Accretion-Regulated Type Transitions in Changing-Look AGNs: Evidence from Two-Epoch Spectral Analysis

TL;DR

This study analyzes 203 low-redshift changing-look AGNs ($z<0.35$) using two-epoch SDSS DR16 and DESI DR1 spectra to track optical continuum and broad-line variability across baselines of $\sim$1,000–8,000 days. Spectral fitting decomposes each epoch into AGN continuum, Fe II, host-galaxy light, and emission lines, enabling robust typing (Type 1.0–1.8/2.0) and identification of transitions; the sample splits into Dataset A (minor changes) and Dataset B (significant turn-on/off transitions). The authors find both blueing and reddening trends with brightness and reveal strong correlations between log($\mathrm{H}\beta/\mathrm{[O III]}$) and log($L_{\mathrm{H\alpha}}$) as well as log($L_{\mathrm{bol}}/L_{\mathrm{Edd}}$), particularly in Dataset B, implying that accretion-rate variations drive BLR ionization changes that trigger type transitions. These results support accretion-regulated unification of AGN types and highlight the role of SMBH accretion in shaping observed CL-AGN diversity, while acknowledging substantial intrinsic scatter and pointing to future multi-wavelength, time-resolved studies to map detailed SED evolution.

Abstract

The changing-look active galactic nucleus (CL-AGN), an extraordinary subpopulation of supermassive black holes, has attracted growing attention for understanding its nature. We present an analysis of the spectral properties of 203 low-redshift CL-AGNs ($z<0.35$) using two-epoch spectra from SDSS DR16 and DESI DR1 with time baseline ranging from $\sim$1000 to 8000 days, based on spectral fitting and decomposition. The sample consists of 11.3\% Type 1.0, 26.6\% Type 1.2, 43.1\% Type 1.5, and 19\% Type 1.8/2.0 AGNs. The total sample is divided into two datasets: Dataset A (110 objects) with minor spectral type variations, likely general AGN variability, and Dataset B (93 objects) showing significant type transitions and characteristic turn-on or turn-off behavior. Our results reveal clear optical continuum and emission-line variability, showing both bluer-when-brighter and redder-when-brighter trends. A strong correlation between the broad H$β$/[O~{\sc iii}] ratio and broad H$α$ luminosity ($L_{\rm Hα}$), ${\rm log(Hβ/[O~III])}=(0.63\pm 0.07){\rm log}(L_{\rm Hα})-(26.49\pm2.96)\pm0.48$ for Dataset B, as well as the correlation between H$β$/[O~{\sc iii}] and Eddington ratio ($L_{\rm bol}/L_{\rm Edd}$), ${\rm log(Hβ/[O~III])}=(0.59\pm 0.08){\rm log}(L_{\rm bol}/L_{\rm Edd})+(1.02\pm0.15)\pm0.53$ for Dataset B, suggests that accretion rate variations drive changes in ionizing flux within the broad-line region, thereby triggering AGN type transitions. These findings underscore the critical role of supermassive black hole accretion processes in refining the AGN unification model. Future work should investigate potential connections between stellar evolution in outer accretion disk and the observed scatter in these correlations.

Figures (7)

• Figure 1: An example of spectral fitting and decomposition of the SDSS (left) and DESI (right) spectra for the changing-look AGN J154529.63+251127.86 ($z$=0.117). The host-galaxy starlight of each epoch spectrum was fitted by three host-galaxy templates, they with stellar population age of 11 Gyr and metallicities of 0.008, 0.02, and 0.05 (more considerations refer to Section \ref{['specfit']}), are marked in each panel, respectively. Each rest-frame spectrum (black) was fitted over the rest-frame wavelength range 4500–6900 Å, the total model is shown in red. Fitting components include the AGN continuum (blue), iron multiplets (gray), host galaxy (green), broad Balmer lines (magenta), narrow Balmer lines (brown), broad helium lines (cyan), and other narrow lines including helium lines, [O iii], [N ii], and [S ii] (orange).
• Figure 2: The variation analysis in the spectral index ($\alpha$), optical continuum flux at 5100 Å ($f_{\rm 5100}$), and broad H$\alpha$ flux for the Total sample, Dataset A, and Dataset B. The standard deviations of these variations, including spectral index ($\Delta \alpha$), optical continuum flux (log($f_{\rm 5100,DESI}/f_{\rm 5100,SDSS}$), in magnitudes), and broad H$\alpha$ flux (log($f_{\rm H\alpha,DESI}/f_{\rm H\alpha,SDSS}$)), are labeled in the titles of the two panels for the Total sample. By examining the variation relations between log($f_{\rm 5100,DESI}/f_{\rm 5100,SDSS}$) and $\Delta \alpha$, and between log($f_{\rm H\alpha,DESI}/f_{\rm H\alpha,SDSS}$) and $\Delta \alpha$, both “bluer-when-brighter (BWB)” and “redder-when-brighter (RWB)” trends are found and marked in the figure.
• Figure 3: Relation between variations in [O iii] $\lambda$5007 and broad H$\beta$ fluxes for the Total sample, Dataset A, and Dataset B. For Dataset A and the Total sample, [O iii] $\lambda$5007 emission increases with increasing broad-line emission ($r=0.44$, $p<0.0001$ for Dataset A; $r=0.20$, $p<0.05$ for Total sample), while this trend is not significant in Dataset B ($r=0.14$, $p=0.17>0.05$). Linear regression models illustrate these potential correlations. The scatter in [O iii] $\lambda$5007 fluxes and the correlation coefficients (with $p$-values) are labeled in the figure.
• Figure 4: The relationship between the broad H$\beta$/[O iii] $\lambda$5007 ratio (a proxy for AGN Types) and broad H$\alpha$ luminosity ($L_{\rm H\alpha}$) in logarithmic space for the Total sample (left panel) and Datasets A and B (right panel). Circles indicate measurements from SDSS and DESI epoch spectra. In the left panel, red circles show high-luminosity states, and blue circles show low-luminosity states. The black solid line, with its uncertainties, shows the linear regression result. In the right panel, gray and purple circles represent Datasets A and B, respectively. Using the broad H$\beta$ to [O iii] $\lambda$5007 flux ratio and standard classification criteria (Osterbrock1977Winkler1992), we label AGN Types 1.0, 1.2, 1.5, and 1.8/2.0 in regions separated by horizontal dashed lines. Regression results, including models and uncertainties, are shown for Dataset A (gray) and Dataset B (purple). The Pearson correlation coefficient $r$ and $p$-value are labeled.
• Figure 5: Relation between $\Delta$log(H$\beta$/[O iii]$~\lambda$5007) and $\Delta$log($L_{\rm H\alpha}$). Dataset A is shown in gray, Dataset B is presented in purple.