Optical Variability and Evidence for a Changing-Look Event in the Galaxy Mrk 6 (IC 450)

Abstract

In this work, the light curve of the Seyfert galaxy Mrk 6 constructed from photometric observations in the B, V, and $R_{\mathrm{c}}$ filters over the period from 5 April 2016 to 1 February 2026 is presented and analyzed. Over the entire monitoring interval (2016–2026), the variability amplitude of the light curve reaches $\mathrm{\Delta }B=1.9$ mag, $\mathrm{\Delta }V=1.5$ mag, and $\mathrm{\Delta }{R}_{\mathrm{c}}=1.4$ mag. During 2024–2026, the galaxy exhibits synchronous photometric variability in the B, V, and $R_{\mathrm{c}}$ filters with an amplitude of ∼0.3 mag. The study also uses spectroscopic observations obtained on 15 and 22 November 2025 and 16 February 2026 at the Shamakhy Astrophysical Observatory (Azerbaijan), as well as on 9 January 2026 at the Fesenkov Astrophysical Institute (Kazakhstan). The fluxes in the H$\beta$ emission line were calibrated using the [O III] $\lambda 5007$ Å line, ensuring consistent relative calibration of the spectral data. A comparison of the optical spectra reveals a pronounced transformation of the H$\beta$ line profile between November 2025 and January 2026. The broad component, clearly present in November 2025, becomes strongly suppressed and nearly disappears in January 2026, while the narrow emission lines remain stable. This behavior is consistent with a changing-look transition, indicating a temporary weakening of the broad-line region emission. The radius of the broad-line region $R_{\mathrm{BLR}}$ was taken to be equal to the average time delays (lags), amounting to ≈20 light days for the H$\beta$ emission and ≈28 light days for the H$\alpha$.

1. Introduction

Markarian 6 (IC 450) is an intermediate-type active galactic nucleus (Sy 1.0–1.5) that exhibits pronounced variability in the optical, ultraviolet, and X-ray bands. Sergeev and Doroshenko (2003) [1,2] analyzed the optical spectral variability of the Mrk 6 nucleus during 1970–1991 and showed that the H$\beta$ emission lags behind the continuum by an average of $13\pm 6$ days. The long-term UBV photometric variability of Mrk 6 was investigated by Doroshenko (2003) [2] using combined data from 1970 to 2002, where the variability amplitudes were found to be $U\approx {1.6}^{\mathrm{m}}$, $B\approx {1.1}^{\mathrm{m}}$, and $V\approx {0.8}^{\mathrm{m}}$. In a subsequent large-scale monitoring campaign (1992–2008) based on reverberation mapping, lags of ≈$21\pm 1.9$ days for H$\beta$ and ≈27 days for H$\alpha$ were obtained [3], and the mass of the central black hole was estimated as ∼(1.8–2.0) $\times {10}^8 M_{\odot }$ [3]. Analysis of XMM–Newton data showed that the X-ray absorption has a multi-component and variable nature [4]. Schurch & Griffiths reported significant changes in the absorption between observations and proposed a clumpy or partially ionized structure of the absorbing material [5]. More recent studies have confirmed the presence of ionized X-ray winds (UFO/WA) in Mrk 6 [6], while long-term X-ray monitoring has revealed the multi-timescale temporal and spectral variability of the nucleus [7]. Spectropolarimetric data indicate a short lag between the polarized and total continuum, suggesting a compact scattering region and a complex internal structure of the BLR [8]. VLBA observations show that the galactic nuclear jet has a curved, multi-component morphology, and that the energy of the large-scale radio cavities is driven by AGN jet activity [9]. IFU spatially resolved spectroscopic studies have revealed extended structures of ionized gas and their connection with AGN emission [10], while spectral variations on hourly timescales confirm strong turbulence and inhomogeneous dynamics in the BLR [11,12]. Thus, Mrk 6 is a suitable target for multiwavelength comparative analysis of the BLR structure, the multilayered nature of the absorbing medium, and the optical–X-ray–radio connections. Building on the results of previous long-term monitoring [13,14], this paper presents a comprehensive variability analysis of Mrk 6 using photometric observations starting from 5 April 2016 and supplemented with new data taken in 2026.

