Local reionizations histories with merger tree of HII regions

We constrain the initial stage of the reionization process around progenitors of galaxies, such as the extent of the initial HII region before its fusion with the UV background and the duration of its propagation. We use a set of reionisation simulations with different resolutions and ionizing source recipes. A catalog of the HII regions properties is built thanks to a merger tree of HII regions. We draw local reionization histories as a function of time and investigate variations according to the halo mass progenitors of the regions. We then extrapolate the halo mass inside the region from high z to z=0 to make predictions about the reionization histories of z=0 galaxies. We found that the later an HII region appears, the smaller will be its related lifetime and volume before it sees the global UV background. Quantitatively the duration and the extent of the initial growth of an HII region is strongly dependent on the mass of the inner halo and can be as long as 50 % of the reionization epoch. We found that the most massive is a halo today, the earlier it appears and the larger are the extension and the duration of propagation of its HII region. Quantitative predictions differ depending on the box size or the source model: small simulated volumes are affected by proximity effects between HII regions and halo-based source models predict smaller regions and slower I-front expansion than in models using star particles as ionizing sources. Our results suggests that Milky Way-type halos have a maximal extent of 1.1 Mpc/h for the initial HII region that established itself in 150-200$\pm 20$ Myrs. This is consistent with prediction made using constrained Local Group simulation. Considering halos with masses comparable to those of the Local Group (MW+M31), our result suggests that statistically it has not been influenced by an external front coming from a Virgo-like cluster.
Comment: 16 pages, 9 figures, Accepted for publication in A&A