Abstract. I review work on modelling the infrared and submillimetre SEDs of galaxies. The underlying physical assumptions are discussed and spherically symmetric, axisymmetric, and 3-dimensional radiative transfer codes are reviewed. Models for galaxies with Spitzer IRS data and for galaxies in the Herschel-Hermes survey are discussed. Searches for high redshift infrared and submillimetre galaxies, the star formation history, the evolution of dust extinction, and constraints from source-counts, are briefly discussed.Keywords. infrared galaxies, radiative transfer, star formation
Early work on radiative transfer models for infrared sourcesDuring the 1960s and early 1970s very simplistic models began to be developed for infrared sources associated with circumstellar dust shells, HII regions and star-forming regions. These models assumed that the dust grain density n(r) and temperature T(r) have a power-law dependence on radius r and that the dust is optically thin. Scattering was neglected and so the emission was simply calculated as a sum of n(r)Q ν B ν (T (r)), integrated over r.The first complete solution of the spherically symmetric radiative transfer problem for an optically thick dust cloud was by Leung (1975). He used detailed grain properties and carried out a full solution of the moment equations, including anisotropic scattering. The grain temperature was solved for by assuming radiative balance. Yorke & Krugel (1977) modelled the dynamical evolution of an HII region, including the interaction between gas and dust. Radiative transfer was approximated by diffusion. These two papers marked a tremendous step forward in the quality of models for infrared sources. To take advantage of numerical techniques which had been used in modelling stellar atmospheres, Leung essentially assumed that the radiation field is isotropic through most of the cloud (the Eddington approximation).Rowan-Robinson (1980) carried out a numerical solution of the full radiative transfer equation. The intensity, I ν (θ, r), was calculated throughout the cloud using a grid of values in ν, r, θ. This showed that the Eddington approximation is a poor assumption throughout a dust cloud and revealed the phenomenon of sideways beaming of the radiation field at the inner edge of cloud. Viewed from a distance a dust cloud surrounding a star should show a ring of bright emission in the mid infrared, coming from the inner edge of the dust cloud. A second interesting phenomenon revealed by a full solution of the radiative transfer equation is that of back radiation onto the star. The dust cloud radiates at the star and so the net spectrum of emission emerging from the star is no longer a blackbody, but has absorption bands at the wavelengths of dust emission features (Rowan-Robinson 1982). Rowan-Robinson & Harris (1982 applied this code to all the bright circumstellar dust shells in the AFGL survey.
446at https://www.cambridge.org/core/terms. https://doi