Recent MErcury Surface, Space ENvironment, GEochemistry, and Ranging (MESSENGER) measurements have shown that Mercury's magnetic field is axial‐dominant, yet strongly asymmetric with respect to the equator: the field strength in the Northern Hemisphere is approximately 3 times stronger than that in the Southern Hemisphere. Here we show that convective dynamo models driven by volumetric buoyancy with north‐south symmetric thermal boundaries are capable of generating quasi‐steady north‐south asymmetric magnetic fields similar to Mercury's. This symmetry breaking is promoted and stabilized when the core‐mantle boundary heat flux is higher at the equator than at high latitudes. The equatorially asymmetric magnetic field generation in our dynamo models corresponds to equatorially asymmetric kinetic helicity, which results from mutual excitation of two different modes of columnar convection. Our dynamo model can be tested by future assessment of Mercury's magnetic field from MESSENGER and BepiColombo as well as through investigations on Mercury's lower mantle temperature heterogeneity and buoyancy forcing in Mercury's core.

No abstract

Gravity signatures observed by the Juno and Cassini missions that are associated with the strong zonal winds in Jupiter’s and Saturn’s outer envelopes suggest that these flows extend for several thousand kilometers into the interior. It has been noted that the winds seem to abate at a depth where electrical conductivity becomes significant, suggesting that electromagnetic effects play a key role for confining the winds to the outer weakly conducting region. Here, we explore the possible mechanisms for braking the zonal flow at depth in two model setups with depth-dependent conductivity and forced jet flow, i.e., in axisymmetric shell models and in more simple linearized box models that allow the exploration of a wide parameter range. Braking of the winds directly by Lorentz forces does not reduce their speed in the conducting region enough to be compatible with the inferred secular variation of Jupiter’s field. Stable stratification above the depth where conductivity becomes significant can solve the problem. Electromagnetic forces drive a weak meridional circulation that perturbs the density distribution in the stable region such that the wind speed decreases strongly with depth, due to a thermal wind balance. For this mechanism to be effective, the stable layer must extend upward into a region of low conductivity. Applying the results of the linearized calculations to Jupiter suggests that the dissipation associated with the zonal winds can be limited to a fraction of the internal heat flow and that the jets may drop off over a depth range of 150–300 km.

a b s t r a c tA series of numerical simulations of the dynamo process operating inside gas giant planets has been performed. We use an anelastic, fully nonlinear, three-dimensional, benchmarked MHD code to evolve the flow, entropy and magnetic field. Our models take into account the varying electrical conductivity, high in the ionised metallic hydrogen region, low in the molecular outer region. Our suite of electrical conductivity profiles ranges from Jupiter-like, where the outer hydrodynamic region is quite thin, to Saturn-like, where there is a thick non-conducting shell. The rapid rotation leads to the formation of two distinct dynamical regimes which are separated by a magnetic tangent cylinder -mTC. Outside the mTC there are strong zonal flows, where Reynolds stress balances turbulent viscosity, but inside the mTC Lorentz force reduces the zonal flow. The dynamic interaction between both regions induces meridional circulation. We find a rich diversity of magnetic field morphologies. There are Jupiter-like steady dipolar fields, and a belt of quadrupolar dominated dynamos spanning the range of models between Jupiter-like and Saturn-like conductivity profiles. This diversity may be linked to the appearance of reversed sign helicity in the metallic regions of our dynamos. With Saturn-like conductivity profiles we find models with dipolar magnetic fields, whose axisymmetric components resemble those of Saturn, and which oscillate on a very long time-scale. However, the non-axisymmetric field components of our models are at least ten times larger than those of Saturn, possibly due to the absence of any stably stratified layer.