Assessing Holocene and late Pleistocene geomagnetic dipole field variability

Abstract: Palaeomagnetic studies of continuous geological archives and archaeological artefacts are required to reconstruct geomagnetic field changes beyond the range of historical observations. In this thesis I assess the reliability of sedimentary palaeomagnetic data and Holocene dipole field reconstructions. New palaeomagnetic records are presented from a maar crater lake in New Zealand and a post-glacial lake in Sweden. The fidelity of the data is evaluated through comparisons with sediment parameters sensitive to environmental change and measurement of cosmogenic radionuclides. The results are used to date the sediments and to study regional and global geomagnetic field variations over the past 50,000 years. The Holocene dipole tilt variation is reconstructed by averaging virtual geomagnetic poles from five sedimentary palaeomagnetic records. The data are selected based on palaeomagnetic quality, chronologic constraint and location, with an emphasis on attaining a globally well-distributed data set. The results indicate a cyclical behaviour of the dipole tilt with a period of c. 1350 years. In addition, two preferred states of the dipole axis are identified with north geomagnetic pole longitudes confined to c. 120° West or 30° East. The performance of the dipole tilt reconstruction is compared to more complex geomagnetic field models. The comparison highlights inconsistencies within the sedimentary palaeomagnetic database that may smooth out variations of the dipole component and potentially shift power to higher degree components of the more complex field models. Independent geomagnetic field intensity data from western Eurasia and North America are shown to be consistent with the dipole tilt reconstruction. The cyclical tilt variation and the two preferred states of the dipole axis are interpreted in terms of displacement and/or distortions of high latitude flux lobes on the core mantle boundary. Based on palaeomagnetic directions and palaeointensity data, the two tilt maxima at 2650 and 1200 years BP are tentatively associated with the presence of high intensity magnetic flux beneath Europe. The dipole tilt reconstruction is highly correlated to millennial scale variations in the length of day, reconstructed from ancient records of eclipses, which suggests that the 1350-year cyclicity may constitute an important component of core flow dynamics.