The dependence of quasar variability on black hole mass

Abstract

In order to investigate the dependence of quasar variability on fundamental physical parameters like black hole mass, we have matched quasars from the QUEST1 Variability Survey with broad-lined objects from the Sloan Digital Sky Survey. The matched sample contains approximately 100 quasars, and the Sloan spectra are used to estimate black hole masses and bolometric luminosities. Variability amplitudes are measured from the QUEST1 light curves. We find that black hole mass correlates with several measures of the variability amplitude at the 99% significance level or better. The correlation does not appear to be caused by obvious selection effects inherent to flux-limited quasar samples, host galaxy contamination or other well-known correlations between quasar variability and luminosity/redshift. We evaluate variability as a function of rest-frame time lag using structure functions, and find further support for the variability--black hole mass correlation. The correlation is strongest for time lags of the order a few months up to the QUEST1 maximum temporal resolution of approximately 2 years, and may provide important clues for understanding the long-standing problem of the origin of quasar optical variability. We discuss whether our result is a manifestation of a relation between characteristic variability timescale and black hole mass, where the variability timescale is typical for accretion disk thermal timescales, but find little support for this. Our favoured explanation is that more massive black holes have larger variability amplitudes, and we highlight the need for larger samples with more complete temporal sampling to test the robustness of this result.