Periodical patterns in major flare occurrence and their relation to magnetically complex active regions

A periodical occurrence rate of solar major flares (observed in hard X-rays) of about 24 days (synodic) was first reported by Bai (1987) for the years 1980--1985. Its origin is still far from being understood. Applying wavelet analyses for daily numbers of Hα flare events covering almost four entire solar cycles (no. 19--22) reveals a 24-day period in each of the cycles studied. This can be established primarily in the occurrence rate of major flares but is also seen in subflares. Since large flares occur preferentially in association with active regions of complex magnetic configuration a comparative study of magnetically classified active regions, subdivided into magnetically complex (i.e. including a γ and/or δ configuration) and non-complex (α, β) was performed. A significant relation between the appearance of the 24-day period in major Hα flares and magnetically complex sunspot groups is found, whereas it cannot be established for non-complex groups. From solar rotation studies based on tracing sunspots practically no siderial rotation velocities as high as 16 deg/day (which corresponds to a synodic period of ∼24 days) are reported. Thus the cause of the 24-day period is very likely not related to solar surface rotation. Alternatively it might be due to periodical patterns in magnetic flux emergence which is an important driver of flare eruptions. We test this hypothesis by investigating synoptic maps of magnetograms (National Solar Observatory/Kitt Peak) for selected time ranges in which the 24-day period is revealed for both flares and magnetically complex active regions.