Leif Svalgaard writes in comments:
We plan to submit tomorrow to JGR the following…
http://www.leif.org/research/IDV09.pdf (preprint)
…showing the run of the heliospheric magnetic field since 1835 [not a typo]. I plan to discuss the whole peer-review process here on WUWT, complete with nasty comments by the reviewers and our responses. This will be an illustration of the peer-review process as it unfolds. Should be interesting.
I’ll say. I’ve taken some of the most interesting graphics and put them up for WUWT readers, along with the abstract.
IDV09 and Heliospheric Magnetic field 1835-2009
Leif Svalgaard1 and Edward W. Cliver2
Stanford University, HEPL, Cedar Hall, Via Ortega, Stanford, CA 94305-4085
Space Vehicles Directorate, Air Force Research Laboratory, Hanscom AFB, MA 01731-3010
Abstract.
We use recently acquired archival data to substantiate and extend the IDV index of long-term geomagnetic activity, particularly for years from 1872-1902 for which the initial version of the index (IDV05) was based on observations from very few stations. The new IDV series (IDV09) now includes the interval from 1835-2009, vs. 1872-2004 for IDV05. The HMF strength derived from IDV09 agrees closely with that based on IDV05 over the period of overlap. Comparison of the IDV09-based HMF strength with other recent reconstructions of solar wind B yields a strong consensus between the series based on geomagnetic data, but significant lack of support for a series based on the 10Be cosmic ray radionuclide.
The reconstructed data in the graphic below, from the paper, is quite interesting. Currently, we appear to be at the lowest point in the record.
Click for larger images.
Here’s the comparison with the Be10 isotope record:
dear god people.
“i” before “e” except after “c” or when sounded like “a” as in neighbor and weigh.. err and leif.
Leif Svalgaard (23:13:13) :
A wave can be a one-time thing [e.g. a tsunami], a cycle cannot as it has a continuance. But the other aspect is more important, for a wave there must be something [some physical object] that does the waving. Cycles don’t need that [e.g. my example of a graph of the distance between Mars and Jupiter].
On cycle –wave discussion here’s my ‘topence’ :
Cycle oscillates, wave propagates.
In graph shown here blue line is a cycle but red line is not a wave.
http://www.vukcevic.talktalk.net/SSN_amplitude.gif
Why? Because red line is static, to be a wave, subsequent cycles have to be subject to wave’s action in ‘equal measure’ for a period of time.
The HMF “wave” is definitely exciting, its is more prevalent than the sunspot wave. Whatever you want to call it, it is a predictable gentle up and down flow of solar modulation that repeats over a longer period than 100 years (172 years).
This will challenge the CORE of the Babcock-Leighton theory…but it can still exist without the “random number generator” component.
savethesharks (23:38:45) :
Leif Svalgaard (23:13:13) : “…for a wave there must be something [some physical object] that does the waving.”
Uh huh….and the complex physical processes that “do the waving” in the case of the Milankovitch cycles.
Ice age on. Ice age off.
Something is doing the waving.
Maybe not a “physical object”, but a physical process.
Chris, the ice ages are an effect of the Milankovitch cycles (probably) not the cause.
There is a scientist proposing that the sun itself has a variability which can fall into different cycles at around 100,000 and ~41,000 years, among others. I have been investigating the harmonic resonant periodicities in the solar sytem, particularly the planetary interactions. I have also been investigating numerical ‘coincidences between various planetary attributes such as length of precessionary periods, surface temperature, angular momentum exchanges and spin rates. Some interesting things are popping up.
Leif calls this ‘numerology’ and fair enough, I have no physical explanations as yet. I still think it’s something worth working on though, because I suspect that the sun’s magnetic field and solar wind have a long term influence on planetary dynamics, and that the planetary dynamics in turn feed back to affect solar activity.
If this is the case, then the interacting cyclicities of the sun and planets and the resonant harmonies created by them will affect solar activity and planetary surface temperature.
tallbloke (00:51:53)
You are perhaps referring to this:
http://www.solstation.com/stars/sol.htm [some 6 clicks down]:
<>
If this is correct: What triggers these magnetic instabilities that also governs our climate?
