Sun said to be “bi-modal”
While many, including the IPCC, suggest the modern Grand Maximum of solar activity from 1950-2009 has nothing to do with the 0.4C global warming measured over that time frame, it does seem to be unique in the last three millennia.
from CO2 Science: A 3,000-Year Record of Solar Activity
What was done
According to Usoskin et al. (2014), the Sun “shows strong variability in its magnetic activity, from Grand minima to Grand maxima, but the nature of the variability is not fully understood, mostly because of the insufficient length of the directly observed solar activity records and of uncertainties related to long-term reconstructions.” Now, however, in an attempt to overcome such uncertainties, in a Letter to the Editor published in the journal Astronomy and Astrophysics, Usoskin et al. “present the first fully adjustment-free physical reconstruction of solar activity” covering the past 3,000 years, which record allowed them “to study different modes of solar activity at an unprecedented level of detail.”
What was learned
As illustrated in the figure below, the authors report there is “remarkable agreement” among the overlapping years of their reconstruction (solid black line) and the number of sunspots recorded from direct observations since 1610 (red line). Their reconstruction of solar activity also displays several “distinct features,” including several “well-defined Grand minima of solar activity, ca. 770 BC, 350 BC, 680 AD, 1050 AD, 1310 AD, 1470 AD, and 1680 AD,” as well as “the modern Grand maximum (which occurred during solar cycles 19-23, i.e., 1950-2009),” which they describe as “a rare or even unique event, in both magnitude and duration, in the past three millennia.”
Further statistical analysis of their reconstruction revealed the Sun operates in three distinct modes of activity – (1) a regular mode that “corresponds to moderate activity that varies in a relatively narrow band between sunspot numbers 20 and 67,” (2) a Grand minimum mode of reduced solar activity that “cannot be explained by random fluctuations of the regular mode” and which “is confirmed at a high confidence level,” and (3), a possible Grand maximum mode, but they say that “the low statistic does not allow us to firmly conclude on this, yet.”
What it means
Usoskin et al. (2014) write their results “provide important constraints for both dynamo models of Sun-like stars and investigations of possible solar influence on Earth’s climate.” They also illustrate the importance of improving the quality of such reconstructions, in light of the fact that previous reconstructions of this nature “did not reveal any clear signature of distinct modes” in solar activity.
Unfortunately, it was beyond the scope of this paper to address the potential impact of solar activity on climate. Yet the reconstruction leaves a very big question unanswered — What effect did the Grand maximum of solar activity that occurred between 1950 and 2009 have on Earth’s climate? As a “unique” and “rare” event in terms of both magnitude and duration, one would think a lot more time and effort would be spent by the IPCC and others in answering that question. Instead, IPCC scientists have conducted relatively few studies of the Sun’s influence on modern warming, assuming that the temperature influence of this rare and unique Grand maximum of solar activity, which has occurred only once in the past 3,000 years, is far inferior to the radiative power provided by the rising CO2 concentration of the Earth’s atmosphere.
Reference
Usoskin, I.G., Hulot, G., Gallet, Y., Roth, R., Licht, A., Joos, F., Kovaltsov, G.A., Thebault, E. and Khokhlov, A. 2014. Evidence for distinct modes of solar activity. Astronomy and Astrophysics 562: L10, doi: 10.1051/0004-6361/201423391.
Abstract
Aims. The Sun shows strong variability in its magnetic activity, from Grand minima to Grand maxima, but the nature of the variability is not fully understood, mostly because of the insufficient length of the directly observed solar activity records and of uncertainties related to long-term reconstructions. Here we present a new adjustment-free reconstruction of solar activity over three millennia and study its different modes.
Methods. We present a new adjustment-free, physical reconstruction of solar activity over the past three millennia, using the latest verified carbon cycle, 14C production, and archeomagnetic field models. This great improvement allowed us to study different modes of solar activity at an unprecedented level of details.
Results. The distribution of solar activity is clearly bi-modal, implying the existence of distinct modes of activity. The main regular activity mode corresponds to moderate activity that varies in a relatively narrow band between sunspot numbers 20 and 67. The existence of a separate Grand minimum mode with reduced solar activity, which cannot be explained by random fluctuations of the regular mode, is confirmed at a high confidence level. The possible existence of a separate Grand maximum mode is also suggested, but the statistics is too low to reach a confident conclusion.
