New paper by Love et al suggests no prominent role for solar‐terrestrial interaction in global climate change. I’m providing it here for discussion.
We are not convinced that the combination of sunspot‐number,
geomagnetic‐activity, and global‐temperature data can, with
a purely phenomenological correlational analysis, be used to
identify an anthropogenic affect on climate.
Abstract
Recent studies have led to speculation that solar‐terrestrial interaction, measured by sunspot number and geomagnetic activity, has played an important role in global temperature change over the past century or so. We treat this possibility as an hypothesis for testing. We examine the statistical significance of cross‐correlations between sunspot number, geomagnetic activity, and global surface temperature for the years 1868–2008, solar cycles 11–23. The data contain substantial autocorrelation and non-stationarity, properties that are incompatible with standard measures of cross-correlational significance, but which can be largely removed by averaging over solar cycles and first‐difference detrending. Treated data show an expected statistically significant correlation between sunspot number and geomagnetic activity, Pearson ρ < 10^−4, but correlations between global temperature and sunspot number (geomagnetic activity) are not significant, ρ = 0.9954, (ρ = 0.8171). In other words, straightforward analysis does not support widely‐cited suggestions that these data record a prominent role for solar‐terrestrial interaction in global climate change.
With respect to the sunspot‐number, geomagnetic‐activity, and global‐temperature data, three alternative hypotheses remain difficult to reject: (1) the role of solar‐terrestrial interaction in recent climate change is contained wholly in long‐term trends and not in any shorter‐term secular variation, or, (2) an anthropogenic signal is hiding correlation between solar‐terrestrial variables and global temperature, or, (3) the null hypothesis, recent climate change has not been influenced by solar‐terrestrial interaction.
Citation: Love, J. J., K. Mursula, V. C. Tsai, and D. M. Perkins (2011), Are secular correlations between sunspots, geomagnetic activity, and global temperature significant?, Geophys. Res. Lett., 38, L21703, doi:10.1029/2011GL049380.
Conclusions
One of the merits of using three separate data sets in a correlational analysis is that intercomparisons can be made. After treatment for removal of autocorrelation and nonstationarity through simple averaging and differencing, we find statistically‐significant secular correlation between sunspot number and geomagnetic activity. This is expected,
and it serves as important support for our analysis method. On the other hand, after making the same treatment to the global surface temperature, correlations between temperature and either sunspot number or geomagnetic activity are not significant.
We have not, in this study, considered derived proxy metrics of relevance to climate change, such as reconstructed total‐solar irradiance [e.g., Fröhlich and Lean, 2004] or
interplanetary magnetic field [e.g., Lockwood et al., 1999]. Still, we believe that our methods are general, that they could be used for other data sets, even though our analysis, here, is tightly focused on specific data sets. [15] From analysis of sunspot‐number, geomagneticactivity, and global‐temperature data, three hypotheses remain difficult to reject; we list them.
(1) The role of solarterrestrial interaction in recent climate change is wholly contained in the long‐term trends we removed in order to reduce autocorrelation and nonstationarity. This possibility seems artificial, but we acknowledge that our method requires a nontrivial time‐dependence in the data that is different from a simple trend. Still needed is a method for measuring the significance of correlation between data sets with trends.
(2) An anthropogenic signal is hiding correlation between solar‐terrestrial variables and global temperature. A phenomenological correlational analysis, such as that used here, is not effective for testing hypotheses when the data record a superposition of different signals. Physics is required to separate their sum.
(3) Recent climate change has not been influenced by solar‐terrestrial interaction. If this null hypothesis is to be confidently rejected, it will require data and/or methods that are different from those used here.
Paper: http://www.leif.org/EOS/2011GL049380.pdf
h/t to Dr. Leif Svalgaard
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Dave Springer says:
November 17, 2011 at 8:32 am
That’s naive even for you. The distribution of the energy can vary a great deal and the controlling factors need not be energetic.
Please, that is the lame argument that the energy required to push a button to explode a nuclear bomb is minute. Does not apply to the climate.
Dr. S.
