Thomas Ceulemans

Radiative transfer is essential in astronomy, both for interpreting observations and simulating various astrophysical phenomena. However, self-consistent line radiative transfer is computationally expensive, especially in 3D. To reduce the computational cost when utilizing a discrete angular discretization, we use a comoving frame interpretation of the radiative transfer equation. The main innovation of this paper lies in the novel stabilization method for the resulting numerical discretization. The stabilization method is able to reduce spurious oscillatory behavior in the computed intensities, at the expense of extra boundary conditions which need to be enforced. We also implement an adaptive angular discretization for the ray-tracing implementation, in order to efficiently and accurately calculate the radiation field. Finally, we apply this new numerical method to compute NLTE line radiative transfer on a hydrodynamics model, showcasing its potential improvement in computation efficiency.

Spectral line observations are an indispensable tool to remotely probe the physical and chemical conditions throughout the universe. Modelling their behaviour is a computational challenge that requires dedicated software. In this paper, we present the first long-term stable release of Magritte, an open-source software library for line radiative transfer. First, we establish its necessity with two applications. Then, we introduce the overall design strategy and the application/programmer interface (API). Finally, we present three key improvements over previous versions: (1) an improved re-meshing algorithm to efficiently coarsen the spatial discretisation of a model; (2) a variation on Ng-acceleration, a popular acceleration-of-convergence method for non-LTE line transfer; and, (3) a semi-analytic approximation for line optical depths in the presence of large velocity gradients.

Spectral line observations encode a wealth of information. A key challenge, therefore, lies in the interpretation of these observations in terms of models to derive the physical and chemical properties of the astronomical environments from which they arise. In this paper, we present pomme, an open-source Python package that allows users to retrieve 1D or 3D models of physical properties, such as chemical abundance, velocity, and temperature distributions of (optically thin) astrophysical media, based on spectral line observations. We discuss how prior knowledge, for instance, in the form of a steady-state hydrodynamics model, can be used to guide the retrieval process, and we demonstrate our methods on both synthetic and real observations of cool stellar winds.