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
Vukcevic (2011) wrote:
“The most likely scenario is the existence of a common driver, whereby the atmospheric pressure responds rapidly whilst the ocean’s temperature responds some years later (see addendum).”
http://hal.archives-ouvertes.fr/docs/00/64/12/35/PDF/NorthAtlanticOscillations-I.pdf
Thanks for giving us something to think about Vukcevic.
You’re looking not at a lag but rather the time integral of spatial pattern. I’ve outlined this here:
http://wattsupwiththat.files.wordpress.com/2011/10/vaughn-sun-earth-moon-harmonies-beats-biases.pdf
I can suggest taking a look at the following expositions if you’re unfamiliar with pressure gradient force & thermal wind:
a. http://wattsupwiththat.com/2011/10/15/shifting-sun-earth-moon-harmonies-beats-biases/#comment-769231
b. http://wattsupwiththat.com/2011/11/10/aurora-borealis-and-surface-temperature-cycles-linked/#comment-794233
c. http://wattsupwiththat.com/2011/11/10/aurora-borealis-and-surface-temperature-cycles-linked/#comment-795243
Caution: At centennial timescales the asymmetries in the geomagnetic field are spatially nonstationary relative to ocean-continent heat-capacity asymmetries (since the continents aren’t moving very fast). (I’m left wondering if you’ve overlooked this when interpreting some of your graphs.)
Have you ever made chocolate chip cookies? The chips remain hot enough to burn your tongue after the cookie dough part has cooled. If the chips are clustered on one side, the cookie is spinning, you’re blindfolded, and you take a bite…
Best Regards.
@Boris Komitov
здрасти.
да, планина и море.
And thank you very much for stopping by. Please join our discussions again when you can.
Jeff Glassman says:
November 17, 2011 at 8:06 pm
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.
As support for their methods, Love et al. points out the they do find a significant correlation between solar activity and geomagnetic activity. A correlation we know shows a real physical cause-effect situation. So, their method works for that case.
Paul Vaughan says:
November 17, 2011 at 9:31 pm
………………..
The article you refer to attempts to ‘demystify’ the AMO and show it as an integral part of the events, and not some mysterious natural driver without cause but with a large consequence. Your comment regarding integration is indeed on the right lines, but for time being it was far simpler to dispense with it and just refer to it as a delay.
The article is only one of three or more, the next one (very soon) relates to CET with detailed explanation of
http://www.vukcevic.talktalk.net/CET-NV.htm and
http://www.vukcevic.talktalk.net/HMF-T.htm
then the following one will explain the NAP, its origins and its overriding role in the most of the North Atlantic’s events, and how it is inextricably linked with the geomagnetism .
Dr. Komitov makes lot of important points: there is indeed large N/S asymmetry in many respects, geomagnetic has been totally ignored, and the same is with the GMF – solar activity negative correlation, where the CMEs have a critical role:
http://www.vukcevic.talktalk.net/MF.htm
I hope my efforts in that respect are of some use to anyone interested.
http://www.vukcevic.talktalk.net/FB.htm
Leif Svalgaard, 11/17/11, 10: 58 pm,
As support for their methods, Love et al. points out the they do find a significant correlation between solar activity and geomagnetic activity. A correlation we know shows a real physical cause-effect situation. So, their method works for that case.
What they actually showed was that two modes of solar activity, sunspots and geomagnetic activity, were significantly correlated, and that that correlation survived their binning and differencing. In fact, applying the teaching of Wang, et al., (2005), binning would have caused a correlation. They attribute geomagnetic activity to the Sun’s open magnetic flux and sunspots to the closed magnetic flux, and say,
At least on timescales on the order of or less than a solar cycle, the open flux is rather poorly correlated with sunspot activity, unlike the closed flux that controls the solar irradiance. These differences need to be taken into account when using cosmogenic isotope or geomagnetic activity records to infer changes in solar irradiance. Id., p. 522.
Love, et al., binned into full solar cycles, from minimum to minimum. Their analysis needs to be reworked to determine the effects of their binning, and whether their significant correlation is an artifact or real, as Wang, et al., leave open as a possibility.
