I really wish Gavin would put as much effort into getting the oddities with the GISTEMP dataset fixed rather than writing coffee table books and trying new models to show the sun has little impact.
This paper gets extra points for using the word “robust”. – Anthony
Solar trends and global warming (PDF here)
R. E. Benestad
Climate Division, Norwegian Meteorological Institute, Oslo, Norway
G. A. Schmidt
NASA Goddard Institute for Space Studies, New York, New York, USA
We use a suite of global climate model simulations for the 20th century to assess the contribution of solar forcing to the past trends in the global mean temperature. In particular, we examine how robust different published methodologies are at detecting and attributing solar-related climate change in the presence of intrinsic climate variability and multiple forcings.
We demonstrate that naive application of linear analytical methods such as regression gives nonrobust results. We also demonstrate that the methodologies used by Scafetta and West (2005, 2006a, 2006b, 2007, 2008) are not robust to
these same factors and that their error bars are significantly larger than reported. Our analysis shows that the most likely contribution from solar forcing a global warming is 7 ± 1% for the 20th century and is negligible for warming since 1980.
Received 17 December 2008; accepted 13 May 2009; published 21 July 2009.
Citation: Benestad, R. E., and G. A. Schmidt (2009), Solar trends and
global warming, J. Geophys. Res., 114, D14101,
doi:10.1029/2008JD011639.
hat tip to Leif Svalgaard
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.

Now let’s try some analysis using REAL data instead of model data…
In http://justtotheleftofvenus.izfree.com/htm/soundandfury/220709-analysing_temps.htm I demonstrate that the HadCRU3 temp record is dominated by a 60-year cycle.
According to Timo Niroma http://personal.eunet.fi/pp/tilmari/tilmari5.htm#60 there is a 60-year cycle connecting solar effects to climate.
My analysis shows that this 60-year cycle has a much stronger effect than a naive FFT would suggest.
Maybe the sun affects climate more than Gavin is willing to admit…?
(BTW I’d like it if people would take a look at my analysis and comment on it)
timetochooseagain (16:03:27) :
Note that temperature data end in 2000-look how the models are shooting off at that point. Now, after that, did the models continue to track? No, they didn’t.
Talk about a cherry pick! I think we are back at zero again and – hey – hindcasting isn’t easy! It must take at least 6 years to correctly hindcast real data.
The Iceberg (07:33:45) : Could it be due to … well … ocean currents … or … maybe … cloud cover variation. Oh, of course not – what am I thinking? The behavior of climate rests only on CO2 and a few other GHGs and TSI which is constant.
press (07:52:10) :
Gavin using only TSI and ignoring other parameters such as the magnetic impact […]
What magnetic impact?
Gavin produces and writes an easy study. What he should now do (if this is indeed his way of dismissing other hypothesis, leaving only CO2) is model ENSO affects against the temp rise. To do that he will need to start with the amazing correlation between these two measurements and then model it.
Question, if CO2 is capable of producing some or all of this warming and ENSO is capable of producing some or all of this warming, something in the back of my mind related to two loud sound sources not being additive makes me consider that we can ignore CO2 warming. It is not additive to the affects of ENSO on temperature.
Iceberg, global temps have been going up the last two years? The changes in temps in the last two years can be entirely explained by weather pattern variation. Are you saying that these wriggles are evidence of global warming? If you do, you are dismissing in total ocean temps, currents, trade winds, jet stream behavior, local geographic climate parameters, etc. Yours seems an overly dismissive statement to me.
Jimmy Haigh and jmc,
the GHG hypothesis does NOT violate the second law of thermodynamics. I have read this paper, initially with great expectation, but was disappointed. Mostly it is a textbook on various physical processes including irrelevances like Freznels law. When you finally get to the analysis of the supposed violation of the second law of thermodynamics, there are serious flaws.
The second law states that heat cannot of its own accord travel from a cold place to a warm place. If it did so, the cold place would become yet colder whilst the warm place gets hotter. Yet GHG works on the principal that outgoing LR radiation is absorbed and re-emitted by the GHG and some of this re-emitted radiation will come back down to the ground and have a warming effect. Since the region where the GHG does this is cooler than the ground, this is supposed to violate the second law of thermodynamics.
