A stochastic approach towards understanding the frequency and polarity bias of geomagnetic reversals

Abstract

During time intervals when the geomagnetic reversal process is stationary, spectral analysis shows that harmonic components are absent. Autocorrelation analysis shows that the lengths of successive polarity intervals are uncorrelated. Both tests establish that the lengths of polarity intervals are statistically independent. Stochastic and deterministic models have both been successful in modeling statistically independent reversals that display no polarity bias. To account for the latter, two classes of stochastic models are discussed. In the first, instabilities in the dynamo process which accompany reversals are assumed to occur at a higher rate when the dynamo is in either the normal or reversed polarity state. In the second, it is assumed that the rate at which instabilities occur is constant but different fractions of instabilities are “infertile” in the two polarity states. The cause of polarity bias is attributed to variations in the physical properties of the lower mantle with a spatial distribution that can be described by an odd zonal harmonic.