The following graph shows the yearly and monthly SSN averages (sunspot numbers) for the solar cycles since 1865, using data of the WDC-SILSO of the Royal Observatory of Belgium, Brussels.
WDC-SILSO web-site: http://www.sidc.be/silso/datafiles
data file used in the graph: http://www.sidc.be/silso/DATA/dayssn.dat
The monthly averages show fast drop in radiations at the end of the maximum of every cycle, usually defining a step-change-like variation in intensity.
The data for cycle 24 is lower than the WDC-SILSO and uses the LSC estimate, Layman’s Sunspot Count, from the “landscheidt.info” site: http://www.landscheidt.info/?q=node/50
I believe the LSC estimate for the present cycle correlates better with the Geomagnetic AA-index recorded by the UK Solar System Data Centre– since 1868– than the WDC-SILSO data, but I still have no explanation of why it does so.
The correlation becomes clear if we use the trimester and yearly averages of the aa-index data,
The first 7 years of the present cycle are similarly low as cycle 14, but the trend then was to increasing geomagnetic perturbation since the beginning of the cycle, while now we seem to have reached a maximum in 2011-12 and have been declining ever since.
The correlation also shows very well in the monthly average
From Fig.2 it is clear that the aa-index yearly averages were very low up to the beginning of cycle 14, with values rarely going above 20 nano-T, but in 1904 the index started to increase steadily, in correspondence with the increasing intensity of the solar cycles, in that period. The acceleration stopped in the strong cycle C19, which registered the highest daily SSNs of the last century.
From the years of the WWII to approximately 2004, the yearly average reached it’s maximum and defined a plateau where it rarely went below 20 nano-T.
At last, after 2004, another important reversal happened and, in the present cycle, it reached below 10 nano-T for the first time since the early XX century.
In the present cycle the index also has shown the feature of having a peak at the beginning of the period of the solar maximum of the cycle– corresponding to year 2012– and steady decrease since then, despite the recent second peak of the maximum, during 2014.
This seems to correspond closer to the behavior during cycles C12 and C13, instead of C14, during which the index showed steady increase during almost the entire cycle.
This seems to indicate that although cycle C14 was similarly low, as much as the present one, during the initial years of the cycle, it was also the initial cycle of a series of increasingly strong ones, while the present cycle seems to correspond to a period of accelerated decline of intensity of the solar radiations.
During the period of more than 20 years, between the end of cycle C20 and the end of C22, the yearly average stayed always above 20 nano-T, and this measurement coincides with the period of the peak of the grand solar maximum of the XX century, according to the recent study of Lockwood et al.
The link to the UK Solar System Data Centre is: http://www.wdc.rl.ac.uk/Help/aa.html
and the data retrieval form is at the following link
altering the period and the amplitude of the oscillation.