Are "Changing-Look" Active Galactic Nuclei Special in the Coevolution of Supermassive Black Holes and their Hosts? II. The Case of Changing-Look Narrow-Line Seyfert 1 Galaxies

Abstract

The evolutionary role of the so-called ``changing-look'' (CL) active galactic nucleus (AGN), which is characterized by spectral-type transitions within $\sim10$ yr, has been suggested in the past few years. By focusing on CL-AGNs having spectra similar to those of broad-line Seyfert 1 galaxies, some authors have proposed that CL-AGNs tend to be at a special evolutionary stage associated with intermediate-to-old stellar populations. Here we attempt to verify this evolutionary role by extending the sample to CL narrow-line Seyfert 1 (NLS1) galaxies, which are believed to be ``young'' AGNs with a less massive supermassive black hole and high accretion rate. Combining the recent large NLS1 catalog provided by Paliya et al. (2024) and the SDSS-V DR19 spectral survey returns only three CL-NLS1s out of a parent sample of 884 objects, reinforcing the rarity of CL-NLS1s. Subsequent spectral analysis shows that the evolutionary role mentioned above still holds, although CL-NLS1s tend to occupy the young end of the intermediate-old population. Finally, we propose that off-center SDSS spectra caused by the ``fiber drop'' effect have great potential for determining the properties of the narrow-line region of NLS1s.

Figures (7)

• Figure 1: Identification of the three identified CL-NLS1s by comparing their SDSS DR16 and SDSS DR19 spectra. In each panel, the SDSS DR16, SDSS DR19 (in which the multi-epoch SDSS DR19 spectra are combined) and the corresponding differential spectra are shown from top to bottom by the red, blue and black lines, respectively, The differential spectrum is generated by using the SDSS DR16 spectrum as a reference. The SDSS DR16 and SDSS DR19 are scaled to a common level according to their [O3] $\lambda$5007 fluxes before the subtraction. The spectra are binned by a boxcar of 3 Å and shifted vertically by an arbitrary amount for clarity.
• Figure 2: Spectral decomposition and emission-line profile fitting. Left panels: The continuum modeling process. In each panel, the observed rest-frame spectrum (the heavy black curve) is fitted by a modeled being consist of a starlight component (the cyan curve) and an AGN's powerlaw with a free slope (the magenta curve). The blue curve at the bottom is the resulted emission-line spectrum after a subtraction of the best-fit continuum (the red curve). All spectra are shifted vertically by an arbitrary amount for clarity. Middle panels: Line-profile modeling for the the H$\beta$ region in the emission-line isolated spectra (see the left panels). The modeling is based on a a linear combination of a set of Gaussian functions. The observed and modeled line profiles are shown by black and red solid lines, respectively. Individual Gaussian function resulted from the fitting is shown as blue dashed lines. The residuals between the observed and modeled profiles are presented in the subpanel below. Right panels: Identical in layout to the middle panels, but for the H$\alpha$ region.
• Figure 3: The same as Figure \ref{['fig:modeling_off']}, but for the "turn-on" spectra. As shown in the left panels, in addition to the AGN's continuum and the underlying starlight component, the Balmer continuum and high-order Balmer lines (magenta curve), along with the Fe2 complex (the green curve), are required to model the continuum properly in each spectrum.
• Figure 4: Left panel: $D_{\rm n}(4000)$ plotted as a function of $\mathrm{H}\delta_{\mathrm{A}}$ for local CL-AGNs, including both the three CL-NLS1s (the red solid triangles) reported in this study and the CL-AGNs (the black open triangles) compiled by Wang et al. (2023). For the three CL-NLS1s, the size of the solid triangles depends on the value of $\mathrm{R4570 = Fe~II/H\beta}$ reported by Paliya et al. (2024). As a comparison, the dashed black heavy lines correspond to the density contours of distribution of $\sim 80,000$ Seyfert 2 galaxies provided in the MPA/JHU value-added catalog (e.g., Kauffmann et al. 2003; Heckman & Kauffmann 2006). The stellar population evolution locus are over plotted in the plot by the solid line for the single-stellar-population model with solar metallicity, and by the dot-dashed line for the exponentially decreasing star-formation rate $\psi(t)\propto e^{-(t/\mathrm{4~Gyr})}$. Right panel: Distribution on the $L_{\mathrm{bol}}/L_{\mathrm{Edd}}-D_{\rm n}(4000)$ diagram. The three CL-NLS1s are shown by the magenta circles, and the local CL-AGNs compiled by Wang et al. (2023) by the black open stars. The "turn-off" and "turn-on" states are connected by a vertical line for individual CL-AGNs. The local "partially obscured" AGNs studied in Wang (2015) are shown by the small squares for a comparison (red: Seyfert galaxies, cyan: composite galaxies). The best-fit nonlinear relationship to these "partially obscured" AGNs is presented by the dashed line.
• Figure 5: Brightness variation of the three CL-NLS1s in MIR WISE$w1$ and $w2$ bands (the left panels), and in optical bands (the right panels), after binning the measured magnitudes within a couple of days. In each panel, the vertical lines or region mark the epochs when the optical spectra were obtained.