CRASH_alpha: coupling continuum and line radiative transfer

TL;DR

CRASH_alpha is a novel radiative transfer code that efficiently simulates coupled Ly_alpha and ionizing photon propagation in evolving gas, revealing the importance of joint evolution for accurate spectral predictions in cosmology.

Contribution

It introduces the first coupled continuum and line radiative transfer algorithm with a statistical approach, significantly improving computational speed and accuracy in cosmological simulations.

Findings

The code achieves high agreement with detailed Ly_alpha transfer simulations.

Emerging spectra retain memory of ionization history.

Differences in spectra can significantly impact cosmological observations.

Abstract

In this paper we present CRASH_alpha, the first radiative transfer code for cosmological application that follows the parallel propagation of Ly_alpha and ionizing photons. CRASH_alpha is a version of the continuum radiative transfer code CRASH with a new algorithm to follow the propagation of Ly_alpha photons through a gas configuration whose ionization structure is evolving. The implementation introduces the time evolution for Ly_alpha photons (a feature commonly neglected in line radiative transfer codes) and, to reduce the computational time needed to follow each scattering, adopts a statistical approach to the Ly_alpha treatment by making extensive use of pre-compiled tables. With this statistical approach we experience a drastic increase of the computational speed and, at the same time, an excellent agreement with the full Ly_alpha radiative transfer computations of the code MCLy_alpha. We find that the emerging spectra keep memory of the ionization history which generates a given ionization configuration of the gas and, to properly account for this effect, a self-consistent joint evolution of line and ionizing continuum radiation as implemented in CRASH_alpha is necessary. A comparison between the results from our code and from Ly_alpha scattering alone on a fixed HI density field shows that the extent of the difference between the emerging spectra depends on the particular configuration considered, but it can be substantial and can thus affect the physical interpretation of the problem at hand. These differences should furthermore be taken into account when computing the impact of the Ly_alpha radiation on e.g. the observability of the 21 cm line from neutral hydrogen at epochs preceeding complete reionization.