Harald Yndestad: The Lunar nodal cycle influence on the Barents Sea

Posted: November 9, 2015 by tallbloke in Ocean dynamics, solar system dynamics
Tags: , ,

I’m presenting the Doctoral thesis written by Harald Yndestad for his degree as Doctor of Philosophy at the Norwegian University of Science and Technology. This work is highly relevant to our investigation of the effect of Lunar cycles on climatic variation.

The Lunar nodal cycle influence on the Barents Sea
Harald Yndestad
Submitted to Norwegian University of Science and Technology for the degree of Doctor of Philosophy
Department of Industrial Ecology and Technology Management
Norwegian University of Science and Technology


The research for this thesis began in 1996. The purpose was to confirm or reject the hypothesis that the life history of Northeast Arctic cod can be explained as a stationary cycle in a time series. I was rector at Aalesund University College from 1997 to 2000 and my research had to wait. In 2000 and 2001 I developed dynamic models for the most important species in the Barents Sea. The results supported the analysis from my first investigations. The next step was to look for the missing link between the 18.6-year lunar nodal cycle and the identified cycles in the Barents Sea. In 2001 I started to develop new methods to analyze climate indicators. The result was the Arctic Oscillation system theory. Wavelet analysis showed promising results and I started to analyze the biomass time series using the same analysis methods. This opened the possibility of a unified theory to explain the results from all time series.

The research on this subject has not been straightforward, partly because I did not follow the mainstream of published ideas and partly because the idea that the moon has an influence on the climate and biomass fluctuation has not been accepted by most scientists. The major problems have been to reduce complexity, to get access to data that extended over a long period of time, and to understand the complex dynamics. During this research I found that other scientists had studied the same problem earlier in the 19th century. Later, the 18.6-year cycle was estimated using a long time series, but the work failed to explain the reasons that drove this cycle length. This problem made it necessary for me to go back to the basic ideas found in early papers. I studied the early philosophy of science and general systems theory. I developed new analytical methods and new theories to explain these new results. The results were first presented at international conferences for discussion and were later published in international journals. This thesis represents a synthesis of seven papers that describe the different facets of the investigation.

Harald Yndestad

Full text is available here

  1. oldbrew says:

    TB: Dr Yndestad doesn’t mention the full moon cycle as such in his paper.

    The search counter says there are 53 references to the 6.2 year period, obviously 1/3rd of the 18.6 year lunar nodal period.

    The point is there are 33 full moon cycles for every 2 (x 18.6 year) lunar nodal periods.
    So each two LN periods breaks down into six 6.2 year periods like this:
    5.5 – 11 – 16.5 (end of LNP1) – 22 – 27.5 – 33 (end of LNP2)

    Maybe that’s why this breakdown of the cycle into 3 parts is happening?
    Alternate periods will be the inverse of each other in effect as far as FMCs are concerned.

  2. Chaeremon says:

    @oldbrew said … each two LN periods breaks down into …

    Confirming 33 full moon cycles for twice the LN interval, but apsidal and nodal components do advance at different speeds, so IMHO must search for points of alignment between both (e.g. perigee at zero inclination).

    Does Dr Yndestad’s analysis depend on syzygy events?

  3. tallbloke says:

    OK OB, you lost me. If they are 6.2 year periods why are you incrementing by 5.5 years in your timeline?

  4. tchannon says:

    A period around 6.x appears often but almost always fits spurious.

    In the past I have mentioned this period include failing to find a causal. There is more than one candidate. Connecting this to the moon always failed.

    A surprise Roger. Two days ago I started looking at the malagabay problem. Whilst you are unlikely to know this, sea level stand has figured quite often in researches but I rarely write anything. This particular problem is about a specific locality. I took some guesses, looked, plenty appeared, enough for an article, one is partially written.

    During a look around in data a surprise was finding cyclic variation, worrying to me was what it looked like, as in, I know what is similar. Chandler wobble. So I spent some time looking at how these two relate.
    Fun Lissajous figures.
    I am still very surprised at the close commonality. Perhaps there could be a common excitation.
    This does not fit lunar. Guesses from people in the past usually mentioned the inner planet gravitational rate.

    I concluded this is one of those curiosities, far too unsafe to say much, wasn’t even going to mention it. Perhaps if there are more suitable datasets, none I’ve seen, a confirmation match might be there. Even if was that still does not provided the holy grail of causal on anything I’ve found in the past. We need that. There has to be tight phase linkage at a believable phase.

