Guest Post by Willis Eschenbach
I came across a curious graph and claim today in a peer-reviewed scientific paper. Here’s the graph relating sunspots and the change in sea level:
And here is the claim about the graph:
Sea level change and solar activity
A stronger effect related to solar cycles is seen in Fig. 2, where the yearly averaged sunspot numbers are plotted together with the yearly change in coastal sea level (Holgate, 2007). The sea level rates are calculated from nine distributed tidal gauges with long records, which were compared with a larger set of data from 177 stations available in the last part of the century. In most of the century the sea level varied in phase with the solar activity, with the Sun leading the ocean, but in the beginning of the century they were in opposite phases, and during SC17 and 19 the sea level increased before the solar activity.
Let me see if I have this straight. At the start of the record, sunspots and sea level moved in opposite directions. Then for most of the time they were in phase. In both those cases, sunspots were leading sea level, suggesting the possibility that sunspots might affect sea level … except in opposite directions at different times. And in addition, in about 20% of the data, the sea level moved first, followed by the sunspots, suggesting the possibility that at times, the sea level might affect the number of sunspots …
Now, when I see a claim like that, after I get done laughing, I look around for some numerical measure of how similar the two series actually are. This is usually the “R2” (R squared) value, which varies from zero (no relationship) to 1 (they always move proportionately). Accompanying this R2 measure there is usually a “p-value”. The p-value measures how likely it is that we’re just seeing random variations. In other words, the p-value is the odds that the outcome has occurred by chance. A p-value of 0.05, for example, means that the odds are one in twenty that it’s a random occurrence.
So … what did the author of the paper put forwards as the R2 and p-value for this relationship?
Sad to relate, that part of the analysis seems to have slipped his mind. He doesn’t give us any guess as to how correlated the two series are, or whether we’re just looking at a random relationship.
So I thought, well, I’ll just get his data and measure the relationship myself. However, despite the journal’s policy requiring public archiving of the data necessary for replication, as is too common these days there was no public data, no code, and not even a Supplementary Online Information.
However, years of messing around with recalcitrant climate scientists has shown me that digitizing data is both fast and easy, so I simply digitized the graph of the data so I could analyze it. It’s quite accurate when done carefully.
And what did I find? Well, the R2 between sunspots and sea level is a mere 0.13, very little relationship. And even worse, the p-value of the relationship is 0.08 … sorry, no cigar. There is no statistically significant relationship between the two. In part this is because both datasets are so highly auto-correlated (~0.8 for both), and in part it’s because … well, it’s because as near as we can tell, sunspots [or whatever sunspots are a proxy for] don’t affect the sea level.
My conclusions from this, in no particular order, are:
• If this is the author’s “stronger effect related to solar cycles”, I’m not gonna worry about his weaker effect.
• This is not science in any sense of the word. There is no data. There is no code. There is no mathematical analysis of any kind, just bald assertions of a “stronger” relationship.
• Seems to me the idea that sunspots rule sea level would be pretty much scuttled by sunspot cycles 17 and 19 where the sea level moves first and sunspots follow … as well as by the phase reversal in the early data. At a minimum, you’d have to explain those large anomalies to make the case for a relationship. However, the author makes no effort to do so.
• The reviewers, as is far too often the case these days, were asleep at the switch. This study needs serious revision and buttressing to meet even the most minimal scientific standards.
• The editor bears responsibility as well, because the study is not replicable without the data as used, and the editor has not required the author to archive the data.
So … why am I bothering with a case of pseudo-science that is so easy to refute?
Because it is one of the papers in the Special Issue of the Copernicus journal, Pattern Recognition in Physics … and by no means the worst of the lot. There has been much disturbance in the farce lately regarding the journal being shut down, with many people saying that it was closed for political reasons. And perhaps that is the case.
However, if I ran Copernicus, I would have shut the journal down myself, but not for political reasons. I’d have closed it as soon as possible, for both scientific and business reasons.
I’d have shut it for scientific reasons because as we see in this example, peer-review was absent, the editorial actions were laughable, the authors reviewed each others papers, and the result was lots of handwaving and very little science.
And I’d have shut it for business reasons because Copernicus, as a publisher of scientific journals, cannot afford to become known as a place where reviewers don’t review and editors don’t edit. It would make them the laughing stock of the journal world, and being the butt of that kind of joke is something that no journal publisher can survive.
