Abstract:
To model the Ca I 4227 Å line polarization, radiative transfer effects with partial frequency redistribution (PRD) must be taken into account. The numerical solution of the relevant polarized radiative transfer (RT) equations is computationally very demanding. The “last scattering approximation” (LSA) is a concept allowing faster methods to be devised. It is based on the remark that a single scattering of the radiation field is sufficient for creating most of the polarization. Its key ingredient is the anisotropy of the radiation field. If the anisotropy is extracted from the observed center to limb variation of the intensity profile, only the wings of the Q/I spectrum can be modeled (Sampoorna et al. 2009). We show here that the core region may be modeled as well if one takes into account the depth variation of the anisotropy which is obtained from an unpolarized multilevel RT (Anusha et al. 2010). After a validation of the LSA approach by comparison with a polarized RT calculation, we apply both approaches to model recent observations of the Ca I 4227 Å line polarization taken on the quiet Sun. Apart from a global scaling factor, both approaches give a very good fit to the Q/I spectrum for all the wavelengths. As the LSA is 8 times faster than the RT approach, we can recommend it as an efficient method to analyze other strong resonance lines in the second solar spectrum.
