Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

First reverberation mapping of a Bowen fluorescence line

Martin W. Ochmann, Edward M. Cackett, Lukas Diehl, Keith Horne, Malte A. Probst, Wolfram Kollatschny

TL;DR

This study reports the first reverberation mapping of the Bowen fluorescence line OI λ8446 in NGC 4593 using a dense HST/STIS monitoring campaign. OI λ8446 exhibits day-scale variability with a fractional rms of about 4% and shows a clear lag of roughly 2.5 days behind Lyα (a Lyβ proxy), while closely following Hα with ~0.3 days of additional delay; the Lyβ pumping mechanism is identified as dominant. The OI response indicates the emitting gas resides at a BLR distance comparable to the Balmer-emitting region, reinforcing a vertically stratified disk-like BLR geometry. The findings establish Bowen fluorescence RM as a viable tool and point toward dual-driver RM, using both the continuum and the pumping line, for future studies of AGN BLR structure, especially in systems with larger BLRs.

Abstract

Reverberation mapping (RM) is a powerful tool to determine the extent, structure, and kinematics of the broad-line region (BLR) of active galactic nuclei (AGN). So far, RM of the BLR has only been performed for recombination lines responding to the varying ionizing continuum. We tested whether OI 8446, attributed to Bowen fluorescence driven by Ly$β$ pumping, varied on short (day- to week-long) timescales during a 2016 HST/STIS campaign of NGC 4593, and examined how it relates to other emission lines and the ionizing UV continuum. We quantified the variability of OI 8446 by calculating its rms amplitude. We then extracted integrated light curves of OI 8446 and other UV and optical emission lines, and compared them with each other and with the UV continuum light curve using correlation analyses. In addition, we used archival near-infrared spectra to assess the dominant excitation mechanism of OI 8446. We detect, for the first time, variability in OI 8446 on day timescales. The fractional rms amplitude is $\sim 4$% over the 4-week campaign. The OI 8446 light curve reverberates with a delay of $\sim 2.5$ days relative to Ly$α$, used as a proxy for Ly$β$, detected at a false-alarm probability of 0.6% (significance of $\sim 2.8σ$) under our adopted null hypothesis. It closely tracks H$α$ with only a minor additional delay of $\sim0.3$ days, placing its emission region at essentially the same distance as the Balmer-line weighted BLR. Line ratios indicate that Ly$β$ pumping is the dominant excitation mechanism for O I 8446. Our results establish OI 8446 as the first Bowen-fluorescence line reverberation-mapped, responding directly to variations in Ly$β$ flux. We propose that in future campaigns targeting AGN with larger BLRs, O I could enable dual-driver RM using both the continuum and the pumping line as drivers.

First reverberation mapping of a Bowen fluorescence line

TL;DR

This study reports the first reverberation mapping of the Bowen fluorescence line OI λ8446 in NGC 4593 using a dense HST/STIS monitoring campaign. OI λ8446 exhibits day-scale variability with a fractional rms of about 4% and shows a clear lag of roughly 2.5 days behind Lyα (a Lyβ proxy), while closely following Hα with ~0.3 days of additional delay; the Lyβ pumping mechanism is identified as dominant. The OI response indicates the emitting gas resides at a BLR distance comparable to the Balmer-emitting region, reinforcing a vertically stratified disk-like BLR geometry. The findings establish Bowen fluorescence RM as a viable tool and point toward dual-driver RM, using both the continuum and the pumping line, for future studies of AGN BLR structure, especially in systems with larger BLRs.

Abstract

Reverberation mapping (RM) is a powerful tool to determine the extent, structure, and kinematics of the broad-line region (BLR) of active galactic nuclei (AGN). So far, RM of the BLR has only been performed for recombination lines responding to the varying ionizing continuum. We tested whether OI 8446, attributed to Bowen fluorescence driven by Ly pumping, varied on short (day- to week-long) timescales during a 2016 HST/STIS campaign of NGC 4593, and examined how it relates to other emission lines and the ionizing UV continuum. We quantified the variability of OI 8446 by calculating its rms amplitude. We then extracted integrated light curves of OI 8446 and other UV and optical emission lines, and compared them with each other and with the UV continuum light curve using correlation analyses. In addition, we used archival near-infrared spectra to assess the dominant excitation mechanism of OI 8446. We detect, for the first time, variability in OI 8446 on day timescales. The fractional rms amplitude is % over the 4-week campaign. The OI 8446 light curve reverberates with a delay of days relative to Ly, used as a proxy for Ly, detected at a false-alarm probability of 0.6% (significance of ) under our adopted null hypothesis. It closely tracks H with only a minor additional delay of days, placing its emission region at essentially the same distance as the Balmer-line weighted BLR. Line ratios indicate that Ly pumping is the dominant excitation mechanism for O I 8446. Our results establish OI 8446 as the first Bowen-fluorescence line reverberation-mapped, responding directly to variations in Ly flux. We propose that in future campaigns targeting AGN with larger BLRs, O I could enable dual-driver RM using both the continuum and the pumping line as drivers.

Paper Structure

This paper contains 10 sections, 3 figures.

Table of Contents

  1. Introduction
  2. Observations
  3. Results
  4. The variability of Oi$\,\lambda8446$
  5. The response of Oi$\,\lambda8446$
  6. Discussion
  7. The excitation mechanism and origin of variability
  8. The emission region
  9. Conclusions
  10. Significance of the Oi lag detection

Figures (3)

  • Figure 1: Optical mean (blue) and rms (orange) spectra from the HST/STIS campaign. The rms spectrum is scaled ($\times$5) and shifted in flux ($- 0.25 \times 10^{-15}$ erg s$^{-1}$ cm$^{-2}$ Å$^{-1}$) to highlight weak line features. The linear pseudo-continuum beneath the Oi $\lambda8446$ and the Caii triplet complex is shown for orientation. The inset shows this spectral region in greater detail, with the rms spectrum scaled ($\times$15) and shifted ($- 1.5 \times 10^{-15}$ erg s$^{-1}$ cm$^{-2}$ Å$^{-1}$). The Oi$\,\lambda8446$ integration area is shaded in blue (see Sect. \ref{['sec:res_response_OI8446']}).
  • Figure 2: Comparison of normalized light curves (left panels) of the emission lines Oi $\lambda8446$, Ly$\alpha$, and H$\alpha$, together with the UVW2 continuum, with CCFs (right panels) of each pair, showing correlated variability and time delays between the individual emitting components. The CCFs are plotted between 0 and 1. The dashed line indicates $\tau_{\rm cent}$.
  • Figure 3: Distribution of $\tilde{r}_{\rm max}$ for 10 000 realizations without physical correlation between the line light curve and the supposed driving light curve. The maximum correlation coefficient $r_{\rm max}$ obtained from the ICCF analysis (see Sect. \ref{['sec:res_response_OI8446']}) is indicated as a dashed black line.