Guest post by David Archibald
Long suspected, it seems that this has now been confirmed by a paper in Astronomy and Astrophysics with the title “Is there a planetary influence on solar activity?” by Abreu et al that was published on 22nd October, 2012.
From the Discussion and Conclusions section:
The excellent spectral agreement between the planetary tidal effects acting on the tachocline and the solar magnetic activity is
surprising, because until now the tidal coupling has been considered to be negligible. In Appendix A we show that the possibility of an accidental coincidence can be ruled out. We therefore suggest that a planetary modulation of the solar activity does take place on multidecadal to centennial time scales.
The authors note that current solar dynamo models are unable to explain the periodicities in solar activity such as the 88 year (Gleissberg), 104 year, 150 year, 208 year (de Vries), 506 year, 1000 year (Eddy) and 2200 year (Halstatt) cycles. They adopted a different view by regarding the planets and the solar dynamo as two weakly coupled non-linear systems.
The idea that planetary motions may influence solar activity seems to have been initiated by Rudolf Wolf in the 1850s. While energy considerations clearly show that the planets cannot be the direct cause of solar activity, they may perturb the solar dynamo.
Specifically, the authors calculated planetary torque at the tachocline. The tachocline of the sun is a shear layer which represents a sharp transition between two distinct rotational regimes: the differentially rotating convection zone and the almost rigidly rotating radiative interior. The tachocline plays a fundamental role in the generation and storage of the toroidal magnetic flux that eventually gives rise to solar active regions. A net tidal torque is exerted in a small region close to the tachocline due to the buoyancy frequency originating from the convection zone matching the tidal period. The tachocline is thought to be non-spherical – either prolate (watermelon-shaped) or oblate (pumpkin-shaped). The authors’ model describes planetary torques acting on a non-spherical solar tachocline.
Figure 5 from the paper shows the 10Be record, shown as modulation potential, and planetary torque in the frequency domain:
Panel a is the Fourier spectrum of the solar activity quantified by the solar modulation potential. Panel b is the Fourier spectrum of the annually averaged torque modulus. The spectra display significant peaks with very similar periodicities: The 88 year Gleissberg and the 208 year de Vries cycles are the most prominent, but periodicities around 104 years, 150 years, and 506 years are also seen.
The match between theory and the physical evidence is very, very good. As the authors put it,”there is highly statistically significant evidence for a causal relationship between the power spectra of the planetary torque on the Sun and the observed magnetic activity at the solar surface as derived from cosmogenic radionuclides.”
They also advance a plausible mechanism which is that the tachocline, playing a key role in the solar dynamo process, is a layer of strong shear which coincides more or less with the layer of overshooting convection at the bottom of the convection zone. The overshoot layer is thought to be crucial for the storage and amplification of the magnetic flux tubes that eventually erupt at the solar photosphere to form active regions. Small variations in the stratification of the overshoot zone “of about -10-4 may decide whether a flux tube becomes unstable at 2·10-4 G or at 10-5 G. This makes a great difference, because flux tubes that do not reach a strength close to 10-5 G before entering the convection zone cannot reach the solar surface as a coherent structure and therefore cannot form sunspots.” This sounds like an explanation for the Livingstone and Penn effect of fading sunspots.
Figure A.1 from the paper also shows the very good correlation between cosmogenic radionuclides from the period 300-9400 years BP and the model output:
Upper middle panel: 14C production rate derived from the INTCAL09 record
Lower middle panel: solar modulation record based on 10Be records from GRIP
(Greenland) and Dronning Maud Land (Antarctica) and the 14C production rate
Bottom panel: Calculated torque based on planetary positions
If planetary torque modulates solar activity, does solar activity in turn modulate the earth’s climate? Let’s have a look at what the 10Be record is telling us. This is the Dye 3 record from Greenland:
All the cold periods of the last six hundred years are associated with spikes in 10Be and thus low solar activity. What is also telling is that the break-over to the Modern Warm Period is associated with much lower radionuclide levels. There is a solar mechanism that explains the warming of the 20th Century. It is also seen in the Central England Temperature record as shown in the following figure:
Conclusion
This paper is a major advance in our understanding of how solar activity is modulated and in turn its effect on the earth’s climate. It can be expected that planetary torque will progress to being useful as a tool for climate prediction – for several hundred years ahead.
Reference
J.A. Abreu, J. Beer, A. Ferriz-Mas, K.G. McCracken, and F. Steinhilber, Is there a planetary influence on solar activity?” Astronomy and Astrophysics, October 22, 2012
Thanks to Geoff Sharp, the full paper can be downloaded from here.
(Note: This post was edited for title, form, and some content by Anthony Watts prior to publishing)
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
But… but but the sun doesn’t effect the earth weather or climate useless research is useless.