To me, it’s a huge tragedy, for two reasons. One is that I and other skeptical researchers get tarred with the same brush. The media commentary never says “a bunch of fringe pseudo-scientists” brought the journal down. No, it’s “climate skeptics” who get the blame, with no distinctions made despite the fact that we’ve falsified some of the claims of the Special Issue authors here on WUWT.
The other reason it’s a tragedy is that they were offered an unparalleled opportunity, the control of special issue of a reputable journal. I would give much to have the chance that they had. And they simply threw that away with nepotistic reviewing, inept editorship, wildly overblown claims, and a wholesale lack of science.
It’s a tragedy because you can be sure that if I, or many other skeptical researchers, got the chance to shape such a special issue, we wouldn’t give the publisher any reason to be unhappy with the quality of the peer-review, the strength of the editorship, or the scientific quality of the papers. The Copernicus folks might not like the conclusions, but they would be well researched, cited, and supported, with all data and code made public.
Ah, well … sic transit gloria monday, it’s already tuesday, and the struggle continues …
w.
PS—Based on … well, I’m not exactly sure what he’s basing it on, but the author says in the abstract:
The recent global warming may be interpreted as a rising branch of a millennium cycle, identified in ice cores and sediments and also recorded in history. This cycle peaks in the second half of this century, and then a 500 yr cooling trend will start.
Glad that’s settled. I was concerned about the next half millennium … you see what I mean about the absence of science in the Special Edition.
PPS—The usual request. I can defend my own words. I can’t defend your interpretation of my words. If you disagree with something I or anyone has written, please quote the exact words that you object to, and then tell us your objections. It prevents a host of misunderstandings, and it makes it clear just what you think is wrong, and why.
kuhnkat says:
January 21, 2014 at 11:08 pm
Am I going to do them all? No, my stomach is not that strong.
Are they ALL weak? I don’t know. I’ve only looked at four of them, and they were all depressingly bad.
w.
Paul Westhaver says:
January 22, 2014 at 2:05 am
No, they are not “related without a doubt”. What part of an R^2 of 0.13 and no statistically significant correlation seems unclear?
However, you illustrate a problem, one that is in the title of the journal. Humans are pattern recognition machines … to the point where we often see a pattern where none exists. That’s why we have math … and it’s why a paper which makes no attempt to do the math is handwaving rather than science.
w.
Brian H says:
January 22, 2014 at 2:14 am
Since when is requiring a calculation of correlation for a claimed relationship “churlish”? That’s nonsense, it’s a standard scientific requirement to make such a calculation before publication of such a claim.
w.
Hi Willis
First of all I am recognised as a ‘sceptic pseudo-scientist’, although I give primate to my engineering degree to the higher science degree and often say ‘I am an engineer not a scientist’.
Of the ‘oscillating’ things going in and out of phase:
few years ago I looked at the AMO (9yr) and SSN (11yr), a perfectly normal thing:
http://www.vukcevic.talktalk.net/AMO-SSN.htm
Hey, ‘pattern recognition’ not Copernicus but ‘vuk’ style had me occasionally speculate: is there anything more to it, and speculation lead to a highly speculative outcome as registered here , where everything fell neatly into phase.
Sometimes in the remotely distant future we might exchange my speculation and your knowledge and experience of the amazing Solomon Islands.
Agnostic says:
January 22, 2014 at 2:35 am
Sorry for the lack of clarity. The sunspots are on the sun. The ocean is on the earth. Obviously they think that something for which sunspots are a marker is affecting the sea level, which I shortened to sunspots are affecting the sea level.
However, their own chart makes that solar effect on sea level very unlikely, no matter what terms you use to describe it. Sometimes the sun moves first, sometimes the sea moves first, sometimes the sun and sea level move in parallel, sometimes in opposition. The correlation is bad, R^2 = 0.13, p=0.08. In other words, whatever energies sunspots might be a proxy for is not affecting the sea level.
w.
Whilst sunspots are an easy thing to count sea level is not. In fact sea level can change in different ocean basins in different ways so we need to know how this metric was arrived at.
While I agree with Greg Goodmans points to Willis I’ll just say that like Willis, he wasn’t banned from the talkshop for his science.
Goodman teaches Eschenbach manners – more popcorn please.
Siberian_Husky says:
This is why it is an utter waste of time reading blogs like these written by poorly trained pseudo-scientists. The peer review process is there so that crap like this doesnt see the light of day. What an embarassment.
Who are you to judge? You are just trolling, anonymously. Post your name, and readers will see if you have the credentials to pass judgement like that.
You would love it if you could censor articles like this. That’s what you wrote here, isn’t it? FYI, this site does the job that your pal reviewed ‘science’ shirks.
I’m a huge Willis fan, but on this issue I’m a bit – shall we say – sceptical.
The first problem is obvious: by simply eyeballing the graph you can see a striking correlation. Of course, the eye can be easily fooled. One test would be to create a set of similar but random graphs and to estimate what proportion seem to have the same correlation.
The sea level graph has 10 positive peaks. The sun spots very accurately match 5 of the peaks, particularly in term of phase (date of the peak). Several other peaks match fairly well. Strangely, the two worst phase matches are at the start and end of the graph. Coincidence?
There is a dramatic phase mismatch right at the start of the graph. There are two possible explanations for these mismatches:
1. Sunspot data should be accurate (easy to measure) but sea level data is much harder. My guess is that, if Holgate had used a different (but still good quality set), his graph would have been similar overall, but in detail somewhat different. Remember that Holgate’s data set was quite small. A different data set might well change the phase matches.
2. The thing (if it exists e.g. solar wind) that affected the sea level is probably complex. And the relationship between that thing and sunspots is also probably complex.
I think R2 tests may be fairly misleading in some cases. Any test has to do what the human eyeball is very good at: noticing correlations between specific features rather than an overall trend.
As I said, some kind of Monte Carlo test could possibly prove whether the apparent correlation is real or just a trick of random chance: create a set of similar randomly generated graphs and ask this question: what proportion of randomly generated graphs appear to have the same correlation?
Ironically, true believers would kill to be able to show a graph with similar correlation between CO2 and global temperature (obviously, with CO2 leading and not lagging the temperature).
Finally, there is an abundance of evidence that solar activity has a dramatic effect on river flows and lake levels. If solar activity had literally a zero effect on sea levels, that would be quite surprising.
Chris
tallbloke says:
“While I agree with Greg Goodmans points to Willis I’ll just say that like Willis, he wasn’t banned from the talkshop for his science”.
No. I was banned because you are unable to accept criticism and have a tantrum if anyone disagrees with you on your blog. You control your blog space like a petulant child and don’t apply your own blog rules.
Rule (1) There are no rules.
Rule (2) See rule (1)
Rule (3) See rule (2)
http://climategrog.wordpress.com/2013/03/05/talkshop-immoderation/
I would have thought you would have the sense to keep schtum about that outside of a space where you have admin rights. You end up looking about as objective as Grant “Tamino” Forster and Real Climate crew.
@johnmarshall
Whilst sunspots are an easy thing to count sea level is not. In fact sea level can change in different ocean basins in different ways so we need to know how this metric was arrived at.
Agreed – this is the reason I am skeptical of drawing a link between sunspots as a metric for solar activity and sea level rise. I don’t think it is necessary to go as far as an R^2 – I’d have questioned the sea level data. And R^2 of harmonically related trends don’t automatically discount a relationship. If you were to describe π in binary you wouldn’t get a statistically significant result but it doesn’t mean that it wasn’t meaningful.
I think the reasoning behind the graph is reasonable, but I doubt there is sufficient accuracy in the sea level data to draw much in the way of conclusion from it.
I don’t think the paper is particularly strong , it is mainly recapitulation of other work (mainly Scafetta it would seem), however, I have not seen this particular tide series before.
Some of it seems of dubious value, eg:
“The GISP2 may have a timing error of decades and/or
show temperatures out of phase with the global temperature
variation. In Fig. 9 we compare the simulation determined
from the GISP2 data with the HadCRUT4 global tempera-
ture series, and find a good fit if we introduce a shift of 85 yr,
which means the response in the ice core as shown in Fig. 8
is delayed 85 yr compared with the instrumental temperature
record. ”
One half wiggle looks similar to another one if we shift it by an arbitrary 85 years. Hmm.
Comparing frequency spectra in fig. 3a and 3b:
“The dominance of a 22 yr period compared with a 10–12 yr
period can be explained by GCR variations. The 22 yr Hale
period is the Sun’s magnetic period, and represents a polarity
change in the two hemispheres of the Sun. This is observed in
the GCR variations as shown in Fig. 5. ”
This is more like it. What is measured in SSN is a “rectified” version of solar variations.
Climate is very complex and it is not possible to just overlay two time series or do trivial linear regressions. No understanding will be gained by such trivial analysis, nor can the presence of a signal be refuted over simplified correlation tests and statistical significance.
This is the main problem with current climate science. There is a resounding need to apply the wealth of existing knowledge in systems analysis to climate system.
It seems these papers suffered from the rather incestuous circle of authors, editors and reviewers and could have been strengthened by applying the publishers rules. There is plenty of expertise in relevant techniques that can be called on, even if its application to climate is lacking.
@Vukcevic:
Nice graph. Nine years is one of the lunar cycyles. I would look into lunar tidal cycles and orbital resonance periods with the gas giant planets for the “connection”. Both for your work and for the interesting graph in the posting.
@Willis:
Your analysis is statistically sound for two matched signals, but will fail on two resonant signals of common origin but different harmonics… A 3:2 frequncy ratio will give shifting phase, yet is a relationship. It speaks to something deeper to find.
Is the article trash? Depends on the authors’ claims in the text, not just the stats on the graph. Taking your word for it that strong correlation was claimed without the math: if so, that is very weak. Ought to have shown the math basis for the claim. IMHO, and also shown the correlation stats for harmonic relations that do match, if any.
But in any case, to look at tides and NOT look at orbital periods of the moon and sun is a big weakness. There are 3, 9, and 18 year lunar orbital cycles, and more, and an 11 year Jovian jiggle. Ignoring them in a lunar solar cycle paper is silly.
Is this just coincidence?
–
>NASA Finds Sun-Climate Connection in Old Nile RecordsIs solar variability reflected in the Nile River?<
Alexander Ruzmaikin, Joan Feynman, and Yuk L. Yung
JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 111, D21114, doi:10.1029/2006JD007462, 2006
http://trs-new.jpl.nasa.gov/dspace/bitstream/2014/39770/1/06-1256.pdf
http://onlinelibrary.wiley.com/doi/10.1029/2006JD007462/abstract
http://trs-new.jpl.nasa.gov/dspace/handle/2014/40231
–
By the way, Joan Feynman is Richard's Feynman sister.
http://en.wikipedia.org/wiki/Joan_Feynman
NASA Finds Sun-Climate Connection in Old Nile Records
http://www.jpl.nasa.gov/news/news.php?feature=1319
Hi Chiefio, this look at lunar cycles in Arctic ice data may interest you.
http://climategrog.wordpress.com/?attachment_id=756
It’s also a good example of the detailed analysis needed to detect signals in spectral data.
Many things in climate lead to one effect modulating another, this leads to splitting of peaks into triplets where the central value may be nearly non existent. Failure to understand how this shows up in spectral data will lead the false conclusion that a particular frequency is not present.
I’d also been surprised by the lack of a clear 29.5 day signal in this data until recently though there was a broad spread in that part of the spectrum there was nothing relating to the usual figure of 29.53 days as synodic lunar period.
I recently discovered (thanks to a link of cheifio’s site) that there is quite a spread in length of the visible lunar cycle. It turned out that the extreme values are clearly present whereas the average is not.
None of this is simple and trivial analysis is not sufficient to either detect or refute the presence of natural cycles.
Similar processing finds a notable anomalistic month signal (distance of moon) . Equally looking for the signal directly you would conclude there is nothing at 27.5545d , however, the combined energy of the triplet makes it clearly significant against the background noise.
http://climategrog.wordpress.com/?attachment_id=757
I also have to say hats off to NOAA for the quality of this data. It really says something for the signal to noise ratio when you can pull out this kind of detail.
Well done again, sir! It never ceases to fascinate me to think how easy it is to fool oneself.
@Siberian_Husky
You should do a little research before you imply that someone is a fool. Especially when your accusation exposes your glaring lack of understanding of simple math terms. Autocorrelation means that a signal correlates to itself (auto = self, correlation = the state of being correlated) . The two datasets are indeed highly autocorrelated since they are not only cyclical, but also quite periodic. Here is the Wikipedia article regarding autocorrelation.
The Merriam-Webster definition of autocorrelation is:
In conclusion, it would appear that autocorrelation does not mean what you seem to think it means.
A recurring problem with “open review” is when the reviewers reach beyond their own grasp (and refuse to admit it). See figure 1 — when there is coupling, linear correlation is for political distortion artists:
http://www.uvm.edu/~cdanfort/csc-reading-group/sugihara-causality-science-2012.pdf
http://climategrog.wordpress.com/?attachment_id=755
The “9.1” year peak found by N Scafetta (as referenced in the Solheim paper) and in the recent BEST land temp study is also found in cross-correlation of N. Pacific and N. Atlantic SST.
Closer inspection reveals that it is likely to be 9.05 , the result of interference between lunar declination by lunar perigee (tidal force) .
The super-position of the two close cycles results in 9.05 year cycle modulated by 356 years. Only the 9.05 is readily detectable in available records due to limited data length. This kind of analysis can reveal longer periods.
Data disclosure and personal statement on peer review by Professor Jan-Erik Solheim:
In my paper “The sunspot cycle length – modulated by planets?” the section 2. Data and methods, contain links to the data:
http://www.ngdc.noaa.gov/stp/space-weather/solar-data/solar-indices/sunspot-numbers/cycle-data/table_cycle-dates_maximum-minimum.txt.
(here I made contact with the person responsible – who promised to keep this address alive)
and to the program package I use for analysis:
http://www.astro.univie.ac.at/dsn/dsn/Period04/.
This is a thrusted period search code used for variable stars – based on Discrete Fourier Analysis
For my second paper “Signals from the planets, via the Sun to the Earth” the same period search code is used
Regarding peer review:
I have done most of my research in the field of close binary stars and pulsating white dwarfs. Our communities are quite small, so we prefer to be anonymous in our referee work. I think this is the best practice.
Normally there is only one referee in Astronomy and Astrophysics publications. Even if I had my last publication in that field in 2010, I still occationally referee for Monthly Notices of the Royal Astronomical Society and The Astrophysical Journal.
The last years I have had a number of referee tasks for Climate research journals.
Regards
Jan-Erik
[snip – you are welcome to resumbit without the ad homs – mod]
To the Moderator, Which ad homs are you talking about?
I am resubmitting with some editing.
Willis criticizes without reading a paper first.
The same graph discussed above by Willis was discussed with Anthony’s great approval here:
http://wattsupwiththat.com/2009/04/15/the-oceans-as-a-calorimeter/
and here
http://wattsupwiththat.com/2009/04/07/archibald-on-sea-level-rise-and-solar-cycles/
The graph is essentially taken from Shaviv (2008):
Shaviv, N. J.: Using the oceans as a calorimeter to quantify the solar radiative forcing,
J. Geophys. Res., 113, A11101, doi:1029/2007JA012989, 2008.
which has been properly referenced by Solheim and where a lot of details are present. Solheim does not need to repeat the details given the fact that those are in the referenced work.
The uncertainty noted by some above is also discussed in Solheim’s paper
http://www.pattern-recogn-phys.net/1/177/2013/prp-1-177-2013.pdf
and it is due to the fact that the decadal climatic cycle may be due to a soli-lunar oscillation at about 9.1 year and to the 10-12 year solar cycle. This is clearly show in numerous figures of the paper such as figure 7 which is taken from one of my papers, which are these
Scafetta, N.: Solar and Planetary Oscillation control on climate change hind-cast, forecast and an comparison with the CMIP5 GCMs, Energ. Environ., 42, 455–496, 2013a.
Scafetta, N.: Discussion on climate oscillations: CMIP5 general circulation models versus a semi-empirical harmonic model based on astronomical cycles, Earth-Sci. Rev., 126, 321–357, 2013b.
Willis, unfortunately, does not seem to me to know how to read a scientific paper.
Willis, sounds like you are arguing the “a trend is not a prediction” story. The article is pointing out a trend, from which one can reasonably conclude that somewhere in the background there might be at least a co-determinant.
Kicking a dog when its down doesn’t make you a great hunter.