2. Materials and Methods

Photometric observations of the Seyfert galaxy Mrk 6 were carried out at the Tien Shan Astronomical Observatory (TShAO) of the Fesenkov Astrophysical Institute using the Zeiss-1000 (Carl Zeiss Jena, GDR) “East” telescope ($F=6600$ mm, $D=1$ m). The angular field of view of a single CCD frame is ${19}^{\prime }\times {19}^{\prime }$.

Since 2016, observations and image acquisition have been performed using the Alta F16M CCD camera (4096 × 4096 pixels, 9 μm pixel size), which was operated until November 2021. From mid-2022 onward, observations have been conducted with the U9000D9 CCD camera in combination with Astrodon $BV{R}_{\mathrm{c}}$ filters. The angular scale of the CCD frames was 0.56″/pixel after 2016 and 0.752″/pixel after 2021.

The spectra of Mrk 6 were obtained with the 1.5 m AZT-20 telescope at the Assy-Turgen Observatory of the Fesenkov Astrophysical Institute. Its spectrograph is equipped with volume phase holographic gratings and an EMCCD camera. Additional spectral observations of the galaxy were carried out with the 2 m telescope at the Shamakhi Astrophysical Observatory. Wavelength calibration was performed using comparison lamp spectra of He, Ne, and Ar. Absolute flux calibration relied on spectra of standard stars with known energy distributions. A spectral data reduction was performed using the IRAF (Image Reduction and Analysis Facility) software package [15] developed at the National Optical Astronomy Observatory (NOAO).

Photometric observations of Mrk 6 were obtained during 2016–2026. The galaxy brightness was measured using differential aperture photometry with nearby comparison stars. The photometric reduction was performed using the MaximDL Pro 6 [16] software package with an aperture radius of 6″. Instrumental magnitudes were transformed to the standard B, V, and $R_{\mathrm{c}}$ photometric system using the standardization equations described in Shomshekova et al. (2017) [17].

For each observation night, 5–7 CCD frames were obtained in each filter, and the instrumental magnitudes and their uncertainties were calculated as mean values over the corresponding series of frames. The resulting photometric errors are therefore small and are listed in

Table 1. New photometric observations of Mrk 6 obtained in 2024–2026.

Date	JD–2,400,000	$\mathit{B}\pm {\mathit{\sigma }}_{\mathit{B}}$	$\mathit{V}\pm {\mathit{\sigma }}_{\mathit{V}}$	${\mathit{R}}_{\mathbf{c}}\pm {\mathit{\sigma }}_{{\mathit{R}}_{\mathbf{c}}}$
14 November 2024	60,629	15.613 ± 0.016	14.667 ± 0.006	14.238 ± 0.004
2 December 2024	60,647	15.639 ± 0.023	14.659 ± 0.005	14.226 ± 0.004
24 December 2024	60,669	15.618 ± 0.011	14.650 ± 0.004	14.217 ± 0.004
10 January 2025	60,686	15.602 ± 0.013	14.644 ± 0.005	14.220 ± 0.003
4 March 2025	60,739	15.605 ± 0.014	14.660 ± 0.006	14.231 ± 0.004
6 December 2025	61,016	15.949 ± 0.027	14.923 ± 0.010	14.495 ± 0.006
22 December 2025	61,032	15.876 ± 0.022	14.823 ± 0.007	14.390 ± 0.004
5 January 2026	61,046	15.816 ± 0.020	14.776 ± 0.005	14.349 ± 0.003
9 January 2026	61,050	15.843 ± 0.011	14.796 ± 0.004	14.370 ± 0.003
20 January 2026	61,061	15.867 ± 0.021	14.836 ± 0.007	14.405 ± 0.004
28 January 2026	61,069	15.864 ± 0.015	14.812 ± 0.005	14.386 ± 0.004
1 February 2026	61,073	15.867 ± 0.023	14.796 ± 0.008	14.373 ± 0.004

.

The root–mean–square (RMS) error was calculated for each image and is also presented in Table 1. The maximum magnitude variations in each filter were determined as the difference between the observed maximum and minimum brightness values on the corresponding light curves; no averaging near the extrema was applied.

