Guest post by David Archibald
With respect to the month of minimum, it is very likely that Solar Cycle 24 has started simply because Solar Cycle 23 has run out. Most solar cycles stop producing spots at about nineteen years after solar maximum of the previous cycle. Solar Cycle 23 had its genesis with the magnetic reversal at the Solar Cycle 22 maximum. As the graph above shows, Solar Cycle 23 is now 19 years old. Only 9% of the named solar cycles produced spots after this.
The graph also shows the position of Solar Cycle 24 relative to its month of genesis. Solar Cycle 24 is now the second latest of the 24 named solar cycles. January is 105 months after the Solar Cycle 23 maximum. Only Solar Cycle 5, the first half of the Dalton Minimum, is later. This lateness points to Solar Cycle 24 being very weak.
This graph shows the initial ramp ups of six solar cycles that were preceded by a vey low minimum. The ultimate trajectory of Solar Cycle 24 should be apparent by late 2009. If Solar Cycle 24 is going to be as weak as expected, the monthly sunspot number should remain under 10 by the end of 2009.
Leif Svalgaard (20:20:52) :
to
vukcevic (14:39:56) :
And if you do it from 1964 the correlation jumps to 0.8.
Thanks for the effort. I have done something similar before, breaking whole thing into blocks. With changed free parameters, everything except two basic frequencies, then 1910.5 – 2008.5 gives correlation = 0.8636 or 1922.5 – 2008.5 correlation = 0.8784. For this period there is a strong phase lock with average cycle length well below 11yr.
I have called this “optimum synchronisation period” which produced current ‘grand maximum’.
For scatter plots with the altered parameters’ values see:
http://www.geocities.com/vukcevicu/syncblock.doc
I have done similar exercise for block 1824-1900 but it was far less successful, and abandon whole idea. I think this may be what you call “shoehorning”, so is it proper science?
Changing the amplitude ratio for 2 Cos functions (in above case 1:0.63, raises levels at 1800 and 1910, but introduces extra cycle at each point. On positive side there is no need for phase change at 1810. Either way no good. [Does Excel calculate R and Rsq., and gradient function y=f(x)]
I shall take your advice and see what happens.
Thanks again.
There is no point wasting time looking for regular patterns, as I have said, it depends on variables, and if its variable its not regular. Take a good look at the Sun over 11000 yrs and its painfully obvious…when the conditions are right the Sun sleeps long and hard, but when times are easy we can have many centuries without major slowdown. You are attacking the problem from the wrong end.
http://landscheidt.auditblogs.com/files/2009/01/c14results3.jpg
Leif, the only problem I have with your posts is that it’s often difficult to tell where the quoted portion ends and your response begins 😉 Maybe use blockquote instead of italics?
Jeff Alberts (07:17:02) :
Leif, the only problem I have with your posts is that it’s often difficult to tell where the quoted portion ends and your response begins 😉 Maybe use blockquote instead of italics?
‘blockquote’ is a long and tedious word to type compared to ‘i’
Leif Svalgaard (08:19:39) :
‘blockquote’ is a long and tedious word to type [twice!] compared to ‘i’. And I even screwed it up…
Just want to point out one more time, FFT works.
http://virakkraft.com/SunspotFFT.jpg (there are of course more cycles)
lgl (10:08:19) :
FFT works.
Indeed, but your curves do not line up, regardless.
lgl (10:08:19) :
Just want to point out one more time, FFT works.
http://virakkraft.com/SunspotFFT.jpg (there are of course more cycles)
Thanks Igl. I tried your numbers and got very similar result, with slightly different phase shift since I used 11.86 and 19.86; what are 10 and 11 supposed to represent. I am trying to simulate SSN phase and if possible amplitude by using only the known quantities.
It also suffers from a phase shift (1900-1930).
http://www.geocities.com/vukcevicu/Igl.gif
I have similar problem but at 1810. I suppose this could be (as Dr.S. said) ‘shoehorned’ into 1800, but I am still dubious of introducing 10 an 11 yr components.
Thanks anyway, I’ll keep it in intray, for possible future consideration.
Leif
Did you miss the last sentence? They are not supposed to line up when not all of the cycles are included, and I have not used exact amplitudes. But this is sufficient to assume we are entering a new grand minimum, if not the coming decades behaves totally different from the last 300 years.
lgl (11:45:15) :
Did you miss the last sentence? They are not supposed to line up when not all of the cycles are included, and I have not used exact amplitudes.
If I take all the peaks in the FFT spectrum and use the amplitudes and phases for each, I get something that is very different from the SSN, because the Cosine waves go as negative as they go positive. If one tries to circumvent that by taking the Absolute values (as Vuk) or by squaring the values, the curve now varies twice as fast [5.5 years between max], so one has to use periods [as Vuk] that are twice as long [20, 22, 23.6]. You used the other trick of the book by only plotting the positive part of the sum.
