Leif Svalgaard: Solar-Terrestrial Manifesto – Flatter Sun – Flatter Earth

Posted: November 12, 2012 by tallbloke in Measurement, methodology, propaganda, Solar physics

Over on WUWT, Leif Svalgaard has published his solar-terrestrial manifesto. I’ve left a couple of comments…

Here’s what Leif plans for the future of solar science – A newly homogenised and flat solar history:

“new reconstructions of Total Solar Irradiance also contribute to our improved knowledge (or at least best guess) of the environment of the Earth System, with obvious implications for management of space-based technological assets or, perhaps, even climate.”

Well since Leif is endeavouring to fit the sunspot and TSI data to the carbon agenda by ironing it as flat as possible, I guess we all know what the subtext of this passage means.

“A strong check and confirmation of the effect of the weighting have been carried out the past year by the Locarno observers, counting both with and without weighting, so the magnitude of the effect of the weighting is now established and is no longer an open issue.”

Attempting a calibration and imposing it on long term historical sunspot records at a time when the sunspot count is anomalously diverging from other solar metrics such as the f10.7 flux is not a valid procedure IMO.

The point was that the Group sunspot number is flawed and must be abandoned. The hard work by Hoyt and Schatten has not beem in vain because the data they dug up will be incorporated in the revised sunspot number, just calibrated correctly.

Since we don’t know much about how the Sun operates over long timescales, it would seem prudent to continue to monitor all of the longer datasets and update them. If we are seeing less groups but more spots, that in itself may be important, we don’t know. To “abandon” metrics maintained over centuries because they don’t fit the agenda is a dereliction of scientific duty.

I agree we should endeavour to ‘calibrate correctly’. ‘Adjusting’ dispararate metrics so they ‘all sing off the same hymn sheet’ may not be the way to do that however. The Sun is still mysterious and since Leif has just abandoned 50 years of deep dynamo theory (Which he previously assured us was “well grounded in solid physics”) in favour of a ‘shallow dynamo’ hypothesis, we should treasure our historical records and not allow them to be manipulated by someone who has strong opinions on the ‘correct’ theory.

James Hansen and NASA GIStemp springs to mind.

Metrics should be maintained and calibrated by impartial bodies whose principle remit is the custodianship of data, not its application. Allowing the definition and calibration of the metrics to be in the same hands as those writing new theory is a recipe for bias. We should not repeat the mistakes of the past so quickly.

Comments
  1. oldbrew says:

    Stating the obvious, but one thing nobody can do is alter the actual behaviour of the sun. If we do see a cooling period sometime in the next 10-20 years certain theories are going to look increasingly out of step with reality.

  2. tallbloke says:

    True. I’d rather see due caution being applied to fiddling with historical datasets in the meantime though. As I said to Leif some time ago, I have no problem with him putting up his IDV metric alongside other solar metrics for the use of researchers. Using it to homogenise all the other metrics is a step too far, given the uncertainties involved.

  3. Paul Vaughan says:

    TB wrote: “To “abandon” metrics maintained over centuries because they don’t fit the agenda is a dereliction of scientific duty.”

    Ethically heavy stuff.

    One of the last exchanges I ever had with Svalgaard — the one where I decided to write him off permanently as quantitatively untrustworthy — involved 2 graphs of the exact same data presented in 2 different formats. He agreed with the first and objected strongly to the second. Too creepy.

    TB or anyone else: Can you point to a plain-text webpage listing the “Ironed Flat” TSI? There’s a test I would like to perform on the ironed record…

  4. tallbloke says:

    Paul, here you go.

    http://www.leif.org/research/TSI-SSN-Guess.xls

    Just say if you need the data pasting into notepad. It’s only three columns.

    Flatter than a flat thing with a puncture it is. :)

  5. Craig M says:

    This quote of yours appears to have been overlooked by the cries of `just an IT consultant may not question the higher order';

    “Metrics should be maintained and calibrated by impartial bodies whose principle remit is the custodianship of data, not its application. Allowing the definition and calibration of the metrics to be in the same hands as those writing new theory is a recipe for bias. We should not repeat the mistakes of the past so quickly.”

    That should be an oath repeated daily by custodians. Sadly they prefer this;

    “Who controls the past controls the future. Who controls the present controls the past.”

  6. tallbloke says:

    Craig, thanks for taking the trouble to come over here and say that. It should be a no brainer for sceptics to be demanding this kind of rigorous, standards based system for data collation and verification. Ravetz and I have been pushing for it since before Lisbon.

    No-one is going to take climate science seriously until it starts behaving like a grown up discipline.

  7. tallbloke says:

    If this plot from my work on Ray Tomes’ Z-axis solar-planetary theory is correct, the SIDC sunspot dataset doesn’t need tweaking much to ‘calibrate’ it anyway.

    Though this older plot I did in 2009 using 24 yr averaged annual data might indicate the Waldemaier onwards data does need 20% lopping off as Leif says. That’s not so much of an issue.

  8. GAI says:

    Tallbloke, In your comments on WUWT do not forget Judithgate. http://www.klimaskeptik.cz/news/judithgate-ipcc-consensus-was-only-one-physicist/

    There are letters from “the best world astro-physicists” Dr Doug Hoyt and Dr. Richard C. Willson to Dr. Nicola Scafetta for his inclusion in “Climate Change and Its Causes – A Discussion About Some Key Issues” that was presented at EPA, Feb 2009.

    To put it bluntly I do not trust Leif Svalgaard, Steve Mosher, or Ferdinand Engelbeen who profess to not agree with CAGW but each in their own way support different parts of the message.

  9. GAI says:

    Paul Vaughan says:
    November 12, 2012 at 1:23 pm

    …. TB or anyone else: Can you point to a plain-text webpage listing the “Ironed Flat” TSI? There’s a test I would like to perform on the ironed record…
    ____________________________________________
    Tallboke posted a link at WUWT:
    Leif, where can I download the time series data for the reconstructed sunspot numbers after all the corrections,

    Here you go Ray.

    http://www.leif.org/research/TSI-SSN-Guess.xls

    Or their is Lief’s answer

    For all the whiners:
    You can yourself make the comparison Wolf-Wolfer. Here is how:
    Take a random year, e.g. 1886 and proceed as shown in [I have replaced the -99 for missing data with a blank cell]
    http://www.leif.org/research/Wolf-Wolfer.xls [or if you don’t have Excel:]

    http://www.leif.org/research/Wolf-Wolfer.pdf

    Source: http://wattsupwiththat.com/2012/11/11/solar-activity-past-present-future/#comment-1144737

    Also look at the Judithgate scandal

    The link I posted above is very good. http://www.klimaskeptik.cz/news/judithgate-ipcc-consensus-was-only-one-physicist/

    Motl, a Physicist, comments here: http://motls.blogspot.com/2010/06/judithgate-ipcc-relied-on-one-solar.html

    The Hocky Schtick: http://hockeyschtick.blogspot.com/2010/06/ipcc-consensus-on-solar-influence-was.html

    It is curious that I can find absolutely NOTHING about the scandal using [ “Judith lean” IPCC 2010] [Judithgate] or anything else I can think of.

    Nasif Nahle’s comment at the the Hocky Schtick is quite interesting.

    Through WUWT and Climate Audit, I criticized Lean’s work at its moment because she had changed her original database, based on proxies and sunspots together, to another database based only on sunspots. Lean’s first database, published by NOAA in 2001, showed a clear increase in the Sun’s radiation intensity in the last 400 years. However, her new graphs show a plain baseline which hides the increase of the amplitude of the solar activity. Her first database shows an amplitude of the solar activity of 3.3 W/m^2 with a minimum of 1363.4 W/m^2 in 1659, and a maximum of 1366.728 W/m^2 in 1981; on the other hand, her new database displays only 1.2 W/m^2 with a minimum in 1902 and a maximum in 1981….

    http://hockeyschtick.blogspot.com/2010/06/ipcc-consensus-on-solar-influence-was.html

  10. ed says:

    Interesting, I had downloaded Leifs TSI reconstruction spaghetti plot (TSI Reconstructions) during the weekend, and noticed a curious datapoint at 2007.5, it shows a strangely low TSI value at 2007.5, as if something is amiss (cal wrong?), otherwise all of the high amplitudes seems to match his TSI-SSN-GUESS (other than an offset). Wonder why/how the anomolous datapoint at 2007.5 was adjusted. Maybe that’s why his plot is now labeled “guess”?

  11. ed says:

    You have to delete all of the other reconstructions to see it…

  12. oldbrew says:

    GAI says: ‘It is curious that I can find absolutely NOTHING about the scandal using [ “Judith lean” IPCC 2010] [Judithgate] or anything else I can think of’

    Try Bing (bing.com) :-)

  13. tallbloke says:

    Thanks GAI and welcome.

    A little bird tells me Hoyt’s letter earned him an abusive email from Leif Svalgaard:

  14. tallbloke says:

    And the letter from Willson went into greater detail:

  15. tallbloke says:

    ed: Welcome. PMOD shows no anomaly at 2007.5, so I’m not sure what you are seeing.

    http://woodfortrees.org/plot/pmod/from:2005

    Can you upload an image to a free hosting site and link it here?

  16. tallbloke says:

    Bruckner8 says:
    November 12, 2012 at 10:07 am
    My take on the entire article: “Here’s why fudge factors x, y, z all make sense and thus have a consensus. (wringing hands) Now…can we take all these FFs, stuff ‘em into the black box, adore their purpose, and agree on the beauty of their utility/outcome? Pretty Please?”