3. Results

The results of our new photometric observations obtained between 14 November 2024 and 1 February 2026 in the B, V, and $R_{\mathrm{c}}$ filters are presented in Table 1. The observations show that the variability of the galaxy in the B, V, and $R_{\mathrm{c}}$ filters is synchronous in nature. Over this interval, the brightness amplitudes are 0.35 mag in the B filter, 0.27 mag in the V filter, and 0.28 mag in the $R_{\mathrm{c}}$ filter. Over the entire monitoring period from 2016 to 2026, the total variability amplitude of the light curve reaches 1.9 mag in the B filter, 1.5 mag in the V filter, and 1.4 mag in the $R_{\mathrm{c}}$ filter. The maximum brightness of the galaxy was recorded on 17 January 2017.

For completeness of the analysis, we used all photometric data obtained starting from 5 April 2016 together with the new observations. Figure 1 presents the light curve constructed in three photometric filters over the period from 5 April 2016 to 1 February 2026. The light curve includes data from Shomshekova et al. (2019) [13], covering the interval 2016–2019, where a decrease in the object brightness by approximately 0.9 mag was reported during 2017–2019. Photometric observations presented in Shomshekova et al. (2025) [14] for the period from November 2019 to February 2024 confirmed this trend: after reaching maximum brightness in early 2017, the object faded by approximately 1 mag by the beginning of 2018. In addition, comparison of archival and recent spectral data revealed that the additional component in the blue wing of the H$\alpha$ line, detected in the 1976 spectra, is also present in the 2024 spectra, with a radial velocity of about 2450 km s⁻¹ (Shomshekova et al., 2025) [14].

For an independent verification of the variability amplitude, data from the ASAS-SN and ZTF survey programs were used. According to the ASAS-SN data in the V filter for the period from 25 January 2014 to 29 November 2018, the variability amplitude is about 0.6 mag. The ZTF photometric data in the g and r filters were transformed to the V system using the relation

$$V=g-0.578 (g-r)-0.003.$$

(1)

After the transformation, the variability amplitude of Mrk 6 during the period from September 2018 to 22 November 2023 was found to be about 0.7 mag, as shown in Figure 2.

Figure 3 presents the color indices $(B-V)$ and $(V-R_{\mathrm{c}})$. The mean color indices are ${(B-V)}_{\mathrm{mean}}=0.94$ and ${(V-R\mathrm{c})}_{\mathrm{mean}}=0.41$. All values are given without correction for interstellar reddening.

Results of Spectroscopic Observations

Spectroscopic observations were carried out at the Shamakhy Astrophysical Observatory (ShAO) on 15 November 2025 (exposure time 2400 s) and 22 November 2025 (exposure time 3600 s). The spectra were obtained in the wavelength range $\lambda \lambda 4100$–7000 Å with a spectral resolution of $R\approx 1170$. The dispersion was 5.6 Å mm⁻¹ with $3\times 3$ binning. The spectrum of the Mrk6 is shown in Figure 4.

Additional spectroscopic observations were performed on 9 January 2026 at the Assy-Turgen Observatory. In the H$\beta$ region, a diffraction grating of 2400 lines mm⁻¹ was used, providing a spectral resolution of $R\approx 5235$ with $1\times 1$ binning and an exposure time of 600 s. In the H$\alpha$ region, a grating of 1800 lines mm⁻¹ was used, yielding a spectral resolution of $R\approx 2078$ with an exposure time of 600 s. The spectral profiles are presented in Figure 5.

For each observation, three spectra of the standard star were recorded with short exposures (4–10 s). The spectrophotometric standard HD 33541 was observed at approximately the same zenith distance as the target galaxy to minimize atmospheric extinction effects. The energy distribution of the standard star spectrum was taken from the Gaia DR3 catalog [18].

All spectra used to determine the fluxes in the broad emission lines were processed following standard reduction procedures, including correction for atmospheric absorption, Galactic interstellar extinction, redshift correction, and absolute flux calibration using spectrophotometric standards within the IRAF software package. The full width at half maximum (FWHM) of the emission lines was determined by Gaussian fitting using the splot task in IRAF. The results are presented in

Table 2. Measured parameters of emission lines in the spectrum of Mrk 6.

