Monday
Sep052011
by 
Bishop Hill
Bishop Hill Santer says
Sep 5, 2011 Climate: Models Santer et al have a new paper out on trends in the tropospheric temperature.
Because of the pronounced effect of interannual noise on decadal trends, a multi-model ensemble of anthropogenically-forced simulations displays many 10-year periods with little warming. A single decade of observational TLT data is therefore inadequate for identifying a slowly evolving anthropogenic warming signal. Our results show that temperature records of at least 17 years in length are required for identifying human effects on global-mean tropospheric temperature.
Please keep comments to the subject matter of the paper.
Reader Comments (163)
Spence, I can see the whole paper if I click on 'pip pdf' in the left panel.
PaulM
OK thanks, I just get an AGU login if I click on "PIP PDF". I assume you are either a member or have access from your work/faculty? It means to see it I will probably have to travel to my local tech library.
What was the period used by Prof. Phil Jones in his recent paper? 15 years?
The number of years before a temperature trend derived from an ensemble of model outputs becomes significant is itself a model output and must be adjusted accordingly.
Robinson
CO2 forcing fits very well with the empirical data. This is apparently what Santer et al. have attempted to demonstrate in their study.
All I am suggesting is that those who disagree that CO2 can warm the climate system must first explain what is wrong with the RTEs in order to 'get rid' of any CO2 forcing.
Second, they need to provide evidence for an alternative forcing sufficient to explain the modern warming.
BBD,
Do you know what caused the medieval warming?
Does any climate scientist know?
Let us start from there, and provide answers for why every warming or cooling episode occurs shall we?. When we get to the current period, we can think of explanations and alternative explanations.
Geronimo
I agree with you wholeheartedly on the disastrous effect on energy policy. Fairy dust will not power an industrialised economy. I agree that green activists have lied and lobbied against nuclear very effectively, for which hopefully one day they will be held to account.
Obviously the right question. And the search has been on for decades - with nothing found sufficient to explain most of the warming since the mid-1970s except the increasing amount of atmospheric CO2.
Santer et al. seem to be attempting to show the steady, underlying effects of CO2 on T by comparing longer time series with CO2 forced and unforced runs of a multi-model ensemble.
Of course they could do the same by altering the solar forcing (DSW), but observations show no significant changes post-1979, so there would be little point.
BBD, Breathe of Fresh Air (at 9:38 AM) made a good observation that I've yet to see anyone address. So, I ask you directly, can CAGW theory provide a consistent explanation for the temperature rise in both periods?
Shub
You could argue that what caused the MWP is largely irrelevant to the effects of CO2 on modern temperatures. Were an MWP-equivalent event occurring today, we would presumably be able to determine the cause with modern instruments. As no alternative explanation has emerged from several decades of observations, the finger is pointing very firmly at CO2.
RE: BBD
"This leaves us with the rigorously tested and widely accepted radiative transfer equations which demonstrate the mechanism by which CO2 warms the atmosphere."
While this is true the effect is too small to explain the temperature rise. I think even Singer (who is an atmospheric physicist and arch-skeptic) has always supported this physical argument but points out the effect is not one to worry about because the doubling in CO2 will likely only lead to a temperature rise of 0.6 degC over 100 years.
The problem with the AGW theory is that it requires a further mechanism of positive feedback to amplify this. Support for this additional physical mechanism is rather thin and is one of the reasons I do not agree with AGW theory. Additionally I think there are far too many other compounding factors and mechanisms, some of which we know little or nothing about, and the temperature records are far too short for us to be able to say that "it must be CO2".
Dave Salt
First, there's no such thing as 'CAGW theory'. It's AGW. The consequences are not currently under discussion (off topic).
Second, the causes of the 1910-1940 warming are not required to have any bearing on the causes of warming from the mid-1970s.
BOFA is making a rhetorical, rather than a scientific argument.
As are you. Why should the cause of early C20th warming make any difference to the effects of CO2 on modern climate?
Thinkingscientist
The no-feedbacks forcing of T from a doubling of CO2 is estimated as + 1.2C. Singer may be confusing transient effects with equilibrium climate sensitivity. And he may be doing so deliberately.