So, the Sun does not add up the Cosine waves. What does it add up?
But this is sufficient to assume we are entering a new grand minimum, if not the coming decades behaves totally different from the last 300 years.
As my little grandson Peter remarked when he saw http://sidc.oma.be/html/wolfaml.html
And, the coming decades will behave different from the last 300 years in the details [I don’t know if one can use ‘totally’ here – what does that mean? That the values match up hour for hour, minute for minute?]
Leif Svalgaard (12:07:15) :
If I take all the peaks in the FFT spectrum and use the amplitudes and phases for each, I get something that is very different from the SSN
Except the DC-component and all the dozens of little peaks below the noise level, of course.
Leif
So, the Sun does not add up the Cosine waves. What does it add up?
Who has ever counted a negative number of sunspots.
Or inverse of the operation you used to get from page 1 to 3 perhaps.
That the values match up hour for hour, minute for minute?]
More like 20 years average with my coarse values.
If I take all the peaks in the FFT spectrum and use the amplitudes and phases for each, I get something that is very different from the SSN
Then something is not done right. If you take the strongest peaks you get the envelope and that’s all you need to determine grand minima.
lgl (13:26:40) :
“So, the Sun does not add up the Cosine waves. What does it add up?
Who has ever counted a negative number of sunspots.
Or inverse of the operation you used to get from page 1 to 3 perhaps.
Explain, makes no sense as it stands. SSN = SUM (wave1,wave2,wave3) or what?
If you take the strongest peaks you get the envelope and that’s all you need to determine grand minima.
If the individual cycles don’t compare, there is no reason the Grand Minima would be right: GIGO.
Do I see a cycle 24 speck? Wonder if they will count this tiny tim but not the cycle 23 tiny tim.
Pamela Gray (17:26:22) :
Do I see a cycle 24 speck? Wonder if they will count this tiny tim but not the cycle 23 tiny tim.
They [NOAA] didn’t count the SC23 speck [didn’t live long enough]. Keep an eye on the SC24 speck.
Pamela Gray (17:26:22) :
Do I see a cycle 24 speck?
They saw it in Brussels:
http://sidc.oma.be/images/last_ORBdrawing.jpg
Leif Svalgaard (20:20:52) :
to
vukcevic (14:39:56) :
And if you do it from 1964 the correlation jumps to 0.8
Have added RSQ calculations to Annual and monthly analyses.
Annual: 1910-2008 RSQ = 0.75, 1922-2008 RSQ = 0.77, 1824-1900 RSQ = 0.60
Monthly 1812-2008 RSQ = 0.55
I think these are good numbers considering nature of the two events compared.
http://www.geocities.com/vukcevicu/syncblock.doc
Dr. Svalgaard
In your article related to polar fields, you commented:
The annual modulation is probably caused by a bunching
up of flux due to the meridional circulation and may be a
proxy of the strength of the circulation.
Is there anywhere else more on this subject ?
Thanks.
vukcevic (04:59:20) :
bunching up of flux due to the meridional circulation and may be a proxy of the strength of the circulation. Is there anywhere else more on this subject ?
Lots more. Here is a good starting point:
http://articles.adsabs.harvard.edu/cgi-bin/nph-iarticle_query?1992ApJ…392..310W&data_type=PDF_HIGH&whole_paper=YES&type=PRINTER&filetype=.pdf
and
http://www.iop.org/EJ/article/0004-637X/580/2/1188/56248.web.pdf?request-id=f4e53d62-d1e3-48dc-aa91-c730bef7a37c
or
http://www.agu.org/journals/gl/gl0005/1999GL010759/1999GL010759.pdf
from which I quote:
To understand these results, we note that the presence of a strong poleward flow, by concentrating the flux in the polar caps, prevents the opposite-polarity elds in the two hemispheres from dif using across the equator and canceling each other. As demonstrated analytically by Sheeley et al. [1989], an equilibrium state is reached in which the poleward convection of flux balances its equatorward dif usion, and there is effectively no further decay of the axisymmetric fi eld. Once established, the polar fields and their open flux can only be removed by switching of the flow (as in our second simulation) or by erupting active regions with reversed axisymmetric dipole moments (as occurred during the rising phase of cycle 23).
one more:
http://articles.adsabs.harvard.edu/cgi-bin/nph-iarticle_query?1989SoPh..119..323S&data_type=PDF_HIGH&whole_paper=YES&type=PRINTER&filetype=.pdf
You’ll get information overload soon.
Leif Svalgaard (09:46:21) :
http://articles.adsabs.harvard.edu/cgi-bin/nph-…………
You’ll get information overload soon.
Thanks for the info. No danger of that, to overload it has to be absorbed first, my cerebral capacity is getting more limited by the day. I am only looking for a very specific information. (since Google Scholar didn’t come up with it).
Leif
What if the Sun can go ‘quieter’ than 0 sun spot count. Negative number of sun spots is non sense so the negative portions of the waves don’t count.