Conclusions. The Sun is shown to operate in distinct modes – a main general mode, a Grand minimum mode corresponding to an inactive Sun, and a possible Grand maximum mode corresponding to an unusually active Sun. These results provide important constraints for both dynamo models of Sun-like stars and investigations of possible solar influence on Earth’s climate.
Leif Svalgaard says:
August 7, 2014 at 11:42 am
Different workers come up with different dates for the Dalton. Recent estimates are1790-1830 and 1796-1820. The troughs come in around 1808-10, as per you.
But why use AD 2025? She and her co-author think it’s happening again right now, about 100 years after 1910 and 200 after 1810. The solar cycles might also be lengthening a bit during the Modern Warm Period, getting closer to 23 years than 22. No doubt these variations (eg nine to almost 14 years for the “11-year” cycle) are old hat to you:
http://www.ips.gov.au/Educational/2/3/7
Anyway, there’s no reason IMO to be wedded to a precisely 88-year Gleissberg. As you say, it used to be around 100 years and obviously can change and range a bit.
sturgishooper says:
August 7, 2014 at 12:00 pm
But why use AD 2025? She and her co-author think it’s happening again right now, about 100 years after 1910 and 200 after 1810.
I don’t think this cycle is the bottom of the Gleissberg ‘through’, possibly the next cycle, that is why. But since it is not a ‘real’ cycle, it doesn’t matter much.
@Leif – trough.
Matt Marler. The Lockwood paper:
http://onlinelibrary.wiley.com/doi/10.1002/2014JA019973/pdf )
was not designed to discus the sun – climate connection but to show the OSF peak in the late 20th century – (solar maximum) However the climate implications of Fig 6 are very clear. Compare the low OSF during the Maunder LIA versus the late 20th Century. Also look at the sharp drop off in solar activity -especially the sharp decline in the cycle minima since the cycle 21/22 minimum.
The really sharp secular break ( which strengthens my belief in a coming cooling ) took place in about 2005-6.See the Ap index in Fig 13 and the historic high neutron count in 2009 in Fig 14 at
http://climatesense-norpag.blogspot.com/2014/07/climate-forecasting-methods-and-cooling.html
Leif Svalgaard says:
August 7, 2014 at 12:06 pm
Your colleagues disagree with your assessment of its reality, but that’s science. Absolute certainty except in certain well established cases (such as the earth going around the sun) however isn’t scientific, IMO.
Thanks for your answer, though.
Leif Svalgaard says:
August 7, 2014 at 11:46 am
Tom O says:
August 7, 2014 at 11:34 am
only God can say “there is no modern grand maximum” since only God has infinite knowledge.
For most things one does not need infinite knowledge. Eg I know that there were no big earthquake in San Francisco yesterday.
You know probably also what it was 100 years ago?
This pattern works best.
http://www.vukcevic.talktalk.net/PF-latest.gif
vukcevic says:
August 7, 2014 at 11:50 am
There is no evidence that you looked at the evidence.
The video is muddled, but why don’t you tell us in clear, easy to understand words what you think the evidence is. Or better: point us to a paper that lays out the evidence.
Dr Norman Page says:
August 7, 2014 at 12:11 pm
but to show the OSF peak in the late 20th century – (solar maximum)
Lockwood agrees that activity [OSF] in the 19th century was on par with that in the 20th.
The really sharp secular break ( which strengthens my belief in a coming cooling ) took place in about 2005-6.See the Ap index in Fig 13
This is Ap since 1844 http://www.leif.org/research/Ap-1844-now.png nothing secular about 2005.
