Spatially resolved circumnuclear coronal [Fe VII] λ6087 emission in nearby Seyfert galaxies

S. Comerón; A. Prieto; P. Dabhade

doi:10.1051/0004-6361/202453159

Home

All issues

Volume 696 (April 2025)

A&A, 696 (2025) A137

Full HTML

Open Access

Issue		A&A Volume 696, April 2025


Article Number		A137
Number of page(s)		19
Section		Extragalactic astronomy
DOI		https://doi.org/10.1051/0004-6361/202453159
Published online		10 April 2025

A&A, 696, A137 (2025)

Spatially resolved circumnuclear coronal [Fe VII] λ6087 emission in nearby Seyfert galaxies^⋆

S. Comerón¹^,2^⋆⋆, A. Prieto²^,1 and P. Dabhade³^,2^,1

¹ Departamento de Astrofísica, Universidad de La Laguna, E-38200 La Laguna, Tenerife, Spain
² Instituto de Astrofísica de Canarias, E-38205 La Laguna, Tenerife, Spain
³ Astrophysics Division, National Centre for Nuclear Research, Pasteura 7, 02-093 Warsaw, Poland

^⋆⋆ Corresponding author; lsebasti@ull.edu.es

Received: 25 November 2024
Accepted: 21 February 2025

Abstract

Coronal lines are forbidden emission lines with a ionisation potential χ ≳ 100 eV. They are linked to energetic phenomena triggered by active galactic nuclei (AGNs) in the circumnuclear medium. We present the first high-angular-resolution integral-field analysis of the [Fe VII] λ6087 coronal line in a sample of four nearby low-inclination Seyfert galaxies of which three are of Type 1 and one of Type 2. The data were obtained with the adaptive-optics-assisted mode of MUSE, and have angular resolutions of 0.″06 − 0.″18, allowing us to probe regions down to a few tens of parsecs in size. In three of the objects, we find a resolved coronal emission in a relatively compact configuration (∼200 pc in radius in the most extended case). The coronal emission is smooth and symmetric with respect to the centre of the galaxy, except for one object where an off-nucleus clump of emission is detected. Through the use of spectroastrometry we find that the [Fe VII] λ6087 outflow of the Type 2 AGN host has a redshifted and a blueshifted component whose centroids are separated by ∼20 pc. We interpret this as evidence that some of the coronal emission comes from the inner part of a biconic outflow, also seen in low-ionisation lines such as [O III] λ5007. Similar [Fe VII] λ6087 properties are found in two of the Type 1 AGN hosts, but with a much smaller separation between the centroids of the lobes of the outflow (< 7 pc). This small projected separation compared to that in the Type 2 host could be due to the foreshortening of the axis of the bicone in Type 1 objects. We also studied the spectrum of the unresolved nuclear source and found that in three out of four galaxies a fraction of at least ∼60% of the [Fe VII] λ6087 emission has kinematics similar to those of [O III] λ5007. We conclude that part of the coronal emission within the inner few tens of parsecs is co-spatial and shares kinematics with the outflows as traced by lower-ionisation lines.

Key words: galaxies: active / galaxies: ISM / galaxies: jets / galaxies: nuclei

^⋆

Based on observations made at the European Southern Observatory using the Very Large Telescope under programmes 60.A-9100(K), 60.A-9493(A), 0103.B-0349(A), 0103.B-0908(A), and 0106.B-0360(A).

© The Authors 2025

Open Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This article is published in open access under the Subscribe to Open model. Subscribe to A&A to support open access publication.

1. Introduction

Coronal emission lines are high ionisation forbidden lines whose presence is an indication of energetic phenomena. Indeed, their ionisation potential is χ ≳ 100 eV, which usually precludes photoionisation from young massive stars as their trigger. Therefore, coronal emission is considered to be a tracer of active galactic nuclei (AGNs; see e.g. Negus et al. 2021, 2023). Coronal lines associated with AGNs have been known at least since 1956 (see comments about the coronal lines in NGC 4151 in Oke & Sargent 1968).