    (Pertinent to the malagbay article? No, too small.)

  5. oldbrew says:

    TB: the 5.5 is full moon cycles not years. 33 FMC = 2 x 18.6 years.

    5.5 FMC = 6.2006 sidereal years.

  6. oldbrew says:

    Yndestad page 53:
    Lunar nodal cycle phase-reversal

    The lunar nodal spectrum has a stationary cycle time but its amplitude and phase are not stationary. In most time series the amplitude of the 4T0=74.4-year cycle introduces a 1.0π (rad) phase-reversal in the harmonic cycles of T0=18.6, T0/3=6.2, 4T0/3=24.8 and 4T0/9=8.3 years. The phase-reversal property introduces a new type of uncertainty in forecasting. The lunar nodal cycles may still be deterministic, but to make long-term forecasts we must know more about the timing of the next phase-reversal.

    The physical process behind this phase-reversal is still unclear.’ [bold added]
    The harmonic cycles of
    T0=18.6, = 16.5 FMC
    T0/3=6.2, = 5.5 FMC
    4T0/3=24.8 = 22 FMC (16.5 + 5.5)
    4T0/9=8.3 years = 22/3 FMC

  7. oldbrew says:

    Wikipedia’s full moon cycle page says:

    A good longer period spans 55 cycles or rather 767 synodic months, which is not only very close to an integer number of synodic and anomalistic months, but also when reckoned in synodic months is close to an integer number of days and an integer number of years:
    767×SM = 822×AM = 22650 days = 55×FC + 2 days = 62 years + 4 days


    So 55 full moon cycles is about 4 days more than 62 years
    One full moon cycle = ~412 days
    412 / 4 (days) = 103
    103 x 62 years = 6386y

    So only one full moon cycle should be ‘lost’ every 6400 years or so, in other words this should be a very reliable cycle.

  8. Chaeremon says:

    @oldbrew said … in other words this should be a very reliable cycle.

    Indeed, take the recent Sept 28 full-moon (perigee + eclipse) as example, 22650.0 days before that was full-moon in also perigee. But this is not aligned with the nodes, for that you have to go 3 more years (1/2 of the 6 years apsidal/nodical wobble) into the past (eclipse @ Sept 26, 1950).

  9. tallbloke says:

    OB: So only one full moon cycle should be ‘lost’ every 6400 years or so, in other words this should be a very reliable cycle.

    I think this comment would be more relevant on your lunar cycles post, where 6400 or so isn’t all that far from the 6441yr period you’ve been working with.

  10. rishrac says:

    Just to add my 2 cents, in 2005, the chandler wobble underwent 180 degree phase shift. ( as the standard its global warming that’s doing it .. sarc) And has done in the past. I think this is very complex problem. There is another issue that is out there, and that is the distance spacecraft loose on long distance voyages. It’s not much, but they aren’t exactly were they should be. I’m sure (fib on accident ,fibonacci number, auto correct, )would be in there. The distance traveled by the moon minus whatever the slowdown maybe related over 6400 years.
    If you want to see something really interesting, throw a top. I thought it was interesting.

  11. Ian Wilson says:


    Here is something that I have noticed that is extraordinary. I call it the 372 year weave cycle. The extraordinary feature starts at 93 tropical years and it shows hour precisely the Synodic, anomalistic and Draconic cycles weave about a precise 31/62/93/186 tropical year lunar tidal pattern before converging at 372 tropical years:

    Mean Parameters used (J2000):

    Synodic month = 29.5305889 days
    anomalistic month = 27.554550 days
    Draconic month = 27.212221 days
    Tropical year = 365.242189 days

    31 Tropical years

    383.5 Synodic months = 11,324.980843 days = 31.006771 tropical year (error + 2.47 days)
    411.0 anomalistic months = 11,324.920050 days = 31.006604 tropical years (error + 2.41 days)
    416.0 Draconic months = 11,320.283936 days = 30.993911 tropical years (error – 2.22 days)

    N.B. 2.3642837 days is 1/12th of the average of Synodic and Draconic months.

