How reliable is Zeeman Doppler Imaging without simultaneous temperature reconstruction?
Aims: The goal of this study is to perform numerical tests of Zeeman Doppler Imaging (ZDI) to asses whether correct reconstruction of magnetic fields is at all possible without taking temperature into account for stars in which magnetic and temperature inhomogeneities are spatially correlated. Methods: We used a modern ZDI code employing a physically realistic treatment of the polarized radiative transfer in all four Stokes parameters. We generated artificial observations of isolated magnetic spots and of magnetic features coinciding with cool temperature spots and then reconstructed magnetic and temperature distributions from these data. Results: Using Stokes I and V for simultaneous magnetic and temperature mapping for the star with a homogeneous temperature distribution yields magnetic field strengths underestimated by typically 10-15% relative to their true values. When temperature is kept constant and Stokes I is not used for magnetic mapping, the underestimation is 30-60%. At the same time, the strength of magnetic field inside cool spots is underestimated by as much as 80-95% and the spot geometry is also poorly reconstructed when temperature variations are ignored. On the other hand, the inversion quality is greatly improved when temperature variations are accounted for in magnetic mapping. When using all four Stokes parameters the reconstructed field strength inside cool spots is underestimated by 30-40% but the spot geometry can be recovered very accurately compared to the experiments with circular polarization alone. Conclusions: Reliable magnetic field reconstruction for a star with high-contrast temperature spots is essentially impossible if temperature inhomogeneities are ignored. A physically realistic line profile modeling method, which simultaneously accounts for both types of inhomogeneities, is required for meaningful ZDI of cool active stars.