The Hα luminosity function of galaxies AT z ~ 4.5

Abstract

The Luminosity Function (LF) is one of the most fundamental observables to study galaxies; moreover, choosing a luminosity tracer that strongly correlates with the Star Formation Rate (SFR) allows us to study the evolution of star formation across cosmic time. Thus, the relationship between luminosity and SFR must be as direct and reliable as possible. The UV luminosity is a direct tracer of SFR but it is very sensitive to dust, so corrections are large and uncertain. A better alternative is to use the rest-frame optical Hα emission line at 6563 Å, which is one of the most used SFR estimators up to z ≲ 3. However, Hα emission spectroscopy at high redshift (z ≳ 4) will only be available with the first data from the James Webb Space Telescope (JWST). Despite the existing observational difficulties, it is possible to use an indirect method based on deep Spitzer/IRAC near-infrared photometry to determine the Hα flux at z > 3. The redshifted Hα emission contributes to the flux measured in one of the IRAC bands at specific redshifts, while the other bands sample strictly the stellar continuum. This color offset can be used to estimate the flux of the Hα line. However, even though this method has been used for almost a decade, a systematic study of the Hα luminosity function at high redshift (z ∼ 4.5) is still lacking. In this thesis, we present the Hα Luminosity Function (Hα LF) derived from a large sample of Lyman Break Galaxies at z ∼ 4.5. This study makes use of the deepest Spitzer/IRAC [3.6] and [4.5] imaging to date from the GOODS Re-ionization Era wide-Area Treasury from Spitzer (GREATS) program, reaching up to 250 hrs of integration. The Hα flux is derived from the offset between the continuum flux estimated from the best-fit Spectral Energy Distribution (SED), and the observed photometry in IRAC [3.6]. Moreover, we study the evolution of the Hα LF providing the best constraints at high redshift. We find that SFRs derived from Hα are higher than those derived from rest-frame UV for low SFR galaxies but the opposite happens for the highest SFRs. This could be explained by lower mass galaxies (also lower SFR) having, on average, rising star formation histories (SFHs), while at the highest masses the SFHs may be declining. It could also be explained by the different timescales of SFR that Hα and UV luminosities trace. The SFR function is steeper and the star formation rate density estimated from Hα is higher than the previous estimates based on UV luminosities. Compared with previous works at lower redshifts, the Schechter parameterizations of the Hα LF show a decreasing normalization factor Φ* with redshift, the characteristic luminosity L* becoming brighter at higher redshifts, and no significant evolution in the faint-end slope α at high-z.