SIMBAD references

Near-infrared observations of Lyman break galaxies at redshifts of z∼3 are beginning to provide constraints on ages, star formation histories, dust content, metallicities, and stellar masses. At present, uncertainties of more than an order of magnitude are typical for many of these parameters. It is nonetheless interesting to ask what the stellar population models imply for the existence and luminosities of Lyman break galaxies at higher redshift. To this end we examine the inferred star formation rates in two well-studied samples of galaxies as a function of redshift out to z=10 for various best-fitting and limiting cases. Taken at face value, the generally young ages (typically 10^8±0.5 yr) of the z=3 Lyman break galaxies imply that their stars were not present much beyond z=4. By z=6 the cosmic star formation rate ρ{dot}_SFR from the progenitors of these galaxies is less than 10% of ρ{dot}_SFR at z=3±0.5, even for maximally old models, provided the derivative of the star formation rate SFR(t) is monotonic. The escaping Lyman continuum radiation from such galaxies would be insufficient to reionize the intergalactic medium. Thus, other sources of ionizing photons (e.g., very massive Population III stars) may be needed, and the more normal Lyman break galaxies may be a phenomenon confined to redshifts of z≲4. This conclusion changes if SFR(t) was episodic, and we examine the parameters of such bursty star formation that might be consistent with both the z=2-4 luminosity functions and the z∼3 spectral energy distributions.