2025A&A...699A.202G

Context. Accurate black hole mass (M_BH) measurements in high-redshift galaxies are difficult yet crucial to constrain the growth of supermassive BHs, and to discriminate between competing BH seed models. Recent studies claimed the detection of massive BHs in very distant AGN, implying extreme conditions. However, these estimates are usually obtained by extrapolating indirect methods that are calibrated for moderately accreting, low-luminosity AGN in the local universe.
Aims. We want to assess the reliability of the single epoch method (SE) in the distant universe.
Methods. We computed the M_BH values for a sample of hyperluminous distant quasars (the X-WISSH sample) and a sample of highly accreting AGN (X-HESS) using the X-ray scaling method.
Results. We first verified that the X-ray scaling method yields reliable M_BH values also for distant highly accreting objects. Then, we carried out a systematic comparison with the SE method and found that these two indirect methods are fully consistent over a broad range of luminosities, intrinsic absorption, and accretion rates. The only discrepancies were associated with AGN that are substantially absorbed (underestimated by the SE method) and AGN accreting well above the Eddington limit (overestimated by the SE method). The latter result casts some doubts on the claim of overmassive BHs in highly accreting AGN in the early universe. Our study also reveals that one of the frequently used AGN catalogs consistently underestimates M_BH by a factor of 2.5. Although this factor is on the order of the uncertainty associated with the SE method, we demonstrated that using these underestimated values can result in potentially misleading conclusions. Specifically, for this AGN sample we confirmed strong positive correlations for Γ versus λ_Edd, for the X-ray bolometric correction versus λ_Edd, and for Γ versus the soft excess strength, at odds with the conclusions inferred using underestimated M_BH values.