The apparent correlation between sunspots and the aa index is almost a text book example of the error in your statement that correlation we know shows a real physical cause-effect situation. Two independent phenomena with a common cause may be correlated, or not, without a cause and effect relationship between them. Where, as here, the processes tend to be cyclic, an effect may be uncorrelated with its cause with a quarter cycle lag. Correlation raises a suspicion that a physics-based, C&E model might have predictive power, but the statistic needs the full treatment of the cross correlation function, which neither Love, et al., nor IPCC appear to do.
Jeff Glassman says:
November 18, 2011 at 5:29 am
……………..
Two independent phenomena with a common cause may be correlated, or not, without a cause and effect relationship between them.
Absolutely agree.
Here is a good example of two independent phenomena with a common cause strongly correlated
http://www.vukcevic.talktalk.net/theAMO.htm
but there is no direct cause – effect relationship.
Jeff Glassman says:
November 18, 2011 at 5:29 am
The apparent correlation between sunspots and the aa index is almost a text book example of the error in your statement that correlation we know shows a real physical cause-effect situation. Two independent phenomena with a common cause may be correlated, or not, without a cause and effect relationship between them.
In this case, there is a strict causal relationship between the two. The aa index directly related to the Sun’s magnetic field which the sunspots are but a visible proxy of.
@Jeff Glassman (November 18, 2011 at 5:29 am)
Simple cross-correlation analysis (including the multiscale time-integrated variety) has a role during preliminary exploration, but it should then be immediately & patently obvious to any sufficiently cognizant explorer that it alone canNOT finish the job because of the nonstationarity.
@M.A.Vukcevic (November 18, 2011 at 12:51 am)
average ≠ gradient
blend ≠ contrast
–
Looking forward to your articles Vukcevic. With your web links you succeed at stimulating appreciation of nature where mainstreamers fail. A sterilized medium can’t grow ideas. Extension of trust to the audience to judiciously interpret uncensored raw material is one of few means available to overcome the gravity, friction, & activation energy needed to lift our collective desire for awareness over the mountains into the neighboring basin of superior conception. Refinement will be a breeze for mainstreamers in the valleys & sea of fertility. Guiding flow from the desert demands sacrifice.
–
a + b ≠ a – b
They’re different variables.
Leif Svalgaard says:
November 18, 2011 at 6:34 am In this case, there is a strict causal relationship between the two.
Are you sure?
Leif Svalgaard, Edward W. Cliver :
The IHV index is used to successfully reconstruct yearly-averages of the range indices AM, AP, and AA from 1959 through 2000.
Hathaway:
These indices are derived from magnetometer data recorded at two points on opposite sides of Earth: one in England and another in Australia. IHV data have been taken every day since 1868.
The amount of geomagnetic activity now tells us what the solar cycle is going to be like 6 to 8 years in the future. We don’t know why this works. The underlying physics is a mystery. But it does work.
http://science.nasa.gov/media/medialibrary/2006/12/21/21dec_cycle24_resources/hathaway1_strip2.jpg
As we are currently witnessing it does NOT.
The obvious lack of correlation cost the top NASA’s solar scientist his reputation.
@Jeff Glassman (November 17, 2011 at 8:06 pm)
Your comments about differencing, binning, & filtering are neither sensible nor well-founded. Your suggestion about PDFs is blindly naive (the kind of ignorance-based uncertainty promotion one expects from Climate Etc.). The climate pie is big enough that we can each have a piece of it. I can suggest that you focus on gradients rather than averages when brainstorming physical models. EOP (Earth Orientation Parameters) inform us clearly about SIMPLE climate asymmetries. Leave the data exploration to capable parties.
Regards.
@dallas
As Tomas Milanovic has patiently explained at Climate Etc., Tsonis, Swanson, & Kravtsov (2007) were NOT doing chaos theory. They were doing statistics. The online climate discussion has become polluted with gross misinterpretations of that timely, stimulating paper. It appears that it will take years – perhaps decades – to straighten out the misinterpretations.
