Bishop Hill Shaviv on models and sensitivity
Jan 11, 2012 Climate: Models Richard Betts joined in the conversation about climate models today, making some interesting comments on validation:
As I've mentioned before, the earlier climate models used in the 1970s were used to make estimates of warming over the next 30 years which were fairly close to what happened ... BH asks for tests of the projections made 10 years ago, but the problem is that with internal variability in the system you need longer than that to test the models, unless you specifically initialise the models with the conditions of (say) 2001 using data assimilation techniques, and that kind of thing was not available then, we only started doing it 5 years ago.
So yes, out of sample testing on timescales relevant to GHG rise is an important point but by definition difficult with the latest models!
One minor point is that I had said I would have been more convinced had the story of model versus data in the last ten years been different - I agree with the 30 years figure for falsification. However, more interesting is a point made in a recent post by Nir Shaviv:
From the first IPCC report until the previous IPCC report, climate predictions for future temperature increase where based on a climate sensitivity of 1.5 to 4.5°C per CO2 doubling. This range, in fact, goes back to the 1979 Charney report published by the National Academy of Sciences. That is, after 33 years of climate research and many billions of dollars of research, the possible range of climate sensitivities is virtually the same! In the last (AR4) IPCC report the range was actually slightly narrowed down to 2 to 4.5°C increase per CO2 doubling (without any good reason if you ask me). In any case, this increase of the lower limit will only aggravate the point I make below, which is as follows.
Because the possible range of sensitivities has been virtually the same, it means that the predictions made in the first IPCC report in 1990 should still be valid. That is, according to the writers of all the IPCC reports, the temperature today should be within the range of predictions made 22 years ago. But they are not!
Go and take a look at the graph at Nir's site. This seems a reasonable point to me.
Reader Comments (130)
Roger Langstaff: today I read this paper which appears to be the root paper justifying the theory used by the climate modellers: www.gps.caltech.edu/~tapio/ese101/Papers/manabe64a.pdf
As you can see, it is a time-marching, finite-difference principle which is set to give radiative equilibrium assuming 100% thermalisation. The lapse rate comes in when the critical lapse rate is exceeded.
I am very unhappy about this. I think there is a fundamental flaw in the assumption of 100% IR thermalisation by GHG molecules except at high pressures. It could be that it is mostly at second phases. Now I could be wrong but it seems no-one can definitively tell me why. Where are these basic data? Here is one argument: http://www.radix.net/~bobg/climate/halpern.on.radiation.html
However, as with almost all such arguments, the guy makes the elementary mistake which is to concentrate on looking at the molecule which absorbed the IR then the probability it transfers that extra energy to molecular motion by collision. In reality, you must treat all the molecules as interchangeable, i.e. another excited molecule will emit a photon in a random direction restoring the system to its previous temperature and thermal excitation of intramolecular vibrations. This is the Law of Equipartition Energy.
If you persist in looking at individual molecules you get the Gibbs Paradox: http://en.wikipedia.org/wiki/Gibbs_paradox
Note for Hengist and Richard; there has to be a very basic flaw in the assumptions made by climate science even worse than the other mistakes. This is because the climate models have been shown by experiment not to be able to predict climate AND I have worked out how you can account for much post 1980s warming without any CO2-AGW. This is a spectacular failure of science.
It's interesting that the graphs of the Met Office, Lucia and Nir Shaviv all seem to be broadly in agreement that the observations are on the brink of falling off the lower end of the model predictions.
The response to this will be interesting. There is a worrying suggestion in the climategate email that Nir Shaviv quotes, where Trenberth says "the data are surely wrong". If the response is to go back and adjust the data so that it fits the models, I personally will blow a fuse and I suspect I won't be the only one. There are hints that this is being done, see this Met Office document about HADCRUT4 ("The inclusion of new land station data at high latitudes and in Russia has resulted in a warming of years in the late 20th century/early 21st century.")
[Roger L, if there is a simple one-paragraph argument that 'proves' that CO2 has no effect, you can be certain that the argument is incorrect. Briefly, this argument tells you about the temperature gradient in the atmosphere, but not the actual temperature.]
