Author note (2026-11-02). The paper cited here, arXiv:2603.01883 (“Geomagnetic signatures of the slurry F-layer inferred from dynamo simulations”), is by Ludovic Huguet, Thomas Frasson, Souvik Naskar, Stephen J. Mason, Andrew T. Clarke, Hannah F. Rogers et al.. The name(s) this essay attributes it to (Maybank) do not appear among the listed authors, so the attribution below is wrong or refers to someone else's earlier work; I have not worked out which. Found by a mechanical check of each essay's named authors against the arXiv author list.
Author note (2026-11-08): the essay credits 'Rogers et al.', but Rogers is author 6 of 8 on arXiv:2603.01883; the first author is Ludovic Huguet.
Earth's core has a snow layer. It leaves a magnetic fingerprint.
At the bottom of the liquid outer core, seismic observations reveal a density-stratified region — the F-layer — that defies the classical picture of vigorous convection all the way down to the inner core boundary. The layer shouldn't be there if the dynamo operates by simple thermal and compositional buoyancy from inner core growth.
The best current model: the F-layer is a slurry. Solid iron particles nucleate in the liquid core above the inner core boundary and snow downward through a quasi-static iron-oxygen liquid on the liquidus. The snow provides a physical mechanism for the observed stratification.
Maybank et al. (arXiv:2603.01883) present the first dynamo simulations that include this slurry F-layer and ask: does it have a detectable magnetic signature? It does. The stratified layer acts as a barrier to the columnar quasi-geostrophic flows that normally span the entire core. These flows, deflected by the stratification, shift their peak radial and azimuthal speeds to lower latitudes.
In the magnetic field, this manifests as polar minima in the radial component — the field weakens over the poles as the stratification strengthens. The signature is most visible in time-averaged field maps resolved to spherical harmonic degree 5 or higher, and in specifically negative zonal Gauss coefficients at degrees 3 and 5.
The constraints are tight: an F-layer thicker than 600 km is incompatible with observed geomagnetic field morphology. The data favor weak stratification and a layer less than 400 km thick. Seismology says the layer exists. Geomagnetism tells you how thick it can be. Two completely different observational windows constraining the same hidden structure.
Maybank et al., "Geomagnetic signatures of the slurry F-layer inferred from dynamo simulations," arXiv:2603.01883 (2026).