Discovery of a Luminosity-dependent Continuum Lag in NGC 4151 from Photometric and Spectroscopic Continuum Reverberation Mapping

TL;DR

Continuum reverberation mapping of NGC 4151 across three campaigns reveals a $\tau \propto \lambda^{4/3}$ lag consistent with a thin accretion disk, but with lags larger by $\sim$6.6× and significant diffuse continuum (DC) contamination from the BLR indicated by Balmer/Paschen jumps. The data show negligible X-ray–UV/optical correlation, suggesting X-ray reprocessing is not the dominant driver of variability. A non-monotonic lag–luminosity trend across campaigns implies a luminosity-weighted mix of disk and DC contributions, with the DC component potentially following an intrinsic Baldwin effect. These findings highlight the need for multi-season, broad-wavelength RM to accurately interpret disk structure and SMBH mass measurements in AGNs.

Abstract

Accretion onto supermassive black holes (SMBHs) powers active galactic nuclei (AGNs) and drives feedback that shapes galaxy evolution. Constraining AGN accretion disk structure is therefore essential for understanding black hole growth and feedback processes. However, direct constraints on disk size remain rare -- particularly from long-term, multi-season spectroscopic reverberation mapping (RM), which is critical for isolating the intrinsic disk response from the broad-line region (BLR). We present results from an intensive multi-wavelength RM campaign of NGC 4151 during its brightest state in nearly two decades. This represents the third high-cadence monitoring over the past decade, capturing accretion states spanning the transitional regime between thin and thick disks, making NGC 4151 the only AGN with continuum RM observations across such a wide range in accretion states. Combining spectroscopy from the Lijiang 2.4 m telescope with coordinated Swift UV/X-ray monitoring, we measure inter-band continuum lags from UV to optical. The wavelength-dependent lags follow a tight $τ\propto λ^{4/3}$ relation, consistent with reprocessing in a thin disk, but exceed theoretical predictions by a factor of 6.6. Our lag spectrum reveals clear excesses near the Balmer and possibly Paschen jumps, confirming diffuse continuum (DC) contamination from the BLR. By comparing the three campaigns, we discover a non-monotonic lag-luminosity trend ($>3σ$), which cannot be explained by DC emission alone. We propose the lags reflect combined disk and BLR contributions, and present the first evidence that the DC component follows an intrinsic Baldwin effect. These results offer new insights into SMBH mass measurements and theoretical models of AGN inner structure.

Figures (6)

• Figure 1: Light curves and CCF analyses of NGC 4151. The left panel shows the photometric light curves from Swift and Lijiang observations, where the subscript S represents Swift data, LP represents Lijiang photometric data, and LS represents synthetic photometric data derived from Lijiang spectra. The right panel displays the light curves at different wavelengths measured from the line-free regions of the spectra. On the right side of each plot are the corresponding CCF and CCCD results.
• Figure 2: Wavelength-dependent lags and test of the line-free region. Left: The top panel shows lags for NGC 4151 derived from Swift photometry (blue dots), Lijiang photometry (green dots), synthetic photometry (cyan dots), and spectroscopy (red squares). The black solid line and red dashed line represent fits to $\tau = \tau_{0}[(\lambda / 1922 {\rm \AA})^{4/3} - 1]$, using all data points and spectroscopy-only data, respectively. Hollow points were excluded from the fits. The bottom panel shows lags for NGC 4593 from Cackett2018. Right: The mean spectrum with Lijiang filter transmission curves (top) and the lags measured at different wavelengths (bottom). The spectrum and transmission have been corrected to the rest frame. Gray dots indicate lags measured from the spectrum at every 30 Å, while the red squares and red dashed line are the same as those in the left panel. The green shaded regions mark the line-free wavelength ranges chosen for the analysis.
• Figure 3: Long-term Swift/UVOT light curve of NGC 4151. The black points represent observations with a frametime of 3.6 ms, while gray points correspond to observations with an 11 ms frametime. The blue points highlight the observation period of our monitoring program. The purple, green, and red shaded regions indicate the durations of three RM campaigns conducted by Edelson2017, Zhou2025, and this work, respectively.
• Figure 4: Lag vs. luminosity for the three campaigns. Black and red represent the measured time delays and the predictions from X-ray reprocessing in a standard thin accretion disk model, respectively.
• Figure A1: Same as Figure \ref{['fig2']}, but with the CCFs calculated using the 5100 Å light curve as the reference, and all other light curves interpolated to match the cadence of the 5100 Å data.