We all know that feedbacks are required to amplify the CO2 forcing to the estimated + 3C per doubling, with water vapour plaing the main role. You say that support for this additional physical mechanism is rather thin. This is incorrect. There is a large body of evidence that supports it. There is a detailed review here, which I hope you will find useful.
BBD, why should a theory that explains the modern warming but cannot similarly explain the previous warming be taken seriously?
Also, when you say there's "no such thing as 'CAGW theory'. It's AGW", are you saying that there's no difference between the basic radiative forcing of CO2 and the enhanced forcing due to net positive feed-backs?
I rather think BBD has taken the Trenberth anti-scientific policy to heart; reverse the null hypothesis.
David Salt
Because in all probability we are talking about two or more different factors forcing climate at different periods. I can only understand you refusal to accept this as tactical. It is certainly not a logical position.
Phillip Bratby, excellent points you have made - saved me a lot of typing!
This has degenerated into waffle.
RE: BBD
"Were an MWP-equivalent event occurring today, we would presumably be able to determine the cause with modern instruments."
That statement positively drips with hubris. Why should we know the cause with modern instruments? Instruments just record and measure, and when it comes to temperature, even that is hard to do properly. It needs thought and careful experiment to attribute cause and effect - something sadly lacking in climate science. I am not convinced we have even a basically reliable temperature record. We don't know what caused the MWP and I am pretty sure we don't really know what is causing the recent warming. And what about explaining the cooling 1940 - 1975 - lots of arm waving been going on over that for at least 10 years, none of it very convincing.
The problem with climate science is it doesn't try very hard. It must be a truly dull area of research to already know that every change is caused by CO2. However, try applying the CO2 argument to explaining cooling from interglacial to ice age, the younger drayas - or even the MWP. Epic fail, yet apparently CO2 is all we need to know to explain a small temperature rise over a 15 year period at the end of the 20th Century.
RE: BBD
"We all know that feedbacks are required to amplify the CO2 forcing to the estimated + 3C per doubling"
I believe I am right in saying that in the IPCC they claim this sensitivity, but only use models (with one exception) and then invert the error terms and thus exaggerate the sensitivity. The only experimental estimate of this in the IPCC suggests climate sensitivity is much lower (maybe 1.0?) and a number of other atmospherric physicists agree that the climate senstivity much be low - some as low as 0.6.
RE: BBD
"Because in all probability we are talking about two or more different factors forcing climate at different periods."
What factors would they be then? And at what probability level are you talking? 50% support for and against your argument of two completely independent mechanisms, or "very likely" 95% support for two different mechanisms?
You cannot simply wave away the argument that if the MWP happened, the same factors might be influencing climate today. If you don't know what caused the MWP you cannot dismiss it as being the current cause.
The reply above by BBD to Robinson is quite interesting I think:
"Robinson
Don't be so stupid. In order to debunk one theory, you simply have to show that it doesn't fit with the empirical data. You don't have to propose and validate an alternative theory.
CO2 forcing fits very well with the empirical data."
The comment by Robinson is true: a valid model must explain the emprical or measured data. However, the response by BBD is not true - the validity of a model is NOT demonstrated by its fit to measured data.
To quote from George Bernard Shaw:
"Or, to take another common instance, comparisons which are really comparisons between two social classes with different standards of nutrition and education are palmed off as comparisons between the results of a certain medical treatment and its neglect. Thus it is easy to prove that the wearing of tall hats and the carrying of umbrellas enlarges the chest, prolongs life, and confers comparative immunity from disease; for statistics show that the classes which use these articles are bigger, healthier, and live longer than the class which never dreams of possessing such things. It does not take much perspicacity to see what really makes this difference is not the tall hat and umbrella, but the wealth and nourishment of which they are evidence, and that a gold watch or membership of a club in the Pall Mall might be proved in the same way to have the like sovereign virtues. A university degree, a daily bath, the owning of thirty pairs of trousers, a knowledge of Wagner's music, a pew in a church, anything, in short, that implies more means and better nurture than the mass of labourers enjoy, can be statistically palmed off as a magic spell conferring all sorts of privileges."
Or in my words, wrong model.