Line ratios indicate that the coronal-line-emitting gas has electron temperatures of at least between T_e ≈ 10⁴ K and T_e ≈ 2 × 10⁴ K (Mazzalay et al. 2010), but other authors find values of up to T_e ≈ 2.3 × 10⁴ K (Negus et al. 2021), or even T_e ≈ 7 × 10⁴ K (Erkens et al. 1997). The coronal-line-emitting gas might have electron densities compatible with those of the narrow-line region, that is between n_e ≈ 10² cm⁻³ and n_e ≈ 10³ cm⁻³notesep=; (Oliva et al. 1994; Negus et al. 2021; Dabhade et al. in prep.). Other authors find coronal-emitting gas with higher densities. For example Moorwood et al. (1996) estimated n_e ≈ 5000 cm⁻³ for the Circinus galaxy, and Erkens et al. (1997) found an average density of n_e ≈ 10⁶ cm⁻³ for a sample of 15 Seyfert galaxies. Some models of nuclear spectra indicate that coronal lines can also originate in even denser n_e ∼ 10⁻⁷ cm⁻³ gas located at a few parsecs of the supermassive black hole (SMBH; Kraemer et al. 1998).

The coronal-line-emission region is sometimes unresolved, but other times extends well into the narrow-line region (NLR) of the galaxy. The unresolved central peak has been hypothesised to be due to from the inner edge of the dusty torus surrounding the central engine (Pier & Voit 1995; Murayama & Taniguchi 1998; Rose et al. 2011, 2015; Glidden et al. 2016). However, a recent study of NGC 3783 by the GRAVITY Collaboration (2021) indicates that at least for the lowest-ionisation coronal lines the emitting region is well beyond the torus. Part of this more extended emission might come from cloudlets that have been ejected from the torus by the intense radiation field emanating from the central engine (Mullaney et al. 2009).

The instances of extended coronal activity are often aligned with the NLR, as well as with radio emission, and might come from AGN-driven outflows (see e.g. Erkens et al. 1997; Reunanen et al. 2003; Prieto et al. 2005; Riffel et al. 2008; Müller-Sánchez et al. 2011; Mazzalay et al. 2013; May et al. 2018; GRAVITY Collaboration 2021; Fonseca-Faria et al. 2023) or a circumnuclear disc (Müller-Sánchez et al. 2011). In both cases the excitation mechanism might be photoionisation by the hard AGN spectrum (Ward & Morris 1984; Mazzalay et al. 2010) and/or shocks (Prieto et al. 2005; Müller Sánchez et al. 2006; Rodríguez-Ardila et al. 2006; Rodríguez-Ardila & Fonseca-Faria 2020; Fonseca-Faria et al. 2023). The emission from ions with the highest ionisation potentials tends to be more compact than those with lower potentials (Gelbord et al. 2009). In radio galaxies, which commonly host powerful radio jets, coronal emission can extend out to several kiloparsecs (Dabhade et al. in prep.).

Because of the association of coronal lines with AGNs, there are ongoing efforts attempting to link their properties to those of the central engine, including the SMBH mass (Prieto et al. 2022). The enticing possibility of accurately determining the SMBH mass without resorting to expensive reverberation mapping is an excellent driver for further studies of these lines.

A technological breakthrough enabling us to delve into the properties of coronal lines is the advent of adaptive-optics-assisted (AO-assisted) integral-field spectrographs. Indeed, those are fundamental to glimpse the distribution of the otherwise often poorly resolved coronal-line region. The first coronal line studies made with these marvelous tools used near-IR data from SINFONI for the Circinus galaxy (Müller Sánchez et al. 2006), and from NIFS for NGC 4051 (Riffel et al. 2008; Storchi-Bergmann et al. 2009) and NGC 1068 (Mazzalay et al. 2013). The AO module GALACSI (Stuik et al. 2006) makes it possible to obtain data with a resolution of $\sim 0 \overset{″}{.} 1$ ${\sim}0{{\overset{\prime\prime}{.}}}1$ in the optical with the Multi Unit Spectroscopic Explorer (MUSE; Bacon et al. 2010).

In this paper we examine AO-assisted MUSE data of four nearby low-inclination Seyfert galaxies, namely Mrk 1044, NGC 3783, NGC 4593, and NGC 7130. We have focused on the [Fe VII] λ6087 line, whose ionisation potential is χ = 99.1 eV (De Robertis & Osterbrock 1984). We also examine the lines in the region of the Hβ plus [O III] complex. The exquisite angular resolution of the instrumental setting is crucial to resolve the details of the coronal-line region.

