The sun seems not to be in cooperative mood again this month. It has gone blank again.
And from SWPC, the brief upticks of April were not repeated in May:
More info at the WUWT solar page
The sun seems not to be in cooperative mood again this month. It has gone blank again.
And from SWPC, the brief upticks of April were not repeated in May:
More info at the WUWT solar page
Watch out for a head long rush to impose green taxes before the climate cooling implications become obvious and the news gets out to the general public. At least this new tax revenue will be warm comfort in the face of headwinds from unsteady control of the official AGW message. The whispers will forewarn politicos that something is up and they better make their final push.
= = =
Leif,
Thanks for your very comprehensive reply. It is sincerely appreciated.
It might be awhile before I can conceptualize and visualize “the convective collapse of magnetic fibrils”.
Quite challenging task for me . . . . but I will work on it.
John
Leif
And the correlations are not good at all. If they were, there would be no discussion about this.
If the correlations were not good, there would be no discussion about this.
http://virakkraft.com/EMB-AM.png from Semi
1. Coincidence
2. Planetary motions drives solar cycle
3. Solar cycle determines planetary positions.
4. Something drives both solar cycle and planetary positions.
Which one?
Leif Svalgaard says:
June 15, 2011 at 7:35 am
So, you think it is better to force all that software [some has even turned into non-changeable firmware] to be changed.
If you want to call it a sunspot count, yes. Better to call it Svalgaards geomag reconstruction of solar magnetism otherwise.
tallbloke says:
June 14, 2011 at 10:29 pm
shows that something non-linear happens when the Sun goes into a major minimum. Recent measurements such as those made by Livingstone and Penn and your radically changing F10.7 flux to sunspot number ratio confirm this.
First, the Dalton was not a major minimum [it was on par with 1900 where your curve does go down. And by that argument the coming minima should also show a gross disagreement. Bottom line: the fit is not so good
And then further down your reply you admit there is no good data for cycles 5 and 6. i.e. the Dalton Minimum. So your assertion that Dalton was like 1900 is, well, just an assertion. Secondly, we will have to wait and see if there is a major excursion from the curve, but I note that the curve is on the downswing anyway, so it may not be so obvious near the start of the downturn.
On the contrary, my correllation improves further with the removal of the Waldmeier step and your increases to the sunspot record prior to 1825.
My New Numbers show the mid 1800s to be on par with second half of 20th, but your curves show different Bz. Bottom line: fit not so good.
Your new numbers are your new numbers. I’ll stick with SIDC and a Waldmeier correction for now thanks.
@lgl has to be 2 or 4, but don’t expect a coherent answer from his nibs. He’d rather eat his own eyeballs than look at these correlations properly.
lgl says:
June 15, 2011 at 1:03 pm
Which one?
Extend is back to 1600 AD and ask again.
tallbloke says:
June 15, 2011 at 1:15 pm
If you want to call it a sunspot count, yes.
The programs won’t change because to change what you call the thing they work with. You have to supply a number to the program. and that number better not be on a changed scale.
And then further down your reply you admit there is no good data for cycles 5 and 6. i.e. the Dalton Minimum.
The fit must to the data we have, good or bad.
Your new numbers are your new numbers. I’ll stick with SIDC and a Waldmeier correction for now thanks.
But did you apply the Waldmeier correction for your plot. I don’t think so from the looks of it. And you become a victim of Leif’s law: the data that fit must be the better ones.
But did you apply the Waldmeier correction for your plot. I don’t think so from the looks of it. And you become a victim of Leif’s law: the data that fit must be the better ones.
Anyone looking at the plot can see that once the Waldmeier correction has been made, the fit will improve. Your upward adjustments to the earlier record are much more than 20% and so until I get a satisfactory explanation I’ll adjust his count down rather then use your ‘new numbers’ to adjust everything else up. Incidentally, once that is done, the divergence downwards from ~1993 will become obvious too, showing the descent into the anomalous minimum your colleague’s ‘three lines of evidence’ indicate. I created the graph before the waldmeier correction was known to me, so there is no application of ‘Leif’s law’ here, and your attempted slur fails.
Regarding Waldmeier, are you sure his numbers need a 20% downward correction across the board? Is it possible he made less of an overcount when there were less spots to overcount during cycle 20?
tallbloke says:
June 15, 2011 at 9:58 pm
Anyone looking at the plot can see that once the Waldmeier correction has been made, the fit will improve.
