Scientists find errors in hypothesis linking solar flares to global temperature
From Physorg.com. h/t to Leif Svalgaard who offers this PDF with this diagram that makes it all clear.
In contrast to a previous analysis, a new study has shown that the distributions of (a) the global temperature anomaly by month since 1880 and (b) the solar flare index by day over a few solar cycles are fundamentally different. One feature the detrended data do have in common is self-similarity: the probability density functions are the same on different time scales, which means that neither can be described as Lévy walks. Image credit: Rypdal and Rypdal.
(PhysOrg.com) — The field of climate science is nothing if not complex, where a host of variables interact with each other in intricate ways to produce various changes. Just like any other area of science, climate science is far from being fully understood. As an example, a new study has discredited a previous hypothesis suggesting the existence of a link between solar flares and changes in the earth’s global temperature. The new study points out a few errors in the previous analysis, and concludes that the solar and climate records have very different properties that do not support the hypothesis of a sun-climate complexity linking.
In a handful of studies published in Physical Review Letters between 2003 and 2008, a team from Duke University and the Army Research Office including Nicola Scafetta and Bruce West analyzed data that appeared to show that solar flares have a significant influence on global temperature. Solar flares, which are large explosions in the sun’s atmosphere that are powered by magnetic energy, vary in time from a few per month to several per day. Although solar flares occur near sunspots, their frequency variation occurs on a much shorter time scale than the 11-year sunspot cycle. In their studies, the researchers’ results seemed to show that data from solar flare activity correlates with changes in the global temperature on a short time scale. Specifically, their analysis showed that the two time records can both be characterized by the same Lévy walk process.
However, in the new study, which is also published in Physical Review Letters, Martin Rypdal and Kristoffer Rypdal of the University of Tromso in Norway have reexamined the data and the previous analysis and noticed some shortcomings. One of the biggest causes of concern is that the previous analysis did not account for larger trends in factors that affect solar flares and global temperature. For instance, the solar cycle has its 11-year periodic trend, where periods of lots of sunspots cause larger numbers of solar flares. Likewise, the global temperature anomaly has numerous other factors (a “multi-decadal, polynomial trend”) that impacts global temperature fluctuations. By not detrending this data, the analysis resulted in abnormally high values of certain variables that pointed to Lévy walk processes. By estimating the untrended data, Rypdal and Rypdal hypothesized that the solar flare records might be described by a Lévy flight, while the global temperature anomaly might obey a distribution called persistent fractional Brownian motion.
Read the entire article here at Physorg.com
A preprint of the paper is available here
Practice making your own Levy walks here
oneuniverse (06:01:52) :
For two correlated variables A & B, either…
As long as R^2 is of the order of 0.04, so explanation is needed.
jinki (04:55:30) :
I gave up on you long ago.
Faced with an R^2 of 0.04 and the glaring discrepancies [green ovals], it seems prudent to give up. So, at least, we are making some progress…
To take a broader view:
One issue that sticks out: Proxy reconstructions are problematic.
There are enough variables (many with their own problems of reconstruction) that a clever person can reconstruct the climate or the Sun’s activity level to suggest whatever “outcome” one is predisposed to find, whether consciously or unconsciously.
Either side to the issue at hand can take most any study and poke holes in it.
It’s a demonstration that “science by proxy” ends up being a field day for whichever side has the most clever individuals supporting it. In my opinion that’s not good Science.
Good Science is not consensus, but consensus in this kind of battle usually wins the war and the victors write the history and thus the consensus ends up appearing as if there was the only reasonable analysis & interpretation of the scientific evidence.
When, in fact, there were other reasonable alternative analysis & interpretation of the evidence — in fact, it’s entirely reasonable and possible that the consensus is wrong — and in terms of past climate-Sun interactions, quite possible Science will never know what happened with a high degree of accuracy or certainty.
Perhaps, this comment will make neither side happy.