Paul,
"Briefly, this argument tells you about the temperature gradient in the atmosphere, but not the actual temperature"
I see where you are coming from, and this worried me for a while. However I think I rationalised it by considering the entropy of the planet as a whole, which includes the conductive effects of surface - gas boundary, and requires all calculations to be performed within a boundary high above the top of the atmosphere.
Richard Betts
Hi Richard, I have just read through our exchanges from last night and I must apologise. I seem to have had a "politeness" failure, some of it was failed attempts at humour.
Sorry there was no need for it.
Regards
GS
"The changes bring absolutely nothing other than obfuscation, why do it?"
That was sarcasm, yes?
tty
"That was sarcasm, yes?".
No, not really. However I should have made it clear that I was referring to the format not the actual forecast.
I don't understand why the latest forecast chart now starts in 1950 when it only relates to model forecasts from 1985 onwards, thus truncating the relevant period. Also why now include data from GISS and NCDC when the forecast is against Hadley I cannot see how this adds clarity? There appears to be some changes in the values of the previous forecasts, but that is a different issue.
The changes seem to make the forecast chart more difficult to assess/compare with its predecessor. Or maybe I am just missing something, which is far from being a rare occurrence
One minor point: the Pinatubo cooling is incorrectly arrowed on the Shaviv graph. However, what is clear is that there is a shed-load of (unprecedented) catch-up needed to get within IPCC bounds which, by itself, suggests falsification. It is time to reconsider sensitivity before we further impoverish ourselves.
Mydogsgotnonose,
Thank you for the reference to the Manabe and Strickler paper. Do you think that this is where the GCM methodology originated?
And just to remind ourselves - sensitivity is an estimate based on the fact that CO2 alone cannot explain warming trends but is a multiplier of its calculated effect based on a controversial GHG theory.
I haven't read the thread, unfortunately. Hope this hasn't been asked already...
Can someone explain why you can't run the models to create forecasts for every year 1900-1980, and check whether the forecasts are accurate?
If they can't predict the cooling 1940-1980 or whatever, then there would be serious questions to answer.
Dear Dr Betts,
On 12th Jan at 1.16 am, you quote from the Met Office website:
"Global average temperature is expected to rise to between 0.36 °C and 0.72 °C (90% confidence range) above the long-term (1971-2000) average during the period 2012-2016, with values most likely to be about 0.54 °C higher than average (see blue curves in the Figure 1 below).
From 2017 to 2021, global temperature is forecast to rise further to between 0.54 °C and 0.97 °C, with most likely values of about 0.76 °C above average."
1. If a period of 30 years is "long-term" how does the Met Office describe 300 years ? Or 3000 ? Or 300,000 ?
2. "Global temperature" does not exist (although "global temperature range" does), whereas "global average temperature" can be estimated. I trust that you agree.
3. "... global temperature is forecast to rise further ..."
a) "Forecast" is prediction, isn't it ? Or simply projection ? (The Met Office recently clarified its use of these terms, I recall).
b) What practical purpose is served by forecasting global average temperature, e.g. during the period 2012-2016 ? Whom does it help ? Does it help me, as a farmer or forester, for example ? Surely the only forecast that might be of use to me concerns that tiny part of the globe where I work ? And even then what use is a forecast of average temperature alone, ignoring other key factors such as rainfall and periods of frost and/or heatwave ?
Please help me to appreciate the value of such forecasts.
RL: it was given to me by the person who flamed me on ClimateEtc as exactly that. Bear in mind that I am not a physicist but that peculiar hybrid called a Metallurgical Engineer taught things from Chemical Engineering to nuclear physics; aka Renaissance Man [whose core science is chemical thermodynamics so Gibbs holds no fears for me].
Over many decades running big programmes I have developed an unerring knack of posing the question the zealots least like to answer. In this case it's the 100% thermalisation, the 'back radiation' perpetual motion machine, the 33 K claimed GHG warming which conflates lapse rate and cooling by polluted clouds, the theory of which fails every time you look at a cloud with a bimodal droplet size distribution.