Thinkingscientist
When DSW from Milankovitch forcing begins to reduce after the peak of the ca 100ky eccentricity cycle, T falls. As you say, this happens slowly, and less rapidly than expected. Why?
- the initial Milankovitch forcing (DSW) causes GHGs to increase (ocean outgassing of CO2 and methane from expaning wetlands, warmed tundra etc)
- this slows the cooling
The abrupt coolings of the Younger Dryas and 8.2ky event appear to have been caused by large injections of cold, fresh water into the N Atlantic as the N American ice sheet melted. This stalled the thermohaline converor and initiated an abrupt and severe cooling.
CO2 is one among many factors that can cause climate change (all ultimately governed by the energetic balance within the climate system). So while RF from DSW runs the show, RF from DLR (influenced by GHGs) contributes to net changes in total RF at the surface. This can and does force T.
Other factors like the THC are hugely important. None are incompatible with the role of GHGs in changing T. CO2 forcing has always been a component of determining T. All that has happened recently is that concentrations have risen sufficiently for CO2 to become a dominant influence through its effect on DLR and net RF at the surface.
What epic fail are you referring to?
"This has degenerated into waffle."
So how about you get it back on track and give some substantive responses rather than an armwaved "what else can it be?"
To put some structure to it, start with a referenced exposition of how the RTEs apply to CO2 in the atmosphere along with a quantified and verifiable result and we'll go from there.
17 authors for this paper - and they come up with a span of 17 years for observations to be valid.
Hm.
Will the next paper have 20 authors to insist on a twenty-year span for valid observations?
Thinkingscientist
If climate sensitivity is low, this means that the climate system sheds energy efficiently into space.
How then does enough energy accumulate in the climate system under the modest DSR change from Milankovitch forcing to instigate glacial terminations?
Claims for a low or very low climate sensitivity are incompatible with known paleoclimate change.
BTW I've been through this in great detail on the NIPCC thread (a couple of pages back) and it is getting rather OT here.
This is absurd. Because of the type of noise: 1/f, the length of time you look makes very little difference to the signal/noise ratio. Indeed, arguably it gets worse the longer you look (noise goes up with longer periods).
The question is not about climate variability, but about short term variability. I.e. at what point does increasing numbers of samples reduce the NONE CLIMATE, or annual noise to become comparable with long term noise.
Now we all know that annual noise has a very distinct frequency, and anyone with even an ounce of intelligence can work out that it is quite feasible to ignore any signal with a period of 1year. After that, we are only left with instrumentation error (which they deny) and the signal spectrum or weather/climate.
Now half the variability of the climate occurs at frequencies greater than a decade and half below. This is just a convenient point at which to divide the spectrum between the long term and short term and sufficiently high to filter out annual cycles - 100 years of data, which means the bandwidth of 1year noise will be about 1/100 of a year. Which means 10years is 1000x the bandwidth of the annual noise. There really should not be a problem removing annual noise from changes occurring over 2 years, so 10years would have a massive safety margin.
In short, they are talking politically motivated claptrap (although one has to read the paper to see if the contents match the hyperbole)
I have a few questions concerning models for those with ready answers (though I accept that LMGTFY would be a justified response!):
Do climatologists still indulge in post hoc elimination of model runs on the basis of their judged "physicality"?
Do the models still incorporate 'flux adjustments' and similar contrivances?
Do they still exhibit laughable artefacts like the 'cold equator problem'?
How is it possible to have 20 equally valid models with different parameterisations?
Do the models explain anything about the climate other than its 'changeology'?
Thinkingscientist
I'm not 'waving it away'. I'm saying that despite the best efforts and modern instruments, there is no evidence for any forcing sufficient to explain most of the recent warming except CO2.
Stanley Unwin would be proud of them!
RE: BBD
"I'm saying that despite the best efforts and modern instruments, there is no evidence for any forcing sufficient to explain most of the recent warming except CO2."
Or in other words, the science is settled. I don't think so. Science is never settled and I think the sceptics are going to have the last laugh.
Apocryphal quote but still relevent:
"Everything that can be invented has been invented."