If you do the strongest peaks right you will get close to reproducing the right envelope, like I have showed, simply because there is not enough power left to alter it significantly. I didn’t make any attempt of doing it right, using appr. values and didn’t bother with phase at all.
If you have the tools, why don’t you do a high resolution FFT and then an inverse FFT extending it a few decades into the future.
The fact that there are clear peaks in the FFT proves that there are cyclic components in the solar activity, and the most likely developement is that these will look nearly the same in the near future.
vukcevic (12:05:40) :
I am only looking for a very specific information. (since Google Scholar didn’t come up with it).
Today I was going trough your files with references to polar field. I have failed to find quote I put up on WUWT:
“The annual modulation is probably caused by a bunching
up of flux due to the meridional circulation and may be a
proxy of the strength of the circulation.”
But I came accros elaborate analysis in
Polar Fields and Nobeyama 17 GHz Radio Flux.pdf (work in progress)
which I must have missed before :
“About three years before the minimum the polar fields show a
strong annual modulation (by a factor of two) lasting until the
next reversal. This is due to the field being concentrated near
the pole combined with the annual “tipping” of a pole to and
fro by the 7 degree tilt of the sun’s rotational axis. In March
we see the South polar region best and in September we see
the North polar region.”
This is what I was searching for. This surely must have been considered before ? (analoguos to IHV yearly modulation).
Am I correct that theory for bunching up of flux due to the meridional circulation has been abondoned in favour of the optimum visilibility factor?
I would appreciate your comment, privately or on WUWT, whichever you prefer.
There are several issues with this:
1) What the magnetograph measures [using the Zeeman effect] is the line-of-sight component of the magnetic field [more correctly: the magnetic flux, B, through the aperture of the instrument], in short: the projection [B * cos(angle)] of the field lines onto a line connecting the observer and the place on the sun he is looking at.
2) Imagine that the Sun’s axis did not its 7 degree tilt towards the ecliptic plane where the terrestrial observer sits, and that the polar fields were radial, then a field line would go out from the pole and an angle of 90d with the line-of-sight and not be observable at all [cos 90d = 0]. Tipping the sun just a little bit [e.g. 7d] would create a projected field of B*cos(7d), so as we watch the Sun through a year where the tipping varies from -7 to +7d, we should see a yearly variation of the observed field from the field line sticking out right at the pole. We can’t observe that field though: what we observe is the total projected flux from all the field lines over the whole of the polar cap [because the spatial resolution of the Stanford Magnetograph – WSO – is such that the polar cap down to latitude 55d just fits in the aperture – not a coincidence, BTW]. If the polar field was uniform over the polar cap, there would not be any yearly variation because as the pole tips towards or away from us another field line takes its place. We do observe a yearly variation [by a factor of two]. This means that the polar fields are not uniform over the polar cap but are strongly concentrated near the very pole.
3) so what concentrates the field? it is thought that the meridional circulation does that by sweeping flux from lower latitudes polewards.
So, both effects are there. None has been abandoned for the other. But, I’m not clear as to what your problem here is.
lgl (02:14:47) :
Negative number of sun spots is non sense so the negative portions of the waves don’t count.
Are you saying to removing the negative parts before you add the waves or only from the final sum? Vuk, any comments on that?
Leif Svalgaard (09:29:56) :
vukcevic (12:05:40) :
then a field line would go out from the pole and an angle of 90d with the line-of-sight and not be observable at all [cos 90d = 0]. Tipping the sun just a little bit [e.g. 7d] would create a projected field of B*cos(7d)
should be cos(90-7d), of course.
lgl (02:14:47) :
If you have the tools, why don’t you do a high resolution FFT and then an inverse FFT extending it a few decades into the future.
The fact that there are clear peaks in the FFT proves that there are cyclic components in the solar activity, and the most likely developement is that these will look nearly the same in the near future.
Well, this plot:
http://www.leif.org/research/R%20and%20FFT%20positive%20only.png
shows ‘observed’ sunspot numbers R since 1700 [red] and reconstructed from the seven strongest peaks [with amplitude more than 5 sunspot count] of a high precision FFT using the correct phases [black]. The light blue curves show the individual ‘waves’. All negative values have been set to zero. Since the sum [black] was generally above 30, while R [red] dips to zero, 30 was subtracted from the sum. Some people would say that the red and the back match quite well [although there are problems – e.g. 1775-1795, and small cycles being very short]. Anyway, the black extends to 2050, thus predicting the next few cycles.
Of course, all this machinery is totally unnecessary. In calculating FFT you take a set of values [from 1700 through 2008], and assume that that block of values repeats indefinitely both before and after the block. If you then use the calculated values, it will be no surprise [if you did it right] that the ‘predicted’ values also repeat indefinitely on both sides, so 2009-2050 will be exactly like 1700-1741, as my little grandson remarked when looking at http://sidc.oma.be/html/wolfaml.html