and the historic high neutron count in 2009
Oulu shows an increasing count, Thule [Greenland] shows no trend, and South Pole shows a decrease:
ST12-05-D3-PM2-CD-004 (ST12-05-A011) AOGS 2014 Sapporo, Japan, July 30, 2014
Long Term Decline of South Pole Neutron Monitor Counting Rate – A Possible Magnetospheric Interpretation
Paul EVENSON#+, John CLEM
University of Delaware, United States
#Corresponding author: evenson@udel.edu +Presenter
“The neutron monitor at the Amundsen Scott Station, located at the geographic South Pole, has operated with some interruptions since 1964. The neutron counting rate follows an 11-year cycle with maxima at times of low solar activity, but over the entire interval exhibited a steady decline, totaling approximately 10% by 2013…”
If you think the ‘Grand Maximum’ supports your 1000-yr cycle, then you have a problem as Usoskin claims it was a unique event, thus not repeating every 1000 yrs. You can’t have it both ways.
sturgishooper says:
August 7, 2014 at 12:12 pm
Your colleagues disagree with your assessment of its reality, but that’s science. Absolute certainty except in certain well established cases (such as the earth going around the sun) however isn’t scientific, IMO.
You are too hung up on the ‘absolute certainty’ thing. When things are compelling enough [5 sigma, 6 sigma, or some other high threshold] they are accepted as facts. And it is not that ‘my colleagues’ disagree. A few people have still not caught up, is all. It usually takes a solar cycle for the slow ones to see the light.
From Bob Weber on August 7, 2014 at 8:56 am:
It is indeed well hidden, as SSN is showing half of an about 100yr possible cycle peaking around 1985 while HadSST3 is showing a trough-to-peak transition point of a different long possible cycle.
HadSST3 from 1960 with 15-yr trends starting every 5 years.
International SSN from 1960 with same trend pattern.
Well, that’s definitely a SSN peak circa 1985 matched to a SST transition between about 1985 and 1990, but SSN could be something other than 100yr.
So what was hidden, a very slow negative feedback where greater solar activity yielded a gradual deceleration of the warming rate?
ren says:
August 7, 2014 at 12:33 pm
This pattern works best. http://www.vukcevic.talktalk.net/PF-latest.gif
The lower graph has already failed and will fail even more for the polar fields in the next cycle.
Thanks Ren, the rain stopped for now, the company I work for calls wanting to know when/if I’m coming back. Just have to tell them that house preparation for an ice age takes a lot of work and it doesn’t help when it rains so much. Should have been done already. Ran around this morning in my skivies putting stuff up and a few odds and ends things I painted yesterday still hadn’t dried.
Leif Svalgaard says:
August 7, 2014 at 12:34 pm
That’s pretty funny, but who would have thought that there are so many slow physicists?
To quote Lord Rutherford, “If your experiment needs statistics, you ought to have done a better experiment”.
I’m not hung up on absolute certainty, but appreciate the limitations of computing confidence levels. (The earth going around the sun is no longer an hypothesis but an observation, so is absolutely certain.) Based upon the greater precision of Ptolemaic tables over Copernican c. 1600, had statistical analysis existed then, the geocentric model would have been favored at a high sigma threshold.
sturgishooper says:
August 7, 2014 at 12:48 pm
That’s pretty funny, but who would have thought that there are so many slow physicists?
Not funny, as there are actually few of them.
I’m not hung up on absolute certainty, but appreciate the limitations of computing confidence levels. (The earth going around the sun is no longer an hypothesis but an observation, so is absolutely certain.)
The Earth does not go ‘around the Sun’ in an absolute way as the orbit is not quite closed. To a high approximation it can be said to go around the Sun. And solar activity is also an observational fact. What is not an observational fact is the splicing of the two records that Usoskin did. That is based on an assumption [and some fiddling]. With any observation there is always an error bar.
Leif Svalgaard says:
August 7, 2014 at 1:23 pm
Variations in earth’s orbit don’t change the fact that the sun doesn’t go around the earth, as in the geocentric model. The actual observation now is that the sun is at one focus of the elliptical, slightly variable orbit around the sun of the barycenter of the earth-moon system.
sturgishooper says:
August 7, 2014 at 1:34 pm
Variations in earth’s orbit don’t change the fact that the sun doesn’t go around the earth, as in the geocentric model.