Leif Svalgaard, 11/18/11, 6:34 am
You say, In this case, there is a strict causal relationship between the two. The aa index directly related to the Sun’s magnetic field which the sunspots are but a visible proxy of.
First, I don’t know what might mean by a strict causal relationship, but it seems to contradict your claim that the relationship is a proxy.
The next problem is your second sentence is not a strict sentence because it lacks a predicate.
Thirdly, you have lost the thread of the discussion. As I pointed out at 5:29 am, you incorrectly replaced Love’s correlation between sunspots and the aa index with two different variables, correlation between solar activity and geomagnetic activity. Now you have replaced it with yet another pair, the Sun’s magnetic field and sunspots.
Wang, et al., decompose say the Sun’s (external) magnetic field into two components, one of being large dipoles of two varieties: open and closed. They claim the open large dipoles control the aa index and the closed large dipoles give rise to sunspots, and that on short time scales the two are rather poorly correlated.
As you correctly repeat, The aa index is directly related to the Sun’s magnetic field. But you weakly claim that sunspots are only a proxy of the Sun’s magnetic field when you could have asserted strongly that they are also directly related to the same magnetic field. But if the Wang, et al., model is correct, then the Sun’s magnetic field is not as you claim a common cause. That follows because that field comprises poorly correlated components that separately cause the aa index and sunspots.
Correction:
The obvious lack of correlation cost the top NASA’s solar scientist his reputation.
should be:
The obvious lack of causation cost the top NASA’s solar scientist his reputation.
Vukcevic, can you share a direct link to the plain-text raw (not anomaly) Iceland SLP data you’re using? If so, thanks.
Hi Paul
activation energy needed to lift our collective desire for awareness over the mountains into the neighboring basin of superior conception
That sounds almost poetic; you weren’t trekking across Montenegro, were you?
http://static.panoramio.com/photos/original/8275195.jpg
re: gradient
Often it tells story far better, here is a good example:
http://www.vukcevic.talktalk.net/LL.htm
Integration will come into its own too, all in good time.
I only got anomalies data, but if you enter Reykjavik ( 64°10’N, 21°57’W ) in
http://www.esrl.noaa.gov/psd/cgi-bin/data/timeseries/timeseries1.pl
you can get absolute values but only since 1950; then from the known anomalies and the recent absolute values you should be able to backtrack back to 1860s.
If interested, I will email the anomaly data file (I personally got from Dr. J. Hurrell, UCAR), I am sure he will not mind.
“”””” Paul Vaughan says:
November 18, 2011 at 6:54 am
@Jeff Glassman (November 18, 2011 at 5:29 am)
Simple cross-correlation analysis (including the multiscale time-integrated variety) has a role during preliminary exploration, but it should then be immediately & patently obvious to any sufficiently cognizant explorer that it alone canNOT finish the job because of the nonstationarity. “””””
Well I’m a believer that statistics is important; maybe even central to Quantum Mechanics.
Werner Heisenberg freed us all from the dogma of a pre-ordained future.
But no matter how fancy one wants to get with one’s statistical mathematics, it is all for naught, unless it is applied to valid data. Applying statistical mathematics to even a random set of numbers, is perfectly valid mathematically. It just that the results are quite useless; with no validity for any purpose (in the real universe).
So why is it that the disciples of climatism, seem to be woefully ignorant of the rules of sampled data systems.
Long before you get around to statisticating, you have to gather a valid data set to apply that to.
No Central Limit Theorem or othe statistical wizardry, can buy you a reprieve from a violation of the Nyquist Sampling Theorem.
So please; let’s gather valid data, before playing statistical mathematics games. A simple 2x violation of the Nyquist Theorem, renders even the data average invalid, because of aliassing noise folded back to zero frequency (another name for the average).
Pushing a button, and dropping an atom bomb on a city; even doing it twice; is an example of perfectly valid single point observations. But unless you do it in compliance with the Nyquist Criterion for SDS, you can make no future projections or intelligent statements about trends or standard deviations or any other statisticsl witchcraft.