But “happy” is for politics — Science is about trying to answer a question about a physical relationship, whether past, present, or future, but also realizing (particularly past events in the field that weren’t observed & measured in high resolution) there will be times when Science has to frankly admit — we don’t know and possibly or likely will never know.
Beware of those in science who claim absolute certainty in the face of uncertainty about issues that are problematic — it’s likely they are simply attempting to be a shaman or worse, a charlatan.
Trillion dollar questions and the actions and consequences dependent on those trillion dollar questions should not be taken on such flimsy foundations.
The first injuction of medical science is “do no harm”.
For a scientific question that has such huge political and economic magnitude and consequences, “do no harm” has to be the fail-safe position.
That Man-caused CO2 results in dangerous climate warming is an extraordinary claim — therefore, it takes extraordinary evidence to confirm it.
So, far, the “extraordinary evidence” is entirely lacking.
Leif Svalgaard (09:00:23) :
oneuniverse (06:01:52) :
For two correlated variables A & B, either…
As long as R^2 is of the order of 0.04, NO explanation is needed.
Leif Svalgaard (21:25:18) :
Piers Corbyn (21:14:52) :
Yet another ‘NEW’ ‘investigation’ to ‘disprove’ the already clearly demonstrated causal connection between solar activity (particularly particle-magnetic) and Earth’s weather!
As the late Jack Eddy once remarked: “this topic brings all kinds of critters out of the woodwork”.
arrampicandosi sugli specchi
To Leif Svalgaard (08:08:15) :
“For people like S&W that claim that TSI is causatitive, the geomagnetic field is a problem. ”
As usual Leif does his best to mislead people and misrepresent our papers. In our paper we have never stated that the TSI is “causative”. We have always stated that we are using the TSI records as “proxy” for the sun-climate interaction.
This means that according the logic of our papers TSI real effects on climate may be even zero. And what is causing climate change is a mysterious extraterrestrial forcing “X” that for some mysterious reason has geometrical patterns that look quite similar to the patterns found in the TSI records, that is it has a 11-year cycle plus the other modulations found in such records. We simply assume that TSI is an acceptable proxy for the extraterrestrial forcings of climate whatever they might be, and found that the results are quite good, indeed. Indeed, the TSI records we have are “proxy” and we treat them as such!
I ask again Leif to read our papers before criticizing them without having read them.
Ulric Lyons (12:17:11) :
arrampicandosi sugli specchi
You are right. However, solving the problem is not impossible! 🙂
I just ask that people read carefully the papers and try to understand the issues and the reasoning instead of misinterpreting them and then claiming that they have disproved them. The careless behaviors of the critics is what makes these kind of issues more complicated than what they are!
oneuniverse (05:17:19) :
Figs. 2 & 3 of the Kirkby 2007 paper referenced above show better correlations over the last 2k years. The 10Be & 14C proxies are used to determine CR-flux rather than SSN. The former has a more direct physical relationship to the isotope formation process, and therefore has fewer uncertainties in its derivation.
The Kirkby results do show a better correlation, and attempting to discredit the correlation with inappropriate tools would also be pointless. The general trends observed in the NH cave deposits agreeing with the Moberg data.
Here is fig 3 from the Kirkby paper.
http://i44.tinypic.com/negpzm.jpg
Two proxy records in the Kirkby paper in general agreement, that’s all any reasonable person could expect.
Nicola Scafetta (15:27:37) :
“Ulric Lyons (12:17:11) :
arrampicandosi sugli specchi
You are right. However, solving the problem is not impossible! :)”
One may consider the use of acronyms. EUV is duplicitous in its meaning, thus, a ‘mirror’ image of your logic may be apparent to a reader. It’s a good idea to present an ‘acronym’ with the more verbose form in brackets as an introduction of the acronym to your text.
Having said that, this isn’t the current problem.
Best regards, suricat.
Leif Svalgaard (08:08:15) :
Remind me of what that was, so I can comment.