Reduce speed of the North Atlantic drift current, and see the reaction of your Nordic cousins. Tropics wouldn’t get any hotter, but the old Danes would be shaking ice out of their beards.
So it is pure coincidence that the historic turning points in the global temperature curve perfectly coincide with every third solar cycle minimum? The years are 1878, 1912, 1944, 1976 and whatever year we decide was the end of solar cycle 23 and start of solar cycle 24? See diagram at http://www.energiminnesfonden.se/index.php?option=com_content&view=article&id=74:klimatet-och-solcyklerna&catid=44:frittfram&Itemid=71 .
M.A.Vukcevic says:
November 17, 2011 at 9:33 am
Reduce speed of the North Atlantic drift current, and see the reaction of your Nordic cousins. Tropics wouldn’t get any hotter, but the old Danes would be shaking ice out of their beards.
We would then have to find a nice warm place elsewhere to pillage and plunder.
Svempa says:
November 17, 2011 at 9:47 am
So it is pure coincidence that the historic turning points in the global temperature curve perfectly coincide with every third solar cycle minimum?
If the PDO has a sequence of 30-yr interval alternating between warm and cold, and the sunspot cycle is close to 10 years, then three cycles would fit in each interval, so no surprise that they do.
M.A.Vukcevic says: November 17, 2011 at 6:38 am
……
You might find this useful starting point
=======
Heated water from the tropics is circulated northwards and provides a “good” climate for N. Europe.
But only X (TSI) amount of energy heats this whole earth thing. So if there were no N. Atlantic THC current then the energy would be less well distributed. But in total it is still only X. so the warm places get warmer and the cool places get cooler. The difference in energy distribution without the THC would be lessened by the increased evaporation from the tropic seas when at a warmer temp. (keeping the nights warmer and the days cooler – which wins? increased albedo or increase water vapour feeback?)
The N.Atlantic current does not horizontally circulate. It is vertical. Cold water at the pole causes the denser water to sink and from there flow south. But why is is more dense?
1 it is cooler
2 it is more salty (evaporation and sea ice concentrating the salt) (hence the Themo-Haline circulation )
GW may decrease the polar salinity (more freshwater from ice) which in turn may affect the THC. We are currently embarking on a global experiment to prove this theory. But is this a good idea?
Leif, you say:
“If the PDO has a sequence of 30-yr interval alternating between warm and cold, and the sunspot cycle is close to 10 years, then three cycles would fit in each interval, so no surprise that they do.”
Sure thing. But what are the odds that the temperature turning points come exactly at solar minimums? After all, there are thirty years to choose from, so I would put the odds at 1:30 myself. Isn’t that at least a tiny bit significant :-)?
Svempa says:
November 17, 2011 at 10:33 am
Sure thing. But what are the odds that the temperature turning points come exactly at solar minimums? After all, there are thirty years to choose from, so I would put the odds at 1:30 myself. Isn’t that at least a tiny bit significant :-)?
The key word is exactly. Both measures have uncertainties, and solar cycles overlap, each being perhaps 17 years long http://www.leif.org/research/SHINE-2011-The-Forgotten-Sun.pdf . We have sunspot data 400 years back. Try to see how they fit.
Leif Svalgaard says:
“Please, that is the lame argument that the energy required to push a button to explode a nuclear bomb is minute. Does not apply to the climate.”
I’m so relieved that positive feedbacks ‘ don’t apply to the climate’
Ockham says:
November 17, 2011 at 10:54 am
I’m so relieved that positive feedbacks ‘ don’t apply to the climate’
You are welcome. Glad to have helped in this regard.
JJThoms says:
November 17, 2011 at 10:15 am
…………
These are all minor drivers. Major ones are the subtropical and subpolar gyres. For the North Atlantic climate it is the subpolar gyre that does the pumping work. It looks like that the summer months’ deep water convection in the Nordic Seas drives changes in the N.A. SST, or at least that is what data shows:
http://www.vukcevic.talktalk.net/theAMO.htm
Solar heating input for all practical purposes (MWP, LIA and the current warming) is more or less constant. The AMO, PDO and ENSO are not due to the solar variation but to the heat distribution via oceans’ currents (south to north, east to west) and there are solid reasons why the ‘oscillations’ are happening:
http://www.vukcevic.talktalk.net/A&P.htm
You may know differently, but on the other hand I may know of the reasons you do not.