Leif Svalgaard
writes in comments: We plan to submit tomorrow to JGR the following…
GREAT WORK Dr. Svalgaard.
I hope you would not mind if I add few remarks.
Page 9 line 206 – graph for Heliospheric Magnetic Field Strength B (at Earth) inferred from IDV and Observed
Shows faint green line to which you refer to as:
Page 10 lines 208-210:
Figure 7. Yearly average values of the HMF B inferred from the IDV-index (blue curve) compared with in situ measurements (red curve). There is a hint of the ~100 year Gleissberg cycle.
– ‘~100 year Gleissberg cycle’ green line starts at 1835. Since you have not plotted values before 1835, I shall assume they are unknown, and you suggest for period ~ 100 years, than initial ½ cycle (50 years) of this line should be omitted. You suggest cycle of ~100 year, so if you are plotting the line, than trend 1835-1870 should be similar to one1935-1970 i.e. rising trend, not falling as shown here. If values for the IDV were following SS numbers than you would have rising trend, since the cycles prior to1835 were lower.
– You characterize you line as a ‘hint of the ~100 year Gleissberg cycle’.
In many authoritative papers the Galesburg cycle is defined as 75-85 years and not 100 as above, so if it is the Gleissberg cycle, than it should conform to its widely accepted definition. Dr. Landscheidt described a cycle of about 175 years, which is twice the accepted value for the Galesburg’s.
I took liberty to reproduce your graph using your data for compression purposes:
http://www.vukcevic.talktalk.net/IDV.pdf
Upper graph is a direct copy from your paper, the lower one is a reproduction using your data. Instead of arbitrarily defined Gleissberg cycle you have used, I have reproduced formula associated with the SSN amplitude (shown as green line), normalized to your graph’s values. It can be observed that so formulated line from 1860-2009 follows precisely values you termed as the ‘Galesburg cycle’ which of course this is not it, since there is no mathematical definition of the Gleissberg cycle, only a vague period value of 75-85 or so years. Here, in my graph, due to the known SSN values of pre-1835 cycles, the trend for 1835 – 1870 follows its natural progression indicating a period of 107 years.
Details for ‘Vukcevic’ amplitude cycle can be found here (pages 2&3).
http://xxx.lanl.gov/ftp/astro-ph/papers/0401/0401107.pdf
In the graph I have referred to, the amplitude is normalised to Svalgaard et al graph. To reproduce with Svalgaard’s data use MS Excel entry :
= 4.65+0.7*(2+COS(3*PI()/2+2*PI()*(A1-1940.5)/118.628)+0.5*COS(2*PI()*(A1-1940.5)/287.6))
tallbloke (23:06:09) :
And you are told the green curve is the sunspot number.
No, you were told that the green curve was B determined from the sunspot number.
that your formula 4.3+0.3*SQT(Rz) overestimates B for low amplitude cycles and underestimates B for high amplitude cycles. This indicates to me that the relationship is non-linear, even at the second order, and that the formula you present is not based on anything physical, but is an arbitrary ‘best fit’ with your B curve.
No, the formula was determined in http://www.leif.org/research/The%20IDV%20index%20-%20its%20derivation%20and%20use.pdf [section 7; Figure 8] updated with IDV09. There is a good physical reason for the formula:
The main sources of the equatorial components of the Sun’s large-scale magnetic field are large active regions. If these active regions emerge at random longitudes, their net equatorial dipole moment will scale as the square root of their number. Thus their contribution to the average IMF strength will tend to increase as the square root of Rz. (for a detailed discussion, see Wang and Sheeley [2003] and Wang et al. [2005]). We find, indeed, that there is a linear relation between B and the square root of the Rz as shown in Figure 8.”
which agreed with Lockwood’s B amplitude for cycle 14 and the intervening cycles to cycle 19 just as well as yours.
No need to tweak a physical relationship. We know whey Lockwood’s curve for cycle 14 is too low: He uses too few stations for those years and [and this is the bad part] deliberately omits other stations that give higher values so that he can maintain that there is a larger centennial increase in B.
The sticking point is your statement “The sunspot numbers seem to agree better with their version” which I have shown repeatedly is false. So, it is not really about Lockwood. It is about you making statements that have no foundation without making even the simplest analysis.