That was not the issue, which was whether the Earth ‘goes around the Sun’, which it only does to some [albeit high] approximation. Any observation has an error bar [perhaps with the exception of a straight count, as of the amount of loose change in my pocket or of the number of spots on the sun]
The reality is the tide is turning toward solar variability and that variability causing climate change.
http://hockeyschtick.blogspot.com/2014/08/new-paper-argues-current-lull-in-solar.html
http://icecap.us/images/uploads/Solar_Changes_and_the_Climate.pdf
The papers and studies go on and on that support my claims. I could easily post many more studies but I think I am making my point.
Yeah, I would have liked to have been a fly on the wall when you gave that talk…;-)
This fits very well over basically the last 50 years of modern data. This too permits us to reconstuct solar activity — consistently with the first measure — again, back to roughly 1850.
, so when combined with the previous measure one can infer 
(here’s where the talk would have been useful, as I have no idea what “
” is in this context:-).
Without the talk part of the talk, I will have to try to interpret the transparencies as best as I can:
* We have been able to measure, and record, the transverse shift in Earth-surface-local magnetic field due to the Sun’s magnetic field, since roughly the middle of the 19th century.
* This accurately measured and recorded field can be shown to be consistent with many other measures of solar activity taken with modern instrumentation, in particular sunspot number and F10.7 flux.
* Another measure, basically the short term variability of the geomagnetic field, can be shown to be related to solar heliospheric magnetic field B strength. (A connection via fluctuation-dissipation?) The correlation is very high,
* Computing the interhourly variation of the geomagnetic field (again, a connection via fluctuation-dissipation?) one can once again obtain a measure of solar activity, this time with the solar wind, which is in turn a measure of power input from the sun into the ionosphere. However, it yields the product of B and a second measure,
* “Near polar” observatories, manned for at least 80 years, provide evidence for the sun’s polarity reversals (?) and can be fairly accurately connected back to the actual patterns of solar magnetic whorls (including sunspots) on the sun. This is proof that solar polar reversals have been going on well into the past.
* Presuming that they have been going well into the past and that they are correlated then as they are now with 10Be atoms produced by cosmic rays, we obtain a radiogenic measure of solar activity into the more remote past (but probably at a lower time resolution?) .
* The flow of solar wind through the ionosphere over the polar caps induces a Hall potential due to the magnetic field coming up/down there. Electric fields being comparatively easy to measure, they form yet another way to observe a variable from which solar activity can be inferred. This has apparently been measured back to the late 19th century? This gives us three different ways of determining only two variables (B and V), which is overdetermined but doesn’t matter much since they all three give very much the same answers.
* From all of this information, we can compute the radial magnetic flux, with very reasonable confidence, all the way back to the early 19th century — at least 1850, maybe 1830. The linear trend is basically perfectly flat, and the variation is strictly bounded from above and below.
* It is true that the 20th century had two periods — three cycles from 1935 to 1965 (ish) and 1975 to 2005 (ish) where the radial field was consistently above the long term mean, only barely dipping back to or just over the mean. In between these intervals, the activity was flat at the mean, which is unusual across the record. In all of the rest of the record, the radial field is fairly consistently oscillatory. Although there is no trend, there are definite highs and lows, and (for example) the recent solar minimum was as low as it has ever been in the entire record, although there was a very similar pattern observed back in 1890 to 1900.
* This pattern almost precisely matches the pattern observed in sunspots across the same interval, the inferred Ap Geomagnetic index, and the heliospheric magnetic field at the Earth. All of these show that the 20th century was more active across the board (on average) than the 19th century, but they also show that the early 19th century was nearly as active as parts of the 20th.
* I don’t understand the point of the space climate slide, sorry.
* The next few slides are also very confusing. It appears to be relating (cosmic ray?) neutron counts to year all the way back to 1350 or so, but of course I don’t think that they had neutron detectors back then so I’m guessing that this is inferred by means of a radiogenic proxy. It does, however, clearly show the Sporer, Maunder and Dalton minima reflected in the neutron count, and also appears to be rather inconsistent with the pattern in the indices discussed above in overlapping regions. For example, it seems that neutron counts were, indeed, very low in the latter half of the 20th century (compared to the 600 year mean) and very high in the 15th century, but there is a lot of structure with rather variable autocorrelation. I don’t know what to make of it.