    It all might be totally legit in the field of Solar Physics. But I felt a little dirty afterward.

  17. tallbloke says:

    The ACRIM composite (red) compared to the Lean Frohlich model 2004

  18. tallbloke says:

    Things are going downhill fast on the WUWT thread.

    tallbloke says:
    Your comment is awaiting moderation.
    November 12, 2012 at 2:03 pm
    Steven Mosher says:
    November 12, 2012 at 10:51 am
    Geoff Sharp
    Do you think we landed on the moon?

    Ah, Mosh has played the Lewandowsky card.

    And here comes Leif to join him:

    Leif Svalgaard says:
    November 12, 2012 at 1:02 pm
    Dunning-Kruger strikes again: “cognitive bias in which unskilled individuals suffer from illusory superiority, mistakenly rating their ability much higher than average.

    Credibility rating now zero – game over, thanks for playing guys.

  19. GAI says:

    “Things are going downhill fast on the WUWT thread.”

    Yes, it makes me sick to see everyone fawning over Lief. Unfortunately I do not have the technical expertise to tangle with him and people like Dr Doug Hoyt, Dr. Richard C. Willson, and Dr. Nicola Scafetta are either too busy are in a position where they need to remain neutral.

    Lief gets away with his shenanigans because the people who would take him down do not know the subject well enough to do so though Vukcevic usually gives him a run for his money.

    OH and Jeff L just hit Lief with a real zinger at http://wattsupwiththat.com/2012/11/11/solar-activity-past-present-future/#comment-1145079

    Seems a Russian, Habibullo I. Abdussamatov of the Pulkovo Observatory, just came out with a paper

    …saying we are on the cusp on a new little ice age based on expected solar variation. He’s forecasting a drop of 6 w/m^2 in TSI over the next 30 years.

    The paper: http://icecap.us/images/uploads/abduss_APR.pdf

    Abstract
    Temporal changes in the power of the longwave radiation of the system Earth-atmosphere emitted to space always lag behind changes in the power of absorbed solar radiation due to slow change of its enthalpy. That is why the debit and credit parts of the average annual energy budget of the terrestrial globe with its air and water envelope are practically always in an unbalanced state. Average annual balance of the thermal budget of the system Earth-atmosphere during long time period will reliably determine the course and value of both an energy excess accumulated by the Earth or the energy deficit in the thermal budget which, with account for data of the TSI forecast, can define and predict well in advance the direction and amplitude of the forthcoming climate changes. From early 90s we observe bicentennial decrease in both the TSI and the portion of its energy absorbed by the Earth. The Earth as a planet will henceforward have negative balance in the energy budget which will result in the temperature drop in approximately 2014. Due to increase of albedo and decrease of the greenhouse gases atmospheric concentration the absorbed portion of solar energy and the influence of the greenhouse effect will additionally decline. The influence of the consecutive chain of feedback effects which can lead to additional drop of temperature will surpass the influence of the TSI decrease. The onset of the deep bicentennial minimum of TSI is expected in 2042±11, that of the 19th Little Ice Age in the past 7500 years – in 2055±11.

    Habibullo Ismailovich Abdussamatov is a Russian astrophysicist and supervisor of the Astrometria[1] project of the Russian section of the International Space Station and the head of Space research laboratory at the Saint Petersburg-based Pulkovo Observatory[2][3] of the Russian Academy of Sciences. According to WIKI

    In this 2006 paper http://www.gao.spb.ru/english/astrometr/KPhCB22_3_141.pdf
    Abdussamatov had predicted a delayed onset and reduced activity for the present solar Cycle 24.

    Luboš Motl has links to a bunch of these papers: http://motls.blogspot.com/2007/01/peer-reviewed-global-cooling.html

    At this point I think I will believe the Russians. When it comes to cold weather they have a real vested interest in the truth. Also Abdussamatov made a prediction in 2006 and he was correct where NASA was WAY OFF. NASA has saved face by saying it was Hathaway’s prediction now that it has been proved so wrong.

  20. GAI says:

    Given the contempt Lief shows towards other posters one wonders why he even bothers to spend so much time at WUWT.

  21. Paul Vaughan says:

    GAI says (November 13, 2012 at 12:17 am)
    “Given the contempt Lief shows towards other posters one wonders why he even bothers to spend so much time at WUWT.”

    Abuse of university resources and authority to protractedly harass volunteers from the top end of an academic power differential would be considered formally unethical at any university where I have ever worked. One is certainly left wondering about Stanford’s (apparent lack of decent) standards.

    http://en.wikipedia.org/wiki/Peter_Principle

    Thanks for the link. I will draft an article titled:
    The Ironed Sun

  22. Greg Goodman says:

    Since TSI seems to be basically the envelop of the 11y SSN cycle the Lean and Wang reconstructions are just adding it to its own average. (Not sure why they do that from a physical point of view).

    This “average” will undoubtedly be a (monthly?) running average. This is a crude low pass filter which will severely distort just the cycle that you are so interested in studying. This additional will also dilute the short term variation.

    Distortion notwithstanding, this is basically adding it back into itself and on a scale >11ys is just multiplying it by some factor (where Wang is less that Lean).

    Now TSI is too small in itself so needs some amplifying effect. All that changes is the size of the amplification needed.

    As you say above, the 20% is not much of a problem and may be corroborated by the z-axis idea.

    If you want compare long term TSI signal you should still find a similar variation in Svalgaard’s reworked TSI and it does not seem to be the <11y component that matches z-axis anyway.

    I don't like Leif's overconfidence in his own importance, but that does not mean he's wrong.

    I suspect his new TSI is a purer reflection of cycles that various Lean / Wang style ones but just with a stronger proportion of shorter signals due to not adding a poorly filtered long term average back in.

    Also, if you want to look at correlation of cycles in various data DON'T use running mean a filter. It severely distorts the frequency spectrum and you risk messing up a correlation that may be present. Especially if you apply different periods as you do in the SSB v SSN plot above.

    Although the Svalgaard versions looks "flatter" when plotted on the scale it don't think it will degrade cyclic analysis. It is basically the same signal.

    It may be a better reflection of the cycles.

  23. Greg Goodman says:

    See what I mean:

    Leif’s version is showing stronger cyclic variation.

  24. Greg Goodman says:

    Flatter Sun – Flatter Earth , not really.

  25. tallbloke says:

    Hi Greg, and thanks for your observations.
    You assume there are no longer term cycle underlying the succession of 11 year cycles. We think there is a 974 year cycle in solar variation. And a ~2240 cycle. And ~4600. And… longer ones too.

    I agree that running mean filters are problematic. My rationale for the ones I used in the z-axis plots are that a 24 year filter on the planetary data steadies down the Jupiter orbital and Jupiter saturn synodic periods so we can see what else is going on. There may be cleverer ways of doing this but it serves as a rough and ready engineers approximation.

    Cheers – TB

  26. Greg Goodman says:

    “You assume there are no longer term cycle underlying the succession of 11 year cycles.”

    No, I don’t assume that. I’m certain there are longer ones. There is obviously a longer term rise in both TSI series. Like I said they are essentially the same data.

    Without going into all that my point is that they are both essentially the same data and I think Leif’s reworking is showing the cyclic data better. The “sitting on top of” idea seems to be without any particular reason for being and as LS points out it does not work on last decade.

    If anyone wants to persist with adding back in a percentage of the low frequency component is needs to be done properly, with runny means.

    “There may be cleverer ways of doing this but it serves as a rough and ready engineers approximation”

    Well one cleverer way would be a gaussian or binomial filter. Both can be done in a similar way to runny as a weighted running average . In reality this is a convolution of the time series with a kernel (the window weights).

    Convoluting with a rectangle, as a runny mean does, multiplies the frequency spectrum by the sinc function, including the _nagative_ lobes. This is why it is such a disaster for this kind of work.

    I’ll try to post a script for doing a gaussian (who’s freq response is also gaussian). If it makes a mess email me and I’ll send it you.

  27. Greg Goodman says:
    
    #!/bin/awk -f
    
    # pass input through 3 sigma gaussian filter where sigma, if not given, is 2 data points wide
    # usage : ./gauss.awk filename <sigma=2> <scale_factor=1>
    # optional scale_factor simply scales the output
    # use  OFMT="%6.4f"
    # sigma can be compared to the period of the -3dB point of the filter
    # result is centred, ie not shift. dataset shortened by half window each end
    # check whether data is continuous !!
    