M.A.Vukcevic says:
November 17, 2011 at 11:56 am
there are solid reasons why the ‘oscillations’ are happening:
http://www.vukcevic.talktalk.net/A&P.htm
Don’t calculate correlation coefficients or regression lines on moving averages…
Yes Sir. It is physical process which is foundation of science, not the minutiae of a statistical enumeration even when applied correctly, let alone erroneously.
M.A.Vukcevic says:
November 17, 2011 at 1:14 pm
It is physical process which is foundation of science, not the minutiae of a statistical enumeration even when applied correctly, let alone erroneously.
Regression on moving averages may mislead you into believing there is a physical process where there is none.
“”””” Leif Svalgaard says:
November 16, 2011 at 8:12 pm
George E. Smith; says:
November 16, 2011 at 7:33 pm
Well Dr Svalgaard, I’m not trying to raise a donnybrook with you, over some semantics.
Well, you certainly fooled me into thinking so 🙂 “””””
Well the 0.1 deg (or 72mK) I was thinking of (and which I thought YOU were referring to) was the 11 or twice in 22 year cycles of the TSI by about 0.1% p-p which you have explained to us several times, is due to sunspots (including the rim) being slightly higher output than the rest of the solar surface.
I would not expect that to have much influence on earth climate if it can’t be detected as an 11 year Temperature cycle on earth; BUT !! if you are suggesting a 0.1 % or 0.1 Deg C earth Temperature change over a ;longer time scale (100 or 1000 yrs); then of course I would agree that such a long term shift might eventually show up, in the climate.
I wasn’t aware that longer term TSI shifts of that order were occurring. (not talking geologic time scales).
Sorry for the confusion.
PS I believe that water vapor and cloud cover changes affect INCOMING SOLAR ENERGY capture and storage (in the oceans) to a considerably greater extent than any 0.1% ; and in that endeavor, H2O even gets assistance from O3 and CO2, which also absorb solar spectrum energy.
I wish people would stop saying that clouds REFLECT either solar energy or LWIR.
A water droplet takes a near colimated (0.5 degree angle) sunbeam, and converts it into a highly divergent (hemispherical) refracted distribution, which quickly becomes a completely isotropic scattered beam; that isn’t reflection. And the LWIR is totally absorbed by cloud droplets, so any LWIR coming back from the cloud is simply re-emission; so it is an inelastic scattering process; also NOT reflection.
Leif Svalgaard says:
November 17, 2011 at 1:59 pm
……….
In this case it started with physical process, the North Pacific Gateways were plotted some months before it was realised that they correlate with the PDO or SOI.
http://www.vukcevic.talktalk.net/NPG.htm
As you can see only two are critical, while the other two are not, or at least I am not yet aware of it.
Leif Svalgaard says:
November 17, 2011 at 1:59 pm
………..
See my post (nearly a year ago) here
http://wattsupwiththat.com/2010/12/09/the-story-told-by-the-southern-oscillation-index/#comment-546830
and I also mentioned also that these events occur well in advance of the SOI and PDO.
With Leif there so often seems to be a low signal-to-squawk ratio, it’s like a Punch and Judy show and it’s difficult to gauge the signal. I discern three interesting comments which I hope I’ve understood aright.
Alec Rawls explains how the omitted ocean currents make the paper useless
Boris Komitov [bad English, but I hear Russian expertise – solar/climate link well known, mid-latitude aurora significant, too simplistic…]
Piers Corbyn – page 37 tabulates with stunning clarity “what does and doesn’t drive climate” and AFAIK he does better-than-Met-Office forecasts based on correlations with solar activity, magnetic modulations, and lunar amplification, that have no need to understand the physics to help the farmers.
This seems to be an important point that Leif seems unable to grasp. I keep on seeing Leif make rude remarks about the work of Holle and Corbyn although I get the impression that they can both do better forecasts than the Met Office which surely deserves recognition.
