Evidence for Turbulent Magnetic Diffusion in Earth's Core

Fluctuations in the paleomagnetic field suggest that the dipole decay time is shorter than expected, based on current estimates for the molecular magnetic diffusivity in the outer core. Similar behavior is observed in turbulent dynamo simulations, where the short magnetic field decay time cannot be attributed to higher‐order decay modes. We interpret the short decay time as a signature of turbulent diffusion and show that mean‐field theory can quantitatively account for the dynamo results. The predictions depend on the amplitude and length scale of the flow that interacts with the magnetic field. We rely on the pairwise balance between Lorentz/Coriolis and buoyancy/Coriolis forces to identify the relevant part of the flow, and use the resulting flow properties to reproduce results from numerical dynamo simulations within their uncertainties. Upon extending these predictions to the paleomagnetic field, we find that the inferred decay time requires a bulk root‐mean‐square velocity less than 0.8–1.2 mm s−1. Somewhat lower velocities have been estimated at the top of the core from observations of secular variation. These results show that velocities in the interior of the core are constrained by paleomagnetic observations, and that the amplitude of this flow cannot substantially exceed estimates at the core surface. Plain Language Summary Turbulent motion in the liquid core increases the rate of dipole decay by enhancing the effects of magnetic diffusion. This process is known as turbulent diffusion and its magnitude depends on the nature of the turbulent flow. Descriptions of the amplitude and length scale of the turbulent flow are needed to quantify the turbulent diffusion in the standard mean‐field theory. We use dynamo models to identify the turbulent flow associated with magnetic forces and show that its properties may be used to predict the rate of dipole decay. Extrapolating these predictions to core conditions yields a modest increase in the total magnetic diffusivity relative to the molecular value. These results are in broad agreement with estimates of dipole decay from paleomagnetic observations. A quantitative assessment of the paleomagnetic observations places a bound on the time‐ and volume‐averaged mean‐square velocity in the core. Key Points Paleomagnetic observations and dynamo models show evidence for turbulent magnetic diffusion Predictions based on mean‐field theory account for the dipole decay in dynamo models Extensions to the paleomagnet