I spent days thinking about this paradox, and I came to the conclusion that the radiation is NOT heat. It is not heat that is flowing to the warmer area. I would say it is quite acceptable for a molecule of CO2 to absorb a photon of energy and reradiate it. In fact this must happen. Can that photon then be used to warm a surface that is warmer? Yes. If the photon is emitted in that direction then it will be absorbed by the warmer ground. The fallacy lies in the fact that the paper overlooks the fact that this energy came from the warmer ground in the first place. In order to violate the second law, more energy would have to migrate from the cooler area than was being received from the warmer area with the result that the cooler area would cool still more. This does not happen. This area in the mid troposhere does not become cooler, although it does not warm as much as the climate models predict. If this is a violation of the second law, then a thermos flask must also be impossible, because that radiates IR from the cooler reflective surface to the warmer liquid.
My only question on this paper is if a layman like myself can see this is not a violation of the second law, then how can these physicists not see this?
Have they Iceberg? Give stats please ( not GISS )
I don’t think it helps to dismiss the Benestad-Schmidt paper out of hand. For one thing, there is much in conveniently conceded in one place. It is critical of the simple correlation analyses. It acknowledges that the climate senstitivity may be different for different forcings. The value to be derived from the paper is an abandoning of the simple linear chart arguments for the AGW hypothesis by emphasizing the poor value and reliability of linear methods when applied to the complex nonlinear climate system. It may be self-serving of Gavin to be arguing this, because it points to the need for climate models. However, a need for climate models does not mean that the climate models we have are up to the task. Diagnostic studies have show they have errors and correlated biases far larger than the task at hand.
The second part of the paper, appears to be an attempt to debunk the methodology of Scaffeta and West in much the same way as McIntyre and McKtrick did the methodology of Mann, et al. There is even an echoing of the complaints about the lack of availability of data and details on the methods, and the showing of a trend from the S&W method, even when applied to forcing time series without a trend, in this case “constant”, rather than statistical white noise. I would think either the S&W methodology needs to be rehabilitated or the B&G analysis found faulty.
While B&G are noting the inadequacies of linear methods, they still are willing to draw simplistic conclusions about attribution from them, such as emphasizing the lack of a significant solar trend in the latter half of the 20th century, while clearly favoring GHG attribution for 1980 to 2000, even though it matches the temperature trend from 1950 to 1980 worse that solar. The problem is not mere selective focus, but one of semantics. As the climate commitment studies showed, just because a forcing plateaus as solar forcing did circa 1950, doesn’t mean that the temperature will also plateau. The thermal inertia of the oceans delays the response, with most of the temperature response occuring in the first few decades as the ocean mixing layers adjust, but with the deep ocean extending the complete response for millenia or more. However, the temperature DID plateau and even cool a bit in the next few decades after 1950. What gives? It appears that humans did influence the climate via aerosols, interrupting the solar and GHG warming responses and allowing those responses to resume perhaps with a catch up vengenence in the 80s and 90s. The sematics of the issue is whether to conclude that the 80s and 90s warming is a resumption of the climate response to the warming influences of solar and GHG gases or is due to the ending of an aerosol cooling event. There is no way the GHG trend accounts for the steepness of the temperature rise any better than the plateau in admittedly high solar activity does. We need models.
However, while B&Gs use of models may have some validity in evaluating the S&W methodology, B&G did nothing to rehabilitate the models from the wealth of issues found in the diagnostic studies prepared for the IPCC FAR and also published since. It should be assumed that the models used still have a positive surface albedo bias, still fail to reproduce the amplitude of the response to the solar cycle detected in the observations, still have a positive rather than negative tropical cloud feedback, and still reproduce only one half to one third of the increase in precipitation observed in association with the warming. B&G and the peer reviewers of their paper should be faulted for drawing conclusions about the attribution of the recent warming without having satisfactorially addressed these issues.
