Is Mercury's weak magnetic field generated and maintained by a magnetospheric feedback dynamo?

Mercury's global magnetic field — an almost axial dipole of about 190 nT equatorial surface strength, with the magnetic equator displaced about 20% of the planetary radius northward — is one to two orders of magnitude weaker than a classical convective core dynamo is expected to produce. Heyner et al. (2011) proposed and numerically supported a feedback dynamo explanation: the solar-wind-driven magnetospheric field, dominated by anti-parallel Chapman–Ferraro currents, diffuses through the mantle and suppresses dynamo growth, locking the dynamo into a weak-field state with Elsasser number (the ratio of Lorentz to Coriolis forces) far below one. The open problem is whether this mechanism actually operates at Mercury. Three linked issues remain unresolved: whether Mercury's dynamo plausibly started from the weak seed field the mechanism requires (a startup episode after dynamo shutdown by giant impact or subadiabatic core-mantle heat flux, or after a reversal); whether the feedback-stabilized weak-field state survives at planetary Ekman and magnetic Prandtl numbers rather than only at the simulated $E=10^{-4}$, $Pm=2/3$; and whether the predicted surface signatures — a dipole-dominated spectrum with enhanced higher (uneven) harmonics and intense polar flux concentrations — match Mercury's internal field, whose measured quadrupole-to-dipole ratio the current feedback models fail to reproduce, and whose definitive two-hemisphere mapping awaits BepiColombo observations beginning after Mercury orbit insertion in November 2026. The mechanism is neither established nor refuted: the problem is open.

Background

Mercury is the only terrestrial planet other than Earth with a global magnetic field of internal origin. Discovered by the Mariner 10 flybys (1974–1975) and measured in detail by the MESSENGER orbiter (in orbit 2011–2015), the field is dominated by an axial dipole of moment about 190 nT·$R_M^3$ ($R_M = 2440$ km is Mercury's mean radius), tilted by less than about 1° from the spin axis, with the magnetic equator displaced roughly 480 km, about 20% of $R_M$, toward the north (Anderson et al. 2012). The equatorial surface field of roughly 190 nT is only about 0.6% of Earth's, and this weakness is the central puzzle.

The field is attributed to a dynamo: self-excited magnetic-field generation by convective motion of the electrically conducting liquid-iron outer core. In classical dynamo theory a convective dynamo grows until the Lorentz force balances the Coriolis force — a magnetostrophic state measured by the Elsasser number $\Lambda$ (the ratio of Lorentz to Coriolis forces) of order one — which for Mercury's core predicts a surface field of several thousand nanotesla, an order of magnitude or more above the observed value. Several explanations of this weak-field paradox exist: a dynamo operating deep below a stably stratified, non-convecting layer at the top of the core whose electromagnetic damping weakens the field that reaches the surface (Christensen 2006); iron-snow crystallization in mid-core, in which sinking solid iron creates two convecting layers whose dynamos couple with negative sign and yield a weak net field (Vilim et al. 2010); dynamos shaped by prescribed core-mantle heat-flux patterns (Cao et al. 2014; Tian et al. 2015); and, most recently, double-diffusive dynamos — convection driven by two components (heat and light elements) with very different diffusivities beneath a stratified layer — which self-consistently reproduce both the weakness and the northward offset of Mercury's field (Takahashi et al. 2019; Kolhey et al. 2025).

The problem belongs to a different family: external feedback. Glassmeier et al. (2007) noted that Mercury's magnetosphere is small and that its Chapman–Ferraro currents — the currents flowing on the magnetopause boundary that shield the planetary field from the solar wind — generate a field anti-parallel to the internal dipole. Because Mercury's mantle is comparatively thin and weakly conducting, this externally generated field can diffuse down to the core-mantle boundary (CMB) over long times and oppose dynamo amplification. Heyner et al. (2011) tested this idea in fully self-consistent three-dimensional magnetohydrodynamic dynamo simulations (using the spherical-shell convection and dynamo code MagIC) by coupling the internal dynamo to a parameterized magnetospheric response obtained by scaling the terrestrial magnetosphere to Hermean (Mercurian) conditions: the external axial-dipole Gauss coefficient — Gauss coefficients being the spherical-harmonic coefficients of the magnetic field — obeys

$$g^{0}_{1,\mathrm{ext}} = \alpha\,\mathrm{sgn}\!\left(g^{0}_{1,\mathrm{int}}\right)\,\frac{\left|g^{0}_{1,\mathrm{int}}\right|^{5}}{\beta^{5}+\left|g^{0}_{1,\mathrm{int}}\right|^{5}}$$