Frankly, climate science is in a mess. To cope with the back radiation perpetual motion machine, the models use double the real optical depth of clouds and now a variable net AIE to account for present temperature stabilisation. They have no shame in this. In 2007, Jeff Kiehl, a good scientist, wrote that although there is no present evidence above aerosol noise of any CO2-AGW signal, in time it just has to rise to that threshold. My work has shown that the optical physics used to predict the cloud part of that aerosol correction is plain wrong, so there may never be any observable CO2-(A)GW!
You can explain most if not all aspects of climate without CO2-(A)GW. This is a non science, a Norwegian Blue physics, it is nailed to the IPCC perch and well meaning people like Hengist who in time i will convert and R. Betts, who is paid to do a job but has a brain, will have to accept that the IPCC science's dead body is proceeding by momentum not credibility.
But I could be wrong if someone can only point to me the evidence concerning IR thermalisation. Will Happer pointed out in 1993 that this part of the science is bunkum, and he's probably the best IR specialist in the US who refused to lie for Gore. Nassif Nahle has the same thermodynamics' insight as me and is picking over the partial molar Cp data.
mdgnn -
"But I could be wrong if someone can only point to me the evidence concerning IR thermalisation."
Please will you say in simple words exactly what you mean by thermalisation or IR thermalisation.
You frequently use the term and I'd like to understand better what you are saying,.
Hengist McStone
You were given coherent answers. Try reading them, you might learn something.
Martin A.: 'thermalisation' refers to the conversion of the extra intramolecular energy in a GHG molecule which has absorbed an IR photon, to extra average kinetic energy of the gas mixture, hence temperature increase.
The Law of Equipartition of Energy suggests that one of the already thermally excited GHG molecules [~5% in CO2 at room temperature] will emit a photon in a random direction thus returning the gas mixture to its original energy state.
At high pressure, the average time between collisions will fall so much that the probability of such thermalisation may increase because the newly excited molecule has a much higher probability of transferring that energy to molecular velocity.
mdgnn
If I understand you correctly;
Arhenius believed he saw a temperature increase in CO2 gas when in the path of IR which I understand is your use of the term 'thermalisation'. Back radiation theory does not require thermalisation as there is no need of any residency time for the received photon in the CO2 molecule at pressures found in the atmosphere.
Correct?
mydogsgotnonose
Thank you for attempting to help me out. I'm doing my best to understand.
I think you are talking about individual photons - am I right? A single photon is absorbed by a (say) CO2 molecule and its energy may set the molecule spinning or vibrating (or whatever it is that molecules given extra energy do when they haven't changed their velocity). But the velocity of the molecule is unchanged, thus the average temp of the gas is unchanged, even though it now contains a molecule with extra internal energy.
But after some time, the intramolecular energy (as you call it) will be transformed into kinetic energy, the molecule's velocity increasing. At that point the energy of the photon has been thermalised. Have I got it right up to here?
Or perhaps, before that happens, a different molecule emits a photon, having been encouraged to do so by its desire to comply with The Law of Equipartition of Energy. So, in that case, despite the original photon having been absorbed by the gas, the total heat in the gas has not changed. Have I got it right up to here?
But you talk about "the probability of such thermalisation" which I don't get. If the energy of an absorbed photon does not eventually get changed to kinetic energy, what happens to it? It says permanently as intramolecular energy? Or the molecule eventually spits out the energy directly as a photon, without the molecule's kinetic energy ever having changed (hence the energy of the photon was never ever "thermalised" ?
I think I am missing something still. As I see it, (assuming we are talking about gas that is in thermal equilibrium with its radiation environment), it will be continually absorbing photons and continually emitting them. The rate of emission will not necessarily be the same because the emitted photons may have different average energy per photon than the absorbed photons. But the rate of emission of energy will equal the rate of absorption of energy - otherwise, the gas would not be in thermal equilibrium.
So there is no need (that I can see) to be bothered about whether or not the captured photons are converted to heat or not. The temperature of the gas stays constant either way and the rate of emission or energy via emitted photons stays equal to the rate of absorption of energy from captured photons. Am I oversimplifying in some way?
Thanks again for helping my efforts to understand.
ssat: 'back radiation' is bunkum because it is really 'Prevost Exchange Energy' emitted and received by any body above absolute zero, a Law invented in 1791. You prove it by measuring the Up-Down signal as a function of height at night. It falls exponentially to zero above ~80 m, Beer's Law absorption of IR emitted from the ground. This experiment was done in Holland a year or so ago. Net 'back radiation' with no temperature gradient is zero so there is no energy source.