However, given the recency of the publication, B&G should not be faulted for their model’s significant under-representation of the warming contribution of black carbon.
Whoops, I should have been using “B&S”, instead of “B&G”. apologies to Gavin.
sorry Lief was referring to magnetic flux http://www.mps.mpg.de/solar-system-school/evaluation/balmaceda.pdf
Anyhow while you are here. Answer me this. Assuming, and this is accepted that the core output of the sun is a constant ( diameter and pressure calculated).
1. During a solar cycle maximum we are losing energy, as can be measured.
2. During a solar minimum we are losing less energy.
3. During a series of high solar max cycles we should then be losing more energy than the core produces?
OK my point here is. The core output of the sun is constant given the parameters and pressures, however the energy output at the surface is not and there is no dispute here.
So during a period of high solar surface output as has been the case in recent cycles is it not logical that the surface plasma is losing energy, cooling, and this in itself slowing the fluidity and conductivity of the plasma?
Is this not what may be ultimately the reason for cycles?
I will go one further. In engineering ( I am one ) if you cool a surface of a given object it takes time to recover to its previous equilibrium, all factors being equal. Why is the sun any different?
press (09:16:54) : You seem to be laboring under the assumption that the Sun is accurately modeled. I noticed no one accurately predicted the length of the current and continuing minimum. That’s because no one understands in detail how the Sun works, including you
The solar cycle last 11 years. This is the MAIN solar variability. There are ups and downs of global temperature every 11 years? Answer:NO.
The variation of solar activity during the whole 20 th century is negligible in comparison to the 11-year solar cycle (see the year-to-year solar variability graphs).
Obvious conclusion: solar forcing is very small. 20 th century warming CANNOT be caused by this forcing. Specially the post-1980 warming, when solar activity remained nearly constant on timescales longers than the 11-year cycles (actually it REDUCED a little).
Metaphysisist Joe Romm’s top 5 posts or first 5 are all stories of the future. How do we deny a report about the future? It is not falsafiable.
If we are going to talk about the future, we will need perfect math models from schmidt.
Vincent (08:41:58) :
The volumetric heat capacity of air makes it impossible to hold heat as the sea water does, having 3227 times more capacity. Hoping a ball of fire jumping around in the atmosphere?…the only fire to find is their firing imagination.
press (09:16:54) :
sorry Leif was referring to magnetic flux http://www.mps.mpg.de/solar-system-school/evaluation/balmaceda.pdf
It doesn’t load [hangs], so can’t comment.
is it not logical that the surface plasma is losing energy, cooling, and this in itself slowing the fluidity and conductivity of the plasma?
The temperature of the solar plasma [the 99% that is not magnetic] is absolutely constant to our best ability to measure it.
The magnetic activity adds a little extra [0.1%] while it lasts, and it comes and goes with the solar cycle. The magnetic field does drag on the movements of the plasma, but very little, and it is mostly the other way around: the plasma moves the magnetic field around. but, the bottom line is that the effect of solar activity compared to the rest of the output of the Sun is minute, and is in any event cyclical [comes and goes].
I am beginning to think (following C.G.Jung´s psychic energetics) that the trouble behind global warming/ecology/green movements, etc. it is a pernicious displacement of unchanneled libido energy, which it is reaching its critical mass within some of the followers/believers/fanatics…:-)
Yes not quoting GISS, try HADCRU or NCDC. ?
Pamela, There is a close fit between ENSO and Global temps. I’ve modelled it many times. However what is also very evident is the underlying background increase in temperature i.e La Nina 2007, brought temps down by the suggested amount but only to a higher low level than previous due to background warming. Also ENSO is a capacitor of global temperature, a short term varience, it is not a true driver of global temperatures.
I’ve heard the arguement of lags, but have yet to see any evidence of where the excess temperature is stored before and during the lagging effect, there is no evidence in this storage mechanism in ocean currents deep or surface.