Charles H. Duell, Commissioner, U.S. patent office, 1899 (attributed)
Shub, Dave Salt,
The same climate models that simulate warming in the latter part of the 20th Century also produce warming the early part of the 20th Century, and early 20th century warming is simulated even when only natural forcings (solar and volcanos) are included (but this is not the case for the more recent warming, which is only simulated when anthropogenic forcings are included). See the figure here.
The reason that the "natural only" warming is greater in the early 20th Century than later is because the increase in solar irradiance is greater then (early 20th Century).
So the whole thing is completely consistent - the models respond to the historical forcings which have all varied over time, with the natural forcing being greater at first, because of more rapid change in solar irradiance while CO2 and other GHG concentrations have yet to build up, and then anthropogenic forcing being greater later as solar irradiance changes less while CO2 and other GHG concentrations have built up to higher levels. There is no magic change in the physics half-way through the century, it is just the differences in external forcing.
BBD. It looks like you're the only batsman at the crease for warmism today,(although having said that the others like Hengist and ZDB are definitely tail enders in debate), so I'm sorry to fling another one at you:
"there is no evidence for any forcing sufficient to explain most of the recent warming except CO2."
I believe the question that most people would ask on here is "What is the evidence that it is CO2?" By evidence I mean a quantitative analysis of the forcings introduced by the increase in CO2 over time, with a relationship between the forcings and the temperature increase. There is no real relationship between CO2 and temperature in the records, in fact apart from the obvious degassing caused by temperature rises the temperature and CO2 seem to live separate lives. I have been burned (no pun intended) too many times by jumping to what I thought were obvious conclusions, and in this case with so many vested interests wanting AGW to be a fact taking anything at face value isn't recommended.
RE: BBD
"If climate sensitivity is low, this means that the climate system sheds energy efficiently into space.
How then does enough energy accumulate in the climate system under the modest DSR change from Milankovitch forcing to instigate glacial terminations?
Claims for a low or very low climate sensitivity are incompatible with known paleoclimate change."
Conversely, claims for significant positive feedback due to CO2 are incompatible with the ice core data. The 800 year lag between T and CO2 is the same whether in a warming from ice age or cooling into ice age. That is physically impossible if the RC CO2 model for exapl,ing ice ages is followed (which is what you seem to be doing) - starter event from Milankovich cycle, warming enhanced by large positive feedback casued by CO2. So how do you get back to an ice age? If the Milankovich effect is too small to get you warming and you need to invoke CO2, then Milankovich effects could never reverse the CO2 effect to get you back to an ice age. Follow the CO2 model and you get only 1 ice age, not repeated cycles. Logically, CO2 effect must therefore be very small and I suspect probably negligible.
BBD will say it has nor relevance ,i hope, but my theory is that this is a problem of balance between the ocean temperature/solubility phas diagram with CO2 in the atmosphere.
Normally a warming ocean throws more CO2 in the atmosphere
If you throw CO2 in the atmosphere first, then the balance needs to be found the other way round eg you need to add heat to the ocean.
As this happens, the ocean draws heat from the atmosphere, which cools.
So we have proved global cooling.
Which can be measured.
the whole discussion is futile btw.
If we , after many altercations and billions worth of more measurements and theories PROVE 100% that CO2 does not add to Warming, and there is cooling instead,
the whole clerus will do a turnabout and STILL blame CO2, now for cooling.
they just want more funds, for reports and for windmills.
Which we shld allow for , of course, but ONLY funds tapped from the BBC penshun fund. Becus thats the only place where there's still funds.
Why do models and observations disagree?
"...all models may be missing some fundamental climate process such as a nonlinear response to forcing. As discussed by Santer et al. [2005, 2008] it is not clear what this could be or why models and observations agree on short timescales but potentially differ on long time scales, given the same undamental physical processes. There may be natural processes that modulate behavior on decadal timescales that are not captured by any climate models. But with highly uncertain observations it remains most likely that residual observational biases underlie the disagreements with the models. However, if the models lack a basic process, then it urgently needs to be understood and incorporated".
Those darn observations!
Richard, is the Raymond Bradley in the irradiance paper the same Raymond Bradley in MBH1998, and the same Raymond Bradley who accused Prof Wegman of plagiarism in a document where he (Bradley) was cited 18 times in the text and 13 times in the references? Is it the same Raymond Bradley that said he would withdraw the accusation if Wegman took his critical report of MBH1998 from the Congress? I do hope it isn't, because that Bradley is a bad 'un as far as I can see, and isn't to be trusted in anything he says.