No-one who has proper physics or engineering training would make that mistake. Aarhenius didn't have that knowledge and climate scientists aren't taught the detail of radiation theory. They should not be allowed to buy single radiometers, just pairs welded back to back with the difference signal.
As for the re-emission of absorbed IR energy, it will happen with a defined probability for isolated molecules but in an ensemble, molecules cease to have individual identity. For details of how to calculate the emissivity of GHG mixtures, go to the works of Hottell at MIT in the 1950s.
Scientists like Aarhenius make mistakes. Sir James Jeans wrote the textbook on the kinetic theory of gases but failed to understand how that young upstart Knudsen worked out his theory of effusion. This of course is what the Manhattan project was based on, and that worked!
@Patagon
Here it is in graphical form. Pale colors are the estimated for the last years, blue vertical line is the forecast and range. I am sorry to say that it does not look very impressive.
Would you consider the forecast validated if it falls within the range?
Thanks for that graph. It helped so much.
Not that I understand anything about statistics and/or climate science, but I do bet on Lucia's casino, and when I read the figures for 2012, as predicted by the Met Office, of 0.48 C mean with the margins of error of 0.34 to 0.62, I thought, "gee, that looks like a huge margin of error".
Then, I moved on ...until you graphically illustrated how big that margin is.
I don't know whether those Met figures have been produced with 90 or 95 percent confidence interval. Eyeballing the margin of error on your graph, however, I am surprised they didn't predict the 2012 with 99.97 percent confidence.
My crystal ball says that the Met will be proven correct on their 2012 figures at 99.99 percent confidence.
Some falsification!
ndgnn @1:11 PM
"...one of the already thermally excited GHG molecules [...] will emit a photon in a random direction..."
If you accept that then it follows that a percentage will be back radiation i.e. returned toward source. Is there a confusion of terms here?
In other words, as Lucia has been saying for some time, the models are on the verge of being falsified?
Jan 12, 2012 at 8:30 AM | Jonathan Jones
The models have long since been falsified. What Richard Betts et al are doing is re-running the models everything few years to try to keep them in touch with reality.
The met off declared some 2 years or so ago that they use the same, new, wonderful model for daily and seasonal and 10 yr weather forecasts. None of their forecasts has ever been shown to have any skill what so ever. Even after a day they begin to deviate from reality, after 3 they are unreliable for detail although pressure patterns remain "similar" but you will still get wet when you least expect it.
If you accept that then it follows that a percentage will be back radiation i.e. returned toward source. Is there a confusion of terms here?
Here's a thought. H²O is a GHG. It absorps IR with the same energy as that absorped by CO².. So why is it that when the air is saturated with water ( fog) it is always colder under the fog than when the sun comes out. Why am I not warmed by the back radiation. Is it that the IR is absorped before it can reach the ground and therefore cools the ground.?
ssat: if you take a spherical volume of the atmosphere sufficiently far from the Earth's surface that the IR from it contributes very little energy input, the total outwards' flux is exactly balanced by the incoming flux from the surrounding atmosphere.
This is why when measuring unidirectional radiative flux, it only represents temperature convolved with emissivity. The real signal is the difference, non-zero in a temperature gradient, the real energy flow.
This elementary failure of physics' thinking is the main reason why the climate models are wrong. The others are the absurdly exaggerated net AIE/use of double real cloud optical depth as fudge factors to offset imaginary 'back radiation' and exaggerated present GHG warming to calibrate the predictions.
All in all an embarrassing failure only held together because the peer review process has been fixed.
Stephen Richards: taking the concept of negative 'back radiation' to its conclusion, we have an even worse fate to contemplate than CAGW. The atmosphere surrounding the Earth may become a black hole.
mdgnn
So yes, there is a confusion of terms - yours. When you negate back radiation you are talking about flux.
Stephen Richards
Thank you for that. My apparent confusion is not with that which I know but with the argument of mdgnn together with a desire to understand GHG theory for which I can find no concise description that makes any sense to me.
