Solar geomagnetic index reaches unprecedented low – only "zero" could be lower – in a month when sunspots became more active

Mr. Alex (11:01:57) :
The value for December was actually 1.41, still 1 in my books though. 😉 http://www.solen.info/solar/
I don’t know where they get their data from [I can make a guess, SWPC], but the official Ap values for December 2009 were:
3-hour ap indices and daily mean value, Ap
Month: December
Year: 2009
Date ap-indices Ap
01-12-2009 0 0 0 0 0 0 0 0 0
02-12-2009 0 0 2 2 2 0 2 0 1
03-12-2009 0 0 0 0 0 0 0 0 0
04-12-2009 0 0 0 0 0 0 0 0 0
05-12-2009 0 0 4 3 3 3 9 7 4
06-12-2009 9 6 3 2 0 0 2 2 3
07-12-2009 5 0 2 2 4 4 5 2 3
08-12-2009 0 0 0 0 0 2 0 2 0
09-12-2009 0 0 0 2 0 0 0 0 0
10-12-2009 0 0 2 2 2 2 2 0 1
11-12-2009 0 0 0 0 0 0 0 0 0
12-12-2009 0 0 3 2 2 3 4 4 2
13-12-2009 0 0 3 3 4 2 3 3 2
14-12-2009 5 15 7 5 3 4 3 2 6
15-12-2009 0 0 2 0 2 0 3 4 1
16-12-2009 7 6 2 4 5 6 3 2 4
17-12-2009 0 2 2 4 3 4 3 2 3
18-12-2009 2 0 4 5 3 4 2 2 3
19-12-2009 4 3 2 2 0 0 0 3 2
20-12-2009 3 2 0 3 3 5 2 0 2
21-12-2009 4 3 2 2 2 2 7 3 3
22-12-2009 3 0 2 3 2 2 6 5 3
23-12-2009 2 5 5 3 2 2 3 5 3
24-12-2009 3 5 3 0 2 0 0 2 2
25-12-2009 0 0 2 2 4 3 6 9 3
26-12-2009 6 7 3 2 2 5 0 2 3
27-12-2009 2 2 4 3 3 3 2 0 2
28-12-2009 0 0 2 3 2 2 0 0 1
29-12-2009 0 0 2 2 2 0 0 0 1
30-12-2009 0 0 2 0 0 0 0 0 0
31-12-2009 *** *** *** *** *** *** *** *** ***
Provisional Monthly Mean Value = 1.9
——————-
It is time to stop all the ‘theorizing’ and whining. I’ll try to explain what is behind the numbers. First, I’ll not be necessarily immodest: I happen to be a world authority on the subject of geomagnetic indices. So here goes:
The various ax indices [aa, ap, am, …] all derive from a measure introduced by Julius Bartels in the 1930s. This so-called K-index [from 0 to 9] is based on the realization that there are two distinct sources of variations of the geomagnetic field [apart from the very slow changes that are due to processes internal to the Earth], namely solar UV and the solar wind. Since the solar UV never falls to zero and since it only hits the Earth during the day at a given location, there is always quasi-regular variation that begins at sunrise and ends at sunset. The magnitude of this variation is of the order of 50 nT or 1/1000 of the Earth’s main field. but is easily measured [and could be measured 200 years ago – the variation was discovered in 1722]. The variation is not entirely regular as it also depends on winds in the ionosphere which in turn are influenced by upwards travelling waves from the near surface troposphere, so the ‘regular’ variation varies a bit from day to day [even if solar UV were constant or zero]. On top of this ‘regular’ variation are effects stemming from interaction with the solar wind. These effects range from almost zero [1 nT or less] to hundreds [or at some places, thousands] of nT, i.e from 1/100,000 to 1/10 of the Earth’ field. The history of this can be found here http://www.leif.org/research/H02-FRI-O1430-0550.pdf which is an invited talk I gave last summer at a IAGA meeting or here http://www.leif.org/research/IAGA2008LS-final.pdf which is a keynote talk I gave at a IAGA workshop in Golden, CO.
The goal of construction of geomagnetic indices is to separate these two competing [and overlapping in intensity] effects into a solar wind part and a solar UV part. This is a problem that has only recently been partly solved [namely by basing the ‘solar wind index’ on night-time data only]. Early on, Bartels saw that an experienced human observer could judge the run of the regular variation to an accuracy of perhaps 5 nT. He therefore devised the K-index [K from ‘Kennziffer’, a German word for ‘index’] as the deviation of the curves from the irregularly varying ‘regular’ variation over a time scale of three hours [the latter chosen as a compromise between compactness and expressiveness, but also based on the observation that the duration of a typical perturbation is about three hours [at mid-latitudes]. It turns out that the three hours is also the time it takes the solar wind to ‘pass’ the Earth’s magnetic field, so no wonder that is the natural time scale for many perturbations.
As 5 nT was the lower limit of ‘accuracy’ of guessing or ‘estimating’ the regular variation, Bartels designated deviations less that 5 nT as K = 0. Then he doubled up, so that K = 1 would be assigned to deviations between 5 and 10 nT, K = 2 to deviations between 10 and 20 nt, and so on. The doubling up produced too few K = 9 values [effective none], so the ‘pace’ was relaxed a bit and the higher values of K lie between limits as a bit less that double the immediately lower K value. Note that K = 0 does not mean ‘no activity’ or ‘no solar wind’, but just that we have given up putting a number to it and all we know is that it was less than 5 nT.
Since K is thus a quasi-logarithmic index it doesn’t make much sense to compute a daily, monthly, or yearly average. To make this possible, Bartels devised the ‘equivalent amplitude’, a which was defined as the deviation of halfway between two k limits, e.g. K = 2 has limits 10 to 20, so a for K = 2 would be 15 nT, and so on. He knew quite well that this would get him into trouble at the ends where K = 0 and K = 9 [there is no upper limit], but since these were thought to be rare he figured that the error would not be large [and would compensate each other a bit], if he assigned a = 2.5 to K = 0 and a = 666 to K = 9. There are other problems with this, for example the assumption that the distribution of values in each interval was linear, namely that there were equally many [for K = 2] values of 10 and 20, as of 11 and 19, as of 12 and 18, etc. This is a reasonable assumption everywhere, except for K = 0 and for K = 9. As we are not really interested [for this topic] in K = 9 [which occurs only a few times in a solar cycle], we shall limit ourselves to K = 0. Careful studies of the ‘fine structure’ of the distribution shows that the statistical average amplitude for K = 0 is not 2.5, but rather near 3.75 nT, so that would have been a better choice. In the end, the 2.5 was rounded down to 2 [which was another mistake]. Now, you can determine K for several stations [about a dozen for ap and two for aa] and take the average of the corresponding amplitudes and then perhaps have some stations help defining the average more meaningfully. Like, all stations reporting K =2 and all stations reporting K = 3 define levels that are definitely [??] lower and higher than if half of the stations reported K = 2 and the other half K = 3. Most of that added ‘accuracy’ is illusory for K = 0 as we just don’t know what ‘real’ level that was ‘measuring’.
Mayaud realized part of the above and chose 2 nT as the lower value of the am-index, so am cannot fall below 2, while ap [in principle, but not with any meaning] could be zero. Bartels knew that ap was not accurate to a 1 nT, so chose to report the index in 2 nT units [another mistake, experience showed, as people just get confused].
To complicate matters, Bartels and Mayaud [and myself] stress strongly that a human observer that has intimate knowledge of the variation at ‘his’ station must be making the judgement of what the ‘regular’ variation is on any given day, as an average curve [Bartels called it an ‘iron curve’] varies much more from day to day than the differences between K = 0, 1 and even 2, so incorrect assessment would introduce significant extraneous variance in the index. Unfortunately, people nowadays try to develop computer algorithms to make the judgement. This effort has not been entirely successful for the small amplitudes and is unnecessary for the large ones, so we have actually regressed in the goal of deriving good ap and aa values.
It is my judgement that the recent very low values [less than 5 for aa and less than 3 for ap are not ‘correct’ in the sense of representing the influence of solar activity on geomagnetic activity, and that speculations based on them therefore are invalid.

