Abstract:
Kp and Ap indices covering the period 1932 to 1995 are analysed in a fashion similar to that attempted by Bartels for the 1932–1961 epoch to examine the time variations in their characteristics. Modern analysis techniques on the extended data base are used for further insight. The relative frequencies of occurrence of Kp with different magnitudes and the seasonal and solar cycle dependences are seen to be remarkably consistent despite the addition of 35 years of observations. Many of the earlier features seen in the indices and special intervals are shown to be replicated in the present analysis. Time variations in the occurrence of prolonged periods of geomagnetic calm or of enhanced activity are presented and their relation to solar activity highlighted. It is shown that in the declining phase the occurrence frequencies of Kp = 4–5 (consecutively over 4 intervals) can be used as a precursor for the maximum sunspot number to be expected in the next cycle. The semiannual variation in geomagnetic activity is reexamined utilising not only the Ap index but also the occurrence frequencies of Kp index with different magnitudes. Lack of dependence of the amplitude of semiannual variation on sunspot number is emphasised. Singular spectrum analysis of the mean monthly Ap index shows some distinct periodic components. The temporal evolution of ∼44 month, ∼21 month and ∼16 month oscillations are examined and it is postulated that while QBO and the 16 month oscillations could be attributed to solar wind and IMF oscillations with analogous periodicity, the 44 month variation is associated with a similar periodicity in recurrent high speed stream caused by sector boundary passage. It is reconfirmed that there could have been only one epoch around 1940 when solar wind speed could have exhibited a 1.3-year periodicity comparable to that seen during the post-1986 period.