* The final transparency offers evidence that there was significant solar magnetic activity even when there were observationally very few sunspots during e.g. the Maunder minimum. This seems to confound in both directions — one cannot consider sunspot counts to be a completely reliable proxy of solar magnetic activity, and one cannot consider inferred solar magnetic activity to be a completely reliable proxy of sunspot counts? I’m not sure what to make of this.
* The conclusions:
1) We know B, V and n (whatever they are:-) back to maybe the 1830s (presumably with larger errors the further back we go).
2) None of the measures reveal a historically unprecedented period of high solar activity in the late 20th century, i.e. a “Grand Solar Maximum”. And there are quite a few, independent measure. If nothing else, the preponderance of evidence, while allowing for the 20th century to be a period of relatively high solar activity (compared to the entire interval, or even back to the 14th century), does not suggest (as Ushokin has on several occasions) that it is as high as it has been since e.g. 9000 BCE, or that there is anything like a third “mode” for the Sun. (To be honest, speaking for myself, there isn’t much reason in the data to consider the sun to be honestly bimodal. Variable, yes, but I can see little evidence in the data to suggest any sort of “phase transition” in its internal dynamics — I think this is all simple variability, “multimodal” if you like but with so many modes that there are none.)
3) FUV varies with sunspot number. I’m not sure what to make of this.
4) Solar cycle variations can be tracked over at least the last few centuries with geomagnetic observations and proxies. Hey, I’m convinced. Some of what the figures show is no-brainer, nearly perfect tracking and of course makes sense — one is nearly directly measuring the fields, presuming only that the Earth’s field is sufficiently slowly varying over the entire interval and not just the recent end of things where we can measure things more accurately.
5) Which is the final point Lief makes — the results do assume that the geomagnetic field’s direct variation (as best we know it now) can be extracted from the data and that it doesn’t “bend” the answers over the extrapolated regions. A very honest disclaimer.
Seems pretty convincing to me. At the very least, for Ushokin’s recent paper to be taken seriously, this all needs to be discussed and directly confronted. In the end, inconsistency might increase doubt in all asserted results, but when there is one result not in alignment with many other, quite independent ones, one has to be very careful about overstepping the bounds of the assertions.
Call me unconvinced at this point that there is either a “grand” minimum or a “grand” maximum in solar activity in the known record. For either term to be apropos, one would (in my opinion) have to show/know something concrete about the internal dynamics of the sun during the proposed “phases” sufficient to support the proposition that they deserve, in fact, to be called phases, in particular some commonality of organization, at least a dynamical phase transition or persistent degree of self-organization. At this point we are at best inferring the possibility of such things from external macroscopic measurements of solar state from far away and long ago. Not much to go on, really.
rgb
(Feel free to correct/enlighten me if I’ve misread anything in the talk, Lief, allowing for the fact that you don’t explain the meaning of all of the variables and I never heard the actual talk and am trying to make sense of the graphs and slides themselves, so any errors are not malicious on my part but merely ignorant.)
The data shows that post 2005 that solar activity has gone from a very active state to a very inactive state.
Until or if this changes this supports those of us who believe solar variability is much greater then what mainstream keeps trying to convey.
I base my thoughts on the data and the data shows me a dramatic change in solar activity post 2005.
rgbatduke says:
August 7, 2014 at 2:16 pm
* We have been able to measure, and record, the transverse shift in Earth-surface-local magnetic field due to the Sun’s magnetic field, since roughly the middle of the 19th century.
Correct, since the 1720s actually [although not mentioned in the talk]
* This accurately measured and recorded field can be shown to be consistent with many other measures of solar activity taken with modern instrumentation, in particular sunspot number and F10.7 flux.
Correct
* Another measure, basically the short term variability of the geomagnetic field, can be shown to be related to solar heliospheric magnetic field B strength. (A connection via fluctuation-dissipation?)
via the Ring Current [Van Allen Belts]
The correlation is very high, R^2 \sim 0.9 This fits very well over basically the last 50 years of modern data. This too permits us to reconstruct solar activity — consistently with the first measure — again, back to roughly 1850.
Correct
* Computing the interhourly variation of the geomagnetic field (again, a connection via fluctuation-dissipation?)