Date	${\mathit{\lambda }}_{\mathbf{center}}$ (Å)	Continuum Flux	Line Flux	EW (Å)	gfwhm (Å)	FWHM (km s−1)
15 November 2025
	4086.3	$9.6\times {10}^{-15}$	$1.8\times {10}^{-13}$	19	53	3868
	4342.5	$1.3\times {10}^{-14}$	$3.4\times {10}^{-13}$	26	38	2634
	4858.6	$1.1\times {10}^{-14}$	$3.6\times {10}^{-13}$	32	22	1368
	4957.9	$1.2\times {10}^{-14}$	$8.9\times {10}^{-13}$	75	17	1043
	5006.0	$1.1\times {10}^{-14}$	$2.5\times {10}^{-12}$	218	19	1122
	6297.8	$9.0\times {10}^{-15}$	$2.1\times {10}^{-13}$	24	23	1088
	6560.8	$1.2\times {10}^{-14}$	$3.1\times {10}^{-12}$	264	54	2465
	6717.1	$1.0\times {10}^{-14}$	$4.7\times {10}^{-13}$	47	27	1202
22 November 2025
	4087.494	$6.8\times {10}^{-15}$	$1.6\times {10}^{-13}$	24	56	4134
	4343.973	$9.9\times {10}^{-15}$	$2.7\times {10}^{-13}$	27	39	2681
	4858.609	$9.3\times {10}^{-15}$	$2.9\times {10}^{-13}$	31	22	1353
	4958.104	$9.6\times {10}^{-15}$	$7.1\times {10}^{-13}$	74	18	1075
	5005.929	$9.1\times {10}^{-15}$	$2.1\times {10}^{-12}$	230	19	1136
	6297.865	$7.6\times {10}^{-15}$	$1.7\times {10}^{-13}$	22	21	1013
	6560.97	$9.5\times {10}^{-15}$	$2.7\times {10}^{-12}$	281	53	2409
	6717.499	$8.5\times {10}^{-15}$	$3.8\times {10}^{-13}$	45	27	1185
9 January 2026
	4862.402	$1.1\times {10}^{-14}$	$1.2\times {10}^{-13}$	11	12	721
	4961.252	$9.9\times {10}^{-15}$	$4.7\times {10}^{-13}$	47	13	785
	5006.963	$9.7\times {10}^{-15}$	$1.2\times {10}^{-12}$	126	12	743
	6567.251	$3.1\times {10}^{-14}$	$4.2\times {10}^{-12}$	137	45	2071
	6724.208	$2.4\times {10}^{-14}$	$8.7\times {10}^{-13}$	37	26	1160
16 February 2026
	4348.272	$6.5\times {10}^{-15}$	$1.6\times {10}^{-13}$	25	42	2874
	4860.209	$7.6\times {10}^{-15}$	$1.6\times {10}^{-13}$	21	22	1383
	4960.229	$7.1\times {10}^{-15}$	$4.6\times {10}^{-13}$	56	19	1173
	5008.325	$6.5\times {10}^{-15}$	$1.2\times {10}^{-12}$	183	17	1013
	6300.006	$5.2\times {10}^{-15}$	$7.4\times {10}^{-14}$	14	21	995
	6367.047	$5.5\times {10}^{-15}$	$2.4\times {10}^{-14}$	4	24	1146
	6561.164	$6.8\times {10}^{-15}$	$1.2\times {10}^{-12}$	182	47	2166
	6715.397	$6.7\times {10}^{-15}$	$2.6\times {10}^{-13}$	39	27	1198

.

In Table 2, the flux in the [O III] $\lambda 5007$ line was used as an internal calibrator in the analysis of the H$\beta$ line variability, while in the red spectral region the fluxes in the [S II] $\lambda \lambda 6717, 6731$ lines were used to normalize the fluxes in the H$\alpha$+[N II] complex. We find that the total H$\beta$ flux decreases by a factor of ∼2–3 between November 2025 and January 2026 (see

Table 3. Fluxes in the H$\beta$ line and their normalization using the [O III] $\lambda 5007$ internal calibrator.