DDB: "If they can't predict the cooling 1940-1980 or whatever, then there would be serious questions to answer."
http://rankexploits.com/musings/wp-content/uploads/2009/03/temperatures_absolute.jpg
Getting the average global temperature correct would seem kind of fundamental to me. I didn't realise that "basic physics" could give so many different answers to the same question.
Stephen Richards Jan 12, 2012 at 4:19 PM
"Is it that the IR is absorped before it can reach the ground and therefore cools the ground.?"
Isn't it simply that, from above, the fog looks like a nice white cloud and reflects a good proportion of the sun's radiation upwards, to be lost for good rather than making you fell warm?
=====
Is the reason a cloudless night in winter generally seems colder than a cloudy one that some of the IR radiated from the ground is reflected back down?
mdgnn
Still at it with the dodgy physics, I see. Despite the recent debunking you got on the Cosmos and consensus thread (see mdgnn @ Jan 6, 2012 at 6:07 PM and my follow-up to the end of the thread).
Predictable enough, I suppose, now that you’ve realised how few people there are here with the background to call you out over your nonsense. Very naughty, all the same.
BBD: unlike you, I learn from my mistakes. If you think I am wrong in questioning the assumption of 100% thermalisation of absorbed IR energy, please point me to definitive proof.
As for 'back radiation' and the claim of 33 K present GHG warming, you're wrong and as it is being realised by the rest of the zealots, you probably know it, otherwise why the desperate flogging of a dead horse?
I don't expect you to be up to speed with the aerosol optical physics because I have not disclosed the mechanism prior to publication. So, go away and have a lie down: for you ze War is over......;o(
mdgnn
If you learn from your mistakes, then why - just to pick a single example from many I highlighted on the other thread - have you not admitted that you got the phytoplankton/cloud wrong?
No!
Easily proved.
mdgnn
Well go on then.
Just as on the previous thread, you assert continuously but 'prove' nothing. Except that you are unable to provide evidence for your claims. You are a time-waster.
BBD
"You are a time-waster."
I may concur (my post at 4:40PM) but for the small problem with GHG theory currently being discussed at Talbloke's and WUWT. Many people have a problem making sense of it so anyone willing to put in some time, such as mdgnn, may yet come up with a concise explanation. It could, however, remain inexplicable and therefore beg the bigger question.
What frustrates me is that people can argue away the apparent failure of the models to predict future temperatures by suggesting a missing process (e.g. aerosols, as suggested by Hansen) but then balk at ideas like Svensmark's, which would imply a reduce CO2 sensitivity. I'm quite willing to believe there's more to be understood, especially the influence of clouds, but fail to see why only things that invoke an increased CO2 sensitivity are acceptable to a certain group of 'scientists'.
I like Shaviv's article -- the original subject of this thread -- because it asks some very basic questions and provides some easily understood data that makes you appreciate that there's quite likely something seriously wrong with the orthodox narrative (i.e. CAGW).
So, to clarify, taking January each year as a starting point, we have 80 starting points if we initialise a model for 1900-1980.
Is this regression test done for these models, and is the result published?
If the results are published, do the models predict a cooling period 1940-1970 and a peak followed by a flat period 1998+? If they can't predict the cooling in particular, then essentially they only predict warming.
Just saying.
SSAT
My best advice is to take it to a serious forum - eg Science of Doom. Email SoD and request that he opens a discussion. Perhaps then some clarity may emerge.
Dave Salt
I posted a brief comment on Shaviv's blog about this, but didn't deal with the Shaviv/Svensmark GCR-cloud hypothesis.
An important point here is that there is no evidence at all for a climatologically significant link between cloud and GCR. Whereas there is mounting evidence that Hansen is right about the recent flat trend in GAT and aerosol negative forcing (eg Vernier et al. 2011).
Later on the thread Shaviv references his own 2008 paper, but it is not as convincing as he would have us believe. The relationship between solar cycle, GCR flux at surface and OHC is far from clearly established, and in particular the reliance on tide gauges as a measure of OHC (thermal expansion) is tenuous.
To be clear, I'm not saying he is wrong. But he says 'unequivocally demonstrates' and that is a gross overstatement. I certainly object to that. It also makes me highly sceptical of Shaviv's objectivity in general.
