Exploring the nature of UV-bright z ≳ 10 galaxies detected by JWST: star formation, black hole accretion, or a non-universal IMF?

Abstract

We use the Cosmic Archaeology Tool (CAT) semi-analytical model to explore the contribution of Population (Pop) III/II stars and active galactic nuclei (AGNs) to the galaxy ultraviolet (UV) luminosity function (LF) evolution at 4 ≤ z ≤ 20. We compare in particular with recent JWST data in order to explore the apparent tension between observations and theoretical models in the number density of bright galaxies at z ≳ 10. The model predicts a star formation history dominated by UV faint (M_UV > -18) galaxies, with a Pop III contribution of $\lesssim 10~{{\ \rm per\ cent}}$ ($\lesssim 0.5~{{\ \rm per\ cent}}$) at z ≃ 20 (z ≃ 10). Stars are the primary sources of cosmic reionization, with $5~{{\ \rm per\ cent}}\!-\!10~{{\ \rm per\ cent}}$ of ionizing photons escaping into the intergalatic medium at 5 ≤ z ≤ 10, while the contribution of unobscured AGNs becomes dominant only at z ≲ 5. The predicted stellar and AGN UV LFs reproduce the observational data at 5 ≲ z ≲ 9-10. At higher redshift, CAT predicts a steeper evolution in the faint-end slope (M_UV > -18), and a number density of bright galaxies (M_UV ≃ -20) consistent with data at z ~ 10-11, but smaller by 0.8 dex at z ~ 12-13, and 1.2 dex at z ~ 14-16, when compared to the values estimated by recent studies. Including the AGN emission does not affect the above findings, as AGNs contribute at most to $\lesssim 10~{{\ \rm per\ cent}}$ of the total UV luminosity at M_UV < -19 and z ≳ 10. Interestingly, considering a gradual transition in the stellar initial mass function, modulated by metallicity and redshift as suggested by recent simulations, the model agrees with JWST data at z ~ 12-13, and the disagreement at z ~ 14-16 is reduced to 0.5 dex.