There is some preliminary discussion about this paper from last month here.
http://tallbloke.wordpress.com/2012/10/25/j-a-abreu-et-al-is-there-a-planetary-influence-on-solar-activity/
What about a outside influence that synced both over time. Pulsar, binary star…
I have to admit, I had not given this hypothesis any credit what so ever. I suspect a few astrophysicists are now walking around campuses knocking on the doors of Archeology and Meteorology Departments to see if they can come up with a coincidental weather pattern.
By the way, Astronomy and Astrophysics made the paper freely available themselves here:
http://www.aanda.org/articles/aa/pdf/forth/aa19997-12.pdf
Geoff Sharp’s work gets a mention in this very informative and easy to read paper by an actuary who has been doing some intensive cross-disciplinary research…
http://tallbloke.wordpress.com/2012/11/10/brent-walker-extra-terrestrial-influences-on-natures-risks/
This covers some of the solar-planetary theory and much else of interest besides.
What did I tell you7
Leif? Comments?
I believe that tidal influence has been known. But do these influences change any other solar metric outside of variation noise? That is the question even CO2 enthusiasts deal with. And so far, all these external and anthropogenic influences do not rise above intrinsic natural variation noise.
One may wonder how effective peer-review of this paper has been. “Received 17 Mai 2011 Accepted 17 Mai 2011”. Usually, such instant acceptance is reserved for climate papers from the “Team”.
In the 1980’s -90’s it was theorized that solar, lunar and planetary gravitational effects influenced earth’s volcanic activity and subsequently climate. Dr. Iben Browning was a noted supporter of this idea and predicted global cooling as a result.
So low solar activity causes the climate to cool. Who woulda thunk?
Does that mean I can sell my warm coat or do I need to look out my fur-lined walking stick?
Rubbish. They are finding coincidence. Too many peaks, and amplitudes mismatch. Correlation doesn’t prove causation.
If you believe the interpretation, make a prediction. Try to get the right phases to go with the amplitudes. Unfortunately, it will take years to get results, just need to keep up the funding….
If they got such a good match then weren’t they just a little bit curious to extend it forward in time ?
204 years ago we had the Dalton, and 500 years ago the Maunder was about to get into its stride. The next few years might be interesting. Presumably the authors felt that we wouldn’t be interested, or perhaps they thought that it would be better if we didn’t know.
Oh, I know why they didn’t, co2 has already saved us.
tallbloke says:
November 10, 2012 at 9:26 am
Geoff Sharp’s work gets a mention in this very informative and easy to read paper
If the Abreu torque-mechanism is working, Geoff’s Angular Momentum [AM] ideas are moot. The AM ‘mechanism’ would not have worked anyway because the Sun is in free fall.
My sarc tags disappeared, guess I can’t use ><
/sarc Oh, I know why they didn’t, co2 has already saved us. /sarc
It’s clear the solar specialists do not know what the H they’re talking about.
I like the fact that this paper connects (at least arguably) plausible physical mechanisms to observations. What it is basically saying is that things that happen within the thin shell of the thermocline — in particular the formation of defects, convective rolls — are the proximate cause of surface solar activity. Defects that form there nucleate magnetic flux tubes that much later emerge as sunspots and solar surface magnetic activity. Even small degrees of modulation of the nucleation process can make big differences at the surface, just as only micro-droplets of water nucleated around particles or aerosols reach the critical size necessary to grow into actual clouds or raindrops.
In the case of rain, it is surface to volume plus some surface chemistry that largely determines the success of the process initially, but tiny modulation in terms of aerosol concentration makes a big difference in the probability of cloud formation because there are strong positive feedbacks once a cloud reaches the point where it starts reflecting sunlight and differentially cooling the interior and lower layers.
Two things I would like explained in more detail before I completely buy the argument, though, are:
* The forces/torques involved are very, very small — I would like to see some sort of numerical magnetohydrodynamic computation that made plausible assumptions about the viscosity, density, lapse rate and UNinfluenced nucleation process (plausible enough to reproduce at least approximately observed surface phenomena) that, when the very weak planetary tidal torque was turned on altered the flux-tube nucleation rate as is asserted by the paper. Or COULD alter it, for some not unreasonable values of the parameters, across the critical boundary where the grow if the tidal forces are large, fail to grow if they are small. It’s one thing to say “this could be happening”; another to say “when we solve the equations of motion for a plausible model system, we observe that this happens”.
Lacking this, their argument is much weaker.
