2023MNRAS.521.3108S

Proximity zones of high-redshift quasars are unique probes of supermassive black hole formation, but simultaneously explaining proximity zone sizes and black hole masses has proved to be challenging. We study the robustness of some of the assumptions that are usually made to infer quasar lifetimes from proximity zone sizes. We show that small proximity zones can be readily explained by quasars that vary in brightness with a short duty cycle of f_duty ∼ 0.1 and short bright periods of t_on ∼ 10⁴ yr, even for long lifetimes. We further show that reconciling this with black hole mass estimates requires the black hole to continue to grow and accrete during its obscured phase. The consequent obscured fractions of ≳ 0.7 or higher are consistent with low-redshift measurements and models of black hole accretion. Such short duty cycles and long obscured phases are also consistent with observations of large proximity zones, thus providing a simple, unified model for proximity zones of all sizes. The large dynamic range of our simulation, and its calibration to the Lyman-α forest, allows us to investigate the influence of the large-scale topology of reionization and the quasar's host halo mass on proximity zones. We find that incomplete reionization can impede the growth of proximity zones and make them smaller up to 30 per cent, but the quasar host halo mass only affects proximity zones weakly and indirectly. Our work suggests that high-redshift proximity zones can be an effective tool to study quasar variability and black hole growth.