2007/08/28 by Hagai Netzer, P. Lira, Paulina Lira +3 · 158 citations
Mathematics · Physics and Astronomy · #Accretion (finance) #Active galactic nucleus #Astronomy #Astrophysical Phenomena and Observations #Astrophysics #Black hole (networking) #Estimator #Galaxies: Formation, Evolution, Phenomena #Galaxy #Gamma-ray bursts and supernovae #Luminosity #Mathematics #Physics #Redshift #Reverberation mapping #Spectral line #Statistics #astro-ph
paper · pdf · doi:10.1086/523035
published in The Astrophysical Journal 671(2), 1256-1263 (IOP Publishing) · 8 pages (emulateapj), 4 figures. Accepted for publication in ApJ
arxiv created 2007/08/28 · openalex publication_date 2007/12/20 · arxiv updated 2009/12/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/06
We present new H - and K -band spectroscopy of 15 high-luminosity active galactic nuclei (AGNs) at redshifts 2.3-3.4 obtained with Gemini South. We combined the data with spectra of an additional 29 high-luminosity sources to obtain a sample with 10 45.2 < λ L λ (5100 Å) < 10 47.3 ergs s −1 and a black hole (BH) mass range, using reverberation mapping relationships based on the Hβ method, of 10 8.8 -10 10.7 M ☉ . We do not find a correlation of L / L Edd with M BH , but we do find a correlation with λ L λ (5100 Å) , which might be due to selection effects. The L / L Edd distribution is broad and covers the range of ~0.07-1.6, similar to what is observed in lower redshift, lower luminosity AGNs. We suggest that this consistently measured and calibrated sample gives the best representation of L / L Edd at those redshifts, and note potential discrepancies with recent theoretical and observational studies. The lower accretion rates are not in accord with growth scenarios for BHs at such redshifts, and the growth times of many of the sources are longer than the age of the universe at the corresponding epochs. This suggests earlier episodes of faster growth at z ≳ 3 for those sources. The use of the C IV λ1549 method gives considerably different results and a larger scatter; this method seems to be a poor M BH and L / L Edd estimator at very high luminosity.