The paper is organised as follows. We describe the galaxies that constitute our sample in Sect. 2, and we give details about the MUSE datasets in Sect. 3. We describe the procedures used to derive physical properties from the MUSE datacubes in Sect. 4. We summarise our results in Sect. 5 and discuss them in Sect. 6. Throughout this paper we assume a flat-Universe cosmology with a Hubble-Lemaître constant H₀ = 70.5 km s⁻¹ Mpc⁻¹, a matter density Ω_m, 0 = 0.27, and a cosmological constant (parameters inferred from a combination of data from WMAP, Type Ia supernovae, and baryonic acoustic oscillations; Hinshaw et al. 2009).

2. The sample

Our sample is composed of four nearby low-inclination Seyfert disc galaxies with science-ready AO-assisted datacubes available in the ESO archive. In our previous studies, we have utilised these datacubes (Comerón et al. 2021; Mulumba et al. 2024; Dabhade et al. in prep.) and noticed the [Fe VII] λ6087 emission, prompting further investigation. The four galaxies are nearly face-on (largest inclination of i = 36° according to HyperLeda¹; Makarov et al. 2014).

DESI Legacy Survey DR10 (Dey et al. 2019) r-band images of the sample galaxies are shown in Fig. 1. Some of the properties of the sample galaxies are gathered in Table 1, and the references are given in the subsections below. The distances in the table were calculated from recession velocities (redshifts) obtained as described in Sect. 4.1. Another magnitude included in the table is the Eddington ratio, λ_Edd ≡ L_bol/L_Edd, where L_bol is the bolometric luminosity, and L_Edd the Eddington limit. We used the classical Eddington (1926) limit

$\begin{matrix} L_{Edd} = 1.47 \times 10^{38} (\frac{M_{∙}}{M_{⊙}}) erg s^{- 1}, \end{matrix}$ $\begin{aligned} L_{\rm Edd}=1.47\times 10^{38}\left(\frac{M_\bullet }{M_\odot }\right)\,\mathrm{erg}\,\mathrm{s}^{-1}, \end{aligned}$ (1)

Fig. 1.

DESI Legacy Survey DR10 r-band images of the sample galaxies. The axes are in arcseconds. North is up and east is to the left. NGC 4593 is interacting with PGC 42399, which is located in the leftmost edge of the corresponding snapshot.

Table 1.

Properties of the sample galaxies.

which comes from assuming a Thomson (1906) scattering caused by completely ionised material with a hydrogen mass fraction X = 0.7. The SMBH masses of Mrk 1044, NGC 3783, and NGC 4593 are based on reverberation mapping and the errors have been propagated to obtain the error bars in λ_Edd.

2.1. Mrk 1044

Mrk 1044 is an SAB0/a galaxy (Ohta et al. 2007) that hosts a narrow-line Seyfert 1 (NLSy1) AGN (Goodrich 1989; Véron-Cetty & Véron 2010). Mrk 1044 is in a pair with a galaxy of comparable luminosity, namely NGC 960 (Tully 2015).

The AGN in Mrk 1044 is powered by a SMBH with a mass of M_⚫ = 2.8 ± 0.8 × 10⁶ M_⊙ (Du et al. 2015). Owing to the NLSy1 nature of the object, the circumnuclear gas emits in a plethora of Fe II lines, which have been known at least since Rafanelli & Schulz (1983). The bolometric luminosity of the AGN is L_bol = 1.42 ± 0.14 × 10⁴⁴ erg s⁻¹ (Husemann et al. 2022). The SMBH is likely accreting at close-to-Eddington rate (Du et al. 2015; Husemann et al. 2022), and indeed the abovementioned SMBH mass and bolometric luminosity estimates yield λ_Edd = 0.34 ± 0.10.