    
    BEGIN { OFMT="%6.4f"
    # ARGV[1]=filename; argv[0] is script name, hence ARGC>=1
      pi= 3.14159265359811668006
    
      if  ( ARGC >3 ) {scaleby=ARGV[3];ARGV[3]=""} else {scaleby=1};
      if  ( ARGC >2 ) {sigma=ARGV[2];ARGV[2]=""} else {sigma=2};
    
      print "filtering "ARGV[1]" with gaussian of sigma= ",sigma
      root2pi_sigma=sqrt(2*pi)*sigma;
      two_sig_sqr=2.0*sigma*sigma;
    
      gw=3*sigma-1;  # gauss is approx zero at 3 sigma, use 3 sig window
    # eg.  window=2*gw-1 -  5 pts for sigma=1; 11pts for sig=2; 3 sig=17
    
    # calculate normalised gaussian coeffs
      for (tot_wt=j=0;j<=gw;j++) {tot_wt+=gwt[-j]=gwt[j]=exp(-j*j/two_sig_sqr)/ root2pi_sigma};
      tot_wt=2*tot_wt-gwt[0];
      tot_wt/=scaleby;
      for (j=-gw;j<=gw;j++) {gwt[j]/=tot_wt};
    
    # strip off last .xxx part of file name
    # improve this  (doesn't work with paths like ../filename)
    
      split(ARGV[1],fn,".");
      basename=fn[1]
      gsfile=basename"-gauss"sigma".dat";
    
      print "# ",gsfile >gsfile;
      ln=-1;
    }
    
    ($0 !~ /^#/)&&($0 != ""){
      xdata[++ln]=$1;
      ydata[ln]=$2;
    
      if (ln>2*gw)
      {
        gauss=0
        for (j=-2*gw;j<=0;j++) {gauss+=ydata[ln+j]*gwt[j+gw]}
        print NR,xdata[ln-gw],gauss
        print xdata[ln-gw],gauss >> gsfile;
      }
    else
      {
    #    print  $1,$2;
    
      }
    }
    
    
    END {
      print "#gausssian window width = "gw+gw+1",done"
      print "#output file = "gsfile
    
    }
    
    
  28. Greg Goodman says:

    Hmm. looks OK.

    You should find that controls your JS frequencies with introducing distortions across the board. It takes wider window to do the job properly but it really will stamp out a lot of stuff that gets through a runny mean.

    The main thing is that it will not invert/truncate peaks and bend them sideways. That is a real killer when you are looking correlations.

    Since the impulse response is a gaussian bell there no “cut-off” frequency like a RC filter (which not that pointy either). I’ve never been too sure what the -3dB point is. I tend to give it an argument of 12 for monthly data but it may not be -3dB at one year.

    I hope that’s of use.

    Greg.

  29. Greg Goodman says:

    OH yes, the reason I popped back was this from the paper covered WUWT this week. http://www.nature.com/ngeo/journal/vaop/ncurrent/carousel/ngeo1626-f3.jpg

    They seem too busy with the “unprecedent” “pollution” BS to note their significant find.

    It seems that 10.7 UV is blasting CO2 and splitting it like it does O2. The Co2 and CO are prefect complements, but CO is not a GHG.

    I assume the residual O is making ozone but I can’t find any info on such a process. It may well be happening in upper tropo as well.

    Now since CO2 absorption is NOT saturated in the stratosphere the difference is probably in the “linear” absorption range (which is log BTW , not linear, since absorption is log(conc) it’s the log plot which is linear ;) ).

    Now there is a direct, easily calculable modulation of incoming solar that can be linked to SSN.

    Note the inverse action of CO2 as detailed in that paper. More 10.7 , less CO2 blockage, more LIR input.

    They also note CO2 increasing by about 2ppm / y , pretty much like tropo. levels. In view of the near saturation in troposphere, that could turn the whole CO2 argument on its head !

    Apparently all this was “unexpected” since it’s not in the models.

  30. tallbloke says:

    Greg, thanks, I’ll get Tim C to look at that with me. He’s the expert in signal processing round here.

    There are four main issues with Svalgaard’s reworking of the data.

    1) The removal of the underlying GSN mean ‘carrier’ – OK, I can see that one.
    2) The use of a short term much massaged TSI signal model as a trend setter for distant past sunspot data. – We won’t let him get away with this one. The ACRIM team know it’s a lie.
    3) The K factor corrections. The dog ate the homework in Ken Schatten’s case (memory loss). Geoff Sharp has things to say on this subject.
    4) The reworking of old geomag records. – This needs oversight.

  31. tallbloke says:

    Greg Goodman says:
    November 13, 2012 at 11:29 am

    OH yes, the reason I popped back was this from the paper covered WUWT this week.

    Have you a link to the WUWT article, I missed this one.

  32. Greg Goodman says:

    change in stratospheric radiative effect of CO2 per decade:
    ln(26/380)=6.6% per decade.
    Don’t know who has worked out how much it radiates or blocks in watts, it could be piffling.

  33. tallbloke says:

    Thanks again,
    You have mail by the way

  34. GAI says:

    Greg Goodman says:
    November 13, 2012 at 5:04 am

    …I don’t like Leif’s overconfidence in his own importance, but that does not mean he’s wrong….
    _____________________________________
    It is not his overconfidence but his refusal to answer sticky questions when I brought up the ACRIM work or the fact NASA says that while the TSI is not too variable the distribution of wavelengths is. Given the oceans only absorb visible light and shorter wavelengths it is not TSI so much but the wavelength distribution that is going to matter.

    You have now brought up “that 10.7 UV is blasting CO2 and splitting it like it does O2. The Co2 and CO are prefect complements, but CO is not a GHG.” Again it is not TSI that counts but the distribution of wavelengths.

    That change in wavelength distribution and other solar changes are the peas under the shell that Lief is hiding at all costs by emphasizing that TSI is constant and therefore the sun has no role in changing the short term climate.

    That emphasis just does not agree with other data.

    NASA: Cosmic Rays Hit Space Age High (Sept 2009)

    “In 2009, cosmic ray intensities have increased 19% beyond anything we’ve seen in the past 50 years,” says Richard Mewaldt of Caltech. “The increase is significant…

    “We’re experiencing the deepest solar minimum in nearly a century,” says Dean Pesnell of the Goddard Space Flight Center, “so it is no surprise that cosmic rays are at record levels for the Space Age.”

    the Solar Dynamics Observatory Mission News: EVE: Measuring the Sun’s Hidden Variability

    …”Solar minimum is a quiet time when we can establish a baseline for evaluating long-term trends,” he explains. “All stars are variable at some level, and the sun is no exception. We want to compare the sun’s brightness now to its brightness during previous minima and ask: is the sun getting brighter or dimmer?”

    The answer seems to be dimmer. Measurements by a variety of spacecraft indicate a 12-year lessening of the sun’s “irradiance” by about 0.02% at visible wavelengths and 6% at EUV wavelengths….

    And there is this study: NASA Finds Sun-Climate Connection in Old Nile Records

    Long-term climate records are a key to understanding how Earth’s climate changed in the past and how it may change in the future. Direct measurements of light energy emitted by the sun, taken by satellites and other modern scientific techniques, suggest variations in the sun’s activity influence Earth’s long-term climate….

    a group of NASA and university scientists has found a convincing link between long-term solar and climate variability in a unique and unexpected source: directly measured ancient water level records of the Nile, Earth’s longest river….

    Alexander Ruzmaikin and Joan Feynman of NASA’s Jet Propulsion Laboratory, Pasadena, Calif., together with Dr. Yuk Yung of the California Institute of Technology, Pasadena, Calif., have analyzed Egyptian records of annual Nile water levels collected between 622 and 1470 A.D. at Rawdah Island in Cairo. These records were then compared to another well-documented human record from the same time period: observations of the number of auroras reported per decade in the Northern Hemisphere. Auroras are bright glows in the night sky that happen when mass is rapidly ejected from the sun’s corona, or following solar flares. They are an excellent means of tracking variations in the sun’s activity….

    The researchers found some clear links between the sun’s activity and climate variations. The Nile water levels and aurora records had two somewhat regularly occurring variations in common – one with a period of about 88 years and the second with a period of about 200 years.

  35. tallbloke says:

    It’s a pity ACE was launched in 2003 just as solar cycle 23 was waning. They might have seen an opposite trend in thermospheric co2 as it strengthened.

  36. tallbloke says:

    GAI: Yes, Leif is in de-Nile on that one. The Gleissberg and de Vries cycles are evident.

    Looking at the ACRIM plot and the deep dive TSI took at SC23/24 minimum, the Sun is a good deal more variable than Leif wants us all to think it is.

  37. GAI says:

    tallbloke says:

    GAI: Yes, Leif is in de-Nile on that one. The Gleissberg and de Vries cycles are evident.

    Looking at the ACRIM plot and the deep dive TSI took at SC23/24 minimum, the Sun is a good deal more variable than Leif wants us all to think it is.
    _______________________________
    If the Russian Solar Physicist is correct the West is going to be in very deep… snow rather shortly. I would not want to be a politician when the fecal material hits the rotating blades. 28gate will take wings at that point.

    They can only keep covering this up for so long but the incredible stupidity of the sheeple sometimes makes me wonder.

  38. ed says:

    Here is the plot I was referring to with what looks like cal issues.

    This is Leifs TSI reconstruction spaghetti plot:

    http://s852.beta.photobucket.com/user/etregembo/media/LeifTSI2008.png.html?sort=3&o=2

    After you remove the long term reconstructions, you can his reconstruction from 2008 quite clearly, and the strangely low value at 2007.5 which is lower than any point during the last 300-400yrs (cal issue to me)

    http://s852.beta.photobucket.com/user/etregembo/media/LeifTSI2008LeifSomeRemoved.png.html?sort=3&o=3

    Here is it zoomed in to see the cal effort. Looks to me that the TSI values were all over the map depending on which you preferred, seems like he preferred bits of each to me.

    http://s852.beta.photobucket.com/user/etregembo/media/LeifTSIScaledOffset2008vsGuessRecent.png.html?sort=3&o=1

    Maybe it’s nothing, or an artifact of smoothing, but it looks to me like he had a hell of a time with the crappy PMOD/ACRIM constructions we have to date, and what parts to cal to, and he picked whatever he wanted in the end, which ended being his “best guess” (with some floor clipping)

    Question for you though…what was the justification for creating a TSI floor for sunspot count equals zero? Couldn’t find it in his “essay”. Is there any data which shows that TSI always drops to the same value for a sunspot equals zero count? I believe the cosmic ray count (oulu) doesn’t support that when comparing solar mins, with the most recent one at an elevated level compared to previous mins. Seems like he snuck that one in there, along with all of the other adjustments.

