State-dependent broadband X-ray timing reconfiguration in the changing-look AGN NGC 1566 2026 State-dependent broadband X-ray timing reconfiguration in the changing-look AGN NGC 1566 – 4
Y. Tao, J. Tang, X. Wei, X.-H. Zhang · 2026
Read State-dependent broadband X-ray timing reconfiguration in the changing-look AGN NGC 1566 2026 State-dependent broadband X-ray timing reconfiguration in the changing-look AGN NGC 1566 – 4 by Y. Tao, J. Tang, X. Wei, X.-H. Zhang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-17290.txt.
Introduction
Tao and co-authors track how the X-ray timing behavior of the changing look galaxy NGC 1566 shifted during its outburst. This galaxy showed dramatic spectral changes, but its timing traits were poorly known. The team joins long term Swift monitoring with XMM-Newton data to build power spectral densities before and during the outburst. Bending power law models fit the quiet and decay states well. The bend frequency shifts down by about two orders of magnitude in the decay state, in both soft and hard bands.
Catalog record
| CREATOR | Y. Tao, J. Tang, X. Wei, X.-H. Zhang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 18960 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.17290
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
State-dependent broadband X-ray timing reconfiguration in the changing-look AGN NGC 1566 2026 State-dependent broadband X-ray timing reconfiguration in the changing-look AGN NGC 1566 – 4 Y. Tao, J. Tang, X. Wei, X.-H. Zhang Abstract NGC 1566 has exhibited dramatic state changes in its X-ray spectrum, but the evolution of its broadband timing properties remains poorly constrained. We combine long-term Swift monitoring with high-time-resolution XMM-Newton observations to model the broadband X-ray power spectral density (PSD) in the dim and bright states. In the hard band, the PSD bend frequency shifts by about 1 dex between the two states, implying a substantially longer characteristic variability timescale in the bright state. The relative timing behaviour of the soft and hard bands also changes with state: in the dim state, the soft-band bend frequency is higher than the hard-band value by ∼ \sim 0.49 dex, whereas in the bright state the two become broadly consistent. The broadband variability evolution of NGC 1566 therefore involves not only an overall shift in characteristic timescale, but also a state-dependent change in the soft–hard timing relation, from a more stratified to a more tightly coupled configuration. Combined with previous spectral results, this supports a genuine reconfiguration of the inner radiative structure during the changing-look transition. 1 Introduction Changing-look active galactic nuclei (CL AGNs) provide a valuable opportunity to probe rapid changes in the inner accretion flow across different radiative states. Multiwavelength observations increasingly suggest that, in many cases, their dramatic transitions are driven primarily by intrinsic accretion-flow changes rather than by line-of-sight obscuration alone ( 11 ; 8 , e.g.) . NGC 1566 is one of the best-studied examples, particularly following its extreme 2018 outburst ( 9 ; 10 ) . Broadband X-ray studies showed that, near the peak of the outburst, the source displayed the spectral properties of a Seyfert 1 nucleus and that its behaviour was most likely associated with intrinsic accretion-flow evolution ( 10 ) . Later work further showed substantial changes in both the soft excess and the broadband continuum, together with systematic evolution in the disc, soft-excess, and coronal properties, supporting a reconfiguration of the inner radiative structure ( 3 ; 14 ) . In contrast, the broadband timing behaviour of NGC 1566 remains much less well constrained. The bend frequency f_{\rm b} in the power spectral density (PSD) defines a characteristic variability timescale, t_{\rm b}=1/f_{\rm b} , so a shift in f_{\rm b} implies a corresponding change in the dominant X-ray variability timescale ( 7 ; 6 ) . Bend frequencies measured in different energy bands can also provide clues to how the variability-producing regions are coupled ( 1 ; 15 , e.g.) . This makes NGC 1566 an especially useful target for testing whether its changing-look transition is accompanied by a measurable evolution in characteristic X-ray timing behaviour. Here we combine long-term Swift monitoring with high-time-resolution XMM-Newton observations to model the broadband X-ray PSD of NGC 1566 in its dim and bright states. We ask whether the bend frequency shifts significantly between states, and whether the soft–hard timing relation evolves as well. We show that the broadband PSD is reconfigured between the dim and bright states, with significant changes in both the bend frequency and its energy dependence. 