there will be a lot of rain on 25 August 2035, we're trying to understand underlying climate trends. They're not chaotic.
Again, you're confusing the boundary conditions that set the underlying climate state, with the chaotic nature of weather.
Again, we're not trying to predict the weather in 20 years time, we're trying to understand the underlying climate trends. We cannot predict the weather in the coming decades, but we can try and understand how our emissions might influence the trends.
I do understand your questions. I don't think they're relevant. You probably need to understand the role of climate models in order to ask questions that are relevant. Pointing things out are essentially impossible, but irrelevant, doesn't invalidate climate modelling.
JAP,
Actually my brain would probably explode if someone here said, "oh, I didn't think of that. Let me give it some thought".
aTTP:
If you are hoping that someone might say that about some point you have made, you will be in for a long wait, as you give us very little to think about, other than to marvel how such a self-proclaimed physicist can be so utterly fooled. However, should you at any point say, “Oh, I didn’t think of that. Let me give it some thought,” there will be many on here who will step back in amazement.Radical,
I don't dispute that.
Try me. Although, if I did ever do that I suspect there would still be some who would argue that I hadn't.
...and Then There's Physics said
"Indeed, but we're not trying to predict if there will be a lot of rain on 25 August 2035, we're trying to understand underlying climate trends. They're not chaotic."
Please explain how the track of the jetstream is not chaotic and how the evolution of the track of the jetstream is not chaotic and hence how the long term climate trends for Europe are predictable.
son of mulder,
I didn't say the track wasn't necessarily chaotic, I'm pointing out that we don't need to know precisely where the jetstream will be in 25 years time, we need to understand how our emissions will influence the general behaviour of the jetstream. Remember, that when we refer to something as being chaotic in this context, we mean that it is deterministic. This means, for example, that it obeys a set of well-defined equations but is very sensitive to small changes in initial conditions. therefore if I run a model in which the processes are chaotic, I may not be able to say precisely how this process will behave at any given time, but I can say something about how it will evolve if I change the underlying conditions. Chaos doesn't mean that it's behaviour is entirely random, it simply means that the precise evolution is extremely sensitive to the initial conditions.
Because the long-term climate trends are not chaotic. You can be fairly sure that it's going to be colder - on average - in December in Europe than in July, but you can't predict what the temperature will be on any given day in 7 years time.
"Because the long-term climate trends are not chaotic."
How do you know?
And does this mean you never got around to reading that Lorenz paper I pointed you at a while back describing how the chaotic track of the jet stream could result in long-term climate trends?
Nullius,
Well, I'm pretty confident that as we move into the latter part of the year we will start to get colder (replace colder with warmer for the Southern Hemisphere) and fairly sure that towards the middle of next year we'll get warmer again. We won't be able to say - with any accuracy - what the temperature will be at midday on 20 July 2015, though.
I do remember reading some of that paper, but have forgotten what it said. Here's a question for you. If I put a double pendulum on the floor and set it in motion, it's behaviour will be chaotic. If I kick it, will it's motion through the air be chaotic, or will it's overall trajectory be parabolic?
aTTP: Your use of analogies suggests that the real answer was "I don't know how."