with $\alpha = 70$ nT and $\beta = 96$ nT, where $g^{0}_{1,\mathrm{int}}$ is the internal axial dipole coefficient. In their simulations (Rayleigh number $Ra = 7.5\times10^{6}$, Ekman number $E = 10^{-4}$, magnetic Prandtl number $Pm = 2/3$, inner-core radius fraction 0.35), a dynamo started from a weak seed field saturates at a volume-averaged Elsasser number below $10^{-5}$ — the flow stays geostrophic, meaning rotation rather than magnetic forces shapes it — instead of the magnetostrophic $\Lambda \approx 1$ state reached by the identical model without feedback. Removing the external field after saturation lets the dynamo grow back to $\Lambda \approx 1$, demonstrating that the feedback, not a numerical artifact, holds the field weak; scaled to Mercury's core size and conductivity, the model predicts a surface dipole of the order of the observed weak field. The mechanism, however, requires the dynamo to be weakly driven and to grow from a weak seed field — for example during a startup episode after a dynamo shutdown event (a giant impact temporarily halting dynamo action, or an epoch of subadiabatic, i.e. below-convection-threshold, core-mantle heat flux leaving a temporarily dead core before inner-core crystallization revitalizes convection) or after a field reversal — and it requires the induced external field at the dynamo region to reach roughly 10% of the internal field. Its testable predictions are intense magnetic flux concentrations near the rotation poles at the CMB and, at the surface, a dipole-dominated Mauersberger–Lowes spectrum (the magnetic energy per spherical-harmonic degree) with relatively enhanced higher, uneven harmonics compared with a strong-field dynamo.

Two further facts define the current frontier. First, the simulations use an Ekman number (the ratio of viscous to Coriolis forces) and a magnetic Prandtl number (the ratio of viscous to magnetic diffusion) that are many orders of magnitude from Mercury's core values ($E \ll 10^{-4}$, $Pm \ll 1$). Second, MESSENGER's orbital data already constrain the low-degree internal field: the observed spectrum features a pronounced axial quadrupole (quadrupole-to-dipole ratios of about 0.2–0.4 across published field models; Anderson et al. 2012; Wardinski et al. 2019) that the feedback models do not accurately reproduce at the parameters explored so far, while the magnetospheric field reaches 10–50% of the dipole at observation locations, making the separation of internal and external contributions the central observational difficulty (Johnson et al. 2012). Mercury's ancient field is separately constrained by crustal remanent magnetization recording a dynamo at 3.9–3.7 Ga (Ga denoting billions of years ago) (Johnson et al. 2015; Oliveira et al. 2019; Narrett et al. 2025). The BepiColombo mission's MPO-MAG magnetometer, following Mercury orbit insertion on 21 November 2026, will map the internal field from a low-altitude polar orbit with, for the first time, good southern-hemisphere coverage (Heyner et al. 2021).

Problem Statement

Determine whether the magnetospheric feedback dynamo of Glassmeier et al. (2007) and Heyner et al. (2011) is the mechanism that generates and maintains Mercury's present weak magnetic field (axial dipole moment about 190 nT·$R_M^3$, equatorial surface field about 190 nT, one to two orders of magnitude below the classical magnetostrophic expectation), by jointly establishing the following three linked components.

1. Startup premise: whether Mercury's dynamo plausibly entered the weak-field branch — growing from a weak seed field during a startup episode after a dynamo shutdown (for example a giant impact temporarily halting dynamo action, or an epoch of subadiabatic core-mantle heat flux leaving a temporarily dead core before inner-core crystallization revitalizes convection) or after a field reversal — as judged against Mercury's thermal-evolution history and the paleomagnetic constraints from crustal magnetization.

2. Parameter robustness: whether the feedback-stabilized weak-field state (volume-averaged Elsasser number far below one, geostrophic flow) persists at control parameters approaching Mercury's core — Ekman number $E \ll 10^{-4}$ and magnetic Prandtl number $Pm \ll 1$, with a magnetospheric response function anchored to the actual Hermean magnetosphere — rather than only at the simulated $E = 10^{-4}$, $Pm = 2/3$.

3. Observational discrimination: whether the model's predicted surface signatures — a dipole-dominated Mauersberger–Lowes spectrum with relatively enhanced higher, uneven harmonics, intense polar flux concentrations, and the predicted weak amplitude, including reconciliation with the measured axial quadrupole-to-dipole ratio of about 0.2–0.4 that current feedback models miss — match Mercury's internal field as determined from MESSENGER data and, with full two-hemisphere coverage and clean internal/external separation, from BepiColombo's MPO-MAG measurements.

These three components are jointly posed: together they answer one question — is the weak field externally regulated by the magnetosphere? — and none of them alone settles it. The question is open in both directions: a verdict that the feedback mechanism cannot operate at Mercury settles it as much as a verdict that it does. Because several other explanations of Mercury's weak field exist (a deep dynamo beneath a stable layer; iron-snow feedback dynamos; heat-flux-pattern dynamos; double-diffusive dynamos), a verdict on the feedback mechanism must remain distinguishable from their predictions.