2 Data and methods 2.1 Data reduction and state selection We combine long-term Swift /XRT monitoring with high-time-resolution XMM-Newton observations to characterise the broadband X-ray variability of NGC 1566. Using the publicly available Swift /XRT 2–10 keV monitoring light-curve products retrieved from the online archive ( 2 ) , we define a dim-state interval spanning MJD 56185–57233 and a bright-state interval spanning MJD 58293–58449. These background-subtracted, exposure-corrected Swift light curves are used directly to constrain the low-frequency variability. For the XMM-Newton data, we reduce the EPIC-pn observations falling within these two state intervals (ObsID 0763500201 for the dim state and ObsID 0800840201 for the bright state) using SAS (v22.1.0), following standard procedures including the generation of calibration files, event reconstruction, flare filtering based on the 10–12 keV background light curve, and event selection with PATTERN <= 4 and FLAG = 0 . We assess pile-up using epatplot and find it to be negligible in the bright-state observation. We then extract soft (0.3–2 keV) and hard (2–10 keV) light curves from a circular source region of radius 36 arcsec and a nearby source-free background region, and apply epiclccorr for background subtraction and instrumental corrections. The resulting XMM-Newton light curves, binned at 50 s, are used to constrain the high-frequency PSD. Together with the multi-year Swift monitoring, they provide complementary coverage of the low- and high-frequency variability for the joint broadband PSD modelling. Figure 1: Swift /XRT 2–10 keV light curve of NGC 1566. Figure 2: XMM-Newton 2–10 keV light curve of NGC 1566. 2.2 PSD modelling and joint constraints We model the PSD with a bending power law, where \alpha_{\rm low} and \alpha_{\rm high} describe the low- and high-frequency slopes, and f_{\rm b} defines the bend frequency. The main parameters of interest are f_{\rm b} , corresponding to the characteristic timescale t_{\rm b}=1/f_{\rm b} , and \alpha_{\rm high} , which describes the suppression of short-timescale variability. For the irregularly sampled Swift light curves, we construct Lomb–Scargle periodograms ( 4 ; 12 ) and quantify the low-frequency PSD constraint using a PSRESP-based forward-modelling approach that accounts for the sampling window, red-noise leak, and measurement noise ( 13 ; 16 ; 17 ) . For the continuous XMM-Newton light curves, we construct periodograms and evaluate the high-frequency constraint on the same (f_{\rm b},\alpha_{\rm high}) grid. At each grid point, the PSD normalization and Poisson-noise level are treated as nuisance parameters and optimised by minimising the Whittle negative log-likelihood in periodogram space. In practice, we restrict the XMM-Newton fit to frequencies below an empirically determined upper limit at which the binned periodogram remains above the estimated Poisson-noise floor ( 18 ) . We then combine the XMM-Newton Whittle likelihood with the logarithm of the Swift PSRESP acceptance fraction on a common (f_{\rm b},\alpha_{\rm high}) grid to define an empirical joint-constraint surface, focusing on the PSD shape parameters rather than on the absolute normalization. Here the Swift acceptance fraction is used as an empirical support term rather than as a formal Bayesian prior, and the resulting normalized joint weights are used only to rank solutions and to define weighted intervals. The Lomb–Scargle periodograms, PSRESP simulations, and joint grid calculations were carried out with custom Python scripts. We do not impose a single low-frequency slope across all states and bands. Instead, we first use single-power-law fits to the Swift PSDs to guide the choice of \alpha_{\rm low} , and then perform the bending-model search. In the 2–10 keV band, both states favour stable solutions with \alpha_{\rm low}=1 , which we adopt throughout the hard-band analysis. In the 0.3–2 keV band, the dim state favours a flatter low-frequency slope, and we therefore adopt \alpha_{\rm low}=0.85 as the fiducial choice while retaining \alpha_{\rm low}=1 as a control. The soft-band bright state is well described with \alpha_{\rm low}=1 . The qualitative conclusion that the soft-band bend frequency is higher than the hard-band value in the dim state, and that the soft–hard bend-frequency separation becomes smaller in the bright state, remains unchanged when