An empirically calibrated model for interpreting the evolution of galaxies during the reionization era

We develop a simple star formation model whose goal is to interpret the emerging body of observational data of star- forming galaxies at z greater than or similar to 6. The efficiency and duty cycle of the star formation activity within dark matter halos are determined by fitting the luminosity functions of Ly alpha emitter and Lyman break galaxies at redshifts z similar or equal to 5-6. Our error budget takes proper account of the uncertainty arising from both the spatial clustering of galaxies and the Poisson contribution. We find that the abundance of luminous Lyman break galaxies in the 500 Myr between z similar or equal to 6 and 10 can be naturally explained by the hierarchical assembly of dark matter halos; there is only marginal evidence for strong physical evolution. In contrast, the first estimates of the abundance of less luminous star- forming galaxies at z = 9-10 are higher than predicted and, if verified by further data, may suggest a top- heavy stellar mass function at these early epochs. Although these abundances remain uncertain because of the difficulty of spectroscopic confirmation and cosmic variance, even a modest improvement in survey capability with present or upcoming facilities should yield great progress. In this context, we use our model to consider those observational techniques that hold the most promise and make predictions for specific surveys that are, or will soon be, underway. We conclude that narrowband Ly alpha emitter surveys should be efficient for searches at z similar or equal to 7-8; however, such conventional surveys are unlikely to detect sufficient galaxies at z similar or equal to 10 to provide useful constraints. For this reason, gravitational lensing offers the best prospect for probing the z similar or equal to 10 universe prior to JWST.