The AGN in Mrk 1044 exhibits optical variability in periods of a year or less (Giannuzzo & Stirpe 1996). It is also variable in X-rays, with periods of tens of minutes (Mallick et al. 2018). The AGN causes a mildly relativistic ultrafast wind detected in X-rays, whose kinetic output might be sufficient to produce a significant effect in the host through feedback (Krongold et al. 2021; Xu et al. 2023). Winkel et al. (2022) deduced from the presence of star formation close to SMBH that these outflows started only recently (less than 10⁴ yr ago; Winkel et al. 2023) and have not affected their surroundings yet. The optical spectrum of Mrk 1044 reveals outflows traced by [O III] emission, with approaching velocities of a few hundred kilometres per second (Winkel et al. 2023).

2.2. NGC 3783

NGC 3783 is an (R′)SB(r)ab (de Vaucouleurs et al. 1991) Seyfert 1.5 (Véron-Cetty & Véron 2010) galaxy. NGC 3783 belongs to a group that might be at the early stages of evolution (Kilborn et al. 2006).

The Seyfert nucleus of NGC 3783 was already identified by Osmer et al. (1974) and is one of the most widely studied in the literature. It is powered by a SMBH with a mass of M_⚫ = 2.34 ± 0.43 × 10⁷ M_⊙ (Bentz et al. 2021). The central mass concentration hosting the AGN is a pseudobulge (Ho & Kim 2014).

The AGN in NGC 3783 exhibits nuclear outflows, as demonstrated by the P Cygni-like profiles of emission lines in X-Rays (Kaspi et al. 2001; Behar et al. 2003; Ramírez et al. 2005). The mass-loss-rate has been subjected to controversy, with some authors supporting a value of Ṁ = 0.01−0.1 M_⊙ yr⁻¹ (Chelouche & Netzer 2005) and others arguing to a much larger value of Ṁ ≈ 1 M_⊙ yr⁻¹ (Behar et al. 2003), or even Ṁ ≈ 2.5 M_⊙ yr⁻¹ (Müller-Sánchez et al. 2011). In the latter two cases, the SMBH would currently be ejecting significantly more material than it is accreting. According to Müller-Sánchez et al. (2011), the kinetic power calculated using the coronal-line-emitting gas is Ė_kin ≈ 7 × 10⁴⁰ erg s⁻¹, which is a tiny fraction of the bolometric power of L_bol ≈ 1.8 × 10⁴⁴ erg s⁻¹ (Prieto et al. 2010). We calculated an Eddington ratio of λ_Edd = 0.052 ± 0.010.

The AGN of NGC 3783 varies quickly with periods of down to a few hours in a variety of bands, including the IR continuum (Glass 1992; Lira et al. 2011), Balmer lines (Menzies & Feast 1983; Stirpe et al. 1988; Evans 1989; Winge et al. 1990, 1992; Winkler 1992), the optical continuum (Winge et al. 1990, 1992; Arévalo et al. 2009; Lira et al. 2011), the UV continuum (Barr et al. 1983; Chapman et al. 1985; Paltani & Courvoisier 1994; Scott et al. 2014), UV emission lines (Barr et al. 1983), the X-ray continuum (Walter & Courvoisier 1992; Green et al. 1993; De Rosa et al. 2002; Netzer et al. 2003; Markowitz 2005; Tombesi et al. 2007; Arévalo et al. 2009; Reis et al. 2012; Scott et al. 2014; Kaastra et al. 2018; Mao et al. 2019; De Marco et al. 2020), and X-ray emission lines (George et al. 1995; Tombesi et al. 2007; Mao et al. 2019). Some of the variability is associated with changes in the absorbers, as shown by the variation or even the appearance and disappearance of UV absorption lines over periods of weeks or months (Maran et al. 1996; Crenshaw et al. 1999; Kraemer et al. 2001; Gabel et al. 2005). These changes are caused by a combination of factors; the first one is the variation of the ionisation state of the gas that obscures the central source due to fluctuations in the photoionising flux (Gabel et al. 2005). The second factor is the varying column density of obscurers as outflowing clumps pass in front of the nucleus (Kraemer et al. 2001; Kaastra et al. 2018; De Marco et al. 2020). The dynamic status of the nuclear outflow has been confirmed by the discovery of decelerating clumps (Gabel et al. 2003; Scott et al. 2014), as well as the appearance and disappearance of blueshifted components of the Balmer lines (Menzies & Feast 1983).