Geoff Sharp (00:50:43) :
solar modulation that repeats over a longer period than 100 years (172 years).
It is plain that on my plot, the ‘period’ is 100 years and not much longer. The same you see here: http://sidc.oma.be/html/wolfaml.html
Vukcevic (05:23:27) :
In many authoritative papers the Galesburg cycle is defined as 75-85 years and not 100 as above, so if it is the Gleissberg cycle, than it should conform to its widely accepted definition.
The Gleissberg cycle is not ‘defined’ but is determined from the data and the data [e.g. http://sidc.oma.be/html/wolfaml.html ] shows that Gleissberg was off for the historical record at least. Anything else is numerology as your post demonstrates so well. It is like the sunspot cycle itself: the long-term average length is ~11 years, but you can find cycles that are longer [like SC23] and shorter. The ’88-year’ Gleissberg cycle has been ~100 years recently, the same way that the 11-yr sunspot cycle has been 13 years recently. The historical record is too short to determine a good value for the Gleissberg cycle length. The 14C record extends over 12,000 years and it’s power spectrum shows it better: http://www.leif.org/research/FFT-INTCAL98-14C.png
(the label on the X-axis is wrong, it should be ‘Period, years’).
So, please, spare us the numerology.
For another view on the Gleissberg cycle [from nitrate deposits in ice] see McCracken et al. http://www.leif.org/EOS/Gleissberg-nitrates.pdf
One of Ken’s conclusions is especially interesting:
“5. The Gleissberg period 1820-1910 was the most prolific
generator of solar proton events followed by the period 1580-
1660. The present Gleissberg period (1910-1985?) is one of
the least effective in large fluence solar proton events observed
at Earth.”
But the whole paper in good food for thought. Read it.
Leif Svalgaard (05:59:30) :
“The Gleissberg cycle is not ‘defined’ but is determined from the data and the data.”
I think most comments about Galesburg cycle in many papers is just a ‘waffle’. Anyone can, and appears you fallen into same trap, ‘I see a cycle and hey, it is a Galesburg’ , regardless what is its period or amplitude is. As I pointed in my previous post, (vukcevic (05:23:27), you can not have a graph with falling trend 1835-1870 and rising trend 1935-1970 and declare it 100 year cycle. It is simply wrong. http://www.vukcevic.talktalk.net/IDV.pdf
People reading this blog may not be all scientists, but they are not totally naïve and illiterate.
What there is, it is a clear ~107 year cycle, as shown in your own FFT Power-Spectrum graph of the SS numbers.
http://www.vukcevic.talktalk.net/FFT-Power-Spectrum-SSN.png
It is time you accepted the fact that I have first succeeded to identify and mathematically define such a cycle.
It controls the envelope of sunspot cycles, as well as clearly identifies the Maunder minimum, and alternative succession of low cycles (25-30% in respect of the neighbouring) and long minima, as shown here:
http://www.vukcevic.talktalk.net/SSNanomaly1.gif
It is pointless denying the obvious, you may not like it but it is reality.
To be scientist, it occasionally means accepting that ‘not everything you happen to believe today is forever’.
My initial paper elaborating on the envelope and anomaly equations is at:
http://xxx.lanl.gov/ftp/astro-ph/papers/0401/0401107.pdf
As an interested layman, I’m thinking that sun-earth heating is analogous to an electric stove. Anyone who has cooked with both gas and electric stove tops knows that electric stoves are tricky because of the time it takes to heat up and cool down. The sun is like the dial, you turn it up, and yes it’s going full blast, but the iron rings (the earth) stay cool for a while, because it takes time to heat up all that iron. And vice-versa, if that metal is red-hot when you turn the power off completely, it will stay hot for while, much hotter (at first) than the first scenario, even though the “power level” becomes much lower in the latter scenario.
Vukcevic (09:06:37) :
My initial paper elaborating on the envelope and anomaly equations is at:
http://xxx.lanl.gov/ftp/astro-ph/papers/0401/0401107.pdf
Of course you are the first to come up with that silly formula.
