Table of Contents
Fetching ...

The Narrow Emission Lines of Seyfert 1 Galaxies: Comparisons with a Large SDSS Sample

Matthew Malkan, Lisbeth Jensen, Lei Hao

TL;DR

This study uses a large SDSS-based sample to dissect the narrow-line emission in Seyfert 1 galaxies by separating broad and narrow Balmer components and mixing AGN NLR emission with host-galaxy HII-region and LINER contributions. A two-component toy model (AGN NLR + HII regions) quantitatively reproduces the observed shifts in BPT and Ke06 diagrams as the broad-line strength decreases, revealing substantial star-formation contamination in weak Seyfert 1s. The authors show that strong Sy1s resemble pure AGN Narrow Line Regions, while weaker Sy1s exhibit higher reddening, denser NL gas, and stellar populations indicative of recent star formation, challenging the simplest Seyfert unification picture. The results imply that unification models must account for SF contamination, reddening geometry, and LINER contributions to accurately characterize NL emission across Seyfert sub-classes, with implications for AGN selection and interpretation of NL diagnostics.

Abstract

We analyzed a large sample of SDSS spectra of Seyfert galaxies, subdividing Seyfert 1s based on their narrow-to-broad Halpha components. Comparing their narrow-lines (NL) to Seyfert2s in line-ratio diagrams, most of the NL of strong Sy 1.0 and Sy 1.2s (with dominant broad lines) are the same as those of pure Sy2s. In contrast, only 25-30 percent of the Sy1.8 and Sy1.9 nuclei (with weak broad lines) are located in the pure Sy2 region, with the rest falling in the composite-star-forming region. We explain these Seyfert-plus-star-formation spectra with a simple model. It shows that 85 percent of NL in Sy1.9 are from HII-regions, while 88 percent of the NL in Sy 1.0 arise from the same NLR as in pure Sy2. About 6 percent of the strong and weak Sy1's have NL dominated by LINER emission, while 15 percent of intermediate Seyferts (Sy 1.5 and Sy 1.6) do. To confirm this Seyfert 1 AGN plus star formation combination, we used stellar absorption-lines to compare their stellar populations. Their Hdelta strengths show that LINERs, pure Sy2s, and also the broad-line dominated Sy1s have old stellar populations. The weak Sy 1s show stronger Hdelta absorption, indicating larger proportions of young stars. About one third of the u band light in Sy1.0 and 1.2 is blended Balmer lines and continuum from the BLR. The NL gas reddening increases as the BLR strength decreases, from Sy1.0 (0.13 mag), to Sy1.9 (0.40 mag), to Sy2s and LINERs both with 0.50 mag. Our data do not support the simplest version of Seyfert 1 and 2 unification, where both AGN classes have identical NL.

The Narrow Emission Lines of Seyfert 1 Galaxies: Comparisons with a Large SDSS Sample

TL;DR

This study uses a large SDSS-based sample to dissect the narrow-line emission in Seyfert 1 galaxies by separating broad and narrow Balmer components and mixing AGN NLR emission with host-galaxy HII-region and LINER contributions. A two-component toy model (AGN NLR + HII regions) quantitatively reproduces the observed shifts in BPT and Ke06 diagrams as the broad-line strength decreases, revealing substantial star-formation contamination in weak Seyfert 1s. The authors show that strong Sy1s resemble pure AGN Narrow Line Regions, while weaker Sy1s exhibit higher reddening, denser NL gas, and stellar populations indicative of recent star formation, challenging the simplest Seyfert unification picture. The results imply that unification models must account for SF contamination, reddening geometry, and LINER contributions to accurately characterize NL emission across Seyfert sub-classes, with implications for AGN selection and interpretation of NL diagnostics.

Abstract

We analyzed a large sample of SDSS spectra of Seyfert galaxies, subdividing Seyfert 1s based on their narrow-to-broad Halpha components. Comparing their narrow-lines (NL) to Seyfert2s in line-ratio diagrams, most of the NL of strong Sy 1.0 and Sy 1.2s (with dominant broad lines) are the same as those of pure Sy2s. In contrast, only 25-30 percent of the Sy1.8 and Sy1.9 nuclei (with weak broad lines) are located in the pure Sy2 region, with the rest falling in the composite-star-forming region. We explain these Seyfert-plus-star-formation spectra with a simple model. It shows that 85 percent of NL in Sy1.9 are from HII-regions, while 88 percent of the NL in Sy 1.0 arise from the same NLR as in pure Sy2. About 6 percent of the strong and weak Sy1's have NL dominated by LINER emission, while 15 percent of intermediate Seyferts (Sy 1.5 and Sy 1.6) do. To confirm this Seyfert 1 AGN plus star formation combination, we used stellar absorption-lines to compare their stellar populations. Their Hdelta strengths show that LINERs, pure Sy2s, and also the broad-line dominated Sy1s have old stellar populations. The weak Sy 1s show stronger Hdelta absorption, indicating larger proportions of young stars. About one third of the u band light in Sy1.0 and 1.2 is blended Balmer lines and continuum from the BLR. The NL gas reddening increases as the BLR strength decreases, from Sy1.0 (0.13 mag), to Sy1.9 (0.40 mag), to Sy2s and LINERs both with 0.50 mag. Our data do not support the simplest version of Seyfert 1 and 2 unification, where both AGN classes have identical NL.

