Interpreting the long-term variability of the changing-look AGN Mrk 1018

Abstract

We present a comprehensive study of the changing-look active galactic nucleus (CL-AGN) Mrk 1018 based on the largest dataset of optical, UV, and X-ray spectro-photometric data ever assembled for this source. Our investigation comprises a detailed analysis of X-ray spectra, broad-band photometry, and optical-to-X-ray spectral energy distribution (SED) fitting, with the aim being to unravel the nature of the changing-look behavior observed in Mrk 1018 between 2005 and 2019. Based on the results of our analysis, we confirm that, in those 14 years, the X-rays from the source underwent a significant spectral variation, with the hardness ratio between the 0.5−2 keV band and the 2−10 keV band increasing from 0.2 ± 0.1 to 0.4 ± 0.1. We also validate the dramatic broad-band dimming, with the optical, UV, and X-ray luminosities decreasing by a factor of > 7, > 24, and ∼9, respectively. We find that the declining UV emission is driving these drops. By describing the X-ray spectra with a two-Comptonization model, with a hot (kT ∼ 100 keV) and a warm (kT < 1 keV) Comptonizing medium reprocessing the photons from the accretion disk, we reach the conclusion that, between 2005 and 2019, the properties of the hot medium remained the same, while the warm component cooled down from a temperature of ∼0.4 keV to ∼0.2 keV. This cooling can be explained by the weakening of the magnetic fields in the accretion disk and is also the source of the UV dimming. We propose that this decline is caused by the formation of a jet, itself originating from the change in the state of the inner accretion flow from a geometrically thin, optically thick structure to a geometrically thick, optically thin flow. Our optical-to-X-ray SED fitting seems to support this conclusion, as the estimated accretion rate normalized to the Eddington rate in the bright state (μ ∼ 0.06) is above the critical value μ = 0.02 for a stable radiative flow, while in the faint state we find μ ∼ 0.01 < 0.02, which is compatible with advective accretion. Instabilities arising at the interface of the state transition are then able to reduce the viscous timescale from ∼105 years to the observed ∼10 years of Mrk 1018 variability, reconciling all the observational properties of this CL-AGN into a complex but elegant physically motivated framework. Finally, we explored a possible mechanism triggering the state transition of the inner accretion flow. Our speculation is that gaseous clouds are pushed onto the innermost regions of the AGN by a galactic (dynamical friction) and/or an extragalactic process (wet merger, cold chaotic accretion). When one of these clouds passes by, it deposits material onto the accretion disk, causing the accretion flow to “puff up”, establishing the state transition. If this scenario is confirmed by future numerical simulations, it will open a new branch of study to place CL-AGN into our current understanding of the feeding and feedback of AGN. We also think that our results can be applied to other CL-AGN as well, and speculate that an accretion rate of μ ∼ 0.02, coupled with minor “disturbances” in the accretion disk, could indeed be the primary factor prompting the complex changing-look phenomenon.