Solar physics

Dynamo theory questioned

Observations of X-ray emission — a diagnostic tool for the mechanisms driving stellar magnetic fields — from four cool stars call into question accepted models of magnetic-field generation in the Sun and stars. See Letter p.526

Every star, including the Sun, hosts a magnetic field. One of the most notable products of stellar magnetic activity is X-ray emission (Fig. 1), most of which comes from active regions — areas on a star's surface where the magnetic field concentrates and whose best-known examples are sunspot groups. The commonly accepted theory of stellar magnetic-field generation is based on a dynamo process of electromagnetic induction: the mechanical energy of internal and surface plasma flows is converted into magnetic energy. According to this theory, less than one-tenth of 1% of the Sun's total luminosity is sufficient to drive the solar magnetic cycle, heat the corona, accelerate the solar wind and power all the eruptive phenomena that collectively make up solar activity.

Figure 1: Short-wavelength emission from the Sun.


In this composite image, extreme ultraviolet (red–yellow) was measured by the Solar Dynamics Observatory; low-energy X-rays (green) were measured by the Hinode spacecraft; and high-energy X-rays (blue) were measured by the Nuclear Spectroscopic Telescope Array. Wright and Drake1 present observations of X-ray emissions from four stars that cannot be explained by currently accepted models of the solar dynamo.

But on page 526, Wright and Drake1 show that the details of the dynamo theory still escape us. They report on the X-ray emissions from four stars detected by the Chandra and ROSAT space observatories. These stars are substantially colder than the Sun and have a different internal structure. If the chain of physical mechanisms that lead to X-ray emission in the Sun conformed to the dynamo theory, we would expect a different pattern of X-ray emission from these stars. Yet, Wright and Drake show that their X-ray emissions have the same pattern as the Sun's.

The solar-surface magnetic field can be observed in great detail because of the Sun's proximity to Earth. Such observations have revealed several spatiotemporal patterns that are driven by the dynamo. Perhaps most notably, polarity reversals occur with a regular 11-year period, and magnetically active regions are found to emerge ever closer to the solar equator as activity cycles unfold2. Conversely, observations of the activity of other stars are usually restricted to surface-averaged, global measurements that lack spatial information.

But, where stars are concerned, what researchers lack in detail is made up for in numbers. Surveys of X-ray emissions from large samples of stars — which, among other things, determine the dependence of the emissions on stellar mass, luminosity and rotation rate — provide information on global stellar activity and, in turn, insight into the underlying magnetic processes.

In Sun-like stars, X-ray emission increases with rotation rate up to a value of a few times the rotation period, and then levels off3. Information about the details of the internal dynamo is thus lost for fast-rotating stars, which fall within this plateau region of emission behaviour. Wright and Drake present an updated version of the emission-behaviour profile (see Fig. 1 of the paper1), based on data for hundreds of stars of varying masses and luminosities.

Within the solar dynamo theory, a key element of the complex causal chain that links the internal dynamo to surface X-ray emission is the tachocline, a transition region between the radiative core of the Sun and its convective outer layer. Solar plasma in the convective zone rotates at different rates, depending on latitude, whereas the radiative core rotates more or less as a solid body. This difference in rotational rate between the two zones produces a strong 'shear' stress in the tachocline, which helps to concentrate the diffuse magnetic field into structures called flux ropes. These magnetic flux ropes emerge at the surface and generate active regions4.

Stars progressively less massive than the Sun have deeper convective envelopes, becoming fully convective at about 40% of the Sun's mass. Such stars no longer have a tachocline, so we should expect some qualitative change in their mode of dynamo action. This long-sought 'dynamo boundary' has not yet been detected. Fully convective stars have been found to emit the same level of X-rays as solar-type stars that harbour a tachocline. However, all previously reported stars of this type are fast rotators, and therefore fall on the plateau region of the diagram that plots the relationship between X-ray emission and rotation rate (see the red circles in Fig. 1 of the paper1), meaning that the details of the internal dynamo action might well be lost.

Enter Wright and Drake. The authors have uncovered the X-ray emissions from four slowly rotating, fully convective stars, and all four fall on the slope part of the X-ray emission–behaviour profile. X-ray emissions therefore scale identically with the Rossby number (the ratio of convective flow speed to rotation rate) in stars with and without a tachocline.The authors thus argue that the tachocline cannot be an essential ingredient for stellar dynamo action, as it is in the currently accepted theory.

There are several possible ways to explain this quandary. Perhaps low-mass stars are not fully convective all the way down to their centres. Or maybe the pattern of stellar X-ray emission is dominated by a re-organization of the magnetic field that occurs in the stellar upper atmosphere5, thus losing its 'memory' of the magnetic field's dynamo origin. And, of course, a tachocline might indeed be non-essential.

Numerical simulations6,7 of global solar convection have demonstrated that solar-like large-scale magnetic fields undergoing regular polarity reversals can be produced wholly within a convection zone, without the need to extend the simulation down to the depth of the tachocline. Some of these simulations even generate rope-like structures of magnetic flux that rise to the top of the simulation domain in a solar-like manner8,9. Wright and Drake's results, together with such simulations, provide an impetus to rethink what we know about the solar and stellar dynamo.Footnote 1


  1. 1.

    See all news & views


  1. 1

    Wright, N. J. & Drake, J. J. Nature 535, 526–528 (2016).

  2. 2

    Hathaway, D. H. Living Rev. Sol. Phys. 7, 1 (2010).

  3. 3

    Noyes, R. W., Hartmann, L. W., Baliunas, S. L., Duncan, D. K. & Vaughan, A. H. Astrophys. J. 279, 763–777 (1984).

  4. 4

    Fan, Y. Living Rev. Sol. Phys. 6, 4 (2009).

  5. 5

    Cheung, M. C. M., Rempel, M., Title, A. M. & Schüssler, M. Astrophys. J. 720, 233–244 (2010).

  6. 6

    Charbonneau, P. Annu. Rev. Astron. Astrophys. 52, 251–290 (2014).

  7. 7

    Hotta, H., Rempel, M. & Tokoyama, T. Science 351, 1427–1430 (2016).

  8. 8

    Nelson, N. J., Brown, B. P., Sacha Brun, A., Miesch, M. S. & Toomre, J. Astrophys. J. 739, L38 (2011).

  9. 9

    Fan, Y. & Fang, F. Astrophys. J. 789, 35 (2014).

Download references

Author information

Correspondence to Paul Charbonneau.

Related links

Related links

Related links in Nature Research

Solar physics: The planetary hypothesis revived

Solar physics: Towards ever smaller length scales

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Charbonneau, P. Dynamo theory questioned. Nature 535, 500–501 (2016).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.