  39. ed says:

    test123…

    can I post?

  40. Greg Goodman says:

    Rog, I think its clear that TSI is a significant driver directly so Svalgaard’s “it can’t be the sun” argument based on this is clearly a deliberate distraction. That means he is quite likely to add the same biased attitude to any adjustments he tries to push through and we don’t need another dataset getting screwed up.

    Apart from that I think getting rid of the carrier will enhance the clarity of cyclic effects and should help detect the cause.

    SSB-z is interesting but fails badly in the earliest peak. That is a big problem, it’s going to need some huge opposing effect to bring it back into line. Like string theory, once the complexity of any model starts getting combersome, it’s usually a good indication that you’re heading in the wrong direction.

    I think the one thing to be saved is the z-axis correlation may indicate that one or more planets with inclined orbits may be important in determining SSN.

    I think the whole idea of the barycentre having a physical effect is fundamentally wrong. The idea of a point centre of mass is only useful in the far field. That means a long way away from all the little bits that make it up. The SSB makes sense from outside the SS not within it !!

    If you want to look at gravitational effects on the sun you need to weight each planet with the square root of its distance, ie you need to work out the centre of _gravity_ at any point in time and not the centre mass.

    This will, primarily, up the effect of J and reduce S with other significant changes. This may tie in the vukevic’s formula which seems a good first approximation and has a dominant J period.

    This will have some similarity to SSB but also significant differences, the net effect may (or may not) give you a correction to phase crisis in the first peak.

    In any case I think you should be using gravity not centre of mass. A lot of people have led themselves astray with this concept of SSB. As Steve Goddard pointed out, the Earth-Moon couple rotate about the Sun not the SSB.

    In fact I don’t think the Sun is the true centre of rotation either but it’s a lot closer to it, than to the SSB.

    I imagine whoever has calculated the z-component of SSB would have no problem finding the vector sum of all the individual gravity vectors.

  41. Greg Goodman says:

    Oops, that should have read: I think its clear that TSI is NOT a significant driver directly so Svalgaard’s “it can’t be the sun” argument based on this is clearly a deliberate distraction.

    It think any solar link has to be via Svensmark effect or vukevic’s magnetic coupling or similar, though SSN does give a handle on solar activity and is the only long record we have.

  42. Brian H says:

    If E and S rotate about a mutual CoG, and one/both are then/also affected by J, how much does it matter that distance from J and relative mass of E and S differ? Their CoG is a local abstraction, not a thing-in-itself.

  43. Greg Goodman says:

    “If E and S rotate about a mutual CoG,” They don’t. The discussion stops there. That kind of argument only applies to a two body system.

    CoG is usually used synonymously with centre of mass. That identity is only valid in a uniform gravitational field (like our earthbound experience) and that ambivalent use of the term CoG is probably the root of much of the confusion that anything rotates about SSB or that SSB can influence anything inside the solar system.

    SSB could be used to characterise the CoG of SS from distances many times larger than the orbit of the major planets ie well outside the solar system itself.

    Inside the SS the “local” variations in gravitational field are precisely what this whole cyclic influence
    is interested in. In that case we need to be looking at vector sum of individual gravitational forces, which implies weighting planetary masses by inv. sqr of distance.

  44. tallbloke says:

    Hi guys. Thanks for keeping this thread ticking along while I’ve been BBC BUSY. :)
    I’ve spent many hours on many occasions visualising the motion of the planets and Sun taking all your points raised into consideration. I came to the conclusion that the inner planets pretty much follow the Sun as the focus of their orbits. The tiny offset due to the distance from the solar centre to their mutual CoG is… tiny.

    But Jupiter and the rest of the jovian gas giants are different. They are much bigger, heavier, and much further away from the Sun’s gravitational source. Jupiter also has a big eccentricity in its orbit. So they have as the focus of their orbits neither the Solar centre nor the (gravitational) solar system barycentre (which is what my JPL ephemeris data supplies, rather than the CoM). Instead, the (ever changing) focus of their individual orbits depends on what the other three gas giants are up to in terms of position at any point in time.

    And because the outer two gas giants are closer to each other than the Sun, they also perturb each others orbits (and therefore their foci) significantly, such that you can’t really talk of their orbits having foci in the Keplerian sense.

    Svalgaard will tell you ‘it doesn’t matter’, because the forces are ‘small’. But small forces often repeated at the right time build up bigger resonant effects, and this is the reason the planets have become organised into the orbits and inter-related timings they possess. It’s also why Earth’s orbit varies in shape over the 100kyr cycle to cause the glacial/interglacial cycle.

    More later, I have to get to the train station.

  45. Greg Goodman says:

    That last point is the good news. Whatever program is currently being used to calculate the SSB could easily be modified by weighting the planetary masses according to (at least the mean) orbital distance.

    However, unlike the combined mass, the magnitude of the net gravitational force will vary somewhat for eccentric orbits. With eccentricities around 0.04 for the giants this will be of the order of a 8% variation for each planet’s orbital period.

    This could produce yet more long period beat frequencies as not only alignment but synchronisation with perihelion come in and out of phase.

    Since looking at J pulling of V-E induced, tidal bulges gives some correlations, that would suggest adding gravitational forces is the way to go.

    This would reunite the (partially working but flawed) SSB and the tidal lines of investigation and should work better than either individually.

  46. Greg Goodman says:

    TB:…(gravitational) solar system barycentre (which is what my JPL ephemeris data supplies, rather than the CoM)

    Sorry, you posted while I was writing. A lot of what I suggested is already done in that case. :(

    Just add to the confusion we now have two meanings for barycentre. Duh.

  47. Paul Vaughan says:

    Greg Goodman (November 14, 2012 at 8:12 am) wrote:
    It think any solar link has to be via Svensmark effect or vukevic’s magnetic coupling or similar, though SSN does give a handle on solar activity and is the only long record we have.”

    It’s a lot simpler than that. It’s just equator-pole gradient modulation, consequent circulatory morphology change, and consequent redistribution of water over the surface. How this got overlooked is not clear, but it’s crystal clear almost nobody is willing to acknowledge the message being shouted by actual data. My theory is that a lot of vocal people want to preserve public perception of immense uncertainty as this facilitates adaptable political twist & spin to suit whatever narrative’s convenient in any passing moment.

  48. tallbloke says:

    Actually, thinking about it more, the JPL ephemeris doesn’t explicitly say that. It simply gives positions objects as x,y,z co-ordinates relative to the barycentre at points in time. So now I need to confirm to myself what they define the barycentre to be – the CoM or the CoG.

    Stand by.

  49. tallbloke says:

    http://ssd.jpl.nasa.gov/?horizons_doc#site

    “Deflecting body:

    This is the Sun PLUS the most massive object in the planet/satellite system (e.g. the system barycenter). These two masses are used to compute the relativistic deflection of light that can change the apparent position of the target body.”

    “LIST OF MAJOR BODY EPHEMERIDES ON-LINE

    The following major body ephemerides are currently on-line. Newly discovered satellites are also available, although they are not shown below. Planet centers are considered the 99th satellite of the system barycenter. Satellites 506-513, 607, 716-721 and 802 do not have defined rotational models in the 2000 IAU report.

    000 Solar System Barycenter
    10 Sun”

    The first bit refers to Mass. The second to ‘orbiting bodies’ – gravity.

    Gerry Pease can help us here. he worked with JPL constructing the ephemeris. I’ll email him.

  50. tallbloke says:

    In the meantime, I’ll address Greg’s points about the z-ais correlation.

    Firstly – be aware that although Ray Tomes has managed to get a 0.6 correlation betrween SSN and unsmoothed z-distance in realtime by introducing a 10.5 yr oscillation for the ‘solar dynamo’, my correlation is apparently unphysical. The two curves have a lag of 45 years, with the sunspots coming first.

    Although at first flush this kills it, it’s worth considering we are dealing with cyclic phenomena, and if the system is a true system, we are dealing with feedbacks in both directions. So to consider an analogy to what might be happening; A more active sun and solar wind ‘pushes’ the orbits, which move outwards and slow down, which then produces a reduction in the solar output and solar wind speed, allowing the planets to fall inwards again. The planets then ‘push’ the sun to become more active again. All systems with cybernetic feedback of this type oscillate about a mean. So having planets ‘leading’ the sun at one epoch and the sun ‘leading’ the planets at another is exactly what we would expect for our solar-planetary ‘governor’ system.

    If the system is very finely balanced near boundary conditions (which it must be for us to be finding such good correlations relying on such apparently tiny forces) then the anomaly distances involved in this ‘hunting’ around the mean orbital radii are not very large, and won’t show up in JPL’s calculations. They probably do show up in other ways though, like the oceanic 60 year oscillations and the chandler wobble and the correlation I found between LOD variation and Z-axis motion is probably related to those, and Paul Vaughan may say some more about that.

