Empirical Modeling of the Redshift Evolution of the [ N II ] /Hα Ratio for Galaxy Redshift Surveys

Abstract

We present an empirical parameterization of the [N ii ]/H α flux ratio as a function of stellar mass and redshift valid at 0 <  z  < 2.7 and 8.5 < 11.0 . This description can (i) easily be applied to simulations for modeling [N ii ] λ 6584 line emission, (ii) deblend [N ii ] and H α in current low-resolution grism and narrow-band observations to derive intrinsic H α fluxes, and (iii) reliably forecast the number counts of H α emission-line galaxies for future surveys, such as those planned for Euclid and the Wide Field Infrared Survey Telescope ( WFIRST ). Our model combines the evolution of the locus on the Baldwin, Phillips & Terlevich (BPT) diagram measured in spectroscopic data out to z  ∼ 2.5 with the strong dependence of [N ii ]/H α on stellar mass and [O iii ]/H β observed in local galaxy samples. We find large variations in the [N ii ]/H α flux ratio at a fixed redshift due to its dependency on stellar mass; hence, the assumption of a constant [N ii ] flux contamination fraction can lead to a significant under- or overestimate of H α luminosities. Specifically, measurements of the intrinsic H α luminosity function derived from current low-resolution grism spectroscopy assuming a constant 29% contamination of [N ii ] can be overestimated by factors of ∼8 at log ( L ) > 43.0 for galaxies at redshifts z  ∼ 1.5. This has implications for the prediction of H α emitters for Euclid and WFIRST . We also study the impact of blended H α and [N ii ] on the accuracy of measured spectroscopic redshifts.