Empirical Constraints on the Star Formation & Redshift Dependence of the Lyman Alpha `Effective' Escape Fraction

TL;DR

This study empirically constrains how the effective escape fraction of Lyman Alpha photons from star-forming galaxies varies with redshift, revealing an increase from 1-5% at z=0 to 30-50% at z=6, and explores models explaining this evolution.

Contribution

It provides the first empirical constraints on the redshift evolution of the Lyman Alpha escape fraction and compares these with theoretical models including winds and variable escape scenarios.

Findings

Escape fraction increases with redshift, from 1-5% at z=0 to 30-50% at z=6.

Models including winds better match the high escape fraction at z=6.

Escape fraction can be constant or decrease with SFR, with models including scatter fitting observations.

Abstract

We derive empirical constraints on the volume averaged `effective' escape fraction of Lyman Alpha (Lya) photons from star forming galaxies as a function of redshift, by comparing star formation functions inferred directly from observations, to observed Lya luminosity functions. Our analysis shows that the effective escape fraction increases from f_esc^eff ~ 1-5 % at z=0, to f_esc^eff ~ 10 % at z=3-4, and to f_esc^eff=30-50 % at z=6. Our constraint at z=6 lies above predictions by models that do not include winds, and therefore hints at the importance of winds in the Lya transfer process (even) at this redshift. We can reproduce Lya luminosity functions with an f_esc^eff that does not depend on the galaxies star formation rates (SFR) over up to ~2 orders of magnitude in Lya luminosity. It is possible to reproduce the luminosity functions with an f_esc^eff that decreases with SFR - which appears favored by observations of drop-out galaxies - in models which include a large scatter (~ 1.0 dex) in f_esc^eff, and/or in which star forming galaxies only have a non-zero f_esc^eff for a fraction of their life-time or a fraction of sightlines. We provide a fitting formula that summarizes our findings.