On the Sun, magnetic fields are most easily seen in the form of sunspots, which were first recorded by ancient Chinese astronomers. They were rediscovered by Galileo and contemporaries at the start of the era of telescopic astronomy. The magnetic nature of sunspots became apparent in the early years of this century, when Hale mapped the solar magnetic field through its effect -the Zeeman effect- on the detailed shape and polarisation state of spectral lines.
The Sun's magnetic field has been mapped on a regular basis ever since Hale's time. We now know that the 11-year sunspot cycle is just a part of an overall 22-year magnetic cycle: the shape of the Sun's magnetic field changes throughout the 11-year sunspot cycle, reverses its magnetic polarity and begins the whole process over again. But how is the field generated? Simple models (called dynamo models) suggest that the interaction of rotation and convection in a stellar envelope generates surface differential rotation, which stretches north-south field lines until they run east-west during the magnetic cycle (see upper picture panel below). Cyclonic convection then somehow regenerates the north-south field, but with a reversed polarity (lower panel).
However, we are very far from understanding in detail how dynamo processes operate within the convective zone of the Sun. To obtain a fully consistent model picture, one indeed needs to describe accurately physical processes operating at very small spatial scales (like turbulence for instance), as well as take into account the feedback effect of the magnetic field on the overall dynamics of the convective zone (which makes the problem highly non-linear).
A possible way out of this problem is to study dynamo processes in
stars other than the Sun: the major obstacle to
understanding the magnetic diseases of the Sun may be the shortage of patients in
which to study them -- only one, in fact.
Using Zeeman-Doppler imaging, we managed to detect and map, at several successive epochs, the surface magnetic topologies of a few extremely active stars of about one solar mass (with ages ranging from a few million to more than ten billion years). The most astonishing and unexpected result is that we systematically find at the surface of these stars magnetic regions in which the field is predominantly azimuthal, i.e. directed along lines of equal latitude. Below is the azimuthal field component (in a flattened polar projection, the equator and pole being depicted by the bold circle and central dot respectively) of the magnetic map we reconstruct for the rapidly rotating K1 subgiant HR 1099 from data recorded on 1995 December. Note the particularly obvious belts of counterclockwise/clockwise magnetic field encircling the star at low/high latitudes respectively. We interpret the observations as the direct detection, at photospheric level, of the toroidal component of the large-scale dynamo-generated magnetic field of HR 1099. We therefore conclude that the underlying dynamo is distributed throughout the whole convective zone in HR 1099, and not confined at the interface with the radiative interior as in the Sun.
We speculate that the higher rotation rate (10 times
solar) and thicker convective zone (almost three times
solar) of HR 1099 are both responsible for the fact that its large scale toroidal
field features two rings of opposite polarities in the
upper stellar hemisphere, as opposed to the Sun where there is only one. In AB Dor,
whose rotation rate is more than 50 times solar, up to three rings of opposite
azimuthal field polarities have been detected in the upper stellar hemisphere.
By monitoring the long-term evolution of the field topology of active stars, we should be able to detect directly the polarity switch of the large-scale toroidal field component, and thus the stellar analogue of a solar magnetic cycle. For instance, we recently detected the appearance of a new azimuthal field polarity at high latitude on the surface of young star LQ Hya, that we interpret as the forerunner of a global polarity switch in the magnetic field of this object. If confirmed, our observations would represent the first direct detection of a magnetic polarity switch (and thus of a magnetic cycle) in a star other than the Sun.
More generally, our aim is to follow the time distortion of stellar magnetic topologies on a typical timescale of a few decades, e.g. by tracking the direction (poleward or equatorward) of the dominant dynamo wave, identifying the main dynamo modes that have beed excited (through their spatial structure in particular), locating the internal regions of the star in which the toroidal and poloidal components are regenerated. The ultimate goal of this study is to set the stage for a new and broader context dynamo theory which could reproduce both solar and stellar observations.
Donati J.-F. , ``Rotation and magnetic fields of solar-like stars'', in: Maeder A., Eenens P. (eds.), IAU Symposium 215 on Stellar Rotation (2003). ASP Conf. Series (in press)
Donati J.-F. , ``Surface magnetic fields and differential rotation of solar-like stars'', in: Arnaud J., Faurobert M., Donati J.-F. (eds.), Proceedings of the workshop dedicated to Jean-Louis Leroy on ``Magnetism and Activity of the Sun and Stars'' (2003). EDP Sciences, EAS 9, 169
Donati J.-F. , ``Magnetic feedback on stellar convective zones'', in: Trujillo Bueno J., Sanchez Almeida J. (eds.), proceedings of the Solar Polarisation Workshop #3 (2003). ASP Conf. Series (in press).
Jardine M.M., Wood K., Cameron A.C., Donati J.-F., Mackay D.H., ``Inferring X-ray coronal structures from Zeeman-Doppler images'' (2002) MNRAS 336, 1364
Budding E., Carter B.D., Mengel M.W., Slee O.B., Donati J.-F., ``A Radio and Optical Study of the Active Young F Star HR 1817 (=HD 35850)'' (2002) PASA 19, 527
Jardine M.M., Cameron A.C., Donati J.-F., ``The global magnetic topology of AB Doradus'' (2002) MNRAS 333, 339
Cameron A.C., Donati J.-F., ``Doin' the twist: Secular changes in the surface differential rotation on AB Doradus'' (2002) MNRAS 329, L23
Donati J.-F. , ``Imaging the magnetic topologies of cool active stars'', in: Boffin H., Steeghs D., Cuypers J. (eds.), AstroTomography, Indirect imaging methods in observational astronomy (2001). Springer, Berlin, p. 207
Kitchatinov L.L., Jardine M., Donati J.-F., ``Magnetic cycle of LQ Hydrae: observational indications and dynamo model'' (2000) MNRAS 318, 1171
Donati J.-F. , ``Surface magnetic fields of late-type stars'', in: Butler C.J., Doyle J.G. (eds.), Brendan Byrne memorial workshop on ``Solar and Stellar Activity: Similarities and Differences'' (1999). ASP Conf. Series, vol. 158, p. 27
Donati J.-F., ``Magnetic cycles of HR 1099 and LQ Hydrae'' (1999) MNRAS 302, 457
Donati J.-F., Cameron A.C., Hussain G.A.J., Semel M., ``Magnetic topology and prominence patterns on AB Doradus'' (1999) MNRAS 302, 437
Donati J.-F., Cameron A.C., ``Differential rotation and magnetic polarity patterns on AB Doradus'' (1997) MNRAS 291, 1
Donati J.-F., Brown S.F., Semel M., Rees D.E., et al., ``Photospheric imaging of the RS CVn system HR 1099'' (1992) A&A 265, 682
© Jean-François Donati, last update on 2003 Nov 5