Paper Structure

This paper contains 28 sections, 11 equations, 12 figures.

Figures (12)

  • Figure 1: Top Two Rows show three examples of Seyfert 2 galaxies with the full spectra in blue and the fitted continuum in red. In the second row of spectra zoomed in on the H$\alpha$/[NII] region, show that the reduction in $\chi^2$ when adding a fourth Gaussian to the emission line fit is not statistically significant. The bottom two rows show three examples of Seyfert 1 galaxies, with the 3-Gaussian fit shown in red and the (preferred) 4-Gaussian fit in blue. The reduction in $\chi^2$ is large enough to require the inclusion of broad components in their H$\alpha$ lines. Based on our quantitative comparison of narrow-to-broad-line H$\alpha$ components, these are classified as Seyfert 1.0, 1.2, and 1.6
  • Figure 2: BPT Diagrams with log([NII]/H$\alpha$) on the x-axis. In all the line ratio classification diagrams, the light Grey dots show the full MPA-JHU SDSS DR7 data. (a) Shaded blue data points are from the Hao AGN sample. We separate the Star-Forming region into three sections: SF-Lower bottom region--light gray, SF-Middle medium--gray, and the Black dots we call SF-Upper. These groups correspond to high, medium and low-metallicity galaxies, respectively. (b) $Red$ and $Orange$ points are Sy 1.0 and Sy 1.2 galaxies. More than 85% of the Sy 1.0 and 77% of the Sy 1.2 are above the Ke01 line, which we consider to be dominated by the AGN NLR. (c) $Purple$ and $Pink$ points are Sy 1.5 and Sy 1.6 galaxies. 65% are found in the "Seyfert" region above the Ke01 boundary line (d) $Light\;Green$ points are the Sy 1.8 NLR, which have similar distribution as in plot (b) but the majority of the Sy 1.9 NLR ($Dark \;Green$ points) lie in the composite region and down towards SF Lower. Only 50% of the Sy 1.8 and 30% of the Sy 1.9 are in the Seyfert-dominated region
  • Figure 3: BPT Diagrams with log([SII]/H$\alpha$ on the x-axis. The color scheme for data points is the same as in Figure \ref{['fig:fig2']}. Galaxies below the short-dash line and to the right of the Ke01 line are LINER-dominated. (a) As before, the Star-Forming galaxy spectra are separated into three sections based on their metallicity. (b) More than 85% of the Sy 1.0 and 77% of the Sy 1.2 are in the Sy 2 "pure AGN" region. We call these the "strong Seyfert 1's". (c) The Sy 1.5 and Sy 1.6 galaxies. 30% are in the SF region. (d) "Weak Seyfert 1's": 40% of the Sy 1.8 and 60% of the Sy 1.9 are in the SF region.
  • Figure 4: BPT Diagram with log([OI]/H$\alpha$ on the x-axis. The color scheme as in Figure \ref{['fig:fig2']}. (a) As before, the Star-Forming galaxy spectra are separated into three (somewhat overlapping) sections. (b) 95% of Sy 1.0 and 88% of Sy 1.2 are in the AGN region. (c) 77% of the Sy 1.5 and Sy 1.6 are in the AGN region. (d) 65% of the Sy 1.8 NLR, and 45% of the Sy 1.9 are in the AGN region.
  • Figure 5: The Ke06 diagram, of log([OIII]$\lambda$5007/[OII]$\lambda\lambda$3726+3729) $versus$ log([OI]/H$\alpha$), adopted from Kewley et al. (2006) to separate AGN, Liners and Star-forming galaxies. The solid black line (AGN/SF) separates Star-forming galaxies and AGNs. The dashed line (Sy/LINER) indicates the demarcation between Sy 2 Galaxies and LINERs. The color scheme is as in Figure \ref{['fig:fig2']}. (a) LINERs, Sy 2-Upper and Lower, and star-forming galaxies are plotted with different colors. $\sim7$% of the Sy 2-Lower are found in the LINER region, while $\sim21$% of the LINERs are above the Sy/LINER boundary. (b) Sy 1.0 and Sy 1.2 NLR. We see the same trend as in the traditional BPT diagrams, that they occupy the same region as the Sy 2 NLR galaxies. (c) Sy 1.5 and Sy 1.6 NLR. There is a small downward and rightward shift as in the other BPT diagrams. There is also a significant rightward shift into the LINER-dominated region. (d) Sy 1.8 and Sy 1.9 NLR. As with the BPT diagrams having log([OIII]/H$\beta$) on the y-axis, the Sy 1.9 are shifted substantially down towards the star-forming region.
  • ...and 7 more figures