    93 Tropical years – out by a quarter of the average of a Synodic month and a Draconic month

    1150.25 Synodic months = 33,967.559882 days = 93.000099 tropical years (err. 0.971 hours)
    1232.75 anomalistic months = 33,967.871513 days = 93.000953 tropical years (err. 8.35 hours)
    1248.25 Draconic months = 33,967.654863 days = 93.000359 tropical years (err. 3.15 hours)

    186 Tropical years – out by a half of the average of a Synodic month and a Draconic month

    2300.50 Synodic months = 67,935.119764 days = 186.000199 tropical years (err. 1.74 hours)
    2465.50 anomalistic months = 67,935.743025 days = 186.001905 tropical years (err. 16.70 hours)
    2496.50 Draconic months = 67,935.309727 days = 186.000719 tropical years (err. 6.30 hours)

    372.0 Tropical years – realigned

    4601.0 Synodic months = 135,870.239529 days = 372.000398 tropical years (err. 3.49 hours)
    4931.0 anomalistic months = 135,871.48605 days = 372.003810 tropical years (err. 1.39 days)
    4993.0 Draconic months = 135,870.619453 days = 372.001438 tropical years (err. 12.60 hours)

    This cycle is way better than the 18.03 year Saros cycle in realigning with the seasonal calendar.

  12. Ian Wilson says:

    Of course,

    372 x 3 = 1,116 tropical years (error for the Synodic and Draconic months still less than 1.5 days)

    1,116 / 18 = 62 tropical years or

    62 x 18.0 tropical years = 1,116 years

    which aligns to a much greater precision with the seasonal calendar than the 18.0 year Saros cycle.

  13. oldbrew says:

    IW says: ‘ I call it the 372 year weave cycle.’

    That’s 5 of Yndestad’s ~74.4 year ‘cycles’ if that’s what they are.

    ’62 x 18.0 tropical years = 1,116 years’ or 60 x 18.6 = nodal cycle

    Yndestad paper: ‘The gravity energy from the Moon introduces an 18.6-year lunar nodal tide in the Atlantic Ocean and the energy introduces an 18.6-year wobbling of the Earth axis (Pugh, 1996).’

    So there will be 31 of the ~6 year lunar wobbles every 186 years.
    That would be 157 Chandler wobbles of about 432.7 days.

  14. oldbrew says:

    ‘aligns to a much greater precision with the seasonal calendar than the 18.0 year Saros cycle’

    Of course the Saros has its own alignment:
    1 Saros = 223 Synodic months = 241.99867 (~242) Draconic months = 238.99215 (~239) Anomalistic months = 15.992 full moon cycles (~16) = 18.99867 Draconic years (~19)

  15. oldbrew says:

    The ratio of Scafetta’s 115 year cycle* to the 186 year cycle is around 1:Phi.
    (*See: https://tallbloke.wordpress.com/2012/03/21/nicola-scafetta-major-new-paper-on-solar-planetary-theory/)

    In 115 years there are about 13 lunar apsidal cycles, and 102 full moon cycles (34 x 3). [13+102=115]
    In 186 years there are about 21 lunar apsidal cycles, and 165 full moon cycles (55 x 3). [21+165=186]

    3,13,21,34 and 55 are all Fibonacci numbers, therefore the ratio of the consecutive pairs (21:13 and 55:34) is close to Phi.

    Some papers refer to a 230 year cycle i.e. double the 115 year period. That would then match up with Ian Wilson’s 372 year ‘weave’ – see above – which is 2 x 186y to give the same Phi ratio.

  16. tallbloke says:

    OB: That all fits together rather nicely!

    Good work.

  17. rishrac says:

    In the ancient Chinese almanac, it’s either 400 years or 396 years that the cycle repeats. Over 5000 years I’m sure they’ve see it before. Note that the years come out pretty close. They also had a cycle that was 28,000 days. (About 60 years). Of course history didn’t exist before 1979.

  18. oldmanK says:

    rishrac says “Over 5000 years I’m sure they’ve see it before.” No sir. 5000 years includes the 2345bce Dodwell datum. The Chinese, and all others, would have then seen more than that.

    Mathematically one can extrapolate a stable dynamic system far back in time (as Milankovitch asserted) but it does not mean nature will then have to oblige. And there is enough evidence that it does not.

    Still the exercise is interesting and for short periods, applicable, and is of utility for civilian and military use.

  19. oldbrew says:

    rishrac says: ‘They also had a cycle that was 28,000 days.’

    That’s a third of 230 years (not 60 years). 68 full moon cycles = 28001.34 days.
    [see comment: November 16, 2015 at 5:36 pm]