When the technique of "Doppler imaging" of stars was first developed in the mid-1980's, it revealed large starspots at or near the poles of rapidly rotating, magnetically active stars. Stars are much too distant to image directly, and this pseudo-imaging method uses the Doppler effect to trace the positions of dark magnetic spots - the stellar analogues of sunspots - as the star rotates. Spots on the side of the star being rotated toward our line of sight lead to blue-shifted features in spectral line profiles, which gradually change to red-shifted features as they rotate around and recede.  

Sunspots all lie at moderately low latitudes, in belts either side of the solar equator from 0 to 30 degrees north and south.  Large polar spots then initially came as a bit of a surprise. It was soon realised though that the Coriolis Force that depends on rotation rate would push magnetic fields deep in the convection zone toward the poles as they moved upward and emerged at the stellar surface. The location of the magnetic spots is important for understanding how the stars spin down and lose angular momentum, and can provide clues as to the workings of the underlying magnetic dynamo.  Spin-down calculations often assume a spherically-symmetric stellar wind. If coronal mass ejections, which originate from magnetic active regions, are the dominant form of mass loss on active stars (see an earlier posting on this) and active regions are polar, then the drag on the stellar spin will be less efficient than assumed because the leverage is much smaller at the poles than at the equator. 

We analysed observations of the young rapidly rotating K dwarf star AB Doradus made by Chandra's High Energy Transmission Grating Spectrometer in search of Doppler signatures of coronal activity associated with surface spots. AB Dor rotates twice per day, with a speed of nearly a million km/hr at its equator.  Not only did we fail to detect any Doppler smearing of the X-ray emission lines from an extended corona, but the smearing was less than would be caused by a low-lying corona surrounding the star.  The only way this can happen is if the corona is concentrated at the poles. This provides direct evidence that the multi-million degree plasma giving rise to coronal activity is associated with the magnetic spots inferred from observations in visible light.  Stellar spin-down models based on spherically-symmetric winds could require revision.  This work was published in the 2015 March 20 edition of the Astrophysical Journal.