“WTF?
First of all clouds don’t absorb they reflect. Second of all it’s not two Watts it’s less than one Watt.”
My! My! I must revise my notes then 🙂 It would seem by the NASA data though that the atmosphere, which I think includes clouds, absorbs ~61Wm-2 versus the surface at 174Wm-2. The clouds reflect ~26% of the SW solar and the Surface ~4% of the SW solar. If the ratio of either absorption or reflection changes, surface temperature changes. There is considerable more than CO2 to consider.
Since I used round numbers that must have thrown you off. The solar was 0.1 TSI impact on surface temperature and 0.05 was the potential surface temperature impact due to UV and near infrared variation that was a little more than expected. The 1 W was at the TOA which is the gross change in solar TSI, from ~1366 to 1367 or there about for solar minimum to maximum TSI change. Only about 0.2 to 0.25 might be felt at the surface, which I believe are the numbers Judith Lean used to estimate the climate impact due to TSI change.
If you compare Dr. Lean’s 2000 reconstruction with Dr. Svaalgard’s TSI reconstruction, Dr. Lean originally thought Solar produced more climate impact than is estimated now. Without some mechanism, it is unlikely her 2000 reconstruction is plausible. Variation in the ratio of surface to atmospheric absorption with prolonged minima, to allow Ocean buffering of higher frequency variation to subside, is somewhat plausible. Proven? Of course not. If a Doubling of CO2 could cause 3 to 5 degrees warming it would be impossible. That degree of warming due to CO2 is looking less likely every day. That makes smaller impacts more significant. 🙂
dallas says:
November 17, 2011 at 4:28 pm
“If you compare Dr. Lean’s 2000 reconstruction with Dr. Svaalgard’s TSI reconstruction, Dr. Lean originally thought Solar produced more climate impact than is estimated now. Without some mechanism, it is unlikely her 2000 reconstruction is plausible.
She has given up on the 2000 reconstruction long ago. Now she is much closer to mine. The old one was based on stars that were thought to be sun-like, but turned out not to be.
Leif, I know she gave up on it. Trying to make direct relationships to climate over the past couple hundred years is not something that I would expect to be successful. What she expected from the sun and stars she tried to relate to past climate. To me, it looks like she was close to what the climate impact was, just not the mechanisms that caused the impacts. There have been several examples in science where the right results came from the wrong relationships.
While the theory of the atmosphere/surface ratio impact, both from solar and radiant impacts of OLR may sound a little bizarre, the increase of vertical convection with CO2 increase isn’t in the models. We have not been through a prolonged solar minimum. but the impact are in the same order of magnitude as the change in vertical convection due to CO2.
It is not a huge impact, just a decent nudge, which has potential feed backs for a little more push.
dallas says:
November 17, 2011 at 6:25 pm
To me, it looks like she was close to what the climate impact was, just not the mechanisms that caused the impacts.
Yes, there may be some circular logic in her reconstruction, namely that a large variation is expected, so the data from the stars is accepted because it fits…. But if that makes it right, I’m not so sure about. In my book that rather goes against her..
What Love, et al., should have discovered. Total Solar Irradiance (TSI) accurately predicts Earth’s Global Average Surface Temperature (GAST) by a simple transfer function. See SGW here . GAST is the HadCRUT3 algorithm, annual from 1850 to 2006. TSI is the Wang (2005) model. The transfer function in two stages has six parameters (in one stage, four parameters), a linear combination of two lagged trend lines. The major trend is a 134 year running average, and the minor a 46 year running average, and the lag 6 years. The data sources are IPCC’s preferred choices in AR4, and the trend lines use IPCC’s method in AR4.
