STiC: A multiatom non-LTE PRD inversion code for full-Stokes solar observations

¹ Institute for Solar Physics, Dept. of Astronomy, Stockholm University, AlbaNova University Centre, 106 91 Stockholm, Sweden
e-mail: jaime@astro.su.se
² National Solar Observatory, 3665 Discovery Drive, Boulder, CO 80303, USA

Received: 19 October 2018
Accepted: 16 January 2019

Abstract

The inference of the underlying state of the plasma in the solar chromosphere remains extremely challenging because of the nonlocal character of the observed radiation and plasma conditions in this layer. Inversion methods allow us to derive a model atmosphere that can reproduce the observed spectra by undertaking several physical assumptions. The most advanced approaches involve a depth-stratified model atmosphere described by temperature, line-of-sight velocity, turbulent velocity, the three components of the magntic field vector, and gas and electron pressure. The parameters of the radiative transfer equation are computed from a solid ground of physical principles. In order to apply these techniques to spectral lines that sample the chromosphere, nonlocal thermodynamical equilibrium effects must be included in the calculations. We developed a new inversion code STiC (STockholm inversion Code) to study spectral lines that sample the upper chromosphere. The code is based on the RH forward synthesis code, which we modified to make the inversions faster and more stable. For the first time, STiC facilitates the processing of lines from multiple atoms in non-LTE, also including partial redistribution effects (PRD) in angle and frequency of scattered photons. Furthermore, we include a regularization strategy that allows for model atmospheres with a complex depth stratification, without introducing artifacts in the reconstructed physical parameters, which are usually manifested in the form of oscillatory behavior. This approach takes steps toward a node-less inversion, in which the value of the physical parameters at each grid point can be considered a free parameter. In this paper we discuss the implementation of the aforementioned techniques, the description of the model atmosphere, and the optimizations that we applied to the code. We carry out some numerical experiments to show the performance of the code and the regularization techniques that we implemented. We made STiC publicly available to the community.