The Solar Dynamics Observatory satellite captured this image yesterday. A massive filament on the Sun erupted in a stunning display as seen here in the videos below.
The giant solar eruption created a long filament of magnetic plasma, which extended an astounding 435,000 miles (700,000 kilometers). This is nearly twice the distance between the Earth and the moon, which is about 238,857 miles (384,403 kilometers)
The location of the eruption and the magnetic field configuration and direction of the eruption suggest the impact of the event on the Earth will be limited. Watch the videos below to see it happen:
Here’s a second video, showing closer detail:
From YouTube: A very long solar filament that had been snaking around the Sun erupted today (Dec. 6, 2010) with a flourish. NASA’s Solar Dynamics Observatory (SDO) caught the action in dramatic detail in extreme ultraviolet light of Helium. It had been almost a million km long ((about half a solar radius) and a prominent feature on the Sun visible over two weeks ago before it rotated out of view. Filaments are elongated clouds of cooler gases suspended above the Sun by magnetic forces. They are rather unstable and often break away from the Sun. Note: the edge of the moon can be glimpsed at 0300 UT during a brief lunar transit.
Wow.
That filiment closeup youtube reminds me of what was captured in this youtube clip:
[youtube=http://www.youtube.com/watch?v=vqgNrj6oEdc&fs=1&hl=en_US]
If it weren’t happening in more or less a vacuum, I wonder what it would sound like !
regarDS
The L&P Effect is showing us that we have more specks. Nothing new here.
In the early days these measurements were done properly by Solanki. He and his students only measured large spots which gave us a good example of the Suns potential magnetic output. L&P today give the same weighting to a large area like 1131 and a small speck that is magnetically irrelevant. By including all the speck measurements (which are increasing) the magnetic value must decrease.
I have measured every sunspot group of SC24 that passes the Layman’s Count threshold (unlike L&P) using a contrast measurement not unlike one of the L &P measures. I also have the luxury of measuring the highest magnetic strength per group which L&P cannot achieve as their data selection is constrained. My results are in complete disagreement and show that the magnetic strength of a weak SC24 is still increasing towards cycle max.
Link to the latest graph HERE.
Christoph Dollis says:
December 7, 2010 at 11:34 pm
the idea that the universe itself may be a lot more pervaded with plasma and electricity than first thought
It has long been known that most of the stuff we can see is in the plasma state. It has also long been known that there are no electric fields in the rest frame of that plasma [it simply shorts out]. And that when plasma meets a magnetic field, transient electric currents are induced with all kinds of effects [e.g. aurorae]. The issue with the EU and PU is what drives what? Large persistent electric currents need a persistent electromotive force to be maintained [otherwise the battery goes flat].
the Sun traversing different regions of our galaxy may cause fluctuations — through some process or other — of the Sun’s luminosity, Sunspot frequency, coronal mass ejection events
The Galaxy is big and the Sun travels very slowly. It takes thousands of years to make traverse or enter a different region. Second, the solar wind blows at a speed 11 times greater than that which electric or magnetic changes can propagate upstream to the Sun.
We don’t come close to understanding all there is to know about the solar system, plasma, comets, or
But we do know quite a lot. ‘All’ is a big word.
for that matter, the Sun (why for example is the center of a Sunspot cooler than the edge and why is the Sun’s surface 6,000 K in temperature, but the corona between 1 and 3 million?).
Those things we do know something about. Sunspot is cooler because its magnetic field slows the convection of heat from the interior. For the corona there are several theories that explain its high temperature. Our problem here is that we do not know which one(s) are dominant. Perhaps they all work to various degrees.
Sir Arthur Eddington — himself said the competing theory was that the Sun drew its energy not from an internal nuclear furnace, but from the surrounding ether, the galaxy, by some unknown process.
Well, perhaps you can show us where he said that in papers published after the nuclear processes have been explained [~1940]. Today we directly observe the expected number of neutrinos generated by the nuclear furnace, so there is no need to invoke ‘unknown forces’.
if the interior is really where all the heat has been generated.
Nobody thinks that ALL the heat is generated in the interior. Some one part in a million is generated in the surface layers.
There are two more movies associated with this filament eruption available
here and
here.
These movies came from a listing of recently flagged events by members of the SDO/AIA team here. Interested parties can subscribe to an RSS feed there as well.