This result has a number of implications, none so important as the fact that AGW is invalid: neither man nor CO2 is the cause of observed GAST variations. Because of the independence of the data records, the result validates both the global temperature reduction by Hadley/CRU and the TSI reconstruction by Wang, et al. as major accomplishments in the field of climate. The result predicts the end of warming circa 2000, confirmed by a best estimate of 2004 in the latest HadCRUT3 data. The sensitivity of GAST to TSI suggests that Earth’s has no radiative equilibrium point in its warm state, but is slaved to the Sun. The lagging response is likely due to the high absorptivity of the ocean and its mechanisms of surface mixing and currents. Governments can restore CO2 at atmospheric concentrations to its proper place as a benign, beneficial greening agent, an optimal effluent in energy usage.
The transfer function amplifies TSI, tending to validate the independent results of Stott, P. A., G. S. Jones, & J. F. B. Mitchell, Do Models Underestimate the Solar Contribution to Recent Climate Change?, J.Clim., v. 16, 4079-4093, 12/15/03. and Tung, K. K., J. Zhou, & C. D. Camp, Constraining model transient climate response using independent observations of solar-cycle forcing and response, Geoph.Res.Lett., v. 35, L17707, 5 pp., 9/12/08. This amplification is likely the effects of cloud cover burn-off, a fast, positive feedback.
To be continued.
Continuing, Why Love, et al., Misfired.
A seeming decrease, over the past few decades, in the correlation between solar‐terrestrial data and global temperature [Solanki and Krivova, 2003; Le Mouël et al., 2005; Lockwood and Fröhlich, 2007] might be interpreted as being anthropogenic. Id.,, p. 4 of 6. Conversely, the close relationship between solar-terrestrial data and global temperature indicates the absence of any anthropogenic signal.
The Wang, et al., (2005) model emulates the Sun’s surface phenomena. Those authors tuned their model to reproduce the records of sunspot and the geomagnetic aa index observations, the two solar parameters Love, et al., relied on for their analysis in comparison to the same HadCRUT3 record. So why didn’t they discover the underlying dependence of temperature on the Sun? There are several reasons.
Love, et al., examined cross- and auto-correlations between the records, between their binned averages, and between first differences between the binned averages. The latter two frequently applied steps, binning and differencing, employ notoriously poor analytical techniques, certain to reduce measured correlations, and to lessen signal to noise ratios. They should be avoided.
Love, et al., mention causal relationships more than several times, but they never reported the lead or lag between temperature and either solar signal essential to causality. The zero lag correlation between these pairs of parameters is irrelevant, especially considering the dominant role of the high capacitance ocean is capturing, storing, and distributing solar energy over the globe. Temperature lags the Sun by as much as several times the length of the Mauna Loa CO2 record, about a dozen 11-year solar cycles.
Analyzing signals in noise is a delicate task, one in which IPCC demonstrates no skill. Cross- and auto-correlations are important tools, but they are not determinative. Ham-handed processing like binning and differencing above can strip natural correlations from signals. Conversely, processes like filtering and smoothing, as practiced by IPCC, can introduce false correlations. Correlations comprising parallel or patterned graphs are visuals that feed a scientists intuition, but they are not science. A correlation function has all the properties of a mathematical function, and at every lag, its value is a correlation number. It is a statistic, a mathematical thing, and at that, not yet science. Modern science requires cause and effect, with causality, and always objectivity.
Characteristics of, and relationships between, parameters need to be reduced by parameterization. See Love, et al. ¶6, p. 4 of 6. The job only begins with correlation studies. And if the correlations turn out to be significantly small, the job is not done; the rest of the job is just that much more difficult. Science calls for transfer functions, probability distributions, and functional equations, well fit to the data and complying with appropriate physical laws.
Leif, said “In my book that rather goes against her..”
I think I would hold it against cyclomania and time series analysis. Seems to be plenty of rabbit holes in statistics. Tsnonis et al, seem to be closer with the chaos approach until they moved into moving averages. I don’t think non-ergodic systems think much of averages. At least with fluid dynamic you can get a rough idea of time delays and scale. Anywho, if the AMO shift does happen, the non-linear relationships with make for interesting conversation. There is even a small CO2 non-linear conductivity relationship at -20C. Totally negligible of course 🙂
http://redneckphysics.blogspot.com/2011/11/radiantion-versus-conductivity.html
That is just speculation on my part, no fishing that day.