Observations indicate that the cosmological reionization history may be quite complex, and to understand at a fundamental level how the universe was reionized and what reionized it, requires a satisfactory understanding of the formation of galaxies during the reionization epoch. Current and future observations are, and will be, detecting many galaxies in the reionization epoch above redshift six, and interpreting this rich dataset will require a comprehensive theoretical framework for galaxies in halos of tens of millions to ten billion solar masses, constructed as much as possible ab initio and from first principles in the standard cold dark matter model. This research program will establish such a framework by carrying out ultra-high-resolution, cosmological, adaptive mesh refinement radiation-hydrodynamic simulations focusing on galaxies in the right mass range, combining a detailed physical treatment with sophisticated post-simulation analysis. The work will significantly advance the fields both of galaxy formation at high redshift and of cosmological reionization.
These research activities will be integrated into the training of graduate and undergraduate students, and will feed into an effective outreach program that improves the scientific understanding of the general public about cosmological research. The demands of carrying out these simulations push technological advances in supercomputers and related hardware and software, and drive the development of national research infrastructure.