The partial quotation from Steinhilber et al. 2010. Please refer to the exchange at 21:03:58, 21:54:17 and 07:00:20 in the earlier thread, and contained in a summary here at 03:19:26.
Leif: The usual excuse is that the CRs that are important for Svensmark are so energetic [greater than 10 Gev] that the dipole changes don’t matter. That introduces another problem, namely that the is very little solar modulation of CRs above 10 GeV.
Although lower energy CRs are present in greater numbers, they’re less able to penetrate the trposphere, and are more strongly modulated by the geo- and solar magnetic fields. For lower portions of the troposphere, higher energy CRs provide a significant amount of the ionisation – see Usoskin, Gladysheva and Kovaltsov (2004), hyperlinked earlier.
The 11-year solar cycle appears to affect low cloud cover :
“Latitudinal dependence of low cloud amount on cosmic ray
induced ionization”
Usoskin, Marsh, Kovaltsov,Mursula, Gladysheva 2004
“We find that the time evolution of the low cloud amount can be decomposed into a long-term trend and inter-annual variations, the latter depicting a clear 11-year cycle. We also find that the relative inter-annual variability in low cloud amount increases polewards and exhibits a highly significant one-to-one relation with inter-annual variations in the ionization over the latitude range 20–55◦S and 10–70◦N. This latitudinal dependence gives strong support for the hypothesis that the cosmic ray induced ionization modulates cloud properties.”
Correction to 06:01:52 : “Therefore either it’s a coincidence, and or some extra-terrestrial factor is influencing B..”:
That should have been “.. some extra-terrestrial factor is influencing A”.
oneuniverse (17:56:36) :
The partial quotation from Steinhilber et al. 2010.
Ah, the one where they say that the climate-induced effects are not negligible.
Although lower energy CRs are present in greater numbers, they’re less able to penetrate the trposphere, and are more strongly modulated by the geo- and solar magnetic fields. For lower portions of the troposphere, higher energy CRs provide a significant amount of the ionization
I think that was precisely what I said: the higher energy CRs that are significant for ionization are hardly modulated at all, so we would not expect any modulation of ionization.
jinki (16:36:39) :
Two proxy records in the Kirkby paper in general agreement, that’s all any reasonable person could expect.
No, these are extraordinary claims, and require extraordinary evidence, not just cherry picked weak correlations.
Nicola Scafetta (15:19:22) :
In our paper we have never stated that the TSI is “causative”. We have always stated that we are using the TSI records as “proxy” for the sun-climate interaction.
In your presentation [and I refer to that one because it is at a more accessible level for most – on purpose, i presume] of 26 Feb. 2009, you refer extensively to TSI (I count 55 times) and refer to ‘thermodynamic’ forcings.
And what is causing climate change is a mysterious extraterrestrial forcing “X” that for some mysterious reason has geometrical patterns that look quite similar to the patterns found in the TSI records
Not a single time do you refer to forcing ‘X’. Now, it is good if you have abandoned the TSI track, as that was clearly not fruitful. I’m not so sure Mystery-X will be more palatable. After all, we are seeking physical causes.
About the Levy walks. The R&R paper was not my paper. I was extracting and perhaps making a bit more accessible the general ideas as expressed by R&R, including the fact that Levy-walks are not scale invariant. You are claiming that R&R do not understand what a Levy-walk is. The Figure 3 I showed [with the spikes] is from a lecture series by the Rypdals http://web.me.com/kristofferrypdal/Publication_Site/Popular_and_media_files/Birkelandseminar%2014.%20januar.pdf
It is in Norwegian [which I understand very well] and may not be easily decipherable, but you can find the discussion on page 30. Their main argument is that you think that the process is a Levy-walk, while it in fact is a Levy-flight.
It is not a valid objection form you that R&R [and by extension I] do not know what Levy-walks are. I think they [and I] do.
Leif Svalgaard (22:35:05): Ah, the one where they say that the climate-induced effects are not negligible.