* I find myself vaguely disturbed by the fourier transforms above. Figure 5 is reasonably plausible — note well that the solar proxy peaks are broad and smeared out, with the long period peaks being much broader than the short period peaks and with a lot of undifferentiated “noise” in the short period high frequency domain. This can be understood very simply — the dataset being fourier transformed has a finite length, a length commensurate with (within an order of magnitude or so) the longest period peak displayed. Consequently those long period peaks are poorly represented in the sample, and the peaks are broad and accompanied by peaks that might well be entirely spurious, artifacts of the length of the dataset and the accidental noise. Basically, the fourier transform itself has artifacts that correspond to (inverse) Gibbs phenomena arising from the de facto decomposition of a Heaviside function representing the length of the data set, which broadens the longer period peaks and introduces irrelevant shorter period peaks both.
The shorter period peaks have many more periods in the integral and hence are much narrower, although they are quite reasonably accompanied by a lot of short period noise because the Sun is a chaotic turbulent magnetohydrodynamic system and probably has internal e.g. breathing mode oscillations with a variety of frequencies that also modulate the phenomenon (if the hypothesis is correct, that small variations in the thermocline can and do produce macroscopically resolvable differences in surface output).
Figure b is similar. Obviously they did a F.T. on the same interval as used in a) so that the peak at e.g. 506 years is similarly broad even though one would rather expect it to be quite sharp given that one is simply doing a fourier decomposition of the vector sum of a set of completely determined torques due to the periodic planetary orbits. If they’d done the F. T. on a much longer time interval (as they easily could have) one would expect the 506 year peak to be much sharper. This also explains why the widths of the 88 year peaks etc scale pretty well between a) and b), and shows that their hypothesis has a very hard time accounting for the largest peaks in the Be proxy in the broad zone between 200 and 500 years. Nor can these peaks in a) be easily explained away as Heaviside/Gibbs artifacts, as the artifacts would be expected to appear in figure b) as well.
Again, perhaps these correspond to breathing mode oscillations, the sun’s internal furnace “chuffing” a bit due to some sort of resonance between gravitational force, fusion rate, and thermal expansion (a feedback loop where the sun contracts slightly, increasing the efficiency of the core fusion process, which heats the core a bit more, which then takes a long time to propagate to the thermocline, which expands the thermocline a bit, which decompresses the furnace so that its output cools/decreases a bit, which propagates outward to cause the thermocline to contract a bit, which recompresses the core a bit, resulting in a wave train of small modulations in output due to coupled breathing mode oscillations of solar density in the “critical” domain of the thermocline). Again, it would be lovely to see a model that reproduces this sort of coupled nonlinear phenomenon even qualitatively.
So figure 5 is moderately convincing. The short period peaks line up very nicely with peaks in Be production, there is at least a peak at 508 years in both (along with a lot of unexplained structure in between), and yes, it could be true that small forces drive relatively big changes if there is any possibility of resonance, where 5a actually provides some evidence of undriven resonances as it is.
What I don’t like so much is figure A.1. To be frank, it looks impossible. The peaks are far too sharp, far too localized, and utterly inconsistent with figure 5. The interval of integration is only around 20x the size of the longest period and yet the 508 year peak is sharp as an arrow in the radionucleotide data. If anything, the SHORT period peaks have widths. I would have rather expected this figure to look a lot more like 5 for all of the proxies, even if they did the FT of the torque over a long enough interval to sharpen up the long period peak(s). This F.T. looks more like a quantum spectrum, where there is actual physics prohibiting most of the possible frequences in the temporal signal, not the FT of a chaotic, noisy process.
Where is all of the noise?
To conclude — it convinces me that planetary tides are a plausible physical mechanism for modulation of the magnetic state, one that is empirically correlated with proxy derived data. Some of the data expressing the correspondence is “reasonable”; other data is rather puzzling although they may have some explanation for it and my intuition of the wrong shape and lack of noise of the FT on a 20x interval in multiple proxies may be wrong. Finally, this is probably not the only important factor — note the other peaks in figure 5 — and the entire argument would be greatly strengthened by even a crude model calculation that can reproduce the result qualitatively and demonstrate that small torques are indeed “capable” of producing the large modulations observed via nucleation and growth in an actual magnetohydrodynamic convective model of the Sun.
rgb
In many discussions with planetary perturbation advocates on these pages, especially involving Nicola Scafetta who no doubt will be making an appearance in 5, 4, 3… ;-), I have agreed that there could possibly be a link due to gravitational gradients, but that I doubt the argument can be made compelling enough for widespread acceptance.
I will also go out on the limb a little and make a perhaps novel suggestion that causation could be the reverse – that solar activity, resulting in increased solar radiation pressure, might perturb planetary orbits. Satellites at geosynchronous orbit drift significantly due to this factor, as well as due to gravitational perturbations from the Sun, Moon, and Earth, and their orbits have to be periodically corrected via stationkeeping maneuvers to maintain position. If solar radiation pressure were the primary driver of the correlations between solar activity and planetary motion, it might explain stochastic variation in the phases and amplitudes of the cycles.