Date	JD–2,400,000	Flux(H$\beta$)	Flux [O III] $\lambda 5007$	Flux(H$\beta$) corr
		(erg s−1 cm−2)	(erg s−1 cm−2)	(erg s−1 cm−2)
15 November 2025	60,995	$3.6\times {10}^{-13}$	$2.5\times {10}^{-12}$	$2.8\times {10}^{-13}$
22 November 2025	61,002	$2.9\times {10}^{-13}$	$2.1\times {10}^{-12}$	$2.7\times {10}^{-13}$
9 January 2026	61,050	$1.2\times {10}^{-13}$	$1.2\times {10}^{-12}$	$1.9\times {10}^{-13}$
16 February 2026	61,088	$1.59\times {10}^{-13}$	$1.20\times {10}^{-12}$	$2.3\times {10}^{-13}$

). The spectra show that the broad component of H$\beta$ becomes significantly weakened and nearly disappears in January 2026, while the narrow component and the [O III] lines remain relatively stable.

As shown in Figure 6, where the spectra are normalized to the [O III] $\lambda 5007$ flux, the broad H$\beta$ component reaches a minimum in January 2026 and then partially recovers by mid-February. Thus, the observed variability is non-monotonic and is primarily manifested as changes in the amplitude of the broad H$\beta$ component, without a significant change in the spectral type (the object remains close to type 1.5). This behavior may resemble that of changing-look AGN. However, the absence of a complete disappearance of the H$\beta$ line and the presence of a partial recovery suggest that this event is more appropriately described as CL-like variability rather than a classical changing-look transition. In terms of the classification by Komossa et al. (2026) [19], it is most consistent with category V (recurrent brightenings and dimmings on timescales from weeks to years), although the observed event occurs on a relatively short timescale. Similar episodes of the temporary weakening of broad emission lines have been reported in a number of objects [19,20,21,22,23,24].

To quantitatively estimate the size of the broad-line region (BLR) at different epochs, we calculated the monochromatic continuum luminosity at $\lambda 5100$ Å ($L_{5100}$) using the luminosity distance to Mrk 6 of $D_L=81$ Mpc and applied the empirical R–L relation from the referenced study. The results are summarized in

Table 4. Continuum flux at 5100 Å, monochromatic luminosity, and BLR size estimates for Mrk 6.

Date	${\mathit{F}}_{5100}$	$\mathit{\lambda }{\mathit{L}}_{\mathit{\lambda }}\left(5100\right)$	${\mathit{R}}_{\mathbf{BLR}}$	Telescope
	(erg cm−2 s−1 Å−1)	(erg s−1)	(Light-Days)
15 November 2025	$1.1\times {10}^{-14}$	$3.96\times {10}^{43}$	22	ShAO, 2 m
22 November 2025	$9.1\times {10}^{-15}$	$3.15\times {10}^{43}$	20	ShAO, 2 m
9 January 2026	$9.7\times {10}^{-15}$	$3.37\times {10}^{43}$	20	FAI, 1.5 m
16 February 2026	$6.52\times {10}^{-15}$	$2.60\times {10}^{43}$	16	ShAO, 2 m

. The characteristic BLR size varies approximately from 16 to 22 light days at different observing epochs.

4. Discussion

The light curve of the galaxy Mrk 6 for the period 2016–2026 is presented and discussed. Based on Figure 1, an estimate of the host-galaxy contribution was performed. As an approximate estimate of the host contribution at the minimum activity state of the active galactic nucleus (AGN), the minimum brightness values in each filter on the light curves were adopted. The minimum brightness was observed on 17 January 2017, when the magnitudes were $B=14.065$, $V=13.433$, and $R_{\mathrm{c}}=13.086$.

The observed color indices in Figure 2 show that $(B-V)$ varies in the range of approximately 0.63–1.07, whereas the $(V-R_{\mathrm{c}})$ index remains significantly more stable and is concentrated around ∼0.43. This indicates that the variability is stronger in the blue part of the spectrum, while the red region varies more weakly.

