Semi-analytic modelling of galaxy evolution in the IR/submm range

This paper proposes a new semi-analytic modelling of galaxy properties in the IRI submm wavelength range, which is explicitly set in a cosmological framework. We start from a description of the non-dissipative and dissipative collapses of primordial perturbations, and add star formation, stellar evolution and feedback, as well as the absorption of starlight by dust and its re-emission in the IR and submm. This type of approach has had some success in reproducing the optical properties of galaxies. We hereafter propose a simple extension to the IR/submm range. The growth of structures is followed according to the standard cold dark matter model. We assume that star formation proceeds either in a 'quiescent' mode, e.g., as in discs, or in a 'burst' mode with 10 times shorter time-scales. In order to reproduce the current data on the evolution of the comoving cosmic star formation rate and gas densities, we need to introduce a mass fraction involved in the 'burst' mode strongly increasing with redshift, probably reflecting the increase of interaction and merging activity. We estimate the IR/submm luminosities of these 'mild starburst' and 'luminous UV/ IR galaxies', and we explore how much star formation could be hidden in heavily extinguished, 'ultraluminous IR galaxies' by designing a family of evolutionary scenarios which are consistent with the current status of the 'cosmic constraints', as well as with the IRAS 60-mu m luminosity function and faint counts, but with different high-z IR luminosity densities. However,these scenarios generate a cosmic infrared background whose spectrum falls within the +/-1 sigma range of the isotropic IR component detected by Puget et al. and revisited by Guiderdoni et al. We give predictions for the faint galaxy counts and redshift distributions at IR and submm wavelengths. The submm range is very sensitive to the details of the evolutionary scenarios. As a result, the ongoing and forthcoming observations with ISO and SCUBA land later with SIRTF, SOFIA, FIRST and PLANCK) will put strong constraints on the evolution of galaxies at z similar to 1 and beyond.