Leif Svalgaard (09:50:31) :
press (09:16:54) :
“sorry Leif was referring to magnetic flux http://www.mps.mpg.de/solar-system-school/evaluation/balmaceda.pdf ”
It doesn’t load [hangs], so can’t comment.
Well, it finally loaded. There are two things wrong with the magnetic flux [apart from it not having any impact on its own – except through TSI]:
1) The reconstructions are based on the Group Sunspot Numbers which are much too low before the 1870s, and
2) on Lockwood’s 1999 reconstruction of the Open flux [used to calibrate (1)]. Not even Lockwood believes that finding anymore. To see what is wrong with the 1999 claim:
http://www.leif.org/research/Reply%20to%20Lockwood%20IDV%20Comment.pdf and
http://www.leif.org/research/Comment%20on%20McCracken.pdf
Roger Pielke Senior Discusses this paper and wonders how Lean and Rind and Schmidt and Benestad can come to such different conclusions:
http://climatesci.org/2009/07/22/new-paper-how-will-earths-surface-temperature-change-in-future-decades-by-lean-and-rind-2009/
And notes that both could easily be wrong because 1. Both use surface temperatures rather than ocean heat content 2. both do not deal with the issue of the effect of circulation changes.
I could scarcely agree more.
Sound and Fury says:
First of all, B&S do not just use “model data”. They use both model data and real data. The advantage of the real data is, of course, that it is real data. However, the advantage of the model data is that you are able to know the “right” answer…i.e., you know how much of the warming in the model is due to changes in solar forcing (because you can run the model turning on and off the various forcings) and hence you can test how well various data analysis procedures give you the correct answer.
Second of all, when doing FFTs, you have to look carefully at the issue of significance. I am very skeptical of how significant your 60-year cycle is since that represents a cycle length that is about half the data record. Hence, there are not very many cycles in the data record…and the error bars should be correspondingly large.
henrychance (09:41:49) :
Metaphysisist Joe Romm’s top 5 posts or first 5 are all stories of the future. How do we deny a report about the future? It is not falsafiable.
If we are going to talk about the future, we will need perfect math models from schmidt.
I’ve just taken a $1000 bet on with Joe Romm that 2010-2019 will be cooler than 2000-2009. He gave me 2:1 odds in my favour as well!
commonsense (09:40:37) : “There are ups and downs of global temperature every 11 years? Answer: NO.”
WRONG.
http://www.sciencebits.com/files/articles/CalorimeterFinal.pdf
“The variation of solar activity during the whole 20 th century is negligible in comparison to the 11-year solar cycle (see the year-to-year solar variability graphs).”
Have you ever heard of a low pass filter? No? Then I can see why you have no clue what you talking about.
press (09:16:54) :
sorry Leif, I was referring to magnetic flux
In spite of the flaws of the paper, their conclusion is in the right direction [just not enough]:
-This first physics-based reconstruction of TSI back to the Maunder Minimum suggests an increase of about 0.80 W/m2 since 1700, with the lower limit being about 0.60 W/m2. This value is much lower than in previous works that may have important implications on Sun-climate relations.
Vincent says:
That’s a very good question. I suppose even people who should know better can delude themselves…or perhaps they do know better but are trying to delude others.
From what I can make out from what your saying, I think your explanation of where G&T are wrong is basically correct but I think it can perhaps be said more simply: G&T assume that because the earth’s surface is warmer in the presence of an IR-absorbing atmosphere than in its absence, this means there is a net flow of heat from that atmosphere to the surface. However, this assumption is not correct. In fact, the net flow of heat is always from the warmer surface to the cooler atmosphere. This may seem counterintuitive since you might wonder how then the earth’s surface ends up warmer! The answer is that in the comparison case, in which there is no IR-absorption, all of the radiation emitted by the earth’s surface escapes into space. Thus, even if an IR-absorbing atmosphere returns only a small part of that emitted radiation (and, in particular, less than it receives) back to the earth, this will still result in the earth being warmer. And, there is nothing particularly mysterious about this…Basically any problem in radiative heat transfer that involves a “heat shield” will show this sort of effect.