|
|
Post by lsvalgaard on Mar 29, 2010 20:42:04 GMT
This is AR9169, the largest region that appeared during Cycle 23. Is it still your contention that activity is "smeared" uniformly around the sphere? Yes, if you plot activity over time with an incorrect longitude system [not matching what the Sun actually does], then it will be smeared. Image a spot that appeared ALL the time but recurring at 27.0000 days rather than at 27.2753 days, then after one rotation it is 0.2753 days ahead, after two rotations 0.5506 days ahead, after n rotations n*0.2753 days ahead. Setting n*0.2753 = 27.0000, you will see that for n = 27.0000/0.2753 = 98. So after 98 rotations or 7.25 years, activity will have appeared all the way around the Sun. |
|
|
|
Post by SDJ on Mar 29, 2010 21:01:22 GMT
But the "smearing" is only the result of possible observational error, which does not affect the actual phenomenon of the activity 93 million miles away.
|
|
|
|
Post by lsvalgaard on Mar 29, 2010 21:44:57 GMT
But the "smearing" is only the result of possible observational error, which does not affect the actual phenomenon of the activity 93 million miles away. No, the smearing is not the result of observational error, but of an error committed by the person trying to make a graph of where active regions are a function of longitude, but using the wrong rotation period. |
|
|
|
Post by SDJ on Mar 29, 2010 23:36:28 GMT
I believe your concept of the "wrong" rotation period would be the observational error, if it is true. The observations were made by people based on the accepted rotation period. I haven't located any datasets yet that are based on your rotation period. Regardless, does the hypothesized "wrongness" render something like this meaningless?  To focus on Cycle 24 spots, I filtered out everything less than 10 degrees from the equator and the southern hemisphere as a whole. I can do the south later. |
|
|
|
Post by lsvalgaard on Mar 30, 2010 1:15:50 GMT
I believe your concept of the "wrong" rotation period would be the observational error, if it is true. The observations were made by people based on the accepted rotation period. I haven't located any datasets yet that are based on your rotation period. Regardless, does the hypothesized "wrongness" render something like this meaningless? To focus on Cycle 24 spots, I filtered out everything less than 10 degrees from the equator and the southern hemisphere as a whole. I can do the south later. There are no observational errors. The longitude is completely arbitrary and has no physical meaning. It is trivial to convert from one rotation period to another. If you used the 'correct' rotation period, the clustering might be even MORE impressive. |
|
|
|
Post by SDJ on Mar 30, 2010 1:31:55 GMT
I believe the notion of Carrington longitude is what is called a "convention". It may be faulty, but it at least provides the basis for discussion.
The freezing/boiling points of water are conventions. They could be redefined as 50F/500C or whatever, they just serve as a basis.
More precise measurements of the rotation period of the Sun would be welcome, but we just have to go with what we have at the moment.
My starting point in all of this is that solar activity is non-uniformly distributed about the sphere, derived initially from observing lots of images that seemed to me to show that the Sun is indeed lopsided. I thought I'd look further into that idea, and I've done what I can. I don't plan to stop.
I appreciate your input, though, and look forward to the availability of more accurate data on the question.
|
|
|
|
Post by lsvalgaard on Mar 30, 2010 3:05:09 GMT
I believe the notion of Carrington longitude is what is called a "convention". It may be faulty, but it at least provides the basis for discussion. The freezing/boiling points of water are conventions. They could be redefined as 50F/500C or whatever, they just serve as a basis. More precise measurements of the rotation period of the Sun would be welcome, but we just have to go with what we have at the moment. My starting point in all of this is that solar activity is non-uniformly distributed about the sphere, derived initially from observing lots of images that seemed to me to show that the Sun is indeed lopsided. I thought I'd look further into that idea, and I've done what I can. I don't plan to stop. I appreciate your input, though, and look forward to the availability of more accurate data on the question. It has been known for a century that the Sun is lop-sided'. It is not a question about 'more accurate data', but about analyzing them correctly. You shouldn't stop, but just do the analysis right. The trick is to calculate the longitude in that other rotation period. Here is how you do it: Given the Carrington rotation number C and longitude L, you do this: 1: D = C * 27.2753 gives you number of days since Carrington started 2: E = (360-L)/360*27.2753 gives you number of days from the start of the rotation [time runs backwards within the rotation, 360 being the start, 180 halfway, and 0 the end] 3: F = D + E is position in days 4: G = INT(F/27) gives you rotation number in the 27-day system 5: H = F - G*27 gives you days from start of the rotation 6: I = 360 - 360*I/27 gives you the longitude in the 27-day system Now you proceed as before counting up spots in longitude bins. By changing '27' to a different number, like N=26.85, you get longitudes in that system, etc. Then you do the analysis for N varying from, say, 26 by 0.001 to 28 and computing a score for clustering for each. See for which N that score is highest. That is the fundamental period [if there is one]. See? |