The nuclear [Fe VII] λ6087 emission in NGC 3783 was first described by Atwood et al. (1982). Ward & Morris (1984) postulated that the line is generated through photoionisation. Rodríguez-Ardila et al. (2006) found that the coronal emission extends to at least out to 400 pc. They also found that the velocity dispersion in [Fe VII] λ6087 is about twice as large as that in lower-ionisation lines. GRAVITY Collaboration (2021) studied the [Ca VIII] coronal line using interferometry and found that the nuclear coronal emission is more extended than the BLR (0.4 pc versus 0.14 pc). At larger scales, the [Si IV] λ1.96 μm line kinematics indicate that a large fraction of the coronal-line-emitting gas is found in an outflow and a subdominant component is in a rotating disc (Müller-Sánchez et al. 2011; GRAVITY Collaboration 2021).

2.3. NGC 4593

NGC 4593 is an (R)SB(rs)b (Buta et al. 2015) Seyfert 1 (Véron-Cetty & Véron 2010) galaxy. NGC 4593 is the brightest member of a group (Kollatschny & Fricke 1985) and is interacting with the 7.7 times fainter galaxy PGC 42399 (Kormendy et al. 2006).

The engine of the AGN in NGC 4593 has a mass M_⚫ = 9.8 ± 2.1 × 10⁶ M_⊙ (Denney et al. 2006). The SMBH sits at the centre of a pseudo-bulge (Kormendy et al. 2006) that hosts a ∼1 kpc-radius nuclear star-forming dusty ring (González Delgado et al. 1997; Comerón et al. 2010). A dusty one-armed pattern departs inwards from the ring and reaches the Seyfert nucleus (Kormendy et al. 2006). The dusty ring and arm are also seen in the CO(1 − 0) line (Mulumba et al. 2024) and are probably tracing gas that is approaching the engine of the AGN.

The nuclear activity of NGC 4593 is variable over a period of the order days in a variety of bands (Pal & Naik 2018), including the optical (Dietrich et al. 1994), the UV (Clavel et al. 1983; Clavel 1983), and the X-rays (Burnell & Culhane 1979; Barr et al. 1987; Ghosh & Soundarajaperumal 1993; Ursini et al. 2016). The bolometric power of the AGN is L_bol = 5 × 10⁴³ erg s⁻¹ (Vasudevan & Fabian 2009), and the nuclear emission is compatible with a two-component accretion mode, where an outer accretion thin disc transitions to an inner and much thicker advection-dominated accretion flow (Lu & Wang 2000; Sriram et al. 2009; Markowitz & Reeves 2009). We find an Eddington ratio of λ_Edd = 0.035 ± 0.007. The outflow of NGC 4593 (Ruschel-Dutra et al. 2021) causes a mass-loss rate of Ṁ ≳ 0.05 M_⊙ yr⁻¹ and a kinetic power of Ė_kin ≳ 4 × 10₃₉ erg s⁻¹ (Mulumba et al. 2024).

2.4. NGC 7130

NGC 7130, also known as IC 5135, is a peculiar Sa object (de Vaucouleurs et al. 1991) with a composite (Phillips et al. 1983; Thuan 1984; Shields & Filippenko 1990; Davies et al. 2014) H II plus Seyfert 1.9 (Véron-Cetty & Véron 2010) nucleus. NGC 7130 is a LIRG (Armus et al. 2009), which further strengthens the case for Seyfert activity and/or a nuclear starburst. The outskirts of NGC 7130 are distorted, which might indicate a past interaction.

In spite of being classified as unbarred by de Vaucouleurs et al. (1991), near-IR (Mulchaey et al. 1997) and Hubble Space Telescope (HST) imaging (Comerón et al. 2021) of NGC 7130 have shown the presence of a bar hosting conspicuous dust lanes. These lanes correspond to the molecular gas traced by the CO(6 − 5) line (Zhao et al. 2016) and extend all the way to the central arcsecond, where they probably contribute to the feeding of the nuclear starburst and the AGN. The bar is surrounded by an inner pseudo-ring (Dopita et al. 2002; Muñoz Marín et al. 2007).

