Optically thick boundary

Background

Non-LTE radiative transfer is a strongly non-local problem. Atomic level populations are not only set by thermal photons, but also from radiation scattered from different parts of the stellar atmosphere. To accurately model the radiation from a star, one must solve the radiative transfer equation simultaneously for the radiation field and the populations. As a first step towards using Monte Carlo methods for solving non-LTE problems in stellar atmospheres, one must identify the regions of the atmosphere where non-LTE is needed, to exclude as much as possible dense regions in LTE where the optical depths $\tau_\nu$ are high. In Monte Carlo simulations, photons tend to get trapped in such regions and significantly slow down the computations.

Goal

To locate the lower boundary region of a stellar atmosphere where $\tau_\nu$ becomes large enough that LTE prevails, considering all wavelengths necessary to model spectral lines of interest. Understand how this boundary location depends on atmospheric and radiation quantities, and how it can be approximated without running a full non-LTE calculation.

Method

1. Starting with the FALC model, use the RH 1.5D code to calculate level populations of a simplified model atom of Ca II, including H & K and infrared triplet lines.
2. Using the output from RH 1.5D, study the dependence of the departure coefficients with geometrical height, column mass, and optical depth.
3. Visualise the total optical depth, photon destruction probability, and mean free path as functions of wavelength for the Ca II K and 854.2 nm lines
4. Repeat the analysis for a 3D MHD atmosphere with flux emergence and flaring.