A new estimate of average dipole field strength for the last five million years

The Earth's ancient magnetic field can be approximated by a geocentric axial dipole (GAD) where the average field intensity is twice as strong at the poles than at the equator. The present day geomagnetic field, and some global paleointensity datasets, support the GAD hypothesis with a virtual axial dipole moment (VADM) of about 80 ZAm². Significant departures from GAD for 0-5 Ma are found in Antarctica and Iceland where paleointensity experiments on massive flows (Antarctica) (1) and volcanic glasses (Iceland) produce average VADM estimates of 41.4 ZAm² and 59.5 ZAm², respectively. These combined intensities are much closer to a lower estimate for long-term dipole field strength, 50 ZAm² (2), and some other estimates of average VADM based on paleointensities strictly from volcanic glasses. Proposed explanations for the observed non-GAD behavior, from otherwise high-quality paleointensity results, include incomplete temporal sampling, effects from the tangent cylinder, and hemispheric asymmetry. Differences in estimates of average magnetic field strength likely arise from inconsistent selection protocols and experiment methodologies. We address these possible biases and estimate the average dipole field strength for the last five million years by compiling measurement level data of IZZI-modified paleointensity experiments from lava flows around the globe (including new results from Iceland and the HSDP-2 Hawaii drill core). We use the Thellier Gui paleointensity interpreter (3) in order to apply objective criteria to all specimens, ensuring consistency between sites. Specimen level selection criteria are determined from a recent paleointensity investigation of modern Hawaiian lava flows where the expected magnetic field strength was accurately recovered when following certain selection parameters. Our new estimate of average dipole field strength for the last five million years incorporates multiple paleointensity studies on lava flows with diverse global and temporal distributions, and objectively constrains site level estimates by applying uniform selection requirements on measurement level data. (1) Lawrence, K.P., L. Tauxe, H. Staudigel, C.G. Constable, A. Koppers, W. McIntosh, C.L. Johnson, Paleomagnetic field properties at high southern latitude, Geochemistry Geophysics Geosystems, 10, 2009. (2) Selkin, P.A., L. Tauxe, Long-term variations in palaeointensity, Phil. Trans. R. Soc. Lond., 358, 1065-1088, 2000. (3) Shaar, R., L. Tauxe, Thellier GUI: An integrated tool for analyzing paleointensity data from Thellier-type experiments, Geochemistry Geophysics Geosystems, 14, 2013