Second question. Can the models actually model the 20th century temperatures? I am assuming your quoting from the 4AR. Don't you have to include aerosols in the mix because the models give higher temperatures than the recorded readings?
Here's the estimable Dr Trenberth on the models in June 2007:
"None of the models used by IPCC are initialized to the observed state and none of the climate states in the models correspond even remotely to the current observed climate. In particular, the state of the oceans, sea ice, and soil moisture has no relationship to the observed state at any recent time in any of the IPCC models. There is neither an El Niño sequence nor any Pacific Decadal Oscillation that replicates the recent past; yet these are critical modes of variability that affect Pacific rim countries and beyond. The Atlantic Multidecadal Oscillation, that may depend on the thermohaline circulation and thus ocean currents in the Atlantic, is not set up to match today’s state, but it is a critical component of the Atlantic hurricanes and it undoubtedly affects forecasts for the next decade from Brazil to Europe. Moreover, the starting climate state in several of the models may depart significantly from the real climate owing to model errors. I postulate that regional climate change is impossible to deal with properly unless the models are initialized."
Me? I'd be careful about quoting the efficacy of models armed with the information above.
Two climatological theories emerge from the discussion:
"The temperature record in years required to identifying human effects on global-mean tropospheric temperature depends linearly on the number of years required to avoid explaining recent temperature declines."
and
"The temperature record in years required to identifying human effects on global-mean tropospheric temperature depends linearly on the number of authors prepared to squander their scientific credibility for the greater good of climatology."
Richard:
So - you're saying the output of the models depends on their inputs ('forcings'). If so - I'd propose that you don't understand the 'forcings' well enough to make predictions.
And by the way - the agreement between the graphs and observations is not good in your figures and should not convince anyone of anything. The noisy correspondence between your inputs ('forcings') and observations could easily be coincidental. Would you take a drug that had been 'proven' to be effective on the basis of such agreement? (As Paul Nurse might ask).
Hi geronimo
Yes it's the same Raymond Bradley on the irradiance paper, but there are other papers giving similar reconstructions (eg: Hoyt and Shatton, cited in Lean et al) - in fact thinking about it H&S would have been more relevant anyway as that's the one actually used in the models, I just remembered Lean et al first. So your statements about Bradley don't affect things.
You're right about aerosols. They are included in the anthropogenic simulations.
On this occasion I agree with Kevin Trenberth's statements above (NB I don't always agree with him...!) but the comments are about regional climate - here I'm talking about global mean temperature only. And in any case I think you have missed my point, as I was merely responding to the questions about possible inconsistency.
I probably should not have bothered to include the model figure, as my point would have been made simply by referring to the external forcings alone! ie: natural forcing greater in early 20th C.
Richsrd Betts: But these are results from models (simulations) which have no validation and are tuned with various assumptions (each model tuned differently, so that only one of them could be correct - which one if any?) to get the desired historical results. All the major uncertainties in the climate are ignored.
MarcH
You're confusing CAGW with AGW
Santer et al only discuss an issue relating to evidence of anthropogenic influence - they say nothing about impacts, catastrophic or otherwise.
Sep 5, 2011 at 11:36 AM | Richard Betts
Since the whole thrust of the alarmist veiw point has always been "its going to be bad, very bad" then surely any argument tryiong to back this premise must relate to CAGW, if its not catastrophic and merely relates to AGW then we're simply stating the obvious, ie man affects the enviroment through his actions. It has always been largely accepted on the sceptic side that there is an element of arthropogenic influence. Our argument is that its not the main forcing, merely a minor variable. Without the C in CAGW our point is made.
BBD writes:
"What Santer et al. and Richard Betts are saying is that comparing each decade to the previous one is more revealing than comparing each decade with itself."
I just love statistics. The claim is very simple: there is something real that increases incrementally but cannot be detected when observed and measured in increments of one year but can be detected if periods of ten years are compared. Obviously, the claim is inconsistent on its face. The only way to make it consistent is with statistical magic. When this is pointed out, what do you get as a response? More statistical magic. BBD, if you cannot tie your claim to something observable in the real world, why do you expect anyone to take it seriously. Your statistical magic shows only that you are truly adept at hiding the pea.
