Numerical simulations of the geodynamo have successfully represented many observable characteristics of the geomagnetic field, yielding insight into the fundamental processes that generate magnetic fields in the Earth's core. Because of limited spatial resolution, however, the diffusivities in numerical dynamo models are much larger than those in the Earth's core, and consequently, questions remain about how realistic these models are. The typical strategy used to address this issue has been to continue to increase the resolution of these quasi-laminar models with increasing computational resources, thus pushing them toward more realistic parameter regimes. We assess which methods are most promising for the next generation of supercomputers, which will offer access to O(10⁶) processor cores for large problems. Here we report performance and accuracy benchmarks from 15 dynamo codes that employ a range of numerical and parallelization methods. Computational performance is assessed on the basis of weak and strong scaling behavior up to 16,384 processor cores. Extrapolations of our weak-scaling results indicate that dynamo codes that employ two-dimensional or three-dimensional domain decompositions can perform efficiently on up to ∼10⁶ processor cores, paving the way for more realistic simulations in the next model generation.

Gravitational coupling between Earth's core and mantle has been proposed as an explanation for a 6 year variation in the length-of-day (ΔLOD) signal and plays a key role in the possible superrotation of the inner core. Explaining the observations requires that the strength of the coupling, Γ, falls within fairly restrictive bounds; however, the value of Γ is highly uncertain because it depends on the distribution of mass anomalies in the mantle. We estimate Γ from a broad range of viscous mantle flow models with density anomalies inferred from seismic tomography. Requiring models to give a correlation larger than 70% to the surface geoid and match the dynamic core-mantle boundary ellipticity inferred from Earth's nutations, we find that 3 × 10¹⁹<Γ<2 × 10²⁰ N m, too small to explain the 6 year ΔLOD signal. This new constraint on Γ has important implications for core-mantle angular momentum transfer and on the preferred mode of inner core convection.

The decline of Mars' global magnetic field some 3.8-4.1 billion years ago is thought to reflect the demise of the dynamo that operated in its liquid core. The dynamo was probably powered by planetary cooling and so its termination is intimately tied to the thermochemical evolution and present-day physical state of the Martian core. Bottom-up growth of a solid inner core, the crystallization regime for Earth's core, has been found to produce a long-lived dynamo leading to the suggestion that the Martian core remains entirely liquid to this day. Motivated by the experimentally-determined increase in the Fe-S liquidus temperature with decreasing pressure at Martian core conditions, we investigate whether Mars' core could crystallize from the top down. We focus on the "iron snow" regime, where newly-formed solid consists of pure Fe and is therefore heavier than the liquid. We derive global energy and entropy equations that describe the long-timescale thermal and magnetic history of the core from a general theory for two-phase, two-component liquid mixtures, assuming that the snow zone is in phase equilibrium and that all solid falls out of the layer and remelts at each timestep. Formation of snow zones occurs for a wide range of interior and thermal properties and depends critically on the initial sulfur concentration, x0. Release of gravitational energy and latent heat during growth of the snow zone do not generate sufficient entropy to restart the dynamo unless the snow zone occupies at least 400 km of the core. Snow zones can be 1.5-2 Gyrs old, though thermal stratification of the uppermost core, not included in our model, likely delays onset. Models that match the available magnetic and geodetic constraints have x0~10% and snow zones that occupy approximately the top 100 km of the present-day Martian core