Via the magnetic field of the currents that cause the aurorae
one can once again obtain a measure of solar activity, this time with the solar wind, which is in turn a measure of power input from the sun into the ionosphere. However, it yields the product of B and a second measure, V^2, so when combined with the previous measure one can infer V (here’s where the talk would have been useful, as I have no idea what “V” is in this context:-).
V is the solar wind speed.
* “Near polar” observatories, manned for at least 80 years, provide evidence for the sun’s polarity reversals (?) and can be fairly accurately connected back to the actual patterns of solar magnetic whorls (including sunspots) on the sun. This is proof that solar polar reversals have been going on well into the past.
This actually holds all the way back to the 1840s
* Presuming that they have been going well into the past and that they are correlated then as they are now with 10Be atoms produced by cosmic rays, we obtain a radiogenic measure of solar activity into the more remote past (but probably at a lower time resolution?) .
Yes, correct, if we calibrate correctly [and there are some problems with that]
* The flow of solar wind through the ionosphere over the polar caps induces a Hall potential due to the magnetic field coming up/down there. Electric fields being comparatively easy to measure, they form yet another way to observe a variable from which solar activity can be inferred. This has apparently been measured back to the late 19th century?
What is measured is the magnetic effect of those currents.
This gives us three different ways of determining only two variables (B and V), which is overdetermined but doesn’t matter much since they all three give very much the same answers.
It is important that the system is overdetermined as that allows us to check the results. [3 eqs with 2 unknowns]
* From all of this information, we can compute the radial magnetic flux, with very reasonable confidence, all the way back to the early 19th century — at least 1850, maybe 1830. The linear trend is basically perfectly flat, and the variation is strictly bounded from above and below.
Correct
* It is true that the 20th century had two periods — three cycles from 1935 to 1965 (ish) and 1975 to 2005 (ish) where the radial field was consistently above the long term mean, only barely dipping back to or just over the mean. In between these intervals, the activity was flat at the mean, which is unusual across the record. In all of the rest of the record, the radial field is fairly consistently oscillatory. Although there is no trend, there are definite highs and lows, and (for example) the recent solar minimum was as low as it has ever been in the entire record, although there was a very similar pattern observed back in 1890 to 1900.
We can find those patterns all over the place, no period is particularly special
* This pattern almost precisely matches the pattern observed in sunspots across the same interval, the inferred Ap Geomagnetic index, and the heliospheric magnetic field at the Earth. All of these show that the 20th century was more active across the board (on average) than the 19th century, but they also show that the early 19th century was nearly as active as parts of the 20th.
The difference is not significant and the 18th century may have been more active than the 20th.
* I don’t understand the point of the space climate slide, sorry.
The point is that we find the same sort of variation in every solar cycle, e.g. that the density is high at solar minimum.
* The next few slides are also very confusing. It appears to be relating (cosmic ray?) neutron counts to year all the way back to 1350 or so, but of course I don’t think that they had neutron detectors back then so I’m guessing that this is inferred by means of a radiogenic proxy. It does, however, clearly show the Sporer, Maunder and Dalton minima reflected in the neutron count, and also appears to be rather inconsistent with the pattern in the indices discussed above in overlapping regions. For example, it seems that neutron counts were, indeed, very low in the latter half of the 20th century (compared to the 600 year mean) and very high in the 15th century, but there is a lot of structure with rather variable autocorrelation. I don’t know what to make of it.
A lot of discussion is needed here. The bottom line is that we can find some of the solar variations in the cosmic ray data, but also that there are problems with the data, e.g. it turns out that the climate plays a role as the deposition of the isotopes also depends on atmospheric circulation and not only on solar activity.
* The final transparency offers evidence that there was significant solar magnetic activity even when there were observationally very few sunspots during e.g. the Maunder minimum. This seems to confound in both directions — one cannot consider sunspot counts to be a completely reliable proxy of solar magnetic activity, and one cannot consider inferred solar magnetic activity to be a completely reliable proxy of sunspot counts? I’m not sure what to make of this.
There is a mystery here. It is very likely that at times [during Grand Minima] the magnetic cycle [and cosmic ray modulation] continues, but for some unknown reason, the magnetic field does not assemble into visible spots.
