Bright and Faint Ends of Lyα Luminosity Functions at z = 2 Determined by the Subaru Survey: Implications for AGNs, Magnification Bias, and ISM H I Evolution

We present the Lyα luminosity functions (LFs) derived by our deep Subaru narrowband survey that identifies a total of 3137 Lyα emitters (LAEs) at z = 2.2 in five independent blank fields. This sample of LAEs is the largest to date and covers a very wide Lyα luminosity range of {log}{L}_{Lyα }=41.7{--}44.4 erg s^-1. We determine the Lyα LF at z = 2.2 with unprecedented accuracy and obtain the best-fit Schechter parameters of {L}_{Lyα }^*={5.29}_-1.13^+1.67× {10}⁴² erg s^-1, {φ }_{Lyα }^*={6.32}_-2.31^+3.08× {10}^-4 Mpc^-3, and α =-{1.75}_-0.09^+0.10, showing a steep faint-end slope. We identify a significant hump at the LF bright end ({log}{L}_{Lyα }\gt 43.4 erg s^-1). Because all of the LAEs in the bright-end hump have a bright counterpart(s) in either the X-ray, UV, or radio data, this bright-end hump is not made by gravitational lensing magnification bias but by active galactic nuclei (AGNs). These AGNs allow us to derive the AGN UV LF at z ∼ 2 down to the faint magnitude limit of M _UV ≃ -22.5 and to constrain the faint-end slope of the AGN UV LF, α _AGN = -1.2 ± 0.1, which is flatter than those at z > 4. Based on the Lyα and UV LFs from our and previous studies, we find an increase of Lyα escape fraction {f}_{esc}^{Lyα } from z ∼ 0 to 6 by two orders of magnitude. This large {f}_{esc}^{Lyα } increase can be explained neither by the evolution of stellar population nor by outflow alone, but by the evolution of neutral hydrogen H I density in the interstellar medium that enhances dust attenuation for Lyα by resonance scattering. Our uniform expanding shell models suggest that the typical H I column density decreases from {N}_{{H}{{I}}}∼ 7× {10}¹⁹ (z ∼ 0) to ∼1 × 10¹⁸ cm^-2 (z ∼ 6) to explain the large {f}_{esc}^{Lyα } increase.