[Paper Review] Taking a Long Look: A Two-Decade Reverberation Mapping Study of High-Luminosity Quasars

S. Kaspi, W. N. Brandt+5 more|arXiv (Cornell University)|Jun 1, 2021

Astrophysical Phenomena and Observations4 citations

TL;DR

This study presents a 20-year reverberation mapping campaign on high-luminosity quasars, measuring time lags between UV continuum and C IV emission lines in three objects and C III] in one. It confirms a BLR size scaling with the square root of UV luminosity over eight orders of magnitude, with significant intrinsic scatter, and tentatively derives black hole masses scaling similarly, extending the virial mass relation to the most luminous AGNs.

ABSTRACT

Reverberation mapping (RM) of active galactic nuclei (AGNs) has been used over the past three decades to determine AGN broad-line region (BLR) sizes and central black-hole masses, and their relations with the AGN's luminosity. Until recently the sample of objects with RM data was limited to low-luminosity AGNs ($L_{ m opt} \lesssim 10^{46}$ ergs s$^{-1}$) and low redshifts ($z \lesssim 0.5$). Here we present results from a reverberation-mapping project of some of the most luminous and highest redshift quasars that have been mapped to date. The study is based on almost twenty years of photometric monitoring of 11 quasars, six of which were monitored spectrophotometrically for 13 years. This is the longest reverberation-mapping project carried out so far on this type of AGNs. We successfully measure a time lag between the CIV$λ$1549 broad emission line and the quasar continuum in three objects, and measure a CIII$λ$1909 lag in one quasar. Together with recently published data on CIV reverberation mapping, the BLR size is found to scale as the square root of the UV luminosity over eight orders of magnitude in AGN luminosity. There is a significant scatter in the relation, part of which may be intrinsic to the AGNs. Although the CIV line is probably less well suited than Balmer lines for determination of the mass of the black hole, virial masses are tentatively computed and in spite of a large scatter we find that the mass of the black hole scales as the square root of the UV luminosity.

Motivation & Objective

• To extend reverberation mapping to the most luminous and highest redshift quasars, where BLR sizes and variability timescales are large.
• To test whether the established BLR size–luminosity relation holds at extreme luminosities (L_opt ≳ 10^46 erg s⁻¹) and high redshifts (z ≳ 0.5).
• To measure time lags between continuum and broad emission lines (C IV, C III]) in high-luminosity quasars using long-term photometric and spectroscopic monitoring.
• To assess the reliability of virial mass estimates in high-luminosity quasars despite potential limitations of the C IV line for mass measurement.

Proposed method

• Conducted nearly two decades of photometric monitoring (1997–2016) of 11 high-luminosity quasars at the Wise Observatory, using B and R bands.
• Performed 13-year spectroscopic monitoring of six quasars to measure time lags between the C IV λ1549 emission line and the continuum.
• Applied cross-correlation techniques (ICCF and ZDCF) to estimate time lags between continuum (B and R bands) and emission-line light curves.
• Used model-independent cross-correlation and model-dependent methods (e.g., JAVELIN, SPEAR) to infer time delays, with careful assessment of significance and uncertainties.
• Combined photometric and spectroscopic light curves, merging R-band data with spectroscopic continuum light curves for six quasars.
• Calibrated magnitudes using non-variable stars from the USNO-A2.0 catalog and averaged nightly observations to reduce noise.

Experimental results

Research questions

• RQ1 Does the BLR size–luminosity relation derived from low-luminosity AGNs extend to the most luminous quasars (L_opt ≳ 10^46 erg s⁻¹)?
• RQ2 Can time lags between the UV continuum and the C IV emission line be reliably measured in high-luminosity, high-redshift quasars despite long timescales and sparse sampling?
• RQ3 What is the scatter in the BLR size–luminosity relation at extreme luminosities, and how much of it is intrinsic to the AGNs?
• RQ4 Can virial black hole masses be meaningfully estimated from the C IV line in high-luminosity quasars, despite its known limitations compared to Balmer lines?
• RQ5 How do the time lags between blue and red continuum bands compare in high-luminosity quasars, and do they align with expectations from reprocessing models?

Key findings

• Three high-luminosity quasars (SBS 1116+603, SBS 1425+606, S5 2017+744) show statistically significant time lags between the C IV λ1549 emission line and the UV continuum, with measured lags of -145⁺⁸⁵₋₆⁵, -264⁺³⁹₋₁₄₆, and -87⁺⁶²₋₅⁸ days (observed frame), respectively.
• One quasar (S5 0014+81) shows a significant time lag of -114⁺⁸⁹₋₂⁶ days between the C III] λ1909 emission line and the continuum.
• The BLR size scales with the square root of the UV luminosity (R_BLR ∝ L^0.5) over a dynamic range of eight orders of magnitude in luminosity, consistent with the standard relation.
• The scatter in the R_BLR–L relation is significant, with a large intrinsic component possibly due to differences in BLR structure or ionization conditions across quasars.
• Virial black hole masses derived from the C IV line show a scaling relation with luminosity (M_BH ∝ L^0.5), though with a large scatter, indicating caution in using C IV for mass estimation in high-luminosity AGNs.
• The R-band light curve lags behind the B-band light curve in several objects (e.g., SBS 1425+606, S5 2017+744), consistent with reprocessing models where red continuum emission lags blue emission due to dust reprocessing or thermal delays.