“rob m. says:
December 7, 2010 at 1:42 pm
Is that first video in real time?”
Based on the time stamps, it looks like 1 sec of video equals about 1 hr of real time.
Leif, does volcanic activity affect geomagnetic aa index?
Paul Vaughan says:
December 8, 2010 at 8:34 am
Leif, does volcanic activity affect geomagnetic aa index?
No.
Not unless a volcano erupts just under either of the two observatories [Hartland, UK; and Canberra, AU], which I don’t see happen as they are both in geologically stable terrains. 🙂
John T says:
December 8, 2010 at 8:14 am
Based on the time stamps, it looks like 1 sec of video equals about 1 hr of real time.
Even without timestamps one can estimate this. The solar radius is 700,000 km. If the ejected matter moves at solar wind speed, 400 km/s [it does not quite as there is further acceleration a bit further out], the radius [which is about the size of eruption] is traversed in 700,000/400 = 1750 seconds [about half an hour].
Re: Leif Svalgaard
Can you think of any reason for clusters of temporally-indexed 1 to 7 year aa index power spikes (complex mexican hat wavelet) to match up with volcanic indices? The pattern, while not stunning, suggests at least some kind of indirect coupling.
Paul Vaughan says:
December 8, 2010 at 10:23 am
Can you think of any reason for clusters of temporally-indexed 1 to 7 year aa index power spikes (complex mexican hat wavelet) to match up with volcanic indices? The pattern, while not stunning, suggests at least some kind of indirect coupling.
The most plausible reason is ‘looking at twenty random things, one of them will be significant at the 95% level’.
There have, of course, been people finding correlations between volcanism and solar activity. E.g. http://lasp.colorado.edu/sorce/news/2008ScienceMeeting/doc/Session4/S4_03_Crowley.pdf
See slide 27.
There are all kinds of hare-brained ‘explanations’: LOD, barycenter, electric forces, ‘unknown forces’, you know the drill. Many of these have been discussed on WUWT over time. I won’t put any credence in any.
It is likely that volcanism influences the deposition of 14C and 10Be and thus creates a seeming solar activity connection. And there is this ‘logic’: volcanoes influence climate, solar activity influences climate, ERGO solar activity triggers volcanoes [or is it the other way around?]. A stone cannot fly, you cannot fly, ERGO, you are a stone! [people with the ‘open mind’ syndrome would counter that stones sometimes fall from the sky … so the premise (that they can’t fly) must be false]. Volcanoes even, at times, fling stones into the sky 🙂
Geoff Sharp says:
December 8, 2010 at 6:22 am
The L&P Effect is showing us that we have more specks. Nothing new here.
The L&P effect is simply that ordinary spots turn into specks.
I have measured every sunspot group of SC24 that passes the Layman’s Count threshold (unlike L&P)
So you bias the selection from the start to the larger spots, invalidating the result.
L&P do not say that the largest spots are decreasing in strength. For your enjoyment I repeat an earlier comment:
The L&P effect [slide 10 and an up-to-date version here http://www.leif.org/research/Livingston%20and%20Penn.png ] is that the average magnetic field strength also seems to have gone down the past 15 years. As the field goes down, it cools the sunspot less and the spots are getting warmer [thus radiating more] and their contrast to the surrounding photosphere is getting weaker, making the spots more difficult to see. When the field strength falls below 1500 or so Gauss, the spots have such low contrast that they are effectively invisible [at least with the small telescopes that are used for the sunspot count] or at least much harder to see, leading to the observer missing many of the smallest spots. The largest spots will always be visible [so there will always be some LARGE spots – during the Maunder Minimum all the spots seen were large spots as far as we (or I) know]. The SSN number is mainly determined by the number of small spots [there are more of those than of large spots]. Here is the observation on 24th Dec. 1957: ftp://howard.astro.ucla.edu/pub/obs/drawings/1957/dr571224.jpg
This is most spotted sun ever observed. In 2003, the L&P effect was beginning to take effect and there were even more smaller spots than large ones: ftp://howard.astro.ucla.edu/pub/obs/drawings/2003/dr031029.jpg
If L&P are correct [and everything so far suggests that – but extraordinary claims require extraordinary evidence, so we need to wait and see until everybody is convinced] even these small spots will disappear and we may be left with only large spots [but few of them] like now: http://www.spaceweather.com/images2010/08dec10/hmi200.gif
The strongest argument for L&P is the growing discrepancy between F10.7 and the SSN [seen in both Japanese and Canadian (as well as combined) observations]. The second strongest argument is the persistence of cosmic ray modulation even when no or few spots were observed [Maunder and Spoerer minima]. And the third strongest is the disappearance of the OH lines. Then comes the actual measurements [and you can disregard the earliest if you like as they don’t matter]. Bill Livingston does not a agenda in this.
