Speaker
Description
Aims. Massive black holes in the early Universe had limited time to assemble. We identify which JWST black-hole candidates place the strongest constraints on early growth models and assess how these constraints depend on mass estimates and growth assumptions. Methods. We compile 350 published measurements at $z\geq4$ and analyse a primary sample of 220 objects with broad Balmer-line masses. An expanded sample adds 13 tentative or conditional Balmer-line candidates and GN-z11, whose mass is inferred from ultraviolet lines. We propagate reported mass uncertainties and test mass-scale shifts and alternative growth assumptions. Results. For $100\,M_\odot$ seeds formed at $z=30$, radiative efficiency $\epsilon=0.1$, and no merger boost, 8 of the 220 objects require a lifetime-average Eddington ratio $\overline{f}_{\mathrm{Edd,req}}>1$. Five exceed unity in at least 95% of Monte Carlo draws using the reported mass uncertainties. For UNCOVER-20466, COSMOS3D-13852, and RUBIES-EGS-55604, at least 95% of draws still require an average Eddington ratio above 1 after reducing each mass by $0.5$ dex (a factor of approximately 3.2). Delaying seeding to $z=20$ gives GN-z11 the highest required average Eddington ratio. At $\epsilon=0.05719$, the ideal thin-disc efficiency for a non-spinning black hole, all 234 central mass estimates are reachable from $100\,M_\odot$ seeds formed at $z=30$ without lifetime-average accretion above the Eddington limit.