Yes – your selective quotation was deceptive, in my opinion. Your quotation established that both a production signal and a climate atmospheric signal exist in the cosmogenic data. The paper goes on to say in the next sentence that the climate signal is not negligable, it is comparatively small compared to the production signal. That’s the very point I was trying to make, and which you were arguing against.
You edited out the salient part of the paragraph – very unscientific.
Leif Svalgaard: I think that was precisely what I said: the higher energy CRs that are significant for ionization are hardly modulated at all, so we would not expect any modulation of ionization.
Modulation does occur, though. The 11-year solar cycle is clearly visible in the instrumental record. The higher energy CRs are modulated less, the lower ones more. Looking at results in Usoskin et al. (2004) paper for polar regions, the lower energy CRs are contributing to troposheric ionisation too, although one would expect the the low energy CR contribution to lessen geomagnetically stronger equatorial latitudes.
Correction to final sentence : “..the low energy CR contribution to lessen at geomagnetically stronger equatorial latitudes.”
Leif Svalgaard (22:35:05) replying to jinki:
No, these are extraordinary claims, and require extraordinary evidence, not just cherry picked weak correlations.
Claims of cherry-picking also require evidence.
There seem to be too many papers finding climate correlations with solar variables for them all to be cherry-picked. It’s possible, but your opinion on the matter is insufficient – provide evidence, please :
– Usoskin, Marsh, Kovaltsov,Mursula, Gladysheva (2004), cited earlier, use the ISCPP D2 cloud cover dataset – are you saying that’s an example of cherry-picking? Is their CR data cherry-picked?
– How exactly are the results of Harrison and Stephenson (2006), also cited earlier, based on cherry-picked data?
Leif Svalgaard (22:35:05) :
as usual you try to not understand the issue. Those spikes emerge in the integrated signal if you look at it at different diffusion times because a Levy walk is constrained.
In any case, in Levy-Walk, what is distributed with an inverse power law is the waiting time between two events. R&R instead took the amplitude of the SFI as the index, which is a measure of the daily energy emitted by the solar fare that we explicitly excluded. SFI amplitude is too a Levy signal, but it is not what we were talking about.
To let you to understand the difference between the waiting time distribution and the SFI distribution of increments as R&R use you need just to consider that a waiting time distribution is defined only for “Delta tau”>0, while for a SFI distribution of increments “Delta X” the distribution is defined for both positive and negative values (that is they are symmetric around zero) which are the ones that R&R shows in their figure 1e and 1d.
This yields to a complete misunderstanding of the math that needs to be used because they have confused the waiting time between the events for the amplitude of the single event. It is like confusing the frequency of a sinusoidal signal for its amplitude!
oneuniverse (03:04:21) :
Yes – your selective quotation was deceptive, in my opinion.
Deception implies intent, so is a severe accusation, but such seems to be the order here.
it is comparatively small compared to the production signal.
The paper was concerned with relatively small variations in climate. The context of our discussion was much longer time scales where much larger climate variation will have occurred. E.g. a glaciation.
Looking at results in Usoskin et al. (2004) paper for polar regions, the lower energy CRs are contributing to troposheric ionisation too, although one would expect the the low energy CR contribution to lessen geomagnetically stronger equatorial latitudes.
Cloud cover in the polar region has very little impact on the climate [albedo is already high because of the snow and ice and solar insulation is already very small]. The important region is precisely the equatorial one.
are you saying that’s an example of cherry-picking?
Perhaps not by them, but certainly by you. How many papers have you cited that show no evidence of CR effect? And there are many, especially newer ones. You have not cited a single one…
Nicola Scafetta (05:12:25) :
as usual you try to not understand the issue
I was explaining the paper by R&R, so you’ll have to claim that they did not understand the issue.
R&R instead took the amplitude of the SFI as the index […] This yields to a complete misunderstanding of the math that needs to be used
The sticking point is not just about SFI [which is a Levy-flight and not a walk], but also whether the global temperature is a FBM process or a Levy-walk. You have to specifically show where their mistake is rather than just claim that they don’t understand anything.