Isn’t it just about the right timing now for some genius to figure out, beyond a doubt, that the odd solar activity REALLY is because the sun is about to go supernova within the next ten years. As a result, there is a mad scramble to build interstellar Noah’s Arks to save the human race, and the mass diaspora begins?
Sorry, just couldn’t resist, must be a result of inhaling all those sci fi novels, libraries worth, as a kid!
“”””” Leif Svalgaard says:
December 7, 2010 at 5:06 pm
George E. Smith says:
December 7, 2010 at 4:03 pm
Question. just how much REAL TIME is in those videos ??
Several hours. Watch the time ticker at the bottom of the frame “””””
Thanks Leif; that puts it in context.
If you ever get invited to go there, I would turn it down; it looks too damn dangerous to me; but spectacular.
Leif, you may recall our discussion about Perry’s mysterious lags (e.g. 32 years, aa & Mississippi flow, 2007 article). I believe his lags will disappear if he looks at the temporal evolution of how El Ninos & La Ninas line up with NH summer/winter (SH winter/summer). Coincident with the 30s Dust Bowl, there’s a sharp switch in summer-winter contrasts. Temperature-precipitation relations vary seasonally & regionally. For example, over large portions of the northern hemisphere, the sign of the relationship is opposite in winter to that in summer (but there are notable regional exceptions). The absolute temperature & pressure (not just anomaly) is relevant for clouds & precipitation and therefore solar-terrestrial relations researchers like Perry might be wise to condition their analyses accordingly. NPI from ~1925-1945 should be a big clue, particularly considering the 0 lag Aleutian Low region mechanism suggested by Perry. (I would toss away all the speculation about 32 year ocean loops and focus instead on on the evolution of 0 lag temperature & pressure by season.)
Thanks for the notes on volcanism/geomagnetism relations (or lack thereof!)
Paul Vaughan says:
December 8, 2010 at 11:59 am
… from ~1925-1945 should be a big clue….
Hi Paul
You mean something like this:
http://www.vukcevic.talktalk.net/CET3.htm
Leif Svalgaard says:
December 8, 2010 at 11:36 am
So you bias the selection from the start to the larger spots, invalidating the result.
Incorrect, only the SINGLE specks/small spots are isolated via the threshold. A measured group that has mixture of spot/speck sizes has the total area measured, the smaller spots influencing the total darkness number but by not measuring each speck on its own the record is not spurious.
The strongest argument for L&P is the growing discrepancy between F10.7 and the SSN [seen in both Japanese and Canadian (as well as combined) observations]. The second strongest argument is the persistence of cosmic ray modulation even when no or few spots were observed [Maunder and Spoerer minima]. And the third strongest is the disappearance of the OH lines. Then comes the actual measurements [and you can disregard the earliest if you like as they don’t matter].
There is always a small discrepancy between F10.7 flux and SSN. High flare activity in 2002 is one cause, even so the Canadian flux record is very close to the SSN record. There is no need to add the Japanese record to skew the results.
Plage and coronal hole activity will still modulate cosmic rays during a solar grand minimum. Your continued reference to a theory that is flawed and without mechanism does not help your reputation. The L&P theory is pseudo-science.
Geoff Sharp says:
December 8, 2010 at 2:23 pm
Incorrect, only the SINGLE specks/small spots are isolated via the threshold. A measured group that has mixture of spot/speck sizes has the total area measured, the smaller spots influencing the total darkness number but by not measuring each speck on its own the record is not spurious.
A single speck/spot has a sunspot count of 11, so removing that has a large effect. The large spot has a large influence on the area, but the magnetic flux comes from not just the large spot but from all the little specks around it [polarities must balance]. Today’s large spot is a good example: http://sdowww.lmsal.com/sdomedia/SunInTime/2010/12/08/l_HMImag.jpg
Your area measure would miss all the white fields to the left of the big black spot.