Leif Svalgaard (11:17:06) :
“http://xxx.lanl.gov/ftp/astro-ph/papers/0401/0401107.pdf
Of course you are the first to come up with that silly formula.”
Do you mind, there are 3 formula’s there, one for periodicity (also provides best available representation of polar fields), one for overall amplitude, and one accurately identifying all recorded anomalies.
Fact that all 3 formulae use same 3 astronomical parameters or their multiples (harmonic oscillations resonance effect) not only confirms that they are rooted in real astronomy, but simplicity and elegance as expressed here:
Y= Sum[Cos 2pi (t-To)/P]
in science usually suggest it is likely to be good.
In this case, your epithet of a ‘silly formula’ is just a sign of inability to overturn its quality, which stems from the imbedded astronomical accuracy.
On the other hand, I am not qualified to judge your paper, but I would suggest to correct green line to have proper ‘~ 100 years cycle’ flow if you are inclined to keep it in.
‘Gleissberg cycle’ as described by McCracken from SC1 to present:
1750-1830 = 80 years
1830-1910 = 80 years
1910-1987 = 77 years.
Good luck with it.
Vukcevic (12:31:46) :
“Of course you are the first to come up with that silly formula.”
Do you mind, there are 3 formula’s there,
They are equally spurious. In the ‘not even wrong category’.
Leif Svalgaard (13:36:09) :
“They are equally spurious. In the ‘not even wrong category’.”
As ‘spurious’ as 3 astronomical numbers used.
Formula matches very well your IDV chart.
Yet another proof, the numbers are good.
See page 2 for a higher resolution graph.
http://www.vukcevic.talktalk.net/IDV.pdf
I have nothing more to add, your IDV graph says it all for me.
Thanks, and good luck with paper.
Leif Svalgaard (05:38:43) :
Geoff Sharp (00:50:43) :
solar modulation that repeats over a longer period than 100 years (172 years).
———————–
It is plain that on my plot, the ‘period’ is 100 years and not much longer. The same you see here: http://sidc.oma.be/html/wolfaml.html
We might be talking the same language, I am talking about from peak to peak. If I overlaid my graph, your HMF graph, the sunspot graph, Scafetta’s solar velocity graph and probably Vukcervic’s graph they would all agree. I have no doubt on the power wave but what will be interesting is whether there is a floor to the wave or does it also modulate like the proxy records suggest.
Vukcevic (14:15:38) :
Formula matches very well your IDV chart.
Yet another proof, the numbers are good.
Since the Gleissberg cycle the past 300 years has been a tad over 100 years long, any curve with that period built in will match. And you don’t seem to understand what ‘proof’ means. It is a fluke that we have had a ~100 period for a short while; in the long run [thousands of years] there is no power at 100 years or 107 years or 108 years [as I mention in my paper] or at 172 years, but at 88 years which then becomes the long-term average length of the ‘cycle’ just as 11 years is for the sunspot ‘cycle’.
The fundamental problem with your stuff is right at the beginning as there is very little similarity to the actual sunspot data which makes all the rest moot [and not even wrong]. Perhaps I should just submit your writing to Ap.J. tonight and see what say about it.
Geoff Sharp (15:15:18) :
If I overlaid my graph, your HMF graph, the sunspot graph, Scafetta’s solar velocity graph and probably Vukcervic’s graph they would all agree.
‘would’? do it, then we can see.
Leif Svalgaard (16:53:07) :
Oh no! then we’ll get Svaalgard, Svalgard, Sualgard, etc, many more changes of getting it wrong. Try to google scholar these misspellings and see how many hits you get 🙂
Svalbard: High arctic Norwegian island territory. Home of snow, ice, rocks, northern lights, polar bears, Musk Oxen, Reindeer, charming cottages, and the world’s highest lattitude university.
Leif Svalgaard (15:53:17) :
Geoff Sharp (15:15:18) :
If I overlaid my graph, your HMF graph, the sunspot graph, Scafetta’s solar velocity graph and probably Vukcervic’s graph they would all agree.
‘would’? do it, then we can see.
and since your period is long, we need several cycles, say the last 3000 years for comparisons [HMF and SSN could be replaced – or spliced to – the 14C and/or 10Be data]. This would be a plot worth seeing, instead of the self-congratulatory bragging we have seen [from several people].