No, if you correct SSN down by 20% for 1950-2000, you make the fit worse. Right now you split the discrepancies with some values higher and some lower. If you adjust SSN down the values all fall below your blue curve, so wore fit. You can repair that somewhat by adjusting your blue curve down too, but then you make the fit worse for the earlier years. I don’t fancy reading off your values. you have the data, compute a goodness of fit [e.g. the RMS value of the discrepancies or the correlation coefficient. Then make the Waldmeier adjust and do it again. Report the results.
Your upward adjustments to the earlier record are much more than 20% and so until I get a satisfactory explanation I’ll adjust his count down rather then use your numbers to adjust everything else up.
As far as the fit is concerned it makes absolutely no difference if you multiply all SSN before 1945 by 1.2 or divide all SSNs after 1945 by 1.2.
The earlier adjustments are indeed much larger than 20%. This is the result of many years of analysis, starting with: http://www.leif.org/research/CAWSES%20-%20Sunspots.pdf then http://www.leif.org/research/AGU%20Spring%202007%20SH54B-02.pdf then http://www.leif.org/research/SSN%20Validation-Reconstruction%20%28Cliver%29.pdf then
http://www.leif.org/research/SH13A-1109-F2007.pdf then
http://www.leif.org/research/AGU%20Spring%202008%20SP23A-07.pdf then
http://www.leif.org/research/Napa%20Solar%20Cycle%2024.pdf then
http://www.leif.org/research/SPD-2009.pdf then
http://www.leif.org/research/Rudolf%20Wolf%20Was%20Right.pdf then
http://www.leif.org/research/SIDC-Seminar-14Sept.pdf then
http://www.leif.org/research/SIDC-Seminar-12Jan.pdf and finally our talk at IUGG [which you’ll have in a few weeks]. Here is the first and last pages:
http://www.leif.org/research/IUGG-preview.pdf
As you can see there has been some evolution, but now we are close to the finish line and ready for formal publication. Note that Ken Schatten [creator of Group Sunspot Number is a coauthor and agrees with our adjustments]. We are having a workshop at Sunspot, NM, in September to get all this straightened up. All the producers of sunspots numbers [e.g. SIDC, NOAA] agree that correction is needed and are sending representatives. So expect something to come from the workshop.
Regarding Waldmeier, are you sure his numbers need a 20% downward correction across the board? Is it possible he made less of an overcount when there were less spots to overcount during cycle 20?
Starting in 1970 his count may be a tad [a few percent] too small which is compensated by SIDC after 1980 being a tad too high, until about 2000 where SIDC is undercounting some 12% until to day. All these fine adjustments are so small that they hardly matter much for the big picture. There are similar periods in the far past where the calibration also changes up and down by a few percent; again not important in the big picture.
Leif Svalgaard says:
June 16, 2011 at 12:26 am
you have the data, compute a goodness of fit [e.g. the RMS value of the discrepancies or the correlation coefficient. Then make the Waldmeier adjust and do it again. Report the results.
I’ll try to make the time to do that.
Leif
When there are sun spots they match. I know you have some solar activity data for the period without sun spots but I can’t find it, link please. Anyway, if they don’t match 1650-1700 you know what I am going to say. http://virakkraft.com/EMB-AM-1600.png ( I think the red is some unreliable C14 stuff)
lgl says:
June 16, 2011 at 5:42 am
When there are sun spots they match. I know you have some solar activity data for the period without sun spots but I can’t find it, link please. Anyway, if they don’t match 1650-1700 you know what I am going to say. http://virakkraft.com/EMB-AM-1600.png ( I think the red is some unreliable C14 stuff)
A main problem is that there is no justification for using a signed sunspot curve [except to improve the fit]. The next problem is to account for the times with no visible sunspots, or more generally for the varying amplitude of the solar cycles. This is separate from the problem that there is no transfer of orbital angular momentum to/from the Sun’s rotational angular momentum [as Shirley points out in http://www.leif.org/research/Spin-Orbit-Coupling-Shirley-JPL.pdf ]. So, I would ascribe whatever match you eyeball to coincidence.
tallbloke says:
June 16, 2011 at 5:05 am
I’ll try to make the time to do that.
You can also directly use my best guess of the ‘New Sunspot Number’
http://www.leif.org/research/The-New-Sunspot-Series.xml
Use column L
lgl says:
June 16, 2011 at 5:42 am
I know you have some solar activity data for the period without sun spots but I can’t find it, link please.
