UPDATE: Some commenters got the wrong idea about this article, see the footnote.
Dear SILSO user,
Mon, Jun 22, 2015 11:42 am
Over the past 4 years a community effort has been carried out to revise entirely the historical Sunspot Number series. A good overview of the analyses and identified corrections is provided in the recent review paper:
Clette, F., Svalgaard, L., Vaquero, J.M., Cliver, E. W.,“Revisiting the Sunspot Number. A 400-Year Perspective on the Solar Cycle”, Space Science Reviews, Volume 186, Issue 1-4, pp. 35-103.
Now that the new data series has been finalized, we are about to replace the original version of our sunspot data
by an entirely new data set on July 1st. On this occasion, we decided to simultaneously introduce changes in several conventions in the data themselves and also in the distributed data files.
There are so many diverse changes that we cannot guarantee that everything will work perfectly on the first try.Our team is too small to make full prior simulations. Therefore, multiple careful consistency checks will be done on July 1st itself, which will slow down the processing. So, please anticipate some delays compared to an ordinary month.
The most prominent change in the Sunspot Number will be the choice of a new reference observer, A.Wolfer (pilot observer from 1876 to 1928) instead of R. Wolf himself. This means dropping the conventional 0.6 Zürich scale factor, thus raising the scale of the entire Sunspot Number time series to the level of modern sunspot counts. This major scale change may thus strongly affect some user applications. Be prepared!
Regarding data files, various files will be replaced by new ones, with new more homogeneous names and new internal column formats. The included information will sometimes change: combining data (e.g. hemispheric numbers together with total numbers), separating data (monthly smoothed numbers in a separate file) or adding new values that were not provided previously (standard errors).
All those changes will be explained in the information accompanying our data, on the web site of the World Data Center SILSO. While the primary files will all be replaced in early July, some other changes will still occur in the next two or three months. During this transitory phase, we thus invite you to visit the SILSO Web site to keep track of the changes, as we are preparing this major transition now scheduled for July 1st, 2015.
An important remark for our faithful observers: the current transition in the sunspot number processing does not change anything to the way you enter your data. So, just proceed as usual on July 1st. Your past k personal
coefficients will simply be recomputed relative to the new re-calibrated sunspot number. We are working on this right now. By the way, the new processing software will open the way towards a better determination of the evolution of each station and so, a better feedback to our observers will become possible in the future.
In the coming weeks, please visit our SILSO Web site:
Royal Observatory of Belgium WDC-SILSO
UPDATE: A number of commenters got the wrong idea about this article, conflating the process with the sort of questionable adjustment techniques For example, Dr. Svalgaard comments:
As the text says all observers should continue the way they have always done. There is no such as ‘the traditional count’ for observers. That concept is completely local to the SIDC [now SILSO]
Frederick Colbourne adds:
These adjustments (at the very least) compensate for the faulty decision by one important observer to use an instrument that did not have sufficient resolving power to count the sunspots properly.
Willis Eschenbach sums it up:
Dear heavens, this resistance to correcting the mistakes of the past is most peculiar. Mosh is quite correct. The sunspot count of the past was differently calculated, due to changes in counting methods which are both well known and well explained.
What they have now done is to use the same methodology from start to finish.
Look, there have been some bogus “adjustments” to climate records by various miscreants. But that doesn’t mean we can just use what we have in front of us in any field. Sunspots are a good example. We know where we changed methodology in the past. We know the dates that calculation method changed, and how the method changed. As a result of the change we have two incompatible sets of numbers.
So should we just continue to use the existing sunspot dataset, which consists of two sets of DIFFERENT NUMBERS which were calculated in DIFFERENT WAYS and then just spliced together? That would be nuts, no?
Instead what we need to do, and what Leif and the others did, was to go back to the underlying observations, and to use a single unified clearly-defined method of counting sunspots from the start of the record to the end of the record. This single internally coherent dataset replaces the SPLICED DATASET of the past.
Anyone who thinks that using the same counting method from start to finish is somehow bad and wrong, well, they’re free to use the old spliced dataset … and if you do, I’m free to laugh at your adherence to past mistakes.
Note well that this says nothing about the endless adjustments to the temperature record, which may or may not be justified in any particular case, and which are nowhere near as clear-cut and clean as the sunspot count.
My viewpoint is that this adjustment corrects a clear mistake, and therefore should be welcomed. – Anthony Watts