The Universe Was Reionized Twice

We show the universe was reionized twice, first at z~15-16 and second at z~6.Such an outcome appears inevitable, when normalizing to two well determinedobservational measurements, namely, the epoch of the final cosmologicalreionization at z~6 and the density fluctuations at z~6, which in turn aretight ly constrained by Lyman alpha forest observations at z~3. These twoobservations most importantly fix the product of star formation efficiency andionizing photon escape fraction from galaxies at high redshift. To the extentthat the relative star formation efficiencies in gaseous minihalos with H2cooling and large halos with atomic cooling at high redshift are still unknown,the primary source for the first reionization could be Pop III stars either inminihalos or in large halos. We show that gas in minihalos can be cooledefficiently by H2 molecules and star formation can continue to take placelargely unimpeded throughout the first reionization period, thanks to two newmechanisms for generating a high X-ray background during the Pop III era, putforth here. Moreover, an important process for producing a large number of H2molecules in relic HII regions of Pop III galaxies, first pointed out byRicotti, Gnedin, & Shull, is quantified here. It is shown that the Lyman-Wernerbackground may never build up during the Pop III era. The long cosmologicalreionization and reheating history is complex. We discuss a wide range ofimplications and possible tests for this new reionization picture. Inparticular, Thomson scattering optical depth is increased to 0.10 +- 0.03,compared to 0.027 for the case of only one rapid reionization at z=6. UpcomingMicrowave Anisotropy Probe observation of the polarization of the cosmicmicrowave background should be able to distinguish between these two scenarios.Comment: submitted to ApJ, 69 pages, substantial revision made and conclusions strengthene