Age dependence of Lyα escape fraction of Lyα emitters and their significant role in cosmic reionization (2506.03242v1)
Abstract: We study the Ly$\alpha$ escape fraction, $f_{\mathrm{esc}}{\mathrm{Ly\alpha}}$ of Ly$\alpha$ emitters (LAEs) identified by Subaru/HSC narrowband imaging at $z = 2.2-6.6$, using publicly available deep imaging data from HST and JWST. We perform SED fitting for 127 LAEs at $0.4 - 5.0\, \mathrm{\mu m}$ to estimate their physical properties robustly, and confirm that two distinct LAE populations exist: young LAEs (< 100 Myr) and old LAEs (> 100 Myr). Young LAEs are characterized by burst-like star formation activity and low dust content, significantly differing from Lyman-break galaxies (LBGs) at the same stellar mass, while old LAEs show similar star formation activity to LBGs, yet with lower dust content and more compact morphology in rest-UV/optical than LBGs. The $f_{\mathrm{esc}}{\mathrm{Ly\alpha}}$ of LAEs is anticorrelated with stellar mass, and this correlation is found to depend on the age of LAEs, such that old LAEs show a weaker anticorrelation than young LAEs, and tend to exhibit higher $f_{\mathrm{esc}}{\mathrm{Ly\alpha}}$ than young LAEs at a given stellar mass. This implies that Ly$\alpha$ photons escape more efficiently from old LAEs, possibly through the low-density channels of HI and dust created by outflows. The average $f_{\mathrm{esc}}{\mathrm{Ly\alpha}}$ of young LAEs remains nearly constant at $\sim40$% at $z = 2.2-6.6$, suggesting that the previously observed evolution of global $f_{\mathrm{esc}}{\mathrm{Ly\alpha}}$ of star-forming galaxies (SFGs) is due to the changes in the LAE fraction among the SFGs. Converting $f_{\mathrm{esc}}{\mathrm{Ly\alpha}}$ to Lyman continuum escape fraction using empirical relations, we demonstrate that LAEs alone can supply the ionizing photons necessary for reionization at $z\sim6$, causing rapid and late reionization.
Collections
Sign up for free to add this paper to one or more collections.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.