Recent comparisons of magnetic field directions derived from maser Zeeman splitting with those derived from continuum source rotation measures have prompted new analysis of the propagation of the Zeeman split components, and the inferred field orient
ation. In order to do this, we first review differing electric field polarization conventions used in past studies. With these clearly and consistently defined, we then show that for a given Zeeman splitting spectrum, the magnetic field direction is fully determined and predictable on theoretical grounds: when a magnetic field is oriented away from the observer, the left-hand circular polarization is observed at higher frequency and the right-hand polarization at lower frequency. This is consistent with classical Lorentzian derivations. The consequent interpretation of recent measurements then raises the possibility of a reversal between the large-scale field (traced by rotation measures) and the small-scale field (traced by maser Zeeman splitting).
We present a detailed study of the Faraday depth structure of four bright (> 1 Jy), strongly polarized, unresolved, radio-loud quasars. The Australia Telescope Compact Array (ATCA) was used to observe these sources with 2 GHz of instantaneous bandwid
th from 1.1 to 3.1 GHz. This allowed us to spectrally resolve the polarization structure of spatially unresolved radio sources, and by fitting various Faraday rotation models to the data, we conclusively demonstrate that two of the sources cannot be described by a simple rotation measure (RM) component modified by depolarization from a foreground Faraday screen. Our results have important implications for using background extragalactic radio sources as probes of the Galactic and intergalactic magneto-ionic media as we show how RM estimations from narrow-bandwidth observations can give erroneous results in the presence of multiple interfering Faraday components. We postulate that the additional RM components arise from polarized structure in the compact inner regions of the radio source itself and not from polarized emission from Galactic or intergalactic foreground regions. We further suggest that this may contribute significantly to any RM time-variability seen in RM studies on these angular scales. Follow-up, high-sensitivity VLBI observations of these sources will directly test our predictions.