* The conclusions:
1) We know B, V and n (whatever they are:-)
[B=magnetic field, V=solar wind speed, n=solar wind density]
back to maybe the 1830s (presumably with larger errors the further back we go).
2) None of the measures reveal a historically unprecedented period of high solar activity in the late 20th century, i.e. a “Grand Solar Maximum”. And there are quite a few, independent measure. If nothing else, the preponderance of evidence, while allowing for the 20th century to be a period of relatively high solar activity (compared to the entire interval, or even back to the 14th century), does not suggest (as Ushokin has on several occasions) that it is as high as it has been since e.g. 9000 BCE, or that there is anything like a third “mode” for the Sun.
That is my conclusion too.
(To be honest, speaking for myself, there isn’t much reason in the data to consider the sun to be honestly bimodal. Variable, yes, but I can see little evidence in the data to suggest any sort of “phase transition” in its internal dynamics — I think this is all simple variability, “multimodal” if you like but with so many modes that there are none.)
Agree
3) FUV varies with sunspot number. I’m not sure what to make of this.
The Far UltraViotet [FUV] varies as the sunspot number, so is known as far back as the SSN.
4) Solar cycle variations can be tracked over at least the last few centuries with geomagnetic observations and proxies. Hey, I’m convinced. Some of what the figures show is no-brainer, nearly perfect tracking and of course makes sense — one is nearly directly measuring the fields, presuming only that the Earth’s field is sufficiently slowly varying over the entire interval and not just the recent end of things where we can measure things more accurately.
That is the point of the talk
5) Which is the final point Lief makes — the results do assume that the geomagnetic field’s direct variation (as best we know it now) can be extracted from the data and that it doesn’t “bend” the answers over the extrapolated regions. A very honest disclaimer.
We have theoretical reasons to believe this to be true, but…
Seems pretty convincing to me. At the very least, for Ushokin’s recent paper to be taken seriously, this all needs to be discussed and directly confronted. In the end, inconsistency might increase doubt in all asserted results, but when there is one result not in alignment with many other, quite independent ones, one has to be very careful about overstepping the bounds of the assertions.
As in my view EVERYTHING must fit together, when something doesn’t fit we can learn something. The famous astronomer Le Verrier [co-discoverer of Neptune] once said: “Tout ecart decele une cause inconnue, et peut devenir la source d’une decouverte”
Call me unconvinced at this point that there is either a “grand” minimum or a “grand” maximum in solar activity in the known record. For either term to be apropos, one would (in my opinion) have to show/know something concrete about the internal dynamics of the sun during the proposed “phases” sufficient to support the proposition that they deserve, in fact, to be called phases, in particular some commonality of organization, at least a dynamical phase transition or persistent degree of self-organization. At this point we are at best inferring the possibility of such things from external macroscopic measurements of solar state from far away and long ago. Not much to go on, really.
Agree, but a lot of progress has taken place the last decade, and we’ll learn a lot the next.
so any errors are not malicious on my part but merely ignorant
My default assumption is that folks are not malicious [until they show they are]. Thanks for slugging through the talk.
rgbatduke says:
August 7, 2014 at 2:16 pm
…………….
There is a problem in trying to ascertain solar activity from the geomagnetic signal. Data used by JPL-NASA and the ‘Institut de Physique du Globe de Paris’ (another solar institution) show that the annual variability in the Earth’s outer liquid core generated magnetic field has a clear and strong 21.3 year spectral component equaling 2x average sunspot period.
Spectrum is shown here
Either way, coming from the Earth’s core or a solar induced these geomagnetic oscillations are far stronger then the the heliospheric magnetic field at the Earth’s orbit.
I have collected my points into a more coherent whole as follows:
Recent advances in reconstructions of solar activity can be described thus [I’ll number them for easy reference. Papers and analyses can be given for each point, but are better presented if and when a point is up for discussion]:
1) Variations of TSI are the result of variations of the Sun’s magnetic field.
2) The sunspot number is a very good measure of solar magnetic fields.
3) Variation of the UV flux is due to variations of the Sun’s magnetic field.