Clive E Burkland (15:38:32) :
Total Solar Irradiance (TSI) is in a nosedive not seen since 2008.23:
http://leif.org/research/TSI-SORCE-2008-now.png
No answer on this one?
Perhaps an equipment malfunction?
No, just a sunspot. A good-size spot is dark, so TSI is smaller. As simple as that. With the spot now gone, TSI will shoot straight up again. The typical variation of TSI during the passage of large spot is that when the spot rounds the East limb we see a lot a faculae, so TSI goes up. As the spot rotates onto the center of the disk, the faculae are less visible and the dark area of the spot causes TSI to go down. Then as the spot goes over the West limb, faculae pick up again and TSI goes high before falling back to background when the spot and its faculae are gone for good.

rbateman (19:44:43) :
How in the heck swpc/noaa managed to sqeeze 80 x 10E6 hemispherical area out of the two little smudgy pores is beyond belief,
They try to correct for projection, so right at the limb that correction factor is infinitely large applied to a zero area. Or just inside the limb perhaps 10, so that an apparent 8 comes out 80 [as it should].

Correction II:
the ‘regular’ variation is on any given day, as an average curve [Bartels called it an ‘iron curve’] varies much more from day to day than the
Should have said:
the ‘regular’ variation is on any given day varies much more from day to day to make the use of an average curve [Bartels called it an ‘iron curve’] meaningless [or at least going counter to the goal of the whole whole thing].

Could this low solar activity be a factor in record cold? Here is a link to some interesting pictures;
http://www.news.com.au/pictures/gallery-e6frflv9-1225817312663?page=1
Note the ones in Brighton, Southern England. I have never heard of nor seen snow settle on the beach at Brighton, not even during the really cold winters of the 70’s. WOW!!

My posting on Ap was [of necessity] simplified to bring out the essential point. If you really want the gory details here is the defining paper: ‘The three-hour range index measuring geomagnetic activity’ J.Bartels, N.H.Heck,H.F.Johnston, Terr. Mag. vol 44, p. 411 [1939], which you can also see here: http://www.leif.org/EOS/TE044i004p00411.pdf
Interesting, but tedious [as the subject is arcane].