To account for the host-galaxy contribution, its flux was estimated from the minimum brightness levels on the light curves in each filter, assuming minimal AGN activity at these epochs. The magnitudes were converted to fluxes using the standard zero-points of the Johnson–Cousins photometric system [25].

As a result, the following host-galaxy fluxes were obtained: ${\mathrm{F}}_{\mathrm{host},B}\approx 1.51\times {10}^{-14}$, ${\mathrm{F}}_{\mathrm{host},V}\approx 1.59\times {10}^{-14}$, ${\mathrm{F}}_{\mathrm{host},R_{\mathrm{c}}}\approx 1.27\times {10}^{-14} \mathrm{erg} {\mathrm{cm}}^{-2} {\mathrm{s}}^{-1}$Å⁻¹.

In this study, we employed a Bayesian statistical approach for the approximation of observed spectra. Unlike the least squares method, this approach allows not only for finding the optimal parameters of emission lines but also for evaluating the adequacy of the spectral description model itself by calculating the ‘Bayesian evidence’. To identify the global maximum of the likelihood function and construct posterior parameter distributions, we utilized the Dynamic Nested Sampling method implemented in the Dynesty package [26]. Non-informative priors were defined over broad ranges of physically grounded values. Optimal parameter values were determined based on the median of the distribution (50th percentile), with uncertainties estimated at the 16th and 84th percentile levels. The total model spectrum included a power-law continuum and a set of emission lines from both the Narrow Line Region (NLR) and the Broad Line Region (BLR). The central wavelengths for the [NII], [SII], [OI], and [OIII] doublets, as well as the Balmer series lines and fixed doublet amplitude ratios, were set according to standard values. Seven model types (M0–M6) were tested, differing in their profiles and number of components: the NLR was modeled using Voigt (M0, M3), Gaussian (M1), or asymmetric Gaussian profiles with an error function (M2, M4, M5, M6). The BLR was represented by either three components (central, ‘blue’, and ‘red’) to account for complex BLR dynamics (M0, M1, M2), two components (M3, M4), or a single broad component (M5, M6). More details can be found in [27].

Spectral analysis indicates that the characteristic gas cloud velocities—estimated from the Full Width at Half Maximum (FWHM) of the Narrow Line Region (NLR) H$\beta$ and H$\alpha$ emission lines—are approximately 1093 ± 40 km s⁻¹. The mean FWHM values for the Broad Line Region (BLR) were determined to be 4630 ± 230 km s⁻¹ for H$\beta$ and 4530 ± 12 km s⁻¹ for H$\alpha$; these were adopted as proxies for the virial velocity of the BLR gas. Based on average time delays (lags) of ≈20 light days for H$\beta$ and ≈28 light days for H$\alpha$, we defined the BLR radius (${\mathrm{R}}_{BLR}$). Adopting a redshift of z = 0.01904 and H₀ = 70 km s⁻¹ Mpc⁻¹, the distance to Mrk 6 is estimated at D ≈ 81 Mpc. Applying a virial calibration factor of f ≈ 1.1 [28,29], the central supermassive black hole mass is calculated as ${\mathrm{M}}_{BH}\approx$ 0.92 × 10⁸ ${\mathrm{M}}_{\odot }$ (from H$\beta$) and ${\mathrm{M}}_{BH}\approx$ 1.24 × 10⁸ ${\mathrm{M}}_{\odot }$ (from H$\alpha$). These results for time delays and velocity dispersions are in good agreement with previously reported values [3,30]. The value of and ${\mathrm{M}}_{BH}$ is approximately two times lower than that reported in [3].

5. Conclusions

New photometric observations obtained between 14 November 2024 and 28 January 2026 are presented, demonstrating the synchronous variability of the galaxy in the B, V, and $R_{\mathrm{C}}$ filters, with an amplitude of about 0.3 mag in all bands. The galaxy light curve over an 11-year monitoring period was also analyzed. For an independent verification of the variability amplitude, data from the ASAS-SN and ZTF survey programs in the V filter were used.

The radius of the broad-line region $R_{\mathrm{BLR}}$ was taken to be equal to the average time delays (lags), amounting to ≈20 light days for the H$\beta$ line and ≈28 light days for the H$\alpha$ line. The obtained results are in good agreement with those reported by other authors.