Leif Svalgaard (17:38:26) :
and since your period is long, we need several cycles, say the last 3000 years for comparisons [HMF and SSN could be replaced – or spliced to – the 14C and/or 10Be data]. This would be a plot worth seeing,
Do you have reliable proxy data that can be used at this scale? To my knowledge the proxy data has only been good enough to show the grand minima/maxima trends, it doesnt get down to the individual solar cycle level going back 3000 years?
I suspect not and it’s probably a loaded question…we only have 300-400 years worth of data that can be used but the wave is certainly evident in that data.
http://www.landscheidt.info/images/Powerwavesm.png
Tip: Be sure to discount the grand minima when matching the wave patterns. They are a result of an unrelated process.
tallbloke (00:51:53) : “Chris, the ice ages are an effect of the Milankovitch cycles (probably) not the cause.”
Could not agree more, Tallbloke. You must have misunderstood my post, or at least the way I worded it.
Humor me for a minute: I still have fixed in my mind the comparison to the steep-walled 1998 global temperature spike and Hotrod’s comparison (and striking resemblance) to the shape of the 100-foot wave that hit the Draupner Rig (Jan, 1994, I think?).
That will always be one of my favorite threads on this site!
When I look at a graph, I can’t help but see waves.
And when they are static, as Vuk mentions, then, even then, I see standing waves.
Admittedly, these “disturbances” are only visible on a graph [for our human temporal perspective], but that does not mean they are not “disturbances”.
Over what continuum, however, I have no idea…
For example, the “disturbance” of say, the AMO “cycle”…propagates over space/time.
Or at least it propagates over an XY graph….LOL.
Or…in the cause of the 1998 temp spike and the Draupner Rig 100-footer, the second one at least being a rogue wave…. such is explained better perhaps by quantum mechanics.
I guess that is what kind of blows my mind: looking at the resemblance of waves, their troughs and crests, to the ones you can measure on a graph over the days….or eons.
Surfs up, dude. Kawabunga.
Chris
Norfolk, VA, USA
Geoff Sharp (20:02:25) :
Do you have reliable proxy data that can be used at this scale? To my knowledge the proxy data has only been good enough to show the grand minima/maxima trends, it doesnt get down to the individual solar cycle level going back 3000 years?
The 14C data has a 10-yr resolution which should be enough to show cycles of 80-170 years
Tip: Be sure to discount the grand minima when matching the wave patterns. They are a result of an unrelated process.
Well, the minima must occur BETWEEN the maxima so there is not much wiggle room. And you have absolutely no reason to assert that they are due to another process. Or are you asserting that the maxima are due to Babcock-like crapshots? [which they probably are].
What you need to make is a graph showing 14C, Vuk’s curve, and whatever curves you might have like AM or whatever you think is doing something. That would give three curves and they should be unadorned with spurious interpretation, just the data.
Leif Svalgaard (21:39:46) :
The 14C data has a 10-yr resolution which should be enough to show cycles of 80-170 years
You well know 1 record every 10 years will not suffice for this type of analysis. Any chance of you going back further with the HMF records?
Well, the minima must occur BETWEEN the maxima so there is not much wiggle room. And you have absolutely no reason to assert that they are due to another process. Or are you asserting that the maxima are due to Babcock-like crapshots? [which they probably are].
Grand minima always occur centered at the top of the wave, so they happen at the same time as maxima, the difference being that sometimes we get longer stretches as we see since the Dalton (210yrs), not all options are used on most occasions which allows the unused AM to be utilized, this causes grand maxima in one form.
You may disagree with the reason but I certainly do have one. It is laid out clearly on my blog as you know.
Lennart S (04:55:51) :
tallbloke (00:51:53)
You are perhaps referring to this:
http://www.solstation.com/stars/sol.htm [some 6 clicks down]:
If this is correct: What triggers these magnetic instabilities that also governs our climate?
Hi Lennart,
I was referring to this article in New Scientist:
http://www.newscientist.com/article/mg19325884.500-suns-fickle-heart-may-leave-us-cold.html
I can’t answer your followup question with any certainty, but I’ve described my line of research and in my response to Chris.