The composite condition of the nucleus of NGC 7130 is known from the mix of narrow lines with velocities matching that of the main body of the galaxy and broader blueshifted and redshifted lines corresponding to AGN-triggered outflows (Busko & Steiner 1988; Shields & Filippenko 1990; Contini et al. 2002). The full complexity of the circumnuclear medium in NGC 7130 was unveiled thanks to AO-assisted MUSE data showing an intricate structure comprising a ionised disc and a biconic outflow made of several components with distinct kinematics (Comerón et al. 2021). Star formation is ubiquitous in the central region of NGC 7130 (see HST images in González Delgado et al. 1998) and some of it is organised in two nuclear rings (Comerón et al. 2021). The outer one is ≈2″ (≈660 pc) in radius. The innermost ring is tiny, with a radius of $\approx 0 \overset{″}{.} 5$ ${\approx}0{{\overset{\prime\prime}{.}}}5$ (≈160 pc), akin to the Ultra Compact Nuclear rings described in Comerón et al. (2008).

The properties of the X-ray emission indicate that the nucleus of NGC 7130 is heavily obscured (Levenson et al. 2005), as expected for a Type 2 AGN. The nucleus has a significant radio emission (Norris et al. 1990). The bolometric power of the AGN is L_bol = 2.5 × 10⁴⁴ erg s⁻¹ (Esquej et al. 2014). The 8.4 GHz data obtained by Thean et al. (2000) and processed by Zhao et al. (2016) show a radio jet aligned with the ionised gas outflow, indicating that the former is likely to be powering the latter (Comerón et al. 2021). The outflow mass-loss rate is Ṁ = 1.5 ± 0.9 M_⊙ yr⁻¹, and the kinetic power is Ė_kin = (3.4 ± 2.5) × 10⁴¹ erg s⁻¹ (Comerón et al. 2021).

No reverberation mapping mass determination of the SMBH of NGC 7130 exists (see the database² by Bentz & Katz 2015). Therefore, we are not estimating the Eddington ratio for this object.

A preliminary study of the [Fe VII] λ6087 emission in NGC 7130 was performed by Comerón et al. (2021). They found that the kinematics of the coronal gas match those of the inner part of the biconic outflow as traced by low-ionisation lines.

3. The data

The galaxies of the sample have good-quality MUSE AO-assisted data. Each of the science-ready datacubes has two extensions, namely one with the spectra, and one with an error estimate for each voxel. The spectral pixel size is of 1.25 Å. The spectral resolution is described by Eq. (8) in Bacon et al. (2017) and is Δλ ≈ 2.9 Å (≈180 km s⁻¹) at the wavelength of [O III], and Δλ ≈ 2.6 Å (≈130 km s⁻¹) at [Fe VII] λ6087. The field of view (FoV) of the AO-assisted mode is $7 \overset{″}{.} 5 \times 7 \overset{″}{.} 5$ $7{{\overset{\prime\prime}{.}}}5\times7{{\overset{\prime\prime}{.}}}5$ . Due to the absence of large dithers in the data observations, this corresponds to the FoV of our datasets.

The spaxels of the datacubes are $0 \overset{″}{.} 025$ $0{{\overset{\prime\prime}{.}}}025$ on a side. The full width at half maximum (FWHM) of the datasets was estimated using the imexamine tool from IRAF (Tody et al. 1986) applied to a white-light image of the central unresolved point source, except for NGC 7130, where the nucleus is obscured and we used star-formation knots instead. The details of the datasets are discussed in the following subsections and are summarised in Table 2. All the datacubes can be downloaded from the ESO archive as Phase 3 products.

Table 2.

Properties of the MUSE datacubes.

3.1. Mrk 1044

The observations of Mrk 1044 were taken between 24 August 2019 and 29 September 2019 within the programme 0103.B-0349(A) (PI B. Husemann). The combined cube is made out of eight 550 s exposures, for a total exposure time of 4400 s.

This dataset has been used to characterise the circumnuclear star formation in the surroundings of the NLSy1 AGN (Winkel et al. 2022), as well as the AGN-powered outflows (Winkel et al. 2023). The point spread function (PSF) of the bright central source has a significant effect on the line emission throughout much of the field of view, as shown by the fact that the broad lines associated with the unresolved BLR can be detected as fas as a few arcseconds from the AGN (Winkel et al. 2022).

