TY - JOUR

T1 - Mass Measurements of AGN from Multi-Lorentzian Models of X-ray Variability. I. Sampling Effects in Theoretical Models of the rms^2-M_BH Correlation

AU - Pessah, Martin Elias

PY - 2006/9/19

Y1 - 2006/9/19

N2 - Recent X-ray variability studies suggest that the log of the square of the fractional rms variability amplitude, rms^2, seems to correlate with the log of the AGN black-hole mass, M_BH, with larger black holes being less variable for a fixed time interval. This has motivated the theoretical modeling of the rms^2-M_BH correlation with the aim of constraining AGN masses based on X-ray variability. A viable approach to addressing this problem is to assume an underlying power spectral density with a suitable mass dependence, derive the functional form of the rms^2-M_BH correlation for a given sampling pattern, and investigate whether the result is consistent with the observations. For simplicity, previous studies, inspired by the similarities shared by the timing properties of AGN and X-ray binaries, have explored model power spectral densities characterized by broken power laws. and ignored, in general, the distorting effects that the particular sampling pattern imprints in the observed power spectral density. Motivated by the latest timing results from X-ray binaries, obtained with RXTE, we propose that AGN broad-band noise spectra consist of a small number of Lorentzian components. This assumption allows, for the first time, to fully account for sampling effects in theoretical models of X-ray variability in an analytic manner. We show that, neglecting sampling effects when deriving the fractional rms from the model power spectral density can lead to underestimating it by a factor of up to 80% with respect to its true value for the typical sampling patterns used to monitor AGN. We discuss the implications of our results for the derivation of AGN masses using theoretical models of the rms^2-M_BH correlation. (Abridged)

AB - Recent X-ray variability studies suggest that the log of the square of the fractional rms variability amplitude, rms^2, seems to correlate with the log of the AGN black-hole mass, M_BH, with larger black holes being less variable for a fixed time interval. This has motivated the theoretical modeling of the rms^2-M_BH correlation with the aim of constraining AGN masses based on X-ray variability. A viable approach to addressing this problem is to assume an underlying power spectral density with a suitable mass dependence, derive the functional form of the rms^2-M_BH correlation for a given sampling pattern, and investigate whether the result is consistent with the observations. For simplicity, previous studies, inspired by the similarities shared by the timing properties of AGN and X-ray binaries, have explored model power spectral densities characterized by broken power laws. and ignored, in general, the distorting effects that the particular sampling pattern imprints in the observed power spectral density. Motivated by the latest timing results from X-ray binaries, obtained with RXTE, we propose that AGN broad-band noise spectra consist of a small number of Lorentzian components. This assumption allows, for the first time, to fully account for sampling effects in theoretical models of X-ray variability in an analytic manner. We show that, neglecting sampling effects when deriving the fractional rms from the model power spectral density can lead to underestimating it by a factor of up to 80% with respect to its true value for the typical sampling patterns used to monitor AGN. We discuss the implications of our results for the derivation of AGN masses using theoretical models of the rms^2-M_BH correlation. (Abridged)

KW - astro-ph

U2 - 10.1086/509611

DO - 10.1086/509611

M3 - Journal article

VL - 655

SP - 66

EP - 76

JO - Astrophysical Journal

JF - Astrophysical Journal

SN - 0004-637X

IS - 1

ER -