    Now, every couple of hundred years, the sun goes into a funk, as we see from the succession of grand minima over the last millennium. Oort – Sporer – Maunder – Dalton – and now.

    So the divergence seen in the plot between ssolar activity and the z-axis curve is the Dalton Minimum, a period of anomalously low solar activity where everything goes ‘non-linear’ for a while.

    The 4 black tick marks above the time axis indicate points where the z-axis motion has a ‘hiccup’ near the solar surface (orange band on the plot is +/- 0.1 solar radii relative to the x,y,z radial distance of the barycentre from the solar centre (which is at around -700,000 km = -70,000km on the left y axis of the plot). Landscheidt noticed that solar activity crashes whenever the radial distance hangs around near the solar surface for more than 3 years or so. (Mechanism unknown).

    Hence the anomalously low activity 1796-1830 and again around 1970 and again about now. During the Maunde there were three such ‘hiccup’ periods over 50 years, so its unlikely to be coincidence.

  51. GAI says:

    ed says:
    …Question for you though…what was the justification for creating a TSI floor for sunspot count equals zero? Couldn’t find it in his “essay”. Is there any data which shows that TSI always drops to the same value for a sunspot equals zero count? I believe the cosmic ray count (oulu) doesn’t support that when comparing solar mins, with the most recent one at an elevated level compared to previous mins. Seems like he snuck that one in there, along with all of the other adjustments.
    No, at least not according to NASA and NOOA.

    This is the TSI at Mauna Loa (above cloud level) from NOAA.

    Notice how the TSI has trended downward in recent years, Also notice the dip around 1965 corresponding to “Global Cooling” panic of the 1970’s. (The sharp dips are volcanoes I think)

    A graph from NASA: Deep Solar Minimum

    This is why looking just at sunspots or Lief’s focus on just TSI is barking up the wrong tree.

    >a href=”http://www.nasa.gov/topics/solarsystem/features/solarcycle-sorce.html”>SORCE’s Solar Spectral Surprise

    Measuring Variation

    As recently as the 1970s, scientists assumed that the sun’s irradiance was unchanging; the amount of energy it expels was even called the “solar constant.” However, instruments similar to TIM and SIM have made clear that the sun’s output actually fluctuates in sync with changes in the sun’s magnetic field.

    Indeed, TIM and its predecessor instruments, whose records of irradiance began in 1978, show that the sun’s output varies by about 0.1 percent as the sun cycles through periods of high and low electromagnetic activity every eleven years or so. In practice, this cycling means the sun’s brightness, as measured by TIM, goes up a bit when large numbers of sunspots and accompanying bright spots called faculae are present on the sun, yet goes down slightly when sunspots and faculae are sparse, like they have been in the last few years as the sun has gone through an unusually quiet period.

    However, there is a critical difference between the SIM and TIM, explains Jerry Harder, the lead SIM instrument scientist and a researcher at the Laboratory for Atmospheric and Space Physics (LASP) at the University of Colorado in Boulder. While the TIM lumps all wavelengths — including infrared, visible, and ultraviolet light — into one overall measurement, the SIM isolates and monitors specific portions of the spectrum.

    Notably, this makes SIM the first space-based instrument capable of continuously monitoring the visible and near-infrared portion, parts of the spectrum that are particularly important for the climate. SIM also offers the most comprehensive view of the individual components that make up the sun’s total solar irradiance to date.

    Some of the variations that SIM has measured in the last few years do not mesh with what most scientists expected. Climatologists have generally thought that the various part of the spectrum would vary in lockstep with changes in total solar irradiance.

    However, SIM suggests that ultraviolet irradiance fell far more than expected between 2004 and 2007 — by ten times as much as the total irradiance did — while irradiance in certain visible and infrared wavelengths surprisingly increased, even as solar activity wound down overall.

    And another article by NASA I linked to in an above comment showed the extreme ultraviolet changed as much as 6%. These are the wavelengths that effect the upper atmosphere, the ozone layer and split CO2 into CO and an oxygen radical http://arstechnica.com/science/2012/11/warming-earth-makes-satellites-move-faster/

    …One of those researchers (along with several of his colleagues) has now published the results of an analysis of CO2 in the thermosphere. It’s a difficult thing to measure, but they used data the Canadian SCISAT-1 satellite collected by observing the setting Sun through the thermosphere above the horizon from 2004 to 2012. To untangle trends from variability caused by solar activity, they combined measurements of carbon monoxide and carbon dioxide. Ultraviolet light splits a portion of the CO2 into CO, which can be oxidized to turn back into CO2.

    Not to mention the fact the Oxygen free radical can combine with the much more abundant O2 to form ozone – but you wouldn’t want to mention that possibility.

  52. Greg Goodman says:

    TB: This is the Sun PLUS the most massive object in the planet/satellite system (e.g. the system barycenter). These two masses are used to compute the relativistic deflection of light that can change the apparent position of the target body.”

    The first bit refers to Mass. The second to ‘orbiting bodies’ – gravity.
    ————

    ” the system barycenter” refers to barycentre of the two bodies (planet and most massive satellite). Orbiting bodies in the context, is simply observed postition , accounting for gravitational deflection of light by the two masses: sun plus planet-satellite common mass at local barycentre. There is no reference to gravitational attraction between planets and/or sun here. It’s all mass.

  53. Greg Goodman says:

    … so observer point 000 (SSB) is the centre of mass. of SS. My initial comments in this seem to be correct. I think you need to use OP 000 coords for each planet and sun, work out distance of planet-N to sun and hence the inv.sqr. weighting and direction of gravitational vector. Finally find the direction and magnitude of the resultant vector of all planetary attractions.

    Off the top of my head it may be quicker to get this as the solar acceleration , deriving that from the ephemeris of the sun.

    Crap! SSB isn’t a fix reference frame anyway, none of this will work. Needs more thought and I have to go out.

    Trouble with the ephemeris is they are all ephemeral !

    They are all about _observable_ positions of moving bodies from a moving point of observation.

    Maybe they also have a table of astronomic positions or a “fixed stars” kind of observation point.

  54. Greg Goodman says:

    OK, solar accel would be a nice short-cut if it was relative to a fixed inertial frame. Otherwise, instantaneous position of a planet and sun, in any frame, enables calculation of distance hence gravity.

    magnitude of net gravitational force and probably its z-axis component is probably what should be plotted.

    Hope that makes sense.

  55. Geoff Sharp says:

    Tallbloke says…

    The 4 black tick marks above the time axis indicate points where the z-axis motion has a ‘hiccup’ near the solar surface (orange band on the plot is +/- 0.1 solar radii relative to the x,y,z radial distance of the barycentre from the solar centre (which is at around -700,000 km = -70,000km on the left y axis of the plot). Landscheidt noticed that solar activity crashes whenever the radial distance hangs around near the solar surface for more than 3 years or so. (Mechanism unknown).

    This is not correct Rog. Landscheidt saw the radial distance that hangs near the solar surface as a boundary condition that enables phase reversal and also influences the human mind, stocks, etc etc. He does not mention a reduction in sunspots.

    http://www.landscheidt.info/?q=node/243

    We now know that when the solar inner loop path travels OUTSIDE the distance of 1 solar radius from the SSB solar slowdown occurs. The greater the distance in this situation the greater the slowdown.

    I also have a new post that explains the 208 year DeVries pseudo-cycle.

    http://www.landscheidt.info/?q=node/293

  56. tallbloke says:

    Hi Geoff and welcome back to the talkshop.
    In the interests of harmony within the solar system I’m going to leave the Landscheidt issue uncontested and move on to your interesting observation about the degree of divergence of the inner loop above the solar surface.

    Regarding the De Vries cycle, Ray Tomes has an interesting analysis of Svalgaards sunspot series here:

    http://cyclesresearchinstitute.wordpress.com/2012/11/13/sunspot-number-reconstruction-by-leif-svalgaard-and-proxy-cycles-of-104-and-208-years/

    I’m going to mull over the inner loop observation for a couple of hours and then comment further.

  57. Geoff Sharp says:

    The observation came to me late last night…strange after starring at these image for the past 4 years. The solar distance/AM graph gave me the key, the further the suns path on the inner loop (ie outside the solar radius) travels away from the SSB the great the perturbation on the AM graph and if the timing is right the greater the solar slowdown. The inner loop for a short time is trying to be an outer loop, this must upset the normal rhythm of the solar dynamo that I think works on some kind of acceleration/deceleration phase associated with each loop. This all working outside of the cycle length and timing which seems to be linked to the torsional oscillation bands and perhaps a JEV tidal influence.

    Charvàtovà.was on the right path and laid the foundation, but did not really understand the complexities of the disordered path, which completely led her astray in the end unfortunately.

    Be ready for the next wave of planetary papers to hit, the ball is really rolling now.

  58. Geoff Sharp says:

    Rays work is interesting but must admit I don’t see how repeating cycles can exist in the solar proxy record. I know this will upset a lot here but the data is strong in this area. Gaps between grand minima are extremely variable because of the multiple options involved, but I can see how one gap can show up more in FFT type analysis, which is quite different from a 208 year cycle. Longer term patterns might be a different argument.

  59. tallbloke says:

    Hi Geoff. OK, you’ve raised several issues here. Let’s have a look at the inner loop distance first.

