Limits on the Sun's core magnetism from solar oscillations

Many years ago Cowling¹ discussed the possibility that the Sun has a significant relic field. This field would have poloidal and toroidal components, with the toroidal component being driven by dynamo action on the poloidal component. The toroidal field would be quadrupole in nature having opposite senses in the upper and lower hemispheres. Subsequently, Dicke ² proposed that the solar quadrupole moment is caused by a strong, inclined toroidal field with a magnitude of ∼6×10⁷ G. Ulrich and Rhodes³ suggested that a poloidal field with a magnitude of 3 × 10⁸ G was required to account for some of the properties of the 5-min period oscillation. Whereas Mestel and Moss⁴ claimed that such fields may not be sufficiently stable to endure. Hill et al. ⁵ argued that solar oscillation data imply that a simple poloidal field is much weaker than 3 × 10⁸ G and Gough⁶ has suggested that the toroidal field may be much weaker than the 6 × 10 ⁷ G postulated by Dicke². Magnetic fields, like rotation, produce a fine structure in solar oscillations. Their effects should be detectable provided the fields are sufficiently intense. Here we perform an analysis of oscillation data due to Hill et al.⁵ to show that limits of a few megagauss can be placed on poloidal and toroidal magnetic fields inside the Sun. A limit can thereby also be placed on the part of the quadrupole moment of the Sun due to magnetism. These fields are too weak to induce a quadrupole moment much larger than that which would result if the Sun rotated rigidly at the observed surface equatorial rate.