A critical examination of Seyfert 1 X-ray spectroscopy

Abstract

The field of high-energy AGN astrophysics is about to be transformed as data of unprecedented quality and quantity will be arriving from the next-generation of X-ray observatories. With the advancement of technology, we must remain critical of our methodologies to ensure the limits of particular techniques are well-defined. For example, the spin of supermassive black holes is of great interest for understanding the physics of accretion and black hole-host galaxy co-evolution. However, consistency and reproducibility of spin constraints have been a challenge even when the observable bandpass is increased with the launch of NuStar. We therefore begin by systematically evaluating the ability of X-ray modelling methods to measure known reflection parameters, like black hole spin, under standard conditions often seen in the literature. We created and fit over 4000 simulated Seyfert 1 spectra each with 375,000[plus or minus]1,000 counts and assessed the success of fits for a range of fixed parameters, reflection fractions, and bandpasses. We discover that most parameters are over-estimated when spectroscopy is restricted to the 2.5 – 10.0 keV regime and that models are insensitive to ionization unless the bandpass is extended. In all cases, we find the spin parameter is only well measured for the most rapidly rotating supermassive black holes (i.e. [alpha]>0.8 to about [plus or minus]0.10). With this knowledge in-hand, we next turned to empirical data by examining the complex short-term X-ray variability of narrow-line Seyfert 1 galaxy Mrk 493 in depth. We first utilized model-independent techniques such as timing studies and PCA as a way to guide the detailed spectroscopy. We were able to isolate two distinct periods in which the spectral variability of Mrk 493 shifts from one dominated by changes in the primary component to one dominated by changes in disk ionization. In addition, both spectroscopic analysis and model-independent tests point to an additional source of variability that is unaccounted for. Different scenarios including spallation, disk hot spot, and hot annulus are tested and discussed.