Here's a question for you BBD. I have been homing in on the reasons why the IIPC climate models are failing to predict correct temperatures and the top of the list is the assumption of 100% thermalisation of Absorbed IR by GHGs.
Can you point to me where this is proven absolutely?
mdgnn
This is a nonsense question and demonstrates your rhetorical, but not scientific, credentials. Point me to anything in science that is 'proven absolutely'.
And you have not yet addressed the problem with your mistake over the phytoplankton/cloud effect. Why is that, I wonder?
BBF: why is my question a 'nonsense question'? It's why Happer decided not to continue in his US government job as Director of research at the DoE. Basically he had to agree to tell what he apparently thought to be a scientific lie, or go back to Princeton: he chose the latter exclaiming he 'didn't need the money'.
So if you are so expert at climate science, answer my question because unlike Happer, I am no an IR spectroscopy specialist. My simple view is that the 100% thermalisation assertion is an assumption and that according to the Law of Equipartition of Energy, most absorbed IR in a GHG mixture should be re-radiated in a random direction, effectively scattered, isothermally.
So, little direct thermalisation implying that the effect is mostly indirect.
BBD: I have made no mistake about phytoplankton. The mistake is yours.
Here is essential reading on the direct thermalisation of absorbed IR energy by greenhouse gases: http://wattsupwiththat.com/2010/08/05/co2-heats-the-atmosphere-a-counter-view/
'These observations demonstrate as expected that CO2 emits the same power as it absorbs and that there is no net energy transfer between the vibrational modes of CO2 and the translational modes of N2 and O2 . If you double the CO2 concentration or make the temperature vary , the observations stay identical showing that the conclusions we made are independent of temperatures and CO2 concentrations .'
There is no direct thermalisation. Just what proportion is indirectly thermalised at say cloud droplets is unknown. It may be most of the relevant energy in a cloudy sky but very little in a clear sky. This is the most basic of the failings of the IPCC models.
mdgnn
No, it is not. You have not shown that it is, nor have you produced evidence that you are correct.
I've drawn you attention to Kruger & Grasserl (2011). So you know that phytoplankton blooms increase cloud albedo and cause cooling. I have reminded you of your error repeatedly since.
Constant repetition of knowingly incorrect, unsubstantiated assertions is only tolerable for so long. You have now crossed a line.
You are lying.
As for Vonk's nonsense - you should read on. You will come to this: A reply to Vonk: Radiative Physics Simplified.
This is one of the problems with 'sceptics'. Incoherence - even amongst yourselves.
Understand this: I have given you the benefit of the doubt and you have repeatedly come up short. I do not take you seriously and I dislike your fundamentally evasive and dishonest approach to discussion.
The only reason I am responding to you is to warn others not to take you seriously either.
My Dear BBD; I knew of this paper but had discounted it. Firstly, because the aerosol optical physics is wrong, the algorithms used to process satellite albedo data create artificially low droplet radius results hence no such satellite data are reliable, something I am writing as a paper.
Secondly, there is a 25% increase of optical depth for clouds with 'drizzle'; for rain clouds it is much more. There really is a BIG PROBLEM with climate science because Sagan got his physics wrong and there are many of us working at bits of the problem, including the top satellite physicists who don't want to be dragged down when climate science goes ti*s up, real soon now.
A tip for Annan.
Online betting is just one more way you can contribute to the solution to global warming and climate change.
Dam! Wrong post!
mdgnn
You say:
How then do you account for the increased upward radiative flux at TOA? From K&G11:
If the aerosol optical physics is 'wrong' as you claim, then the upward radiative flux would be reduced. Not increased.
I'm going to enjoy reading this paper of yours, if it makes it past review.
This is not a trivial problem. In essence, it comes down to whether there is or is not a bimodal droplet size distribution and the efficiency of nucleation of cloud droplets.
You also have to take into account the extra optical backscattering from the bare aerosols [g is very low].
The definitive data would be LWP which like albedo is a pukka result. If LWP is constant, increased albedo could have come from increased precipitation efficiency.
The palaeoclimate data show that phytoplankton blooms in the Antarctic increase CO2 by ~20 ppm, a sign of greater warming, less reflection of solar energy.