@george E. Smith (November 18, 2011 at 10:46 am)
Aliased data is particularly informative if the nature of the aliasing is understood.
@M.A.Vukcevic (November 18, 2011 at 10:12 am)
http://www.cru.uea.ac.uk/cru/data/nao/
NAO
http://www.cru.uea.ac.uk/cru/data/nao/nao.dat
Azores
http://www.cru.uea.ac.uk/cru/data/nao/nao_azo.dat
Gibraltar
http://www.cru.uea.ac.uk/cru/data/nao/nao_gib.dat
Iceland
http://www.cru.uea.ac.uk/cru/data/nao/nao_ice.dat
MSLP (mean sea level pressure) Climatology (average annual cycle) Animation: http://i54.tinypic.com/swg11c.png
Don’t forget about NPI (North Pacific Index), AAM (Atmospheric Angular Momentum), & their integrals (see the link I provided above [ http://wattsupwiththat.com/2011/11/15/are-secular-correlations-between-sunspots-geomagnetic-activity-and-global-temperature-significant/#comment-800503 ]).
Regards.
@ur momisugly Paul VAughan,
You are correct. I tend to incorrectly use Chaos theory since they appeared to be setting up for one approach then fell back into time series statistics. The network portion did look to me to be a valid start to estimating probability states/relationships regionally, but Chaos is not my strong suit by any stretch.
M.A.Vukcevic says:
November 18, 2011 at 8:18 am
“In this case, there is a strict causal relationship between the two.”
Are you sure?
Yes, see below
The obvious lack of correlation cost the top NASA’s solar scientist his reputation.
Completely different thing. Hathaway was assuming that the amount of geomagnetic activity some time before minimum would be an indicator of the strength of the coming cycle. This may still be the case, but the problem occurs when there are several peaks, which one to select. And he went for the wrong one [in 2003]. Had he picked the one in 2008, he would have been right on the mark.
Jeff Glassman says:
November 18, 2011 at 8:56 am
First, I don’t know what might mean by a strict causal relationship, but it seems to contradict your claim that the relationship is a proxy.
Strict casual in this case means the following: we see a big sunspot on the sun, we observe that it flares or triggers a CME [coronal mass ejection] that we can directly observe. We follow the CME as it travels through space, we observe it go by our spacecraft. We observe it hitting the Earth’s magnetosphere [by its effect on the magnetic field on the ground as well as directly by other spacecraft. We observe with spacecraft the compression and perturbation of the magnetosphere and observe the particles accelerated towards the Earth, where finally we measure the magnetic effects of the electric current induced and the aurorae that result from it. As every step on the way we have direct observation and quantitative measurements. This is the strict causal relationship in every single individual case. It works in reverse too, there are no magnetic storms without a solar event causing it.
The next problem is your second sentence is not a strict sentence because it lacks a predicate. Not worth responding to.
You are not understanding what a proxy is. Let me explain: you take your temperature with an old-fashioned thermometer. What you see and measure is a length of a mercury string. That length is a proxy for the temperature. One can calibrate the proxy [the length] in terms of degrees.
Thirdly, you have lost the thread of the discussion. As I pointed out at 5:29 am, you incorrectly replaced Love’s correlation between sunspots and the aa index with two different variables, correlation between solar activity and geomagnetic activity. Now you have replaced it with yet another pair, the Sun’s magnetic field and sunspots.
Solar activity is measured by the number of sunspots, the number of sunspots is a measure of the sun’s magnetic field, and the aa-index is a measure of geomagnetic activity. All of these things are proxies of something, but as with the thermometer, every proxy can be calibrated to something physical [the length of the mercury string to temperature in degrees].
Wang, et al., decompose say the Sun’s (external) magnetic field into two components, one of being large dipoles of two varieties: open and closed. They claim the open large dipoles control the aa index and the closed large dipoles give rise to sunspots
This is a much too simplistic point of view and has to be properly understood. It is must more complicated than that. Not that I shall invoke authority here, but I am one of the foremost researchers in this field. To the next level of complexity, what happens is that the closed fields in sunspots can become ejected into the open field from the large dipole [called a CME]. The CME in turn hits the Earth and induced electric currents whose magnetic effect we measure and call ‘aa’.