Nicola Scafetta (05:12:25) :
as usual you try to not understand the issue.
I think that my extensive comments here are good indicators that I am trying to understand the issue. But on your part, instead of saying that people don’t understand anything, you should better use your energy on furthering their understanding. E.g. one of R&Rs criticisms is that you do not show the PDFs [as they did in their Figure 2]. This would seem to be easily countered: simply show them.
About the SFI and amplitudes vs. frequencies. The amplitude of the SFI by its definition includes a frequency component: If many flares occur on a given day [or month] that is the waiting time is short, the amplitude of the SFI will be high. If anything, that might indicate that the SFI is inappropriate for this kind of analysis because of mixing of amplitude and frequency components.
Leif Svalgaard (06:13:55) :
Ok, now I see that you are trying to understand. You have finally acknowledged that “SFI [which is a Levy-flight and not a walk]”. Because we are talking about “Levy Walk” and not “Levy Flight” R&R took apples for oranges. So SFI is inappropriate for the analysis.
The problem now is that if you think that the phenomenon is described by “apples” you will think that what matters are the local increments and believe that the smooth component of the signal can be removed without problems. This is what R&R did.
If you realize that the phenomenon is described by “oranges” you will understand that what matters is the smooth component, not the increments which can be also random Gaussian noise. So the smooth component of the temperature signal can not be removed without changing the physical properties of the signal.
In conclusion R&R is misleading because their methodology did not disprove what we were talking about, they just misunderstood the physics of our arguments!
The waiting time PDF relative to the solar flare intermittency is discussed in my longer paper on the issue, see figure 5 in http://www.fel.duke.edu/~scafetta/pdf/PRE46203_02.pdf
As you can see the distribution is quite different from what R&R plot.
The PDF cannot be taken for the temperature because you need to study the temporal patterns, not its increments. This is better done in the frequency domain.
suricat (17:12:30) :
… think I get the idea. We’re looking for a frequency that isn’t there in the solar signal, to match an observed temperature anomaly frequency. However, as ocean surface is ~70% of Earth’s surface with an albedo of ~0.9 (ignoring cloud shadow), wouldn’t the major ‘coupling’ be to ocean surface temperatures, or perhaps near [sub] ocean surface temperatures? As the ocean surface coupling would likely be swamped at times in tropical latitudes and result in hurricanes and typhoons (the ‘overheated’ thermostat).
So. Why are we looking for this coupling in land near surface temperatures?
Seasonal changes in global temperatures obviously are caused by changes in the solar radiation received by the different hemispheres. Whilst this is much greater than TSI variation it causes a 40C variation in UK that is exactly synchronized but phase shifted by about a month with the changing radiation patterns. The oceans can act as a low pass filter on the temperature variation. But frequencies of 1 cycle per year are still allowed to give seasons.
TSI is on 11 year cycle I do not see why the low pass filter effect of the oceans should not allow the signal to be visible. It is not however!
\harry
Thanks for your input Harry, but I’m not convinced that we are not “trying to read a multi-meter at far less than half scale needle deflection”! Then again, perhaps we’re both saying the same thing, but in other words.
IMHO, when we see a ‘flight’ in a ‘levy walk’ we can conclude that the sample resolution is lower than the ‘walk’ event on a temporal scale. IOW, the sample rate is good for some of the ‘levy walk’ actions, but too slow to capture the more frequent activity of the ‘walk’ which then becomes ‘a flight’!
Best regards, suricat.
Nicola Scafetta (07:31:29) :
acknowledged that “SFI [which is a Levy-flight and not a walk]“. Because we are talking about “Levy Walk” and not “Levy Flight” R&R took apples for oranges.