    You said:
    We now know that when the solar inner loop path travels OUTSIDE the distance of 1 solar radius from the SSB solar slowdown occurs. The greater the distance in this situation the greater the slowdown.

    Looking at the lower panels of your nice graphic at http://www.landscheidt.info/images/powerwave3.png I can see that in 1791 the inner loop doesn’t actually break out of the solar surface at all, but SC4 morphs into an anomalously long cycle followed by two very low cycles (Dalton Minimum).

    Then in 1827-1832 there is a big breakout getting the furthest distance above the solar surface for the longest time in the whole set. But this coincides with SC7 which is higher than the previous two and is followed by two high solar cycles 8&9.

    On the mid 1860’s-1870’s and mid 1920’s-1930’s occasions I’m not sure why you’ve shaded in the area between the path of the SSB and the solar surface because these are not continuous events. The SSB comes in at a low angle from an outer loop and 10 years later leaves at a low angle, but there isn’t a continuous period where the ssb breaks out of and then re-enters the Sun while traversing the inner loop.

    Maybe I’m thinking around this in a different way to you, so let us know how you’ve arrived at your summary of the observation.

    Cheers – TB

  60. tallbloke says:

    Geoff, you said:
    I can see how one gap can show up more in FFT type analysis, which is quite different from a 208 year cycle.

    The 207-8 year peak in the FFT’s of the proxy data is quite sharp and well defined. However, we know 10Be is affected by terrestrial as well as celestial phenomena, so maybe there is a terrestrial amplification of the celestial signal due to a non-linear factor. In Tim C’s model reconstruction of steinhilber et al’s 2009 result, he found the 208 year cycle peaking in 1580, 1790 and 2000, putting the troughs at 1685 and 1895. How well does this phase with your Uranus-Neptune observations?

  61. Geoff Sharp says:

    tallbloke says:
    November 15, 2012 at 6:36 am

    Great questions.

    Looking at the lower panels of your nice graphic at http://www.landscheidt.info/images/powerwave3.png I can see that in 1791 the inner loop doesn’t actually break out of the solar surface at all, but SC4 morphs into an anomalously long cycle followed by two very low cycles (Dalton Minimum).

    The 1791 event is medium, it differs from the 1970 event by happening earlier. Timing is important which does not really show as much when looking at the orbit diagrams, 1791 starts its perturbation higher than 1970 which means it is disrupting the deceleration stage, in 1970 the deceleration has almost finished so the effect is weaker. The 1791 event does not go outside the solar radius but is still reasonable. It seems in many places across the Holocene that a good disruption at the tail end of one cycle severely effects the next cycle. The theory then suggests the 1st grand minimum cycle experiences a phase catastrophe of the Hale cycle which takes the next cycle to sort out, we might see that in the next year.

    Then in 1827-1832 there is a big breakout getting the furthest distance above the solar surface for the longest time in the whole set. But this coincides with SC7 which is higher than the previous two and is followed by two high solar cycles 8&9.

    The 1830 event is not as strong as the 1650 event but still strong. Once again timing is important but this time it is about what stage the solar cycle is experiencing. The AM disruption or solar path changes I think happen outside of cycle timing, with 1830 the disruption (which is normally only a few years) is happening near cycle max…too late but some damage is still witnessed. This is fairly rare over the Holocene suggesting that cycle timing might be in some sort of sync with the disruptive force.

    On the mid 1860′s-1870′s and mid 1920′s-1930′s occasions I’m not sure why you’ve shaded in the area between the path of the SSB and the solar surface because these are not continuous events. The SSB comes in at a low angle from an outer loop and 10 years later leaves at a low angle, but there isn’t a continuous period where the ssb breaks out of and then re-enters the Sun

    The shading is to show more where the excursion occurs. In those situations the distance from the SSB is not important as the overall path is still toward the SSB (as you suggest), so position of the inner loop excursion is also important. These are weak to very weak disruptions. The main focus is on the inner loop that is trying to be an outer loop for a few years…1650 a great example.

  62. Geoff Sharp says:

    tallbloke says:
    November 15, 2012 at 7:18 am

    The 207-8 year peak in the FFT’s of the proxy data is quite sharp and well defined.

    I think this is because overall the Holocene is filled with mainly weak grand minima, so the 208 gap is the most common because a lot of the “prongs” are weak.

    I have looked back at probably every disturbed orbit for about 5000 years and what becomes clear is that every 172 year average cycle the planet positions evolve in a linear trend. You can see it by comparing the 1st disruption of one epoch (trident) with the following 1st disruptions. But another factor to throw in is the outer planets can be in different positions ie either Neptune or Uranus can lead to get the desired result (waiting for J&S to align), this changes the gap again by actually changing the 172 year cycle…. there are so many permutations that no cycle (trident) would prob look the same. There is a long term 4628 year cycle that is very close but even that goes out of phase after about 3 cycles.

    http://www.landscheidt.info/?q=node/226

    I personally think there are no cycles in grand minima, but at a higher level it may appear that way. Of course this will only apply if my theory is correct.

    I should point out that not every epoch has 3 disruptions or disordered orbits, it can be as high as 5 (generally all very weak) and as now only 2, that is not uncommon. Another part of this is the type A & B disruptions which may be better left for another day.

  63. tallbloke says:

    Thanks for the detailed replies Geoff, I’ll have a think about those after work today. Just a point to make about the overall way it comes across. While it’s fine for us who are into this stuff, and sympathetic to the difficulties, someone like Svalgaard is going to dismiss it as ‘armwaving’ or ‘special pleading’ unless you can find a way to tabulate, analyse and summarise the results from all the ‘AM disturbances’ within the range of interest. I appreciate your comment that ” there are so many permutations that no cycle (trident) would prob look the same.” and understand the reasons for it, but nonetheless, some generally applicable heuristic statements have to be laid out which are not contradicted by more than, say, one event in twenty (95% significance), otherwise it carries no weight for reviewers.

    Cheers – TB

  64. Tim Cullen says:

    Russ Steele has a post that includes a pessimistic view of TSI from:

    Bicentennial Decrease of the Total Solar Irradiance
    Leads to Unbalanced Thermal Budget of the Earth
    Habibullo I. Abdussamatov, Pulkovo Observatory of the RAS

    The diagrams from the Abdussamatov paper are very chilling:

    http://nextgrandminimum.wordpress.com/2012/11/10/russians-scientist-the-next-grand-minimum/

  65. Paul Vaughan says:

    1. Multidecadal Sea Level
    2. Solar Cycle Deceleration — via:
    (a) International Sunspot Numbers
    (b) Ironed Sun Total Solar Irradiance (ISTSI) (Svalgaard)
    3. Sea Surface Temperature
    (a) Pacific Ocean
    (b) Northern Hemisphere

  66. Greg Goodman says:

    TB, what did you conclude about my suggestion that you should be using the centre of gravitational force rather than SSB ?

  67. tallbloke says:

    To be honest Greg, I’ve not had time to think it through. But my good lady assures me I’m of the hook for the rest of this afternoon, so lets discuss it a bit further. Any thoughts on how this could be tested in a spreadsheet with monthly position data? I think you are a bit quicker on the draw with the equations than I am. Presumably its solving a vector addition and translating that back to spherical co-ordinates?

  68. tallbloke says:

    Ray Tomes worked out the relative gravitational effects in the z axis long ago:

    Planet   Mass  Distance  Period  Inclination  Acceler.  Displacement
        M      D         P        I
    
    Mercury  0.056   0.387  0.2408522   3.18       0.021      0.0012
    Venus    0.826   0.723  0.6152078   3.75       0.10       0.039
    Earth    1.012   1.000  1.0000417   7.14       0.13       0.13
    Mars     0.108   1.524  1.880885    5.51       0.0045     0.016
    Jupiter  318.4   5.203   11.86233   6.00       1.228    172.9
    Saturn    95.2   9.538   29.4568    5.45       0.099     86.2
    Uranus    14.6  19.182   84.016     6.36       0.0044    31.1
    Neptune   17.3  30.06   164.802     6.36       0.0021    57.6
    

    So I think we should be able to use these displacement values as a check on whatever we come up with.

  69. Greg Goodman says:

    It seems more a case of getting the right data. As I mapped out previously the first thing you’d need is the position of each planet and the sun in the same reference system (at any one point in time).

    If this is SSB coord system the base will move all the time but as long as each set of readings is in the same base system then the relative distances can be worked out. worry about the trig later.

    For the direction vector of the gravitational force we also need to work out direction from sun to planet. Here it does not matter if SSB moves, as long as reference system does not rotate. This needs to be established. I think the z-axis of SSB system is solar z-axis so that does not complicate things.

    I really don’t work with spreadsheets but the mechanics of doing the numbers should not be a problem.

    So a few prerequisites:

    Firstly is all this available in a fixed inertial coord frame like “relative to fixed stars”? That would be the ideal and save a lot of messing.

    If not, let’s take SSB. Then assuming z-axis is OK we just need to know what determines rotation of XY plane. Is it locked to fixed stars? Not sure what else they could take as ref. but needs to be established. Could be SSB-Sun line . Try looking sun coords in SSB view point. If x or y is always zero, it’s sun line referenced.

  70. Greg Goodman says:

    “Ray Tomes worked out the relative gravitational effects in the z axis long ago:”

    Did he do that for a full timeseries? like that kind of thing every month for centuries ?

