Numerical Geodynamo Simulations Can Capture the Long-Term Variability of the Paleomagnetic Field

Abstract:

Earth’s magnetic field varies on time scales spanning several orders of magnitude, from less than a year to hundreds of million years or longer. These variations arise from the complex magnetohydrodynamic processes at work in the liquid core, which can be studied using global numerical dynamo simulations. Presently these simulations can reproduce several features of the geomagnetic field, including a dipole-dominated field and polarity reversals, aspects of the recent secular variation, and the large-scale properties of the historical and Holocene fields. However, Sprain et al. (2019) recently questioned the ability of numerical geodynamo models to capture the long-term paleomagnetic field behavior. This was demonstrated by defining a set of criteria (Q_PM criteria) to quantify the compliance of geodynamo simulations with the variability of the paleomagnetic field of the last 10 Myr, and by testing 46 simulations against these criteria. Applying the Q_PM criteria to an expanded suite of geodynamo models differing in input parameter values, thermal boundary conditions and convective driving mode, we present here the first numerical simulations known to simultaneously reproduce the salient aspects of the paleosecular variation and time-averaged field behavior of the last 10 Myr. We find that all simulated paleomagnetic observables vary with the relative dipole field strength at the core-mantle boundary (dipolarity). Simulations that best comply with the observed paleomagnetic field variability are driven by purely chemical convection, have an Ekman number (ratio of viscous to Coriolis forces) not larger than 3×10^-4, and present intermediate dipolarity values. We show that our results are compatible with estimates of dipolarity obtained from observation-based models, namely global field model reconstructions and statistical (giant Gaussian process) field models. Our findings confirm that the Earth’s core may lie at the transition between the dipolar and multipolar dynamo regimes, as suggested by previous numerical studies.