In any case, the bottom line is that the cycles exist regardless of the mechanism. The warmist line is that the existence of such cycles must depend on some kind of unknown, and unlikely, deterministic driver, and those pointing to coincidence with planetary cycles are ceding the high ground in the battle to the opposition by accepting this premise. But, the premise is flawed. The danger is that the advocates of planetary influence may find evidence which proves the naysayers wrong, but they may not, and the naysayers will then claim that the hypothesis is disproved when, in fact, the hypothesis of natural cyclical climate behavior does NOT depend on the existence of a deterministic external driver.
Oscillations in natural systems depend only on the ability to store energy in alternating states, e.g., potential to kinetic, and back again. Such oscillations arise frequently and naturally in systems which are governed by partial differential equations on a bounded domain, and can be continuously excited by purely random forcing. Such oscillations are ubiquitous in nature.
The oscillations exist, and we are at the peak of an approximately 60-65 year cycle in the global temperature metric right now. The entire AGW scare was mounted on the back of the preceding upsweep in that cycle, and it is going to fail on the imminent downsweep.
Central to the torque mechanism is that the solar dynamo is working in the overshoot layer just beneath the tachocline. BTW, Abreu et al. accepts that a dynamo is creating the solar cycle.
Central to many dynamo mechanisms is a shear layer across which the solar rotation speed changes. There is such a layer at the tachocline [that is what defines the tachocline which is about 30% of the solar radius below the photosphere]. Another important ingredient is a Meridional circulation that recycle surface magnetic flux back into the interior where it can be amplified for the next cycle. This is the idea of the ‘conveyor belt’. Modern observations seem to indicate that there is no such single large conveyor belt. There is a shallow belt just under the surface where, BTW, there is also a shear layer, so dynamo models may be moving to a shallow dynamo rather than the deep one needed for the torque mechanism to work.
Hoser says:
“Rubbish. They are finding coincidence. Too many peaks, and amplitudes mismatch. Correlation doesn’t prove causation. ”
I read that they developed a theory and a model. and then tested it against the real world data. That in their opinion, is validation of their theory and model. Just where did you get your “coincidence” concept from? Maybe I missed it, please advise.
To me the title of the paper seems somewhat rhetorical IMHO.
Simply put, if gravity is real (which, of course, it is) – the planetary gravitational exchanges and combinations via elliptical orbits MUST have a corresponding change on the gravitational forces experienced by the sun and its surface (if we are simply considering sunspots for example).
Irrespective of the actual suns ‘burning’ processes – simple logic dictates that if its ‘surface’ is subjected to an external variation in gravity – it must likely affect the way that that ‘surface’ emits its radiation/plasma, etc, etc. The gravitational ‘impacts’ will presumably also affect the magnetic fluxes and so on. Perhaps, the gravitational effects of planetary motions can also affect the ‘spin’ of the sun or some of its component parts (does it have a rotating core?), e.g slowing them down or speeding them up.
Following this through, it is quite obvious that the summary solar output/activity WILL almost certainly be affected in some way by planetary gravitational effects (irrespective of what those effects actually are!). I suppose the argument is then about how much of an effect these gravitational changes are on the (largely unknown?) actual internal processes of the suns ‘workings’.
We know that lunar influences cause tides on earth, what’s to say that ‘jovian’ influences don’t have a similar effect on the actual gas/plasma ‘surface’ of the sun? – albeit on a much ‘longer’ timescale (perhaps a bad analogy, but readily understood I think).
Thence, it is a simple deduction to conclude that if the suns output can vary (for whatever reason) and that solar energy is primarily what drives our climate – then logically, changes in solar output can and will likely result in a change to our climate. (I do largely subscribe to the ‘it’s the sun stupid’ meme, because, in the end, even if TSI only varies by 1% – 1% of a lot of energy, is still a signifcant amount! My secondary view is that earth’s biosphere and atmosphere has the ability of adjusting to such solar changes amidst all its complexity, and thus perhaps to mitigate such minor solar effects)
If gravitational forces are also considered as able to cause an effect on the space betwixt sun and earth and indeed affect the earth itself, altering its magnetosphere/heliosphere, etc – then this is another ‘influence’ of planetary motions which can change the actual solar ‘input’ reaching earth?
The Central England Dye temperature graph: if the datapoints were coloured to show decade or greater, you might see an unexpected pattern that indicates time as an important factor, that is, something that is time-related.
Nice theory and observation, now how about some predictions! Yes it will take time to see but we have lots of time.