  71. tallbloke says:

    Hi Greg, the orbits of the individual planets don’t change much on a timescale of centuries, so the displacement of the Sun due to the motion of a planet relative to the solar equatorial plane can be calculated over a single orbit with sufficient accuracy to apply it to centuries long sunspot obs.

    By assuming a simplest case fixed period internal solar oscillation of 10.5 years Ray got a correlation of ssn to z axis motion of R^2=0.66 over the period of the sunspot record. However, he is of the opinion that the solar ‘memory’ has a ‘half life’ of round 70 years, and that the internal oscillation period will vary due to the longer term effects of the planetary imposed oscillations. That seems right to me.

    If you want to see the various options the JPL Horizons ephemeris gives you go here:

    http://ssd.jpl.nasa.gov/?horizons_doc

    The front end of the application is here

    http://ssd.jpl.nasa.gov/horizons.cgi

    It would be great if you’d have a go yourself, as trying to follow someone else’s prescriptions is likely to lead to misunderstanding and error.

    Ray thinks it is important to do this relative to the solar equatorial plane because the sun is tilted at around 7 degrees to the average plane the planets orbit in, and this means that the upward or downward tug of the planets as they orbit the revolving sun will create a toroidal convection in the near surface plasma. This is because of the differential relativistic effect of the planetary gravitational force on the material at different levels in the Sun, producing a motion of the core relative to the surface built up over the years the outer planets remain above or below the solar equatorial plane. This is in contrast to the cancelling of forces occurring every 25 days as the Sun rotates.

  72. Greg Goodman says:

    TB. ” the orbits of the individual planets don’t change much on a timescale of centuries,”
    agreed, that’s more like 10ka time scales.

    TB “Ray got a correlation of ssn to z axis motion of … ”
    that’s motion of centre of mass then.

    TB “Ray thinks it is important to do this relative to the solar equatorial plane because the sun is tilted at around 7 degrees ”
    agreed, that’s why I was stressing the question of solar z-axis in relation to whatever coord system is used.

    TB “This is because of the differential relativistic effect of the planetary gravitational force”
    Do you mean relativistic in Einsteinian terms or relative, here? Is this accepted theory or Ray’s hypothesis?

    TB “This is in contrast to the cancelling of forces occurring every 25 days as the Sun rotates.”
    There is clearly scope for a modulating effect here.

    Could you clarify what Ray has done on this already?

    I don’t have time to do all this from scratch, I’m looking at the SST that all this will eventually be compared to that is taking most of the time I have available now. I was assuming that someone was already familiar the JPL interface , what it offers and how to get stuff out of it.

    If that is Ray , maybe I should be talking directly to him.

  73. tallbloke says:

    Hi Greg,

    Yes, we are talking Einsteinian relativity and Birkhoff, who applied it to radiative components.

    I put it all together in my own model here:

    http://tallbloke.wordpress.com/2010/01/05/my-simple-solar-planetary-energy-model/

    After developing the solar-ocean energy accumulation model I originated here:

    http://tallbloke.wordpress.com/2010/07/21/nailing-the-solar-activity-global-temperature-divergence-lie/

    I developed a more comprehensive SST model here:

    http://tallbloke.wordpress.com/2012/10/17/phil-jones-we-dont-know-what-natural-variability-is-doing/

    Which awaits a full write-up because I’m so busy running this blog

    If that is Ray , maybe I should be talking directly to him.

    Talk to whoever you like. I see you’ve been commenting on his blog recently anyway.

  74. Greg Goodman says:

    In fact my initial reason for commenting on this was to point out that you should probably be looking at gravity not centre of mass (and the usual confusion of what barycentre represents).

    While I think what you are all doing here is a valuable line of research, my main interest at this time is straightening out the terrestrial climate data which are still badly corrupted by sampling issues and to a significant degree by attempts to “correct” the data.

    I don’t want to get diverted into cycle spotting.

    Other than that I would encourage you to look a gravity rather than mass,which appears to be a conceptual mistake here.

    If you have any specific questions you are more than welcome to contact me.
    best regards. Greg.

  75. Greg Goodman says:

    I put it all together in my own model here:

    http://tallbloke.wordpress.com/2010/01/05/my-simple-solar-planetary-energy-model/

    What’s the “plani-Solar t” line here? Looks like it might match some of the data I’m looking at.

  76. tallbloke says:

    Greg, I do value your input here, thanks for sparing some of your time.

    No-one here thinks the barycentre ’causes’ anything. It’s a ‘shorthand’ for the collective disposition of solar system masses. But remember this: When Jupiter and Saturn are together, the distance from them to the Sun INCREASES rather than DECREASES. So although their collective gravitational pull on the Sun is stronger than when they are opposed to each other, the moments of inertia and the disposition of the masses dictated by the conservation of Angular Momentum actually cause them to be further apart from the Sun. This is counter-intuitive and many people don’t understand it.

    I am interested in your idea of calculating the ‘centre of gravity’, though again, this is a woolly concept, since the sun is an extended body, and its surface layers plasticly deform in ways unlike the incompressible fluid of Earth’s oceans.

    It could be that the effects are predominantly electro-magnetic rather than gravitational anyway, and that the tidal calcs and barycentre-AM-torque calcs are vague proxies for what is actually happening. We don’t know yet.

    “plani-solar-t” is my model temperature derived from the addition of LOD data to SSB-Z data, integrated as a proxy for ocean heat content. The LOD data is itself (I believe) a proxy for the longer term solar system cycles as well as being coincident with the oceanic oscillations.

    Peace – Rog

  77. Greg Goodman says:

    TB “So although their collective gravitational pull on the Sun is stronger than when they are opposed to each other, ”

    That would seem to be obviously wrong, perhaps it does not come over as you intended. Clearly, when the two are in opposition, their gravitational forces will oppose and to a large degree cancel. If you are thinking of tidal bulges, remember that the far-side bulge is centrifugal, not tidal. In opposition these will also cancel.

    TB “… It’s a ‘shorthand’ for the collective disposition of solar system masses.”
    Fine, so what kind of effect will a mass have? In terms of angular momentum it may be a useful quantity. But as soon as you start considering tidal bulges etc , it is not appropriate. So, doing a gravity calc may help you to see whether the dominant effect is gravitational or inertial.

    The two will have some similarity but the relative influence of J,S and J,V,E will change quite a bit. That may explain some deviations in SSB-z idea if, in fact, it is predominantly gravitational. A negative result would also be informative.

    TB “The LOD data is itself (I believe) a proxy for the longer term solar system cycles as well as being coincident with the oceanic oscillations.”

    LOD is an odd one. It seems ridiculous at first to suggest such a minuscule change could change anything but the strong negative correlation to SST forces one to consider it indicates something. But that’s a whole other subject.

  78. tallbloke says:

    Greg,
    Yes, sorry, too late at night. I meant the Sun retreats further to the other side of the barycentre when J+S align.

    I’m uncertain about the tidal situation. On Earth you get a bulge on the opposite side to the Moon, but I think this is due to the incompressibility of water. I doubt solar plasma behaves the same way, though I could be wrong.

    Another factor in your gravity calc idea is that Jupiter’s orbital eccentricity is high, so the gravitational effect on the Sun will be different depending on whether Jupiter is near perihelion of aphelion when Saturn swings by.

    I’ll have a further think on it and see if I can work out how to get a gravity model together. Thanks for your further thoughts.

  79. Greg Goodman says:

    TB : Another factor in your gravity calc idea is that Jupiter’s orbital eccentricity is high, so the gravitational effect on the Sun will be different depending on whether Jupiter is near perihelion of aphelion when Saturn swings by.

    I mentioned eccentricity early on and suggested gravitation variation of ~8% from back of envelop calcs. N may almost disappear on gravity calcs. I should be able to give you some help on trig and vectors if that’s a weak point.

  80. tallbloke says:

    Hi Greg,

    I’ll have a go at creating the ephemeris for the outer 4 planets with monthly data from 1700AD to 2200AD and then we can take it from there. Any extras you’d like to see?

    Cheers – Rog

  81. Greg Goodman says:

    You’ll need the Sun in the same coord system at the same time points (obs point 000 is as good as any if there is not a “fixed stars” or similar static frame).

    Also some way to know how solar z-axis is oriented in all that.

    Two planets plus sol is enough to establish a method. The rest will be simple duplication.

    Venus is likely a major player as well, comparable to J, due to its proximity.

  82. tchannon says:

    Can I help in some way?

    Some time ago I went through at least some of the exercise including gravity calculation. A quick look at what is on disk shows various files but I cannot remember what they are as such.

    Some of this was reproducing other works. I have a vague recollection of reaching no useful conclusions so I moved on.

    I compute object co-ordinates locally here because it is easier, I do not understand the JPL site.

    Available are

    Topocentric (centred on reference system)
    Heliocentric (sun centre)
    Barycentric

    I don’t understand what it is you are trying to compute. Producing time series of those with orthogonal data probably the one you need. (XYZ distance)

    Gravitational force between the sun and the other bodies is merely a lot of work, have the distance and approximate masses.

    A rotate of reference frame obviously can be done, mostly matrix operations. Same stuff as done all the time in graphics work, including video games.

  83. tallbloke says:

    Tim, I’d be really grateful for your input to this. Given distance, direction and mass, we should be able to do a vector addition which will specify the ‘centre of gravity in x,y,z co-ordinates relative to the Sun’s centre. So, heliocentric might be easiest if your local equations are for that frame of reference.

