The relative stabilities of the reverse and normal polarity states of the earth's magnetic field

Recent analyses of the geomagnetic reversal sequence have led to different conclusions regarding the important question of whether there is a discernible difference between the properties of the two polarity states. The main differences between the two most recent studies are the statistical analyses and the possibility of an additional 57 reversal events in the Cenozoic. These additional events occur predominantly during reverse polarity time, but it is unlikely that all of them represent true reversal events. Nevertheless the question of the relative stabilities of the polarity states is examined in detail, both for the case when all 57 "events" are included in the reversal chronology and when they are all excluded. It is found that there is not a discernible difference between the stabilities of the two polarity states in either case. Inclusion of these short events does, however, change the structure of the non-stationarity in reversal rate, but still allows a smooth non-stationarity. Only 7 of the 57 short events are pre-38 Ma, but the evidence suggests that this is a real geomagnetic phenomenon rather than degradation of the magnetic recording or a bias in observation. This could be tested by detailed magnetostratigraphic and oceanic magnetic surveys of the Paleogene and Late Cretaceous. Overall it would appear that the present geomagnetic polarity timescale for 0–160 Ma is probably a very good representation of the actual history, and that different timescales and additional events now represent only changes in detail