The luminosity history of fading local quasars over 10$^{4-5}$ years as observed by VLT/MUSE

TL;DR

This study probes AGN variability on $t \sim 10^{4-5}$ year scales by exploiting large EELRs in five nearby galaxies observed with VLT/MUSE. Using CLOUDY photoionization models tied to spatially resolved line diagnostics, it derives $Q(H)$ and $L_{bol}$ at multiple radii, mapping these to a luminosity history via light-travel time. The results reveal a consistent fading pattern of $\sim$0.2–3 dex over $4\times10^{4}$–$8\times10^{4}$ years, in line with flickering AGN models and prior observational constraints. They further link this variability to merger-driven gas supply and accretion-disk instabilities, underscoring the potential long-term impact of AGN on their hosts and the need for larger, less biased samples to refine duty cycles and feedback implications.

Abstract

We present a comprehensive study of five nearby active galaxies featuring large (tens of kpc) extended emission-line regions (EELRs). The study is based on large-format integral field spectroscopic observations conducted with the Multi Unit Spectroscopic Explorer (MUSE) at the Very Large Telescope (VLT). The spatially resolved kinematics of the ionized gas and stellar components show signs of rotation, bi-conical outflows, and complex behavior likely associated with past interactions. Analysis of the physical conditions of the EELRs indicates that in these systems, the active galactic nucleus (AGN) is the primary ionization source. Using radiative transfer simulations, we compare the ionization state across the EELRs to estimate the required AGN bolometric luminosities at different radial distances. Then, considering the projected light travel time, we reconstruct the inferred AGN luminosity curves. We find that all sources are consistent with a fading trend in intrinsic AGN luminosity by 0.2--3 dex over timescales of 40,000--80,000 years, with a time dependence consistent with previous studies of fading AGNs. These results support the hypothesis that most AGN undergo significant fluctuations in their accretion rates over multiple timescales ranging from 10,000 to 1,000,000 years, as proposed by existing theoretical models. These results provide new insights into the transient phases of AGN activity at previously unexplored scales and their potential long-term impact on their host galaxies through various feedback mechanisms.

Figures (23)

• Figure 1: Composite color images for the targets analyzed in this work, with red, white, and green pseudo-colors. Images were created using three bands extracted from the data cube. Green was created with a spectral window that encloses the [OIII]5007 emission lines to represent the highly ionized gas; Red was created with a spectral window that encloses H$\alpha$ 6563Å, while white represents the stellar continuum, with a spectral window of 7000-9000Å. In all images, North is up and East to the left.
• Figure 2: Flux, velocity, and velocity dispersion maps for the stellar component of SDSS1510+07. The top panel shows the integrated flux, obtained from collapsing the data cube around 8000 Å. Overplotted in white are the isophote profile fits, which show the extension and inclination of a stellar component, consistent with a disk. The velocity map (middle panel) shows that the kinematics of the stellar component are rotation-dominated. The velocity dispersion map (bottom map) shows higher velocities concentrated in the center, as expected from a rotation-dominated disk.
• Figure 3: Flux (top), velocity (middle), and velocity dispersion bottom) maps for the stellar component of UGC 7342. The integrated flux maps show the extension of the stellar component, which is highlighted with the ellipses fitted to the isophotal profiles. Fitted ellipses are shown as white contours in all the maps. Beyond the centrally concentrated high surface brightness stellar structure, tidal arms dominate the distribution of the stellar component. The velocity map in the middle panel shows a gradient that spatially coincides with the centrally concentrated component, indicating the possible presence of rotation. Beyond the centrally concentrated distribution, the kinematics of the stellar component are clearly perturbed due to the streaming motions. The velocity dispersion map shows a higher velocity concentration in the nuclear region, indicating some possible rotational support.
• Figure 4: Flux, velocity, and velocity dispersion maps for the stellar component of SDSS1524+08. Overplotted in gray are the ellipses fitted to the isophotal profiles. This fit indicates a close-to-face-on disk. Outside the main centrally-concentrated stellar distribution, there are clear tidal tails that dominate the stellar distribution. The velocity map (middle panel) indicates little to no sign of rotation, with an apparent small gradient in velocity in the central region. The velocity dispersion map (bottom panel) shows a high velocity dispersion in the redshifted region NE of the nucleus and seemingly along the tidal tail that goes E to N. This indicates very low rotational support in the stellar distribution.
• Figure 5: Normalized flux (left), velocity (middle), and velocity dispersion (right) maps for SDSS1510+07 based on the measurements of the [OIII] emission line. The black contours represent the values (in km/s) of the parameter W80, above the velocity limit used to fit a secondary Gaussian component to the emission line fit.