Apologies to Richard Betts. BBD associated his name with BBD's claim. I do not.
Richard, thanks for the links in your post (at 3:17 PM) though I have to say that my 'eyeball' assessment of the model behavior in the period 1900-1945 is not very convincing: the difference in general trend (slope) of the Black and Red lines seems quite significant although, absent a decent set of error bars on either, it's rather hard to make a fair judgement.
ZT said to Richard Betts:
Trenberth himself published the following in 2007.
"I have often seen references to predictions of future climate by the Intergovernmental Panel on Climate Change (IPCC), presumably through the IPCC assessments (the various chapters in the recently completedWorking Group I Fourth Assessment report ican be accessed through this listing). In fact, since the last report it is also often stated that the science is settled or done and now is the time for action.
In fact there are no predictions by IPCC at all. And there never have been. The IPCC instead proffers “what if” projections of future climate that correspond to certain emissions scenarios. There are a number of assumptions that go into these emissions scenarios. They are intended to cover a range of possible self consistent “story lines” that then provide decision makers with information about which paths might be more desirable. But they do not consider many things like the recovery of the ozone layer, for instance, or observed trends in forcing agents. There is no estimate, even probabilistically, as to the likelihood of any emissions scenario and no best guess."
Richard Betts Sep 5, 2011 at 3:17 PM
Many thanks for this post, having been told for a number of years that solar irradiance could not be possibly (insufficient energy) have caused the later 20th Century warming I am now met with a statement that it was strong enough earlier in the century to effect a similar occurance of warming. I need to take some time out and study the paper you have referenced.
Hi Phillip
The models are too complex and computationally expensive for us to tune them to get a match to observations in response to a number of time-varying forcings - they take months to run, even on a supercomputer, so it is simply impractical to do what you suggest (even if we wanted to, which we don't because we are not fiddling our results in order to support some political agenda, despite what some may claim.
Indeed - sunderland steve - if you are a regular here you'll have seen that I'm not an "alarmist", I just want to improve our understanding of climate change and variability (arising from whatever cause), mainly in order to inform sensible adaptation.
Dave:
yes you're right, the fit is not great, I'm not claiming the models are perfect and neither are IPCC (not Santer et al), I'm just saying that they give some warming in early 20th C and much of this is due to natural forcings. But like I said above, my point is merely that there is no inconsistency as the changes in relative strength of natural and anthropogenic forcings vary over the 20th C, which addresses the original question.
ZT:
nice analogy! Actually on a daily basis I have to make decisions over amounts of medication under conditions of high uncertainty. Without going into personal details here, I have to hedge my bets on the consequences of errors in either direction. The problem of administering medicine to a complex system such as a human body is an extremely good analogy for what we are doing to the climate - it is impossible to predict with certainty, but sometimes you have to make a decision, and when you do so you make the decision as informed as you can through a combination of scientific process understanding, experience, and a readiness to modify the response in the face of future improved understanding. That's what climate science is trying to do in informing (but not prescribing) policy - on adaptation as well as mitigation.
@ Richard Betts,
I wasn't intending to imply that you are an alarmist and I'm sorry to give that impression.
I am a regular on an irregular basis (if that makes sense) as i has a wife with MS and she has priority on my time. I have read many of your post and find them instructional. I thank you for that.
My point was more aimed at Santer. To wit, if he can only show AGW rather than CAGW then whilst very interesting in of itself it is not the disaster some would have us believe.
I will continue to read your postings with interest.
All the best Steve.
Opps typo, have a wife, not has :)
Can't read the whole thread now but I'd second a request for every AGW attribution paper ever published on the basis of less than 17 years' worth of data, to be listed, retracted, drawn and quartered, and for all those editors to resign and apologize (to me).
Hi Steve
OK, thanks, I assumed too much! All the best to you too.
sunderland steve: Opps typo, have a wife, not has :)
from this do I take it you have replace :) with a wife? Are the two mutually exclusive? Can you not have a wife and have a :) or am I unique to have :) and a wife?