Sun's magnetics coming alive again

vukcevic says:
May 9, 2010 at 11:39 am
That is a ‘much ado about nothing’ ; in some older files there is typo error of pi/3 instead of 2pi/3.
VukGate continues:
So, the sunspot curve has 2pi/3, while the polar field curve uses pi/3 [at latest count]. And since only one of the cosine terms has this offset, there will be a continuously changing phase between the two, belying your claim:

vukcevic says:
May 9, 2010 at 9:48 am
The PF formula in essence, is the same as the sunspot formula with 3-4 year advance, to account for delay of magnetic polarity change between SS and PF .

And

vukcevic says:
May 9, 2010 at 3:01 pm
No need for recalculation, it is the labelling which needs to be adapted.

Well, taking your word for it, we get:
http://www.leif.org/research/Vuk-polar-fields-9.png
Hmmm, it is worse than we thought…
Because:

vukcevic says:
May 9, 2010 at 9:48 am
the Hale Cycle polarity reversals are shown clearly here
http://www.vukcevic.talktalk.net/LFC1.htm

It seems that some handy touch-up work was done with a reversal of the sign in 1921, to make it look like matching the Hale polarities. This shall henceforth be known as “Vuk’s trick”
I would like you to read and understand [and prove that you understood by stating it here in your own words] the 22-year cycle in geomagnetic activity, and how it tells you the polarity of the polar fields to convince yourself that there was no phase reversal of the polar fields at least since the 1840s.
If you refuse to do that, you have just shown yourself to use ostrich-science [“don’t want to know”].

Leif Svalgaard says:
May 9, 2010 at 5:24 pm
Wow, that has jumped up some, its almost at 75%!
Leif Svalgaard says:
May 9, 2010 at 5:35 pm
When in Rome……………………………
pseudo-scientists
[but you have a learning disability, it seems].
the impending falsification of your prediction and the ensuing abandonment of your ideas.

vukcevic says:
May 9, 2010 at 3:01 pm
Green looks pretty good to me, far better then anything Svalgaard, Hathaway & co. can come up with.
In my previous post your sloppiness was contagious and I wrongly stated that there would be a varying phase shift. This is not the case, the difference between 2pi/3 and pi/3 is a slight amplitude modulation and only a phase problem for a minor part of the curve. I think that in all your 2pi/3 to pi/3 machinations you got the sunspot curve offset wrong. It should also be pi/3 [then you have one less free parameter] and then you should do the phase difference between SSN and PF on the time offset 194x.x. Anyway, taking your SSN formula [removing the spurious sign reversal in 1921], your SSN curve [shifted earlier as precursor] and PF curve match very well http://www.leif.org/research/Vuk-polar-fields-10.png . In fact, SSN = 0.6444 PF [or DM as I call it]. This is the Svalgaard/Cliver/Kamide relationship you say is so bad. From our paper http://www.leif.org/research/Cycle%2024%20Smallest%20100%20years.pdf :
“Assuming that Rmax = 0 when DM = 0, we fit a straight line through the origin to the two data points for cycle 22 and cycle 23: Rmax = 0.6286 DM (in mTesla) and compute Rmax from this regression line for cycles 22, 23 and 24″ This ‘fit’ is just for calibration and not for statistical testing, as a relationship is taken as given from theory [meaning, that if the theory is right, then this would be the calibration, which we cannot otherwise compute because the sunspot number is arbitrarily defined].
This is, of course, not a surprise, because the same data [cycles 22 and 23] go into both correlations [yours and ours]. It is not difficult to curve fit over a short interval. One can fit almost anything [US postage cost, e.g.]. The issue comes when the fit is used to forecast [or hindcast] outside of the domain where it is defined. That is where the physics comes in. Your fit fails completely on cycle 20, for example, and requires a sign reversal neat year 1900, which we know from geomagnetism did not happen.
Now, you can recover a bit from the wrong sign disaster, by saying: “OK forget about the sign and use only the absolute value” [as you once claimed one had to do to avoid negative sunspot numbers]. After all, in our paper we only use the magnitude and not the sign. That results in this correlation http://www.leif.org/research/Vuk-polar-fields-11.png with an R^2 below 0.6 [as I found for the differences correlation]. And then the correlation is not REALLY good by any measure.

Leif Svalgaard : May 9, 2010 at 9:17 pm
…..
To look at correlation for the rectified values may be wrong.
As I understand formula r = [sum(i =1 to n) of (xi-x mean)(yi-y mean)]/[(n-1)(dev x*dev y)]
since mean values and standard deviation are altered by rectification process.
Therefore one has to stick to signed values.
The unidirectional meridional flow analysis achieves polarity change. Gravity change and possible magnetic feedback from the magnetospheres also would be a positive only (unipolar) value functions, so Hale cycle could be explained in terms of the meridional flow.

Leif Svalgaard says:
May 9, 2010 at 7:41 pm
After you have examined my findings in detail, It will be apparent to you, as to why I have clearly no intention of abandoning Planetary Ordered Solar Theory, regarding short term Global temperature change, or the cause and nature of the Hale cycle. And good luck with your prediction!

Polarity switch can be achieved by adding the third factor to the equation, but it isn’t such a good fit (SS undulation) and it is a bit messy.
Q.
A. I am still looking for my daughter’s old compass.
My Q: What about BL-Vuk hypothesis?

The return of Murphy`s winter, 1837/8, will be from late January/February 2017, earlier in January will be milder. {+179.05=112 Venus synodic periods}