The mistake is to try to model SFI as a Levy-walk. The SFI is fundamentally memory-less [like the deaths of soldiers from being kicked by horses in 19th century studies of exponential distributions]
Leif Svalgaard (06:13:55): “Perhaps not [cherry-picking] by them, but certainly by you. How many papers have you cited that show no evidence of CR effect? And there are many, especially newer ones. You have not cited a single one…”
You accused the authors of cherry-picking – please provide some evidence instead of changing the subject! You’ve admitted that the two papers I mentioned do not use cherry-picked data and they find significant correlations.
The Kirkby 2008 paper I cited is a review – all you needed to do was to read the papers I cited to get an idea of the range of papers for and against. Most papers tend to review the field, mentioning different or conflicting results.
I don’t notice you following your own advice when you cite papers in the comments, by the way.
Leif: “Cloud cover in the polar region has very little impact on the climate [albedo is already high because of the snow and ice and solar insulation is already very small]. The important region is precisely the equatorial one.”
See Usoskin, Marsh, Kovaltsov,Mursula, Gladysheva (2004) mentioned just earlier: “We also find that the relative inter-annual variability in low cloud amount increases polewards and exhibits a highly significant one-to-one relation with inter-annual variations in the ionization over the latitude range 20–55◦S and 10–70◦N. ”
Harry Lu (18:01:15) : “TSI is on 11 year cycle I do not see why the low pass filter effect of the oceans should not allow the signal to be visible. It is not however!”
White et al 2007 (” Response of global upper ocean temperature to changing solar irradiance”) seemed to find the 11-year solar signal in upper layer ocean temperatures.
In order to preempt another accusation of cherry-picking from Dr. Svalgaard for citing a paper with evidence for some idea, I’d better recommend that you carry out a thorough review of all the pertinent literature in the field before drawing any conclusions from the cited paper.
Mistake in my previous 04:29:36 comment to Harry Lu post – the paper is White et al. 1997, not 2007.
————————————————————
‘………..The 11-year solar cycle appears to affect low cloud cover :
“Latitudinal dependence of low cloud amount on cosmic ray
induced ionization”
Usoskin, Marsh, Kovaltsov,Mursula, Gladysheva 2004
“We find that the time evolution of the low cloud amount…………….
————————————————————
I am Puzzled by the apparent correlation in say, some Amazon Delta rivers and solar activity but not, presumably, in all the Amazon Delta rivers falsifying the correlation. Quite accepted—thanks Leif.
Here’s my question:
Is the earth’s atmosphere acting like a eye-ball and lensing
and focusing a .1K input at the highest level to a 1K affect in a very localised
band according to the density of the air in the various levels of Wilde’s atmosphere?
Just a guess and mean no offence if this is too retarded to post on this impressive thread. But, i’ve not seen this phenomenon mentioned. Just an idea. Never under estimate the power of distributed processing is my defense. Pablo Maus would be delighted if his river analysis wasn’t cracked out of the park with one sentence! 🙂
Thanks for all this posting. In a way, it’s a great public service to the public as I see alot of hard work in put here, to educate us to the intricacies beyond the ‘concencus’.
oneuniverse (04:29:36) :
You accused the authors of cherry-picking
I accused you of cherry picking, not the authors
Most papers tend to review the field, mentioning different or conflicting results.
Now you admit that the papers admit to conflicting results.
I don’t notice you following your own advice when you cite papers in the comments, by the way.
Leif: “Cloud cover in the polar region has very little impact on the climate [albedo is already high because of the snow and ice and solar insulation is already very small]. The important region is precisely the equatorial one.”
It doesn’t matter much for the climate how sensitive the polar regions may or may not be. The climate comes from where the action is: low latitudes.
Now, instead of just reviewing papers or reviewing reviews of papers that review yet other papers, there are scientists out there [ourselves included] that are actually trying to do science and advance the field. I and my colleagues have been involved in the CR modulation field for decades. We just presented a paper at the second workshop on cosmic rays in the heliosphere http://www.issibern.ch/workshops/cosmicrays/
Click on ‘talk’. Then go to Thursday and take a look at ISSI 2010 Wieler GCR long term variations .ppt and there are many other goodies. worth spending your weekend looking at all of them, actually.