On the Coevolution of the AGN and Star-forming Galaxy Ultraviolet Luminosity Functions at 3 < z < 9

Abstract

Studies of the rest-frame ultraviolet (UV) luminosity functions (LFs) typically treat star-forming galaxies and active galactic nuclei (AGNs) separately. However, modern ground-based surveys now probe volumes large enough to discover AGNs at depths sensitive enough for fainter galaxies, bridging these two populations. Using these observations as constraints, we present a methodology to empirically jointly model the evolution of the rest-UV LFs at z = 3–9. Our critical assumptions are that both populations have LFs well described by double power laws modified to allow for a flattening at the faint-end, and that all LF parameters evolve smoothly with redshift. This methodology provides a good fit to the observations and makes predictions to volume densities not yet observed, finding that the volume density of bright (M UV = −28) AGNs rises by ∼105 from z = 9 to z = 3, while bright (M UV = −21) star-forming galaxies rise by only ∼102 across the same epoch. The observed bright-end flattening of the z = 9 LF is unlikely to be due to AGN activity, and rather is due to a shallowing of the bright-end slope, implying a reduction of feedback in bright galaxies at early times. The intrinsic ionizing emissivity is dominated by star-forming galaxies at z > 3, even after applying a notional escape fraction. We find decent agreement between our AGN LFs and predictions based on different black hole seeding models, though all models underpredict the observed abundance of bright AGNs. We show that the wide-area surveys of the upcoming Euclid and Roman observatories should be capable of discovering AGNs to z ∼ 8.