The Spectral Appearance of Primeval Galaxies

The current and forthcoming observations of large samples of high–redshift galaxies selected according to various photometric and spectroscopic criteria can be interpreted in the context of galaxy formation, by means of models of evolving spectral energy distributions (SEDs). We hereafter present stardust which gives synthetic SEDs from the far UV to the submm wavelength range. These SEDs are designed to be implemented into semi–analytic models of galaxy formation. 1. The zoo of high–redshift galaxies The search for primeval galaxies has been one of the long–term programs of observational cosmology, along with the development of sensitive detectors, and is currently receiving an exciting boost with the new generation of 8–metre class telescopes. However, the search itself is somewhat hampered by the fuzziness of the concept of “primeval galaxy”. This is generally understood as being “a galaxy which is captured at the epoch of its formation”. Since there are several competing theories about what a forming galaxy should look like, it is readily possible that we are missing part of the process of galaxy formation because we do not know yet what to search for. More specifically, two general paradigms have been proposed in the last twenty years or so. In the picture of monolithic collapse, galaxies form at a given epoch zfor, when the physical conditions of the universe are favourable, and evolve at different rates which are fixed by the initial conditions. In the picture of hierarchical collapse, there is nothing like a given redshift of galaxy formation. Larger galaxies form from the merging of smaller ones, which on their turn have formed from the merging of still smaller lumps, and so on. The beginning of the process took place at some early redshift z ∼ 30 (when the first objects can cool) and is still going on now. As a result, the “epoch of galaxy formation” can be defined e.g. as the epoch when the first stars formed, when 50 % of the stars have formed, or when the morphology was fixed after the last major merging event. This hierarchical galaxy formation is now modelled in the context of hierarchical clustering where dark matter completely rules gravitational collapse. The models of galaxy formation have to reproduce the wide variety of objects which are now observed at high redshift, probably after strong observational biasses that are not fully understood. These objects are generally selected according to photometric criteria. We now have Luminous Blue Compact Galaxies (LBCGs; z ∼ 1 and IAB < 22.5), Lyman–Break Galaxies (LBGs; z ≃ 3 or z ≃ 4,