References

1. Glassmeier, K.-H., Auster, H.-U., Motschmann, U. (2007). A feedback dynamo generating Mercury's magnetic field. Geophysical Research Letters 34. doi:10.1029/2007GL031662, https://doi.org/10.1029/2007GL031662
2. Heyner, D., Wicht, J., Gómez-Pérez, N., Schmitt, D., Auster, H.-U., Glassmeier, K.-H. (2011). Evidence from Numerical Experiments for a Feedback Dynamo Generating Mercury's Magnetic Field. Science 334(6063), 1690–1693. doi:10.1126/science.1207290, https://doi.org/10.1126/science.1207290
3. Christensen, U. R. (2006). A deep dynamo generating Mercury's magnetic field. Nature 444, 1056–1058. doi:10.1038/nature05342, https://doi.org/10.1038/nature05342
4. Vilim, R., Stanley, S., Hauck, S. A. (2010). Iron snow zones as a mechanism for generating Mercury's weak observed magnetic field. Journal of Geophysical Research: Planets 115. doi:10.1029/2009JE003528, https://doi.org/10.1029/2009JE003528
5. Anderson, B. J., Johnson, C. L., Korth, H., Winslow, R. M., Borovsky, J. E., Purucker, M. E., Slavin, J. A., Solomon, S. C., Zuber, M. T., McNutt, R. L. (2012). Low-degree structure in Mercury's planetary magnetic field. Journal of Geophysical Research: Planets 117. doi:10.1029/2012JE004159, https://doi.org/10.1029/2012JE004159
6. Johnson, C. L., Purucker, M. E., Korth, H., Anderson, B. J., Winslow, R. M., Al Asad, M. M. H., Slavin, J. A., Alexeev, I. I., Phillips, R. J., Zuber, M. T., Solomon, S. C. (2012). MESSENGER observations of Mercury's magnetic field structure. Journal of Geophysical Research: Planets 117. doi:10.1029/2012JE004217, https://doi.org/10.1029/2012JE004217
7. Cao, H., Aurnou, J. M., Wicht, J., Dietrich, W., Soderlund, K. M., Russell, C. T. (2014). A dynamo explanation for Mercury's anomalous magnetic field. Geophysical Research Letters 41, 4127–4134. doi:10.1002/2014GL060196, https://doi.org/10.1002/2014GL060196
8. Tian, Z., Zuber, M. T., Stanley, S. (2015). Magnetic field modeling for Mercury using dynamo models with a stable layer and laterally variable heat flux. Icarus 260, 263–268. doi:10.1016/j.icarus.2015.07.019, https://doi.org/10.1016/j.icarus.2015.07.019
9. Johnson, C. L., Phillips, R. J., Purucker, M. E., Anderson, B. J., Byrne, P. K., Denevi, B. W., Feinberg, J. M., Hauck, S. A., Head, J. W., Korth, H., James, P. B., Mazarico, E., Neumann, G. A., Philpott, L. C., Siegler, M. A., Tsyganenko, N. A., Solomon, S. C. (2015). Low-altitude magnetic field measurements by MESSENGER reveal Mercury's ancient crustal field. Science 348, 892–895. doi:10.1126/science.aaa8720, https://doi.org/10.1126/science.aaa8720
10. Oliveira, J. S., Hood, L. L., Langlais, B. (2019). Constraining the Early History of Mercury and Its Core Dynamo by Studying the Crustal Magnetic Field. Journal of Geophysical Research: Planets 124, 2382–2396. doi:10.1029/2019JE005938, https://doi.org/10.1029/2019JE005938
11. Wardinski, I., Langlais, B., Thébault, E. (2019). Correlated Time-Varying Magnetic Fields and the Core Size of Mercury. Journal of Geophysical Research: Planets 124, 2178–2197. doi:10.1029/2018JE005835, https://doi.org/10.1029/2018JE005835
12. Takahashi, F., Shimizu, H., Tsunakawa, H. (2019). Mercury's anomalous magnetic field caused by a symmetry-breaking self-regulating dynamo. Nature Communications 10, 208. doi:10.1038/s41467-018-08213-7, https://doi.org/10.1038/s41467-018-08213-7
13. Heyner, D., Auster, H.-U., Fornaçon, K.-H., Carr, C., Richter, I., Mieth, J. Z. D., Kolhey, P., Exner, W., Motschmann, U., Baumjohann, W., Matsuoka, A., Magnes, W., Berghofer, G., Fischer, D., Plaschke, F., Nakamura, R., Narita, Y., Delva, M., Volwerk, M., Balogh, A., Dougherty, M., Horbury, T., Langlais, B., Mandea, M., Masters, A., Oliveira, J. S., Sánchez-Cano, B., Slavin, J. A., Vennerstrøm, S., Vogt, J., Wicht, J., Glassmeier, K.-H. (2021). The BepiColombo Planetary Magnetometer MPO-MAG: What Can We Learn from the Hermean Magnetic Field? Space Science Reviews 217, 52. doi:10.1007/s11214-021-00822-x, https://doi.org/10.1007/s11214-021-00822-x
14. Kolhey, P., Heyner, D., Wicht, J., Gastine, T., Glassmeier, K.-H., Plaschke, F. (2025). Dynamo Models With a Mercury-Like Magnetic Offset Dipole. Journal of Geophysical Research: Planets 130. doi:10.1029/2024JE008660, https://doi.org/10.1029/2024JE008660
15. Narrett, I. S., Weiss, B. P., Steele, S. C., Biersteker, J. B. (2025, revised 7 August 2026). Mercury's Crustal Magnetization Indicates a Stronger Ancient Dynamo. arXiv:2510.02003. https://arxiv.org/abs/2510.02003