    Also, take a look at the code Greg posted if/when you find time
    Greg Goodman says:
    November 13, 2012 at 11:03 am

    http://tallbloke.wordpress.com/2012/11/12/leif-svalgaard-solar-terrestrial-manifesto-flatter-sun-flatter-earth/#comment-34937

  84. Greg Goodman says:

    Hi Tim,

    what’s topocentric, is it an intertial frame like fixed stars?

    TC: I compute object co-ordinates locally here because it is easier, I do not understand the JPL site.

    Yeah, JPL looked like hard work. What do you mean by you computute locally? From what?

    TC: I don’t understand what it is you are trying to compute.

    Net gravitational force of all planets on the sun as a time series. Magnitude and direction.

    This should account for inclination and eccentricity.

    There seems to be a fair bit being done on SSB (z-axis, speed with repect to , etc.) and some talk of J pulling on V-E induced tidal bulges. These show some degree of correlation but then go totally out of sync.

    My suggestion was that gravity maybe physically what should be considered.

    TC: I have a vague recollection of reaching no useful conclusions so I moved on.

    I would expect the result to be either an interesting correlation to other data or a non correlation. Either would be informative.

    could you explain in a bit more detail what data you have and how it was obtained?

    Thanks for the input.

  85. tchannon says:

    Okay. Will take about 11 minutes computer time, disk light is flashing.

    DE406 (has a wider time applicability than DE421)

    Time is awkward. Start 1/1/1700, output every 10days, AD2100, make that 60,000 data rows. Use Julian date.

    Heloicentric, light aberration and relativistic off (we ain’t looking at the light)
    Include nutation and precession.

    Only bother about major planets.

    Right, header in Jupiter file with first data row.
    Time, XYZ and distance R (computed from XYZ but no vector angle), V is speed)

    The snag? 9.8M for just one of the files.

    Heliocentric Orthogonal Equatorial Coordinates of Jupiter
    True Equinox of Date (Geometric Positions)
    Time is UT
    T X Y Z R Vx Vy Vz Vr
    2341972.50000 0.929460689453 -4.679021794667 -2.038881481112 5.187887799696 1.102023678996 0.246054903148 0.078588988843 -0.055367851253

    Note well
    Before uploading a 31M zip archive are you sure heliocentric is correct given that there is mention of rotating the system plane? IIRC this demands a re-reference to co-ordinate system zero, rotate and re-reference back to where you started. Alternatively some maths fun.

    I suggest that decimation and to the human calender is a last step even though this is an awful large data to handle.

    I’m going to dig anyway to see whether I have already done the task.

    Do I know this stuff, expert? Ho ho, no way, I find this hard. Have to look things up and try to figure a cross check.

  86. tchannon says:

    Greg,
    “Yeah, JPL looked like hard work. What do you mean by you computute locally? From what?”

    Got a good telescope here and it’s a clear night. Tough on those without a time machine.
    Just kidding, got a DOS Basic program.

    Not kidding that last part but is a serious piece of code where modern compiled Basic is apt, not far removed from Fortran so translating existing subr code might be what was done. Think it runs under Wine too, makes sense.
    Solex 11 written by an Italian Chemistry professor with a long side interest. Underlying is ephemeris.
    I have recently been using this to compute solar azimuth and altitude for photograph dating, a tale to be revealed, maybe with a practical how-to

    topocentric

    http://scienceworld.wolfram.com/astronomy/TopocentricCoordinates.html

    Something like this, written by a maths idiot who has visited the scene and run away

    Do the rotation.
    Rereference to solar centre
    Change units from distance to force via mass/distance calculation.
    Sum the xzy as vector addition
    RMS the xyz, pythagoris giving force
    Tear hair out doing atan() to get angles
    Decimate to a sane sample rate and Gregorian calender. (easy part for me, feed into sausage machine I built earlier)

    or accept xzy as the answer

    Recommend scaling first otherwise floats will explode, some very large numbers involved.

  87. tallbloke says:

    Tim: Wouldn’t monthly be quite adequate (or bartels rotation periods)?
    Sounds like the CPU is getting warm over there!

  88. Greg Goodman says:

    Tim , sounds good.

    “The snag? 9.8M for just one of the files.”
    Peanuts, have you seen the size of HadSST3 gridded data ? Better too much resolution than not enough.

    T X Y Z would be enough for this exercise.

    “Before uploading a 31M zip archive are you sure heliocentric is correct given that there is mention of rotating the system plane? ”

    “Decimate to a sane sample rate”
    Last step. Clean, low pass filter before decimation please (hint Gaussian), prevent aliasing. Basic DSP step in any domain.

    Heliocentric probably saves some effort but I’m not clear on what the xy plane is referred to. Is the coord frame spinning at 1/25 per day like the sun??

    If that’s the case 10d intervals will be all over the place. Needs transposing to a more stable frame. Heliocentric 0,0,0 is fine , but needs a stable x=0 line to a star or something.

    Can you define heliocentric in this context?

  89. Greg Goodman says:

    TC: “Recommend scaling first otherwise floats will explode, some very large numbers involved.”

    Any options on that? Output distances in AU ? Not a worry really.

  90. tallbloke says:

    Agree with Greg, needs fixed star referent. Tilt of Sun important in z-axis so work to solar equatorial plane if possible.

  91. Greg Goodman says:

    I’m fine with heliocentric, it will save some messing, as long as that is clearly defined.

  92. tchannon says:

    Zip archive of Solex planetary files.

    12 hours 1st Jan 1700 through 2100 at 10 day sampling. Heliocentric.
    http://tallbloke.files.wordpress.com/2012/11/orb-1700-2100-10d.zip 31MByte

    Let me know if I have missed out one of the planets on copy to archive. Few asteroids etc. are omitted.

    I have found the calculation I did some time ago, which starts 1400AD though 2030 but does not include solar plane rotate. I don’t like the result, needs discussing, perhaps before bothering to put effort into the above archive data. Put another way, is the result errors/artefacts or not?

  93. tchannon says:

    The above, I point out that

    sqrt(x^2+y^2+z^2) = R, distance is the pythagorean calc, same as RMS and of course is related as a domain dual.

    Rotation stuff is fun. Plenty of math about, same as used in image processing such as games. Usually done using matrix math.

    I suspect a problem might be we don’t really know what rotation to use. Hopefully this is not terribly important, enough can be gleaned without.

    For clever folks
    eg. The Mathematics of the 3D Rotation Matrix

    http://www.fastgraph.com/makegames/3Drotation/

  94. Greg Goodman says:

    Thanks for posting the data.

    TC: “I suspect a problem might be we don’t really know what rotation to use.”

    That is what I’ve been pointing to all along. Until you know the coordinate frame you don’t know what the data is.

    “Heliocentric” tells us the sun is at 0,0,0 but does not define the axes. Does the doc with the program tell us what the output actually means?

    The sun is a ball of plasma with different bits rotating at different speeds, so I imagine that the xy plane is anchored to something external.

    Any idea what?

  95. Greg Goodman says:

    Solex 11 doc says: When the DE406 option is selected, the primary output resulting from the integration gives heliocentric equatorial rectangular coordinates referred to the DE406 reference frame of J2000.

    If someone can explain what the means in English, I’ve just wasted half an hour I didn’t intend to spend and I’m no wiser.

  96. tchannon says:

    Welcome to the mire Greg.

    http://www.iers.org/IERS/EN/Service/IERSComponents/components.html

    and

    “The Jet Propulsion Laboratory DE405/LE405 planetary and lunar ephemerides (usually just referred to as DE405) [18] have been aligned to the ICRS. These ephemerides provide the positions and velocities of the nine major planets and the Moon with respect to the solar system barycenter, in rectangular coordinates. The data is represented in Chebyshev series form and Fortran subroutines are provided to read and evaluate the series for any date and time. DE405 spans the years 1600 to 2200; a long version, DE406, spans the years -3000 to +3000 with lower precision.”

    http://www.usno.navy.mil/USNO/astronomical-applications/astronomical-information-center/icrs-narrative

    Caution is needed over ephemeris, vary in validity, a right can of worms. The newer tend to include lots of tweaks which are fine for active space operations but mess up elsewhere. This is all approximation.

  97. Greg Goodman says:

    TC: This is all approximation.

    yeah, well Solex seems pretty good over a millenium or two. That is good enough to see if there any patterns in climate TSI SSN or whatever.

    BUT , I still don’t know what the coordinate system of the proposed data is.

    Seems like no one else does either.

  98. Max™ says:

    The barycenter is the point everything in the solar system orbits around, it is generally found either slightly below or slightly above the surface of the sun.

    en.wikipedia.org/wiki/Barycentric_coordinates_(astronomy)

    Jupiter and the Sun orbit each other, hence why an honest description of the solar system is: The Sun, Jupiter, and some debris.

    [Reply] A nice simple description. Unfortunately, incorrect. The inner planets more closely orbit the Sun, not the system barycentre.

  99. tchannon says:

    It is aligned to the internationl reference frame.

    We don’t know the current solar tilt, awkward. I suspect the reference is the solar system general orbital plane but that is still only one axis.

  100. […] Recently on the Talkshop a discussion has started about Sol and gravity forces. I stepped up in case I can help. I think this is where the recent discussions commenced. […]