IMPLICATIONS OF WMAP 3 YEAR DATA FOR THE SOURCES OF REIONIZATION

New results on the anisotropy of the cosmic microwave background (CMB) and its polarization based on the first 3 years of data from the Wilkinson Microwave Anisotropy Probe (WMAP) have revised the electron scattering optical depth downward from tau(es) = 0.17(-0.07)(+0.08) to tau(es) = 0.09 +/- 0.03. This implies a shift of the effective reionization redshift from z(r) similar or equal to 17 to z(r) similar or equal to 11. Previous attempts to explain the high redshift of reionization inferred from the r r WMAP 1 year data have led to widespread speculation that the sources of reionization must have been much more efficient than those associated with the star formation observed at low redshift. This is consistent, for example, with the suggestion that early star formation involved massive, Population III stars that early on produced most of the ionizing radiation escaping from halos. It is therefore tempting to interpret the new WMAP results as implying that we can now relax those previous high demands on the efficiency of the sources of reionization and perhaps even turn the argument around as evidence against such high efficiency. We show that this is not the case, however. The new WMAP results also find that the primordial density fluctuation power spectrum has a lower amplitude, sigma(8), and departs substantially from the scale-invariant spectrum. We show that these effects combine to cancel the impact of the later reionization implied by the new value of tau(es) on the required ionizing efficiency per collapsed baryon. The delay of reionization is surprisingly well matched by a comparable delay (by a factor of similar to 1.4 in scale factor) in the formation of the halos responsible for reionization.