3.2. NGC 3783

The data used to construct the cube of NGC 3783 were acquired between 11 January and 6 March 2021 within the programme 0106.B-0360(A) (PI S. Raimundo). According to the abstract of the proposal, the data were acquired to measure SMBH mass from the circumnuclear dynamics of the gas and the stars, but the article has not been published yet. The combined datacube is built from nine exposures with lengths between 1800 s and 2090 s, for a total exposure time of 16 780 s.

3.3. NGC 4593

The data of NGC 4593 were taken within the programme 0103.B-0908(A) (PI J. Knud). The combined dataset sums eight 600 s-exposure cubes totalling 4800 s taken on 28 April 2019. As for Mrk 1044, the central source is so bright that the extended wings of the PSF of the broad-line region (BLR) emission contaminate most of the field of view. This complicates the analysis of the narrow-line emission in the Balmer and neighbouring lines (Mulumba et al. 2024).

3.4. NGC 7130

The cube of NGC 7130 comes from combining two distinct datasets. The first one encompasses three 300 s exposures taken during the commissioning of the AO-assisted mode of MUSE on 19 and 21 June 2018 (programme 60.A-9100(K)). The second dataset is made of two 600 s exposures obtained during the science verification within the programme 60.A-9493(A) (PI M. Seidel). The combined datacube has a total exposure of 2100 s. These data have been exploited in two works unveiling the intricacies of the multi-component circumnuclear medium of NGC 7130 (Knapen et al. 2019; Comerón et al. 2021).

4. Extraction of the spectroscopic information

4.1. Derivation of the redshifts of the galaxies

Owing to the difference of up to a few hundred kilometres per second between determinations of the recession velocity of the galaxies in the literature, we obtained our own redshifts from the MUSE data. In order to avoid being biased by the outflows, we derived the velocities from the stellar kinematics. The AGN emission lines often overshadow the nuclear stellar emission, so we obtained averaged spectra from circular corona with an inner radius of 30 spaxels and an outer radius of 50 spaxels (corresponding to $0 \overset{″}{.} 76$ $0{{\overset{\prime\prime}{.}}}76$ and $1 \overset{″}{.} 27$ $1{{\overset{\prime\prime}{.}}}27$ , respectively).

The spectra were fitted with pPXF³ (Cappellari & Emsellem 2004; Cappellari 2017, 2023) using stellar population templates from the E-MILES library (Vazdekis et al. 2015). We selected the templates with a Kroupa initial mass function (Kroupa 2001) and BaSTI isochrones (Pietrinferni et al. 2004, 2006, 2009, 2013; Cordier et al. 2007). The fit was run between λ = 5100 Å and λ = 6300 Å. We masked the regions with strong emission lines ([N I] λ5198, [N I] λ5200, [Fe VII] λ6087, [O I] λ6300) with 30 Å-wide windows. We further excluded the region around the strong 5577 Å sky line (20 Å-wide window) and the region between λ = 5774 Å and λ = 6054 Å, which is affected by the laser guide stars. The spectral resolution was taken from Eq. (8) in Bacon et al. (2017). The shape of the stellar continuum was modelled with a 16^th-degree additive Legendre polynomial.

We find redshifts of z = 0.01641, z = 0.00966, z = 0.00833, and z = 0.01621 for Mrk 1044, NGC 3783, NGC 4593, and NGC 7130, respectively. These values were used to compute the comoving distances listed in Table 1.

4.2. [O III] λ5007 and [Fe VII] λ6087 kinematic maps

We extracted velocity and velocity dispersion maps of the [Fe VII] λ6087 and the [O III] λ5007 lines with the goal of comparing the properties of the coronal lines with those of lower-ionisation ones. In order to prepare the maps we used the GIST pipeline⁴, an all-in-one modular integral-field-spectrograph data processing software (Bittner et al. 2019). GIST accounted for the spectral resolution using the description given by Eq. (8) in Bacon et al. (2017).

Since many spaxels do not have a large enough signal-to-noise ratio (S/N) to extract good-quality kinematics, we instructed GIST to produce a Voronoi binning with VorBin⁵ (Cappellari & Copin 2003). