The time series of total solar irradiance (TSI) satellite observations since 1978 provided by ACRIM and PMOD TSI composites are studied. We find empirical evi-dence for planetary-induced forcing and modulation of solar activity. Power spectra and direct data pattern analysis reveal a clear signature of the 1.09-year Earth-Jupiter con- junction cycle, in particular during solar cycle 23 maximum. This appears to suggest that the Jupiter side of the Sun is slightly brighter during solar maxima. The effect is observed when the Earth crosses the Sun-Jupiter conjunction line every 1.09 years. Multiple spectral peaks are ob- served in the TSI records that are coherent with known planetary harmonics such as the spring, orbital and synodic periods among Mercury, Venus, Earth and Jupiter: the Mercury-Venus spring-tidal cycle (0.20 year); the Mercury orbital cycle (0.24 year); the Venus-Jupiter spring-tidal cycle (0.32 year); the Venus-Mercury synodic cycle (0.40 year); the Venus-Jupiter synodic cycle (0.65 year); and the Venus- Earth spring tidal cycle (0.80 year). Strong evidence is also found for a 0.5-year TSI cycle that could be driven by the Earth’s crossing the solar equatorial plane twice a year and may indicate a latitudinal solar-luminosity asymmetry. Be- cause both spring and synodic planetary cycles appear to be present and the amplitudes of their TSI signatures appear enhanced during sunspot cycle maxima, we conjecture that on annual and sub-annual scales both gravitational and electro-magnetic planet-sun interactions and internal non- linear feedbacks may be modulating solar activity. Gravitational tidal forces should mostly stress spring cycles while electro-magnetic forces could be linked to the solar wobbling dynamics, and would mostly stress the synodic cycles. The observed statistical coherence between the TSI records and the planetary harmonics is confirmed by three alternative tests.
From the conclusion:
Other researchers have studied the fast oscillations pre- sent in alternative solar indexes and found results compatible with ours. Rieger et al. ( 1984 ) found the quasi 25–35 days solar rotation cycles and major oscillations within 138–168 day period in energetic solar flare events, which approxi- mately corresponds to Mercury–Venus synodic cycle. Pap et al. ( 1990 ) also found evidences for similar oscillations; Caballero and Valdes-Galicia ( 2003 ) reported 58, 78, 89 and 115 day oscillations in cosmic rays time series; Kilcik et al. ( 2010 ) analyzed the solar flare index and reported frequency peaks close to 53, 85, 152, 248 days. More recently Tan and Cheng ( 2012 ) found evidences of planetary fast harmonics in sunspot and solar microwave records. The 1.09-year Earth-Jupiter conjunction cycle appears to dominate in the TSI records, in particular during solar cy- cle maxima. Indeed, Jupiter’s gravitational and magnetic in- teractions with the Sun are the strongest among the eight planets of the solar system. The likely explanation is that the side of the Sun facing Jupiter would be expected to be the locus of maximum interaction and cause this region to be slightly brighter causing satellite instruments to detect a slightly stronger TSI signal when the Earth crosses the Sun- Jupiter conjunction line. This coalesces with strong hot spots observed on other stars with closely orbiting giant planets (Shkolnik et al. 2003 , 2005 ).
Tallbloke's Talkshop 
@Nicola Scafetta
Can this be the occiput of the sun while it’s looking at jupiter (SOHO is at L1).
Graphs of the sunspot numbers show “truncated” peaks and valleys, i.e. a smoothed sinusoidal-like/curivlinear (I don”t know which term you would use) for the cycles shows that the maxima and minima readings are “missing”.
Do these truncations indicate there are negative feedback mechanisms in the sun that either kick in or become signficant at max/min times? Stability functions, essentially. In the case of planetary influences, it would be feedback that moderates the destabilization influences of (presumably) shifting centre of sun-planetary rotation relative to the Sun’s own rotational centre.
Roger
thank you
the paper is quite complex. I hope that people might find it interesting.
At the end of the pdf file that you download there are 2 large figures that may be useful to see details.
You may not need to identify all peaks, but look in particular at the TSI spectral peaks at the band 0.3-0.34 year that contain numerous strong astronomical harmonics, and the peak at 1.09 year which appears to be the Earth-Jupiter conjunction cycle.
As you can see from the three tables there are a lot of planetary frequencies in this range, from 0 to 1.2 year. These harmonics interfere among them and are modulated by nonlinear effect, producing a very complex signal. Thus, it is difficult to do the calculations.
Surely, some people will tell that I am seeing only noise, but if you carefully look at the figures, a coherence can be observed also using the colored bands I added in the figures.
I have looked into this some 3 or 4 years ago, and mentioned few times on this blog.
SC17 offers the best evidence with clear 399 days sub-cycle embedded within the susnspot data. Fact that in the ‘1940 epoch’ the Earth magnetic field became far more unstable (judged by the secular variability) than either in the previous or subsequent decades, in my view is a strong indication that an electromagnetic planetary feedback is in play. Some more details here:
http://www.vukcevic.talktalk.net/Sub-cycle.htm
What am I doing wrong?
tchannon say: July 30, 2013 at 2:35 am
PMOD alters the data significantly in many places and adopts heavy algorithms to correct its own Virgo data.
You need to use PMOD as I say in the paper, that is during solar cycle 23 maximum because before 1996 the data that they use are altered ACRIM and Nimbus records.
Moreover, what kind of algorithm are you using for the evaluation of the periodogram?
In any case, your graph shows a clear major peak at about 1.08 year, which is not truly different from 1.09 year.
Moreover, if you see carefully my tables and figures in particular the black bars between panels 2A and 2B, you will see that here are numerous cycles there. Earth-Jupiter synodic cycle is just the major one, but in addition there are also the synodic period between Earth and Saturn, Uranus and Neptune that create some variability.
Anthony published another piece of junk written by Willis.
It is evident to me that Anthony and Willis are behaving quite dishonestly by trying to defame my research.
I do not know why Anthony is behaving in this way, but I found his behavior highly unprofessional and misleading. Evidently my research is “disturbing” somebody who is benefitting of the current chaos.
My research is moving toward the solution of the problem, and I believe that a lot of people from both side of the debate prefer the actual scientific confusion.
About the new post by Willis, I do not have time to rebut it in details. Just two points that demonstrate how Willis is incompetent on these topics.
The truth is that Willis is trying to criticize my papers, and many times I have denounced that he is not even reading my papers. He is just jumping around without trying to understand thing. But Anthony does not get it.
1)
Let us start from the easy thing.
Look at Willis Figure 9. He claims that it represents the ACRIM composite. However, it is the PMOD composite.
It is easy to figure it out. See here the two composites
http://acrim.com/TSI%20Monitoring.htm
Not only the shape is different, but ACRIM averages around 1361 W/m^2, while PMOD averages around the outdated value of 1366 W/m^2 as in Willis graph.
Thus, it is evident that or Willis does not know much about these data or he is misleading people.
Curiously, Leif did not note the error too.
2)
Now, note his Figure 5. He fully confirmed my result that the 11-year sunspot record is made of three peaks close to 10, 11 and 11.86 year.
He comments, “Now, before anyone points out that 11 years 10 months is the orbital period of Jupiter, yes, it is. But then ten years, and eleven years, the other two peaks, are not the orbital period of anything I know of … so that may or may not be a coincidence. In any case, it doesn’t matter whether the 11 years 10 months is Jupiter or not, any more than it matters if 10 years is the orbital period of something else. Those are just the frequencies involved to the nearest month.”
So, he acknowledges that the 11 years 10 months is the orbital period of Jupiter. However, later he claims that he does not know about the planetary origin of the quasi 10-year cycle. However, as clearly stated in my papers many times, that is the spring tidal period between Jupiter and Saturn.
Also the major cycle at 11 year, can be found as combination of the tides from Venus, Earth and Jupiter.
So, Willis has not read my papers, but he criticized them!
His additional critique that by using short periods “the strength of these cycles waxes and wanes over time”
is not a mystery. The cycles waxes and wanes simply because they are interfering. Moreover, using short periods the error associated to the frequency evaluation increases and it becomes more difficult to separate the frequencies.
In general, to separate close sequences at 10, 11 and 12 years, I need to see at least two beats, which implies that I need to use more than 200 years of data. Because the sunspot record is 262 years long, if I divide it in half, as Willis does, I get 131 year long sequences which are too short to separate the cycles with periodogram like techniques.
The right way to proceed is how I did in my paper:
Scafetta N., 2012. Multi-scale harmonic model for solar and climate cyclical variation throughout the Holocene based on Jupiter-Saturn tidal frequencies plus the 11-year solar dynamo cycle. Journal of Atmospheric and Solar-Terrestrial Physics 80, 296-311.
Click to access ATP3581.pdf
where I build a model based of Jupiter and Saturn two tides and I see how the model performs in hindcasting past solar pattern before 1750. The hindcast tests are fully missed my Willis, but it is these tests that support the interpretation of the cycles.
***
So, tell me, how can Anthony and Willis be trusted further? They are really doing some dirty game to defame my research or they are simply incompetent.
Very likely by listening Willis and Leif, Anthony is making suicide. Somebody needs to inform Anthony before it is too late for him.
To the readers of this blog I say that only by reading my papers one may understand what I write.
**********************************************
Nicola, Stay for the course mate, it is only those people such as yourself doing real science, that advance knowledge. Willis is a very intelligent cowboy and questions everything, he can be a bit abrupt but his heart is in the right place. It is my understanding that our universe from the creation of matter all the way to evolved life forms is a chaotic process that forms harmonics, nodes or if you like fatal attractors. These harmonics rule and are the universal creative force, they also rule the mundane such as our climate and the sun, if you can figure out the basic harmonics and what causes them, the theory can be applied to other branches of science that could cause many sleepless nights for some and a joy for others.
Sir, take no notice of naysayers and continue your work with a passion, the music of the spheres shall yet prevail.
[…] Wayne Job on Scafetta and Willson: Empirica… […]
Nicola says:
“he claims that he does not know about the planetary origin of the quasi 10-year cycle. However, as clearly stated in my papers many times, that is the spring tidal period between Jupiter and Saturn.”
Yeah, I commented on that at 6am this morning UK time
tallbloke says:
July 29, 2013 at 10:56 pm
The 10 year cycle is actually 9,93yrs, half the synodic (i.e. the tidally effective period) of the two largest bodies in the solar system outside the Sun itself. Fig7 doesn’t find these periods in the two halves of the sunspot data because the cycle lengths tends to cluster at around 10.38 (VEJ) or 11,86 (J) and the longer term cycle of this bi-polar behaviour won”t be captured by half the sunspot data length. The 11.07 average is rarely the actual cycle length.
“Look at Willis Figure 9. He claims that it represents the ACRIM composite. However, it is the PMOD composite.”
I’ve commented on that too. Willis munged his link in the references at the bottom of the article, but it is indeed supposed to be the PMOD composite available from the ACRIM website. I think I’d put it down to cockup rather than conspiracy. Leif should have known better though.
” I believe that a lot of people from both side of the debate prefer the actual scientific confusion.”
Well, everyone is pushing their own pet theory and doesn’t want anyone else’s to become dominant. Plus Leif in particular is going to lose a lot of face when it eventually becomes irrefutably obvious to everyone that he has been wrong all the way through about planetary-solar interaction. Willis will just shrug and move onto something else, but Leif will be crushed, and Anthony will be left to pick up the pieces.
“the PMOD composite available from the ACRIM website”
the PMOD composite is not available from the ACRIM web-site. There is only a figure.
Nicola:
Isn’t this it?
http://acrim.com/TSI/composite_d41_62_1204.txt
Yes, I was in error.
I always get the PMOD from the PMOD web-site.
I did not know about that link
I write my own software Nicola, in this case is a windowed z-transform chirp written in C designed for climatic work, why it operates in periods. The example is about as far as things can be pushed given this data on frequency discrimination. Label values are auto computed.
Using other software which is not limited by discrete data, I still can’t get 1.09y.
However, I think there is a tendency to look like 1.09 caused by interference from other components and that might explain what you found. The discrimination of most of these algorithms is basically very poor.
I have gone a little further through to modulated. Incomplete so I’m not showing anything.
In my opinion all the TSI data is in a mess. I used PMOD because it demonstrates regular sampling although this might be fake. Acrim provide irregular data but is broadly the same.
SORCE I already know does not contain 1.09 but as will all the other satellite data it is heavy mangled by modelling platform and orbit movements. The instruments are also not good enough anyway.
The problem is that you posit a connection. If you had merely reported 1.09 I would not have commented but connecting this as factual with a chronometric external factor demands exact coherence. I don’t see that so I speak up.
Okay so what can be done going on the basis there is a brightening to find?
This need synchronous detection a fierce problem I think given very little information is available, the devil to avoid Nyquist. Spinning bodies, orbits, moving around platforms, instrument limitations and variation, etc.
Not sure I want to go there.
PMOD has some problem of stability, and you need to clarify the period you are analyzing.
Moreover, does your algorithm use even intervals or uneven intervals? note that PMOD as well as ACRIM composites are incomplete, so there are missing days all over the records.
About the synchronicity, you find a cycle at 1.08 year, which is not really different from 1.09. This is just one month and the 1.09 cycle tends to disappear during solar minima. So the issue is complex.
The error here is about 0.5 year. It is possible to find some variation and as I sad there are other cycles very close that interfere.
About SORCE, it also shows a signature, but this is more complex I will talk about it in another paper
The way to properly test the coherence issue is not to find a rigorous 1.092 cycle in power spectra evaluations that have error bars, but to use my figures 3-5.
Look for example at 4C, but also the others.
On the WUWT thread:
tallbloke says:
Your comment is awaiting moderation.
July 30, 2013 at 11:42 am
Anthony Watts says:
July 30, 2013 at 11:27 am
What Matthew R. Marler said. Exploring an alternate method isn’t defamation.
If someone searches for a needle in a haystack with a pitchfork and fails to find it, and then declares the needle doesn’t exist, and calls the people who found it using a magnet cyclomaniacs, it might not be defamatory, but it does make them look stupid.
We disagree.
Who disagrees with who about what?
Can I have a side order of why? to go with that please.
Agreeing to differ with NS. Can’t talk and do.
Ah, fair enough. 🙂
Wayne Job (July 30, 2013 at 9:50 am) erred fatally: “[…] his heart is in the right place”
If that’s what you actually think then we know for sure that YOU are part of the problem. I don’t say this lightly and I don’t want to discuss this further.
Sincerely
I did a small mistake above
Tim cycle is at 1.077 year, what I found is closer to 1.092 (+/- 0.5) year
The value varies a little bit also because other close cycles are present and there is an interference.
note that (1.092-1.077)*365.25=5.5 days
So Tim value is just 5 day shorter than mine, not one month as I erroneously wrote above..
Tim has not clarified whether his algorithm uses even or uneven time series. What it is known is that PMOD present missing days.
Tallbloke says,
“If someone searches for a needle in a haystack with a pitchfork and fails to find it, and then declares the needle doesn’t exist, and calls the people who found it using a magnet cyclomaniacs, it might not be defamatory, but it does make them look stupid.”
This is an important issue. Time series analysis of complex systems is not easy. To find specific results one may need to use specific tools of analysis with specific properties.
It is a typical error when people try to “disprove” a specific result by using a poor methodology of analysis unable to find it.
For example, if I say that a specific time series contains two close frequencies, it is an error to claim the result wrong simply because by studying subsets of the same sequence one does not find the two peaks but only one peak that waxes and wanes over time. It is evident that the ability of separating the two frequencies depends on the length of the time sequence and if the time sequence is too short the technique will not see the multiple components but detects the beat waxing and waning pattern.
A specific result can be disproved or by using the same original technique of analysis and show an error in the original study or by using another technique but by first “demonstrating” that the latter is better than the one used in the original study.
However, simply saying that a result must be “wrong” because not found when a simply different methodology is used, is a logical fallacy in general. Those who make this error simply demonstrate not to understand much of time series analysis.
Nicola, I agree. For the latest inappropriate methods from Willis, see here:
And Bart’s replies here:
I didn’t want to take on Willis at the same time as Leif so my reply is on twitter 🙂
@RogTallbloke 2h
Heaven preserve nuanced and delicate data from Willis Eschenbach’s 12 guage Blunderbuss. http://wattsupwiththat.com/2013/07/29/cycles-without-the-mania/#comment-1375846 …
The scattergun approach #meh
Nicola, I estimated 4 days by mental arithmetic, but was too uncertain of my ground to speak up. 🙂
Can you explain how the eccetricity of Jupiter’s orbit affects the calculation. there must be a 11.86 yr cycle in the periodicity of the Jupiter-Earth synodic period.
The 11.86 yr cycle is seen in the long sunspot records (260 year since 1749).
In the TSI it cannot be detected because the record it too short. Around 2000 Jupiter was relatively close so perhaps this is one reason why we see strong 1.09 peaks during solar cycle 23 maximum.
I am developing new ideas to optimize the model, but we may talk about it another time.
Yes, we might expect the Earth-Jupiter synodic period to be fractionally longer when Jupiter is nearer perihelion, because Jupiter is moving faster and and it takes Earth a little longer to ‘catch up’ to form the conjunction.
Anthony would be wise to throw Willis under the bus. Even in his articles on the British Columbia carbon tax — a relatively extremely simple topic — he made a severely ignorant &/or deceptive and patently false assumption about the political context here in BC. If he can’t even be trusted to get the most basic thing about something that simple right, any sensible person with even moderate contextual awareness is going to just stop and become suspicious about why a spotlight’s always pointed at this guy and why he’s artificially propped up on a pedestal. If I wrote the full extent of what I really think about this, it wouldn’t pass moderation anywhere.
[Reply] Thanks for stopping short of the limits on editorial comment about other websites.
Kevin Trenberth has done a piece called “Global warming is here to stay …..” and over at WUWT Bob Tisdale wrote about it. Tisdale shows graphs of Trenberth’s account of SST in the Indian and Western Pacific. He referred to the graphs as Trenberth’s “Big Jumps”. What was not discussed was what I thought was incredibly interesting. All of Trenberth’s big jumps coincided with Jupiter perihelion positions. I was wondering if Dr Scafetta noticed this.
{reply] Looked through Bobs articles at WUWT but didn’t find that one. Got a link?
Guy’s, it would be well to bear in mind the differences of approach that exist on both sides of the Herring Pond. Those North Americans who do not find themselves tenured in a big money earning pillar of academia IPenState) do have a somewhat more Vibrant attitude towards emerging ideas, vibrant to the extent of being a Gang Fight :O It may be a good thing inasmuch as it could promote a more rigorous examination of new theories ? and does have the virtue of the knives being on open display 🙂
Nicola’s investigations look interesting, and I imagine that the Heat being generated will ensure robustness at each step ?
PMOD composite data is presented with a strict daily sample interval. (verified)
Software implements a chirp z-transform on DFFT and the data is windowed, in this case kaiser-bessel b=3 gives good frequency discrimination and is recognised in some specialist circles as about the best compromise. It will leak more noise than many other windows but this is irrelevant since dynamic range is not the problem with this data if only from quantisation.
Test plots are in this PDF, chirp response when applied to a pure 0.5 year test signal. Was not intended for publication, internal use document. (y-axis is power dB, 10x == 20dB)
Interpolation is done if feasible.
The next ploy is try and produce a good decomposition using other software, however, solar is extremely awkward. I’ve never seen satisfactory results but it is nevertheless useful.
A good move is signal process decimate to monthly, reduces the size of the problem and magically interpolates, so now there are no missing data points.
[edit: Looking at that after posting I had bettr point out there are *two* traces, fine blue line down the middle, overlay plot, one from day data, other from month data.
Following is from other software. –Tim]
This looks sensible but is fiction, offers clues for human interpretation, why it is useful.
1.0795497894 0.9953212232
This pair of periods mix to produce a modulated wave.
These could be taken as 1.08 and 1.00, mixing to roughly 12.9 and 0.5
Roughly 0.5 is in there muddled. What might be 1 year ought to be a clue there is something going on given the earth orbit and moving satellites.
At this point I raise the nature of the sun.
This behaves as a blue noise source with an impulse nature, hence this puts high stress on instrumentation: extreme crest factor.
Not quite so simple, UV catastrophe, it is limited but we do not understand this.
A consequence is a likely highly non-linear solar characteristic and need to consider signals as impulses of unknown characteristic but tending to defining by instrumentation characteristics.
The “signals” are unipolar on any stimulation, can only go one way. Trying to decode this using bidirectional is never going to be correct but might approximate.
Salvaging something out of the mess is I think all we can do unless some bright sparks have ideas.
A ploy sometimes used in data analysis is predistorting data to linear because all good methods are linear (our maths system is linear). With solar? Certainly I have used this, interesting without major conclusions.
An example might be using integration and differentiation where an analysis in in the alternate domain. Done this and it cloned, ie. worked perfectly.
Time to shut up, as if the above is makes sense anyway. 🙂
I’ve now carried out a fairly full quick investigation.
Basic answer to whether there is 1.09, no.
What has come out of this is probably quite interesting but putting all I have done in a comment is a step too far.
A lot of trickery and shenanigans. I extracted whatever is there in the frequency band of interest and gone as far as extracting a usable the analytic signal. (phase unwrap fails at one point)
Ploy to get a result with the tools I have.
Use the monthly data, much easier for this work.
Subtract out the longer period components and primary offset using the 13 term approximation. this does not touch the region of interest but makes the signal less stressful for subsequent operations. (there are sane limits on what software can tolerate from noise, we want to look at 1/1000th of the whole signal)
I then produced a bandpass filter version around the frequency of interest.
I forget to mention above that I was correct and there are 3 not 2 terms.
Roughly 1.079 0.995 0.952 where discriminating there is beyond DFT. I have ways (thousands of lines of code and perhaps as many hours).
Those will be B…..x but a vague idea of what makes the shape. Filtering on the other hand has no rigid limits so things can move around.
The analytic signal
[mod: should have added, the ends of the data will be invalid, very hard to get the ends right with a Hilbert Transform, beware –Tim]
I show this because many will not be familiar and it is a bit mind bending. A new signal is derived from the original where all frequencies are phase shifted by 90 degrees, called a quadrature signal . In this case a very limited range of frequencies are present.
If you have the signals in quadrature various magical things become possible, is used a lot in communications signal processing.
A simple procedure is compute the envelope where this is the only method where this can be done perfectly. (unfortunately creating perfect quadrature in the general case is impossible). I won’t go into detail, it is a very simple function of adding the sine() and cosine() of the two.
For completeness
click
An XY plot, are in perfect phase and is a circle.
Now for the fun stuff
click
The phase unwrap algorithm failed but I’ve fixed it up by hand, the glitch you can see. (added PI after that point)
If I have done everything correctly and what I have done is valid (ho ho, fat chance) these two reveal things are not simple nor static.
These show instantaneous phase and frequency (actually period), which means the value at a particular instant. (very hard to do well from noisy data)
According to this the frequency is varying quite widely. Maybe this will be illuminating so if anybody recognises anything please speak up.
I am sure much more could be done but I am no guru and it needs a true expert. Perhaps showing there is a possible route is what I do.
Tim
nice analysis
yes the things are varying and beating because there are numerous close frequencies. The 1.092 year cycle is just one of them, by adding the others one may get a better optimized model by providing some variability that may be up to 40-50 days in 30 years,
I also wrote in the paper that he model is not fully optimized. A task left to another study.
Note also that in physics people need to give an physical interpretation of the results. In general 1.08 and 1.09 are not different cycle considering that there are several reasons why a 1.09 cycle may appear as a 1.08 cycle in a periodogram evaluation. And by changing the analyzed data the peak may change.
Questions:
Could the fact that TSI is measured by a succession of instruments near Earth have anything to do with the fact that the average of Tim’s phase plot seems pretty close to 1? i.e. is there an annual signal here that might be diluting the Earth-Jupiter synodic signal? If there is, what might its cause be? Solar axial inclination to the plane of the ecliptic? Or is that effect the tiny annual double wiggle we can see in Tim’s plot?
How is TSI data corrected for the Earth’s orbital eccentricity, which produces an ~80W/m^2 annual swing?
tallbloke asked: … is there an annual signal …
Rog, what do you expect? If, as Tim says, the data is chopped off once a calendar month, then I wouldn’t be surprised that the (maths in the) analysis tool believes the signal must be dominated by the calendar year.
TB: ‘is there an annual signal here that might be diluting the Earth-Jupiter synodic signal?’
Don’t know the effect, but the Earth-Saturn synodic period of 378.09 days (NASA data) = 1.035154 years, might be a candidate?
The J-S : J-E : S-E synodic ratios to each other are 11 : 200 : 211.
Chaeremon, not writing a book here. There is a huge subject in signals both as-is and the digital number representation of the analogue as-is. It is not statistics nor the maths usually taught, why it is a specialism and mostly dealt with by engineers.
In this case the solar heat as received inside a box (cavity) on a satellite is turned into a stream of numbers. Done correctly this is an exact description of the variation in solar heat, within certain precise limitations.
There no problem from the calendar year until someone makes a fatal mistake.
There are reasons why I am wary of these signals but notice I said “decimate” and mentioned signal processing. Nothing I have done. That’s why I showed the spectra with an overlay of original and decimated. They are actually different after quite a few decimal places.
Very likely this is showing artefacts of the observation platform movement. The whole problem is extremely complicated.
Below is a chosen very clear and simple example of things going bad.
From this thread https://tallbloke.wordpress.com/2012/02/24/solar-cycle-progress-to-maximum/
The WSO (Wilcox Solar Obs.) takes measurements at the solar poles from earth, a difficult measurement. The earth is in an orbit and with orientation which makes the sun seem to bob up and down through the year.
The WSO telescope uses an instrument with a fixed aperture but the view of the solar pole, north and south varies as the bobbing goes on.
In consequence the “data” is mis-sampled and the annual motion modulates the wanted signal. The plot above is close to a classic example of a doublet and with the fundamental missing, 1 year in the middle. (it should be missing, is correct)
I’ve stated many times TSI data is not very good, I am serious.
The similarity between these two solar observation data needs to be considered. I was going to raise the matter of satellite movement etc. anyway.
About the possibility of a residual annual orbital cycle in the TSI measurements there is a need to notice that PMOD, which is used by Tim does present a small annual cycle that might be instrumental. The paper talk about such possibility.
ACRIM2 and ACRIM3 should be more accurate.
In any case, during the solar cycle 23 maximum these oscillations are macroscopic, they are 1.09 year (not 1 year), are observed by multiple instrumentations (ACRIM2, ACRIM3, VIRGO, SORCE) and are in phase with the Earth-Jupiter synodic cycle. See figure 4
So, both ACRIM and PMOD and SORCE agree that these oscillations are not annual residual artifacts but real total solar irradiance components.
However, as my tables show, numerous very close astronomical oscillations exist between 0.95 and 1.1 year.
Note that if these oscillations were instrumental orbital artifacts, as Tim is arguing, they would not disappear during solar minima and appear during solar maxima.
In addition,
there is a need to interpret also the other TSI spectral peaks.
So, a physical interpretation must look at the problem from multiple points of view.
Note that my previous comment about the TSI 1.09 oscillations
“Note that if these oscillations were instrumental orbital artifacts, as Tim is arguing, they would not disappear during solar minima and appear during solar maxima.”
is very important for the interpretation of the data.
On the contrary the WSO record analyzed by TIM presents a quasi regular annual oscillation in the amplitude that suggests its orbital origin. See here (blue record)
Vanishing at zero is exactly what happens with WSO data because it is a Product (amplutide modualtion). Whether this could occur with TSI depends on the details, would occur early on in the chain. Any offset in sampling, improper maths, will make a mess. This is very widespread out there in many fields.
I’ve pointed out a strong resemblance.
I’m sorry Roger for that. I’m always in a hurry. The Trenberth Big Jumps article can be found on Bob Tisdale’s blogsite. It’s really worth looking at. I apologise.
Roger,
The Trenberth article is a June 4 entry at Bobtisdale.wordpress.
Trenberth Article
http://www.rmets.org/weather-and-climate/climate/has-global-warming-stalled
Tisdale response
Thank you RJ Salvador for posting the link to the Tisdale response.
It was figure 8 that caught my eye with its SST spikes at Jupiter perihelion.
During the last four solar cycles, solar minimum has been somewhat in sync with Jupiter perihelion. This can confuse and hide the fact that these temperature spikes(figure 8) are better in line with Jupiter perihelion than with solar minimum.
Bob Tisdale has argued that these temperature spikes are to be expected due to the shifting of warmer waters. This is true.
But what is not explained is why these spikes happened at perihelion.
Seasurface temperatures spiking as the Earth passes between the Sun and a Jupiter perihelion,
this fits well with Scafetta’s new research.
Nicola,
As I have already mentioned ACRIM data is difficult. In this case ACRIM 3 is irregularly sampled, unmentioned missing data.
A quick look with software which handles irregular shows much the same..
The other software is in progress on enhancing handling of irregular where I’d left it in a broken state. (wasn’t happy with it and probably want to switch technique, not an easy problem doubly resampling)
I’ll think about what to do.
Maybe there are two issues going on here. The Jupiter perihelion alignments with SST jumps might be more related to Jupiter affects on our planet from changes in albedo or the like, which is quite different to solar output (esp if spikes are happening at solar minimum).
The TSI spikes at Earth/Jupiter conjunctions do suggest a barycentre function occurring. This could easily be tested if we had a 360 degree sunspot view of the sun, the Stereo Behind image is probably not satisfactory.
This might be of some interest:
Click to access IAU_Vernova.pdf
Longitudinal asymmetry in sunspot activity during the ascending and descending phase
of the solar cycle
E. S. Vernova, K. Mursula, M. I. Tyasto and D. G. Baranov
Multi-Wavelength Investigations of Solar Activity
Proceedings IAU Symposium No. 223, 2004
Hi Geoff, seems to me that if there’s an effect on albedo when Earth is between Sun and Jupiter, it’s as likely something to do with the global electrical circuit. a while back on the Solar experts thread I plotted the motion of the Earth as it passes Jupiter. There is a ‘spike’ of around 8000km in the orbital motion. Given the 186 million mile diameter of earth’s orbit, it’s not much. If cloud is reduced over the tropics, the ocean near surface mixed layer warms pretty quickly.
Hi Rog, Nicola might be onto something, especially the 60 year Jupiter/Saturn cycle. Perhaps one of the true repeating cycles of our solar system that impact our climate.
Geoff, Agreed, whichever or how many of the possible causes are involved. I like to say all are, because keeping them all in mind makes it less likely to miss something. 🙂
[…] Scafetta and Willson: Empirical evidences for a planetary modulation of total solar irradiance and t… […]
Since Nicola likes ACRIM I’ve started to look
Here is an initial look at ACRIM 3
Central here means periodicity restricted by removing both fast and slow parts of the spectrum.
On looking at the data, daily, I find it has missing days (~356 days in a year)
Tim. Nicola is a member of the ACRIM team.
Thanks for the ACRIM 3 TSI chart, Tim. Didn’t realise that there was such a large variance in TSI between strong cycles and weak. No wonder the Maunder Minimum was so cold.
Geoff Sharp says
Maybe there are two issues going on here.
I think you may be right about that Geoff.
SST spiking at Jupiter perihelion suggests a Jupiter/Sun tidal influence on the oceans cycles.
While the Jupiter/Earth synodic period (1.092 years) conforming to TSI suggests a planetary tidal influence on the Sun.
I find this whole subject to be very interesting.
[…] Coda 1 to Scafetta a… on Scafetta and Willson: Empirica… […]
WRT missing days and irregular sampling of ACRIM3 results:
ACRIM3 results are published as daily means. The mean time of acquisition is the time tag for the day. Missing days do occur for various reasons. No modeling or interpolation is applied. The ACRIM3 Composite TSI time series is constructed the same way.
In contrast, The PMOD Composite TSI time series replaces components of the satellite TSI database with modeled results that are significantly different than those originally published by the various experimental science teams. This is done without an instrumental rationale for the corrections applied or reprocessing of the actual data.
Richard: Thanks for your comment, and welcome to the Talkshop. While you’re around, please could you tell us what you’re take is on the current state of TSI measurement and it’s integration with the longer term TSI series maintained by your team. There seems to have been something of a ‘news blackout’ on the TIM instrument on the SORCE platform.
Richard W.,
That’s useful information thank you, adds confidence. I’d guessed something like that was going on, only an issue if the situation is unclear.
Resampling from exact tends to be somewhat hard to do well enough, at least if there is question over the data Nyquist compliance. The alternative of working with irregular is never entirely satisfactory.
In this case with what I have been doing here I doubt it matters.