All of these things are directly and causually connected.
M.A.Vukcevic says:
November 18, 2011 at 9:01 am
Correction: The obvious lack of causation cost the top NASA’s solar scientist his reputation.
There very likely is a causation, it was the correlation that failed [Hathaway picked the wrong peak]. And he has not lost his reputation at all. Making a mistake and acknowledging it is quite OK and does not ruin your reputation. Clinging to a theory long after it has been shown to be a failure and especially if based on shaky physics, is the reputation-killer.
Leif Svalgaard says: November 18, 2011 at 8:06 pm
……………..
Svalgaard:
These indices are derived from magnetometer data recorded at two points on opposite sides of Earth: one in England and another in Australia. IHV data have been taken every day since 1868.
Hathaway:
The amount of geomagnetic activity now tells us what the solar cycle is going to be like 6 to 8 years in the future. We don’t know why this works. The underlying physics is a mystery. But it does work.
http://science.nasa.gov/media/medialibrary/2006/12/21/21dec_cycle24_resources/hathaway1_strip2.jpg
Svalgaard:
This may still be the case, but the problem occurs when there are several peaks, which one to select. There very likely is a causation, it was the correlation that failed.
Since Hathaway does not know why this occurs, do you?
There is a further problem:
Assuming the solar origin only, then perhaps it should correlate better to the previous cycle then the next. I have read (red) (Svalgaard, & Cliver) paper couple of times, to make sure I understand what this is about. Care is taken to eliminate any changes caused by the Earth’s field itself, so any leftovers are assumed to be solar. In which case no geo-event should be preceding and correlating (conditional) with the long term IHV, but that is not the case (remember the ‘cringe’ post), there is also:
http://www.vukcevic.talktalk.net/HMF-T.htm
http://www.vukcevic.talktalk.net/CET-NV.htm
In the above NAP signal (my data is by no means perfect, but it is good for start) is delayed along time axis by number of years, in which case makes good match with both the SSN (since 1870’s, and may even agrees with your hypothesis that last few cycles may be a bit overrated, SC19 is a problem as usual) and the derivative of the CET (blue line), all the way back to start of the data. I think, there are good reasons why it is so, and the article http://www.vukcevic.talktalk.net/theAMO.htm
you dismissed as irrelevant, presents a very important link in the chain.
(Why the ‘average bin’ correlation isn’t representative for these types of data is clearly demonstrated by the illustration at the top of page 4, since no common size of ‘bin’ is suitable for both).
M.A.Vukcevic says:
November 19, 2011 at 2:37 am
Since Hathaway does not know why this occurs, do you?
Geomagnetic activity depends on the magnetic field and solar wind speed in space, which in turn are a reflection of the open magnetic flux which depends on the strength of the solar polar fields, which is thought to be a good precursor of the next cycle. The reason one should look a bit before minimum is that at that time the influence of CMEs and low-latitude solar activity is becoming smaller.
Vukcevic, can you share a direct link to the plain-text raw (not anomaly) “Temperature” data you’re using here [ http://www.vukcevic.talktalk.net/HMF-T.htm ]? If so, thanks.
Leif Svalgaard says:
November 19, 2011 at 7:02 am
………………
Agree with all that, but there are some exceptions (on occasions delay I found is larger then implied by PF-SSN relationship), so I shall leave it to that for time being.
Thanks for the help; it is appreciated despite appearances, which of course are often misleading. See you at the next solar thread.
Paul Vaughan says:
November 19, 2011 at 7:18 am
Vukcevic, can you share a direct link to the plain-text raw (not anomaly) “Temperature” data you’re using here [ http://www.vukcevic.talktalk.net/HMF-T.htm ]? If so, thanks.
Sorry, but I have to say no, anyone interested has to wait for my article to appear on-line. I have emailed data to Dr. Svalgaard (in confidence) since the HMF data (green line) is Svalgaard-Cliver property (available also in their paper).