|
|
|
Post by SDJ on Mar 30, 2010 4:08:01 GMT
I just wanna know when the next sunspot is gonna happen!  Actually, the Debrecen data might help you with your hypothesis about the proper rotation period. They're a bit sloppy in their records, so if it was a large active region that had many spots, they numbered those that day. Then the next day they numbered them again, but apparently did not refer to the previous day's numbering, so 1 becomes 40 and 38 becomes 2. I plan to complain the next time I'm in Hungary. Since they were doing the recording of relative umbra/penumbra sizes, if you could pick out the largest spot in groups by umbra or penumbra size and track it over the period it was visible, and could match that with a reappearance on any subsequent rotations, with their recorded observations of longitude, that might give you some good fuel for nailing down the true rotation period that you're working on. It would take some effort, though. Fortunately, they do have a huge set of photo images available to help with the numbering thing, but those take a while to get. |
|
|
|
Post by lsvalgaard on Mar 30, 2010 4:26:53 GMT
I just wanna know when the next sunspot is gonna happen! Actually, the Debrecen data might help you with your hypothesis about the proper rotation period. They're a bit sloppy in their records, so if it was a large active region that had many spots, they numbered those that day. Then the next day they numbered them again, but apparently did not refer to the previous day's numbering, so 1 becomes 40 and 38 becomes 2. I plan to complain the next time I'm in Hungary. Since they were doing the recording of relative umbra/penumbra sizes, if you could pick out the largest spot in groups by umbra or penumbra size and track it over the period it was visible, and could match that with a reappearance on any subsequent rotations, with their recorded observations of longitude, that might give you some good fuel for nailing down the true rotation period that you're working on. It would take some effort, though. Fortunately, they do have a huge set of photo images available to help with the numbering thing, but those take a while to get. I worked out the period(s) 36 years ago: 26.85 days during the 1st half of the cycle and 27.15 during the last half of the cycle. |
|
|
|
Post by SDJ on Mar 30, 2010 19:06:28 GMT
Here's the matching plot for the one above, for the Southern Hemisphere. Same filtering as before, nothing less than 10 degrees from the equator, other hemisphere also removed.  At first glance, there are fewer spots in the southern compared to the northern hemisphere, and the "clustering" is in different locations. |
|
jinki
Level 3 Rank
Posts: 123
|
Post by jinki on Apr 3, 2010 14:18:59 GMT
I just wanna know when the next sunspot is gonna happen! I worked out the period(s) 36 years ago: 26.85 days during the 1st half of the cycle and 27.15 during the last half of the cycle. Does this statement suggest a varying solar rotation rate over the solar cycle? |
|
|
|
Post by SDJ on Apr 3, 2010 14:41:01 GMT
Today's Carrington Longitude at the centerline is ~250 degrees. That northern hemisphere activity "cluster" in the above plot fits pretty well with that huge, complex coronal hole in the SOHO images.  This region is where the greatest portion of concentrated activity has occurred since the start of Cycle 24, and the scars are there to show for it. Interestingly, it's also the same neighborhood where AR 10981, the first Cycle 24 spot group, appeared. |
|
clive
Level 2 Rank
Posts: 50
|
Post by clive on Apr 3, 2010 15:01:18 GMT
SDJ is on the ball...this longitude has been the strongest during solar cycle 24, producing 1035, 1040 & 1045
|
|
|
|
Post by lsvalgaard on Apr 3, 2010 15:42:17 GMT
I just wanna know when the next sunspot is gonna happen! I worked out the period(s) 36 years ago: 26.85 days during the 1st half of the cycle and 27.15 during the last half of the cycle. Does this statement suggest a varying solar rotation rate over the solar cycle? Yes. Although the rate is only a surface variation. The interior of the Sun rotates [AFAWK] at a constant rate. |
|
|
|
Post by SDJ on Apr 3, 2010 19:03:01 GMT
If anyone is interested in looking at the original data, I just finished adding the Carrington Rotation numbers to the Excel file that contains my collation of the Debrecen data for Jan. 2008 to March 2010. This is the "spots" data, i.e., the daily observations for each individual sunspot, grouped by Active Region number. I listed the data fields a few days ago.
Since it covers a fairly quiet period, the file size isn't that large, 517 KB whole, 155 KB zipped, so it shouldn't do any damage to most email accounts.
Loaded into Excel, you can do any sorting you'd like to do, by latitude, spot size, longitude, etc., and you can use the simple plotting functions that Excel offers to have a look at anything that might interest you.
Just drop me a private message or contact me at my listed email address and I'll get it off to you.
|
|