Here is McCracken and Beer’s 10Be based data: http://www.leif.org/research/McCracken-HMF.xml
The data has been converted to magnetic field in the Heliosphere near the Earth [this should not matter as far as the relative wiggles are concerned]. We believe that they have a calibration error around 1948 so give both series.
Here is McCracken and Beer’s 10Be based data: http://www.leif.org/research/McCracken-HMF.xml
Sorry: http://www.leif.org/research/McCracken-HMF.xls
Thanks Leif, I’ll dig out the relevant files and give it a go. Who knows, we might have a better agreement than you thought. I hope so.
tallbloke says:
June 16, 2011 at 11:54 am
Thanks Leif, I’ll dig out the relevant files and give it a go. Who knows, we might have a better agreement than you thought. I hope so.
If the data fits better, the data must be good 🙂 Leif’s law in action.
Leif
Thanks for the data.
Good to see none of the problems you listed are real.
The ‘real’ solar cycle is 22 years so that’s the one to use.
There are more planets, which can cause amplitude and phase variation.
No AM transfer is something you can’t prove (and I can’t prove there is 🙂
lgl says:
June 17, 2011 at 12:14 pm
The ‘real’ solar cycle is 22 years so that’s the one to use.
No, that is a myth. Each cycle is a unit in itself. That the polarities change between cycles does not connect the two cycles and make them into ‘one’
No AM transfer is something you can’t prove (and I can’t prove there is 🙂
According to the known laws of physics there is no such transfer. Shirley [who is a strong believer in planetary influence] has shown that so clearly: http://www.leif.org/research/Spin-Orbit-Coupling-Shirley-JPL.pdf
Leif
No, that is a myth.
Strange. Magnetic fields are caused by something moving relative to something else. Part1 of the Sun moves faster than part2 for 11 years, then part2 moves faster than part1 for 11 years. That’s one 22 year cycle. http://www.solarstation.ru/TL/PDF/tl_22.pdf
lgl says:
June 18, 2011 at 11:07 am
Strange. Magnetic fields are caused by something moving relative to something else. Part1 of the Sun moves faster than part2 for 11 years, then part2 moves faster than part1 for 11 years. That’s one 22 year cycle. http://www.solarstation.ru/TL/PDF/tl_22.pdf
That is not the way it works. It is the other way around: the magnetic field from the sunspots are slowing down the rotation, e.g. http://www.leif.org/research/ast10867.pdf
The paper you cite has this conclusion: “Thereby, in our opinion, the solar magnetic cycle may be generated due to the interaction between torsion waves and the relic field inside the Sun, rather than to the dynamo mechanism.”
This is very speculative and considered to be ‘fringe’ science by most solar physicists.
Leif
Where in your paper is the evidence for “the sunspots are slowing down the rotation” ?
lgl says:
June 18, 2011 at 12:56 pm
Where in your paper is the evidence for “the sunspots are slowing down the rotation” ?
There are good theoretical reasons for that. Our paper is an observational test of the theories and confirms the general conclusion:
The interaction between differential rotation and magnetic fields in the solar convection zone was discussed by Brun (2004), who describes the three-dimensional numerical simulations of compressible convection under the influence of rotation and magnetic fields in spherical shells. One consequence of the model is that the Maxwell stresses can oppose the Reynolds stresses and in this way contribute to the transport of the angular momentum towards the solar poles. This leads to a reduced differential rotation. So, when magnetic fields are weaker, one can expect a more pronounced differential rotation yielding a higher rotation velocity at low latitudes on the average. The Maxwell stresses are associated with the correlations of the fluctuating magnetic field components which arise from tilts and twists within magnetic structures (Brun, Miesch & Toomre 2004). They originate from the reaction on the flow through the Lorentz forces (e.g. Sturrock 1994; Parker 1996; Mestel 2003).”
Figure 2 is the important one.
Our paper ends with: “A dependence of the solar rotation velocity measured by magnetic tracers and solar activity and interplanetary magnetic field was found. An interplay between the Reynolds and the Maxwell stresses is proposed for the interpretation. As stated by R¨udiger & Hollerbach (2004), the more magnetic the Sun is, more rigid is its rotation”
Leif
Like I suspected, just speculations. There are “observational tests” of planetary influence too, like http://eprints.usq.edu.au/4795/1/Wilson_Carter_Waite_Author%27s_version.pdf but it doesn’t prove anything.
lgl says:
June 19, 2011 at 3:50 am
Like I suspected, just speculations.
The difference is that there is real physics behind it. Physics that explains how it works and why it works.
After you have defined what is real physics.