Dual Signatures of Bursty Star Formation in High-Redshift UV Luminosity Functions

Abstract

The UV luminosity function (UVLF) encodes key information about galaxy formation. The slowly evolving bright-end UVLFs at $z\gtrsim10$ have made UV variability from bursty star formation a promising explanation, but the impact of such variability need not be restricted to the bright end. Motivated by ultra-deep measurements of the $z\simeq7$ UVLF, we investigate whether UV variability can provide a unified interpretation of the low-mass star formation efficiency (SFE) at $z\simeq7$ and the abundance of UV-bright galaxies at $z\gtrsim12$. Using abundance matching, we show that a steep faint-end UVLF at $z\simeq7$ admits a range of interpretations. For modest, mass-independent UV variability, the inferred median SFE starts flattening below $M_{\rm h} \sim10^{10.5} M_{\odot}$. If UV variability instead grows with decreasing halo mass---a possibility motivated by simulations and observations---the inferred median SFE is steeper and lower by a factor of about three at $\sim10^9 M_{\odot}$, consistent with stronger feedback suppression in low-mass halos. Calibrated to the common target $z\simeq7$ UVLF, these two scenarios diverge strongly when extrapolated to $12\lesssim z\lesssim17$ under redshift-independent UV variability and SFE prescriptions. Strongly mass-dependent $M_\mathrm{UV}$ scatter links the steep $z\simeq7$ faint-end slope and high $z\gtrsim12$ bright-end abundance as dual signatures of burstiness in similarly low-mass halos. Stronger variability makes bright galaxies contribute a larger fraction of the UV luminosity density, whereas weaker variability favors faint sources, despite their similar predicted reionization histories. At fixed $M_\mathrm{UV}$, UV variability broadens the halo mass distribution and lowers the clustering bias, which provides a useful diagnostic of burstiness for JWST and Roman.