Probing the Physical Origin of the Balmer Decrement in the Broad-line Region of Nearby Active Galactic Nuclei via Spectral Variability

TL;DR

The paper addresses whether the Balmer decrement in AGN broad-line regions is driven by dust reddening or intrinsic BLR physics. It analyzes 1,116 low-redshift Type 1 AGNs with multi-epoch SDSS spectra, applying careful spectral fitting to measure the broad Balmer decrement and its temporal variability, and contrasts these with continuum luminosity and color. The authors find a near-zero correlation between the mean Balmer decrement and luminosity but a weak anti-correlation with the Eddington ratio; more importantly, the decrement varies in time in anti-correlation with continuum brightness, and extinction-based explanations cannot reproduce the observed color-decrement relation, pointing to optical-depth and collisional excitation in the BLR as the primary drivers. The results inform the interpretation of changing-look AGNs and high-redshift red AGN populations, suggesting that steep Balmer decrements in the early Universe may reflect BLR radiative-transfer effects rather than enhanced dust extinction.

Abstract

To investigate the physical origin of the Balmer decrement in the broad-line region of active galactic nuclei (AGNs), we measure the temporal variability of the fluxes of the broad H$β$ and H$α$ emission lines using multi-epoch spectroscopic data of low-redshift AGNs from the Sloan Digital Sky Survey. The analysis of the mean spectra reveals that the Balmer decrement shows no correlation with AGN luminosity, while it is inversely correlated with the Eddington ratio. However, the temporal variation of the Balmer decrement in individual objects exhibits an even stronger anti-correlation with AGN luminosity, suggesting that the change in AGN luminosity plays a dominant role in determining the Balmer decrement. By comparing the temporal evolution of the Balmer decrement with the continuum color, we find that reddening due to the AGN itself may not be the primary factor. Instead, radiative transfer effects and excitation mechanisms, which deviate from the Case B recombination, appear to be critical for the variation of the Balmer decrement. These results provide useful insights into the underlying physics of changing-look AGNs and high-$z$ AGNs, such as the ``little red dots'', which exhibit extreme values of the Balmer decrement that can be misinterpreted as evidence for dust.

Figures (6)

• Figure 1: Example of the spectral fitting. (a) The black solid line represents the original data in the rest-frame, corrected for the Milky Way reddening. The cyan line denotes the continuum from the accretion disk, modeled with a power law plus polynomial terms, along with Fe2 multiplets, while the magenta line shows the host galaxy component. The red line represents the best-fit model. Spectral decomposition of the (b) H$\beta$ and [O3] and (c) H$\alpha$ and [N2] region. The black histogram shows the residual spectrum after subtracting the continuum fit. The blue lines represent the broad components of H$\beta$ and H$\alpha$, while the green lines represent the narrow components of H$\beta$ and H$\alpha$, with the [O3] doublet and [N2], respectively. The red line indicates the total best-fit model.
• Figure 2: Balmer decrement versus (a) continuum luminosity at 5100 Å and (b) Eddington ratio. The dashed line shows the best-fit linear regression. The density contour is shown on a logarithmic scale.
• Figure 3: Comparison of temporal variation of the Balmer decrement with AGN brightness indicated by the continuum flux at 5100 Å. The density contour is shown on a logarithmic scale.
• Figure 4: Comparison of the temporal variation between the Balmer decrement and the flux ratio of the continua at 3800 and 5100 Å. The various lines represent the temporal variation predicted solely by the intrinsic extinction of AGNs, under the assumption that the accretion disk and BLR share the same extinction. The blue dotted, solid, and dashed lines correspond to the predictions based on the extinction curve of fitzpatrick_1999 with $R_V = 1.5,3.1,$ and $5.5$, respectively. The red and green lines are derived from the extinction curves of star-forming galaxies (calzetti_2000) and AGNs (gaskell_2004), respectively. The black dashed line denotes the orthogonal distance regression of the observed data.
• Figure 5: Comparison of temporal variation between the Balmer decrement and AGN brightness indicated by the flux of broad H$\alpha$ emission. The dashed line shows the best-fit linear regression. The density contour is shown on a logarithmic scale.