In order to derive the stellar and wind parameters of observed OB-stars, state of the art spectral analysis compares synthetic spectra, calculated by means of 1D spherically symmetric NLTE atmosphere codes, with observations. Certain atmospheres, however, show strong deviations from spherical symmetry, and have to be analyzed by means of 3D radiative transfer. Typical examples are magnetic winds of OB stars (that also display a variability of UV lines (Marcolino et al. 2013) and of H_\alpha (Sundqvist et al. 2012), which might be qualitatively explained by magneto-hydrodynamical simulations), and rapidly rotating stars (e.g., VFTS102 with v_{rot} = 500-600km/s, Dufton et al. 2011), which are affected by centrifugal forces and gravity darkening.
The aim of this poster is to present a newly developed 3D radiative transfer code, which calculates the radiation field in the winds of hot stars self-consistently for continuum and line-scattering problems. The code is able to handle arbitrary (non-monotonic) velocity fields and density structures. It is based on a finite-volume method (Adam 1990) and incorporates an accelerated \Lambda-iteration (ALI) by means of a newly developed non-local approximate \Lambda-operator (ALO). We present the basic assumptions of the code, and provide error estimates by calculating 1D spherically symmetric winds within our framework. As a first application, we present UV resonance line-profiles (based on a two-level-atom approach) for a magnetic wind, which is described by a simplified atmospheric model, the 'analytical dynamical magnetosphere' (Owocki et al. 2016).