The $z \sim 2$ $\rm{[O\ III]}$ Luminosity Function of Grism-selected Emission-line Galaxies

Abstract

Upcoming missions such as Euclid and the Nancy Grace Roman Space Telescope (Roman) will use emission-line selected galaxies to address a variety of questions in cosmology and galaxy evolution in the z > 1 universe. The optimal observing strategy for these programs relies upon knowing the number of galaxies that will be found and the bias of the galaxy population. Here we measure the [O III] λ5007 luminosity function for a vetted sample of 1951 m J+JH+H < 26 galaxies with unambiguous redshifts between 1.90 < z < 2.35, which were selected using HST/WFC3 G141 grism frames made available by the 3D-Hubble Space Telescope program. These systems are directly analogous to the galaxies that will be identified by the Euclid and Roman missions, which will utilize grism spectroscopy to find [O III] λ5007-emitting galaxies at 0.8 ∼ < z ∼ < 2.7 and 1.7 ∼ < z ∼ < 2.8, respectively. We interpret our results in the context of the expected number counts for these upcoming missions. Finally, we combine our dust-corrected [O III] luminosities with rest-frame ultraviolet star formation rates to present the first estimate of the SFR density associated with 1.90 < z < 2.35 [O III]-emitting galaxies. We find that these grism-selected galaxies contain roughly half of the total star formation activity at z ∼ 2.

Simulating properties of emission line galaxies from Nancy Grace Roman Space Telescope

“…Bagley et al (2020) construct a sample using WISP survey and forecast the properties of H𝛼 and [OIII] emitting galaxies from future surveys. Bowman et al (2021) adopt a similar method based on the data from HST/WFC3 G141 grism and measure the luminosity function of [OIII] galaxies for both Euclid and Roman. Baronchelli et al (2020) apply a machine learning algorithm on the identification of single emission line which can improve the survey completeness for spectroscopic surveys.…”