4) The F10.7 microwave flux is a very good proxy for the UV flux and it is at the same level at every solar minimum.
5) The variation of the diurnal variation of the geomagnetic field is caused by the UV and is a very good proxy for said UV since 1781 [and with some gaps back to 1722].
6) The solar magnetic field is dragged out into the heliosphere and can be measured directly by spacecraft or almost as accurately by its effect on the Earth’s ring current (Van Allen Belts) whose magnetic effect can be measured on the ground, since 1830s.
7) The magnetic effects caused by the solar wind can also be monitored at auroral latitudes, allowing determination of both the solar wind magnetic field and the solar wind speed. Different research groups agree on these determinations.
8) Cosmic Rays modulation depends [inversely] largely on the heliospheric magnetic field.
9) These various determinations [by several researchers] of the solar magnetic field agree, so we know with good accuracy the solar magnetic field back to at least the 1830s, and hence also TSI.
10) The sunspot number has recently been revised and the result is that solar activity in each of the centuries 18 to 20 is very similar: a minimum about every 100 years near the turn of the centuries and a local maximum in mid-century.
11) There is therefore no Modern Grand Maximum.
12) A result of all of the above is that solar activity reaches almost the same low level at every solar minimum.
13) Early reconstructions of TSI assumed that the solar cycle variation was riding on a varying background level which itself varied as a function of solar activity
14) This background was assumed to be caused by a solar-cycle dependent emergence of small magnetic [so-called] ephemeral regions. Modern measurements show that this assumption is false and that the emergence rate of ephemeral regions is almost constant in time and thus does not vary with solar activity.
15) Thus, reconstructions that show varying background level [e.g. Lean, Krivova, Wang, and others] are not correct, and conclusions based on them are similarly suspect.
16) All our determinations show that solar activity recently is very much the same as a century ago.
17) This means that the decrease of solar activity from the 1870s to the 1910s is very much similar to the decrease from 1980 to now. In particular, TSI now is very likely the same as it was 100 years ago
18) If our climate depends strongly on solar activity [be it TSI, magnetic field, UV, cosmic rays, what-have-you] then our climate the last 30-40 years would be very similar to that a century before [even allowing lags of several solar cycles], and it is not.
I’ll take issue only with this conclusion, which I do not think is strictly justified. If the climate were Markovian (or nearly so), one might expect a strong dependence on solar activity to lead to similar climates at times of similar solar state. But we know that this is not true. The climate depends on an integral over past states of both the earth (and, possibly the sun even if there is a strong dependence) that extends at least centuries into the past, perhaps longer. However, we also know that the climate is almost certainly highly multivariate, and thus far we do not have any convincing model that can explain even meso-scale data from the past — the last 2000 years of climate, for example. In a highly nonlinear, multivariate model, the effect of solar state on the climate might be different one century to the next just because the rest of the Earth’s climate system is very different because of the integrated changes in the meantime.
I would suggest softening it to the more reasonable:
If our climate depends on solar activity [be it TSI, magnetic field, UV, cosmic rays, what-have-you] in a linear response model with no other dynamical influences with longer (or just different) timescales, then our climate the last 30-40 years would be very similar to that a century before [even allowing lags of several solar cycles], and it is not.
This makes the point clear — one cannot just look at solar state locally in time (or integrated as you say over several cycles) and predict what the climate will do with any confidence. Solar state then can take its place in the long line of variables that similarly have turned out, or are turning out, to not be linearizable climate drivers. So far the internal nonlinear dynamics of the climate appear to overwhelm any simple variation in its drivers, making it nearly impossible to extract any sort of linearized response “signal” that can be cleanly associated with a single knob.
Personally, I think we could do much better if we really used the fluctuation-dissipation theorem to analyze the data, but the quality of the data probably still cannot support this, at least not convincingly. But as you say, we are learning more all the time and another decade or three of data taken with modern instrumentation at adequate sampling density might eventually give us a clue from this and other analytic methods.
rgb
Dr Norman Page: However the climate implications of Fig 6 are very clear. Compare the low OSF during the Maunder LIA versus the late 20th Century.
I don’t find them to be clear at all if you look at the complete data.