As I have said earlier, your count is misconceived and badly executed [e.g. uncalibrated] and is basically junk, but may serve your purpose for which it was apparently designed.
There is always a small discrepancy between F10.7 flux and SSN.
I’m talking about a systematic difference. And by definition, there is not a systematic difference as the conversion formula is a least squares fit.
Canadian flux record is very close to the SSN record. There is no need to add the Japanese record to skew the results.
Ken Tapping discovered the difference using the Canadian record. The Japanese record simply confirms his finding.
Plage and coronal hole activity will still modulate cosmic rays during a solar grand minimum.
You only have plages and coronal holes when you have magnetic fields so the dynamo must still be working. If these magnetic fields are strong enough, spots form. Not above 1500 G, no spots form or are visible, simple as that. The fields cannot be much weaker than 1500 G, because then we would lose the modulation.
Your continued reference to a theory that is flawed and without mechanism does not help your reputation. The L&P theory is pseudo-science.
I need no help wrt to my reputation. L&P is not theory, but observational fact so far. You may call it an ‘inconvenient truth’. There are mechanisms if you need some. E.g. sunspots are formed by coalescence of small pores and specks driven together by convective motions, which in turn depends on temperature differences. In a complicated system [also true for the Earth’s atmosphere] natural fluctuations occur stochastically, no more is needed.
Geoff Sharp says:
December 8, 2010 at 2:23 pm
Your continued reference to a theory that is flawed and without mechanism does not help your reputation.
I need no help wrt to my reputation. L&P is not theory, but observational fact so far. You may call it an ‘inconvenient truth’. There are mechanisms if you need some. E.g. sunspots are formed by coalescence of small pores and specks driven together by convective motions, which in turn depends on temperature differences. In a complicated system [also true for the Earth’s atmosphere] natural fluctuations occur stochastically, no more is needed. A recent paper has more on how this can happen:
http://onlinelibrary.wiley.com/doi/10.1002/asna.200911311/pdf
Leif Svalgaard says:
December 8, 2010 at 3:10 pm
Your area measure would miss all the white fields to the left of the big black spot.
As I have said earlier, your count is misconceived and badly executed [e.g. uncalibrated] and is basically junk, but may serve your purpose for which it was apparently designed.
If tiny specks are produced next to a large area they are all included in the darkness measurement, each individual pixel is taken on it merit ie its contrast strength in the green channel. So nothing is missed unless L&P are now measuring plage?
The threshold makes very little difference to the darkness measurement as it only removes SINGLE specks and small spots so your comments are unwarranted. The “junk” counting method just happens to come very close to your perceived value for the sunspot record if you were going to discount it…
Meanwhile the F10.7 flux value stays at a low 84.6 while at the same time the area and darkness measurement is close to a SC24 record. Unipolar groups are dominating, and their affect on flux production might mean we see an opposite divergence this cycle (SSN higher than F10.7). It is certainly starting to show in the SIDC record.
Geoff Sharp says:
December 8, 2010 at 3:47 pm
If tiny specks are produced next to a large area they are all included in the darkness measurement, each individual pixel is taken on it merit ie its contrast strength in the green channel. So nothing is missed unless L&P are now measuring plage?
the total area of the tiny specks is so small compared to the large spot that they don’t matter. For 1131 give us the area of the large black spot and the area of the [white] specks, and you’ll see.
Meanwhile the F10.7 flux value stays at a low 84.6 while at the same time the area and darkness measurement is close to a SC24 record.
So is 84.6, the average for 2010 is a low 79.9. And you miss the point: L&P is about disappearance of the small spots.
Unipolar groups are dominating
There are no unipolar groups. All active regions are bipolar. L&P says that the weakest spots are getting weaker. The leader spot is usually the strongest, so a consequence of L&P is that the weaker follower spots should be lost leaving us with more single large spots. This has a large influence on the sunspot number as most of the count comes from the small spots.
Geoff Sharp says:
December 8, 2010 at 2:23 pm
A measured group that has mixture of spot/speck sizes has the total area measured…
It sounds like you have abandoned the Layman’s Sunspot Count [where you simply subtracted the contribution from groups you didn’t like from the SIDC number] and are now simply measuring the area. This is progress. The best way would be to total the unsigned magnetic flux over the disk. This is now very easy with HMI. MWO, and SOLIS providing that information in standard formats [FITS files]. An example of SOLIS data: ftp://solis.nso.edu/synoptic/level3/vsm/time-series/svsm_m11_absmeanfield_mr.jpg [ignore the single spike]. And MWO: http://obs.astro.ucla.edu/150_data.html#plots [but beware of instrument upgrades: http://www.leif.org/research/MWO%20MPSI%20-%20F107.pdf ]. So, modern instruments have very good means of quantifying solar activity. The challenge is to carry that back in time [centuries and more].
Leif Svalgaard says:
December 8, 2010 at 5:23 pm
It sounds like you have abandoned the Layman’s Sunspot Count [where you simply subtracted the contribution from groups you didn’t like from the SIDC number] and are now simply measuring the area. This is progress.
You are getting confused. There are three metrics to consider.
1. The Layman’s Sunspot Count that sets a sunspot threshold of 333 pixels. Counted pixels have a reading less than or equal to 150 in the green channel. (SDO Continuum image). The SIDC count is discounted by the spots/specks that don’t make the grade.
2. The Darkness Ratio measures all pixels in a group that make the grade. So once a group passes the threshold all pixels in that group are measured. Two readings are taken over the entire group, 0-34 in the green channel and 0-70 in the green channel and then a percentage is taken of the totals (SOHO Continuum images). This allows the less magnetic areas in a group to influence the overall value without skewing the data as seen via L&P.
3. DSN value. This has no threshold but measures the total solar face and records all pixels between 0-150 in the green channel. Another measure is also recorded between 0-132 in the green channel to determine the darkness ratio (SDO Continuum). Then the total area is multiplied by the darkness factor (81% translates into 8.1) to give a value which is then multiplied by a conversion factor to bring it back to the SOHO standard. A bit convoluted but the end result I think showing the true potential of SPOT activity.
For 1131 give us the area of the large black spot and the area of the [white] specks, and you’ll see.
Measuring 1131 at 2:30 UTC 9th Dec the main spot has 7633 pixels (up around 300 pixels from yesterday). There are another 8 pixels representing extremely small pores that surround it. In this example the extra pores would contribute very little to the overall darkness ratio value which is correct. If L&P measured the pores it would give 1 spot with a gauss reading over 3000 and the rest around 1600. Bin them altogether like is seen and the daily value is grossly understated. I have used the SDO Continuum image in this example but could do the same exercise with the SOHO image.
Geoff Sharp says:
December 8, 2010 at 7:27 pm
1. The Layman’s Sunspot Count that sets a sunspot threshold of 333 pixels. Counted pixels have a reading less than or equal to 150 in the green channel. (SDO Continuum image). The SIDC count is discounted by the spots/specks that don’t make the grade.
But you don’t use this for anything.
Measuring 1131 at 2:30 UTC 9th Dec the main spot has 7633 pixels (up around 300 pixels from yesterday). There are another 8 pixels representing extremely small pores that surround it.
Since the magnetic flux must add to zero [both polarities cancel], you are assuming that the 7633 pixels have the same total magnetic flux as the 8 small pores. This is clearly nonsense and shows why your measure is not a good measure of solar activity.
Geoff Sharp says:
December 8, 2010 at 7:27 pm
Measuring 1131 at 2:30 UTC 9th Dec the main spot has 7633 pixels (up around 300 pixels from yesterday). There are another 8 pixels representing extremely small pores that surround it.
There is no need for a new and ‘improved’ measure. We have already the total magnetic flux going forward and two good proxies of that, the F10.7 and the Ca-index going back in time. The reasonable task is to figure out how we can extend those as far back as possible. Here, the sunspot number can be helpful, if we can figure out how it relates to those proxies. And here we must accept that that relationship could be changing with time [as it seems to do]. This includes not trying to prop up a measure designed to support a preconceived idea and agenda.
Leif Svalgaard says:
December 8, 2010 at 9:57 pm
Since the magnetic flux must add to zero [both polarities cancel], you are assuming that the 7633 pixels have the same total magnetic flux as the 8 small pores. This is clearly nonsense and shows why your measure is not a good measure of solar activity.
I think you are still confused. I am not measuring the magnetogram. I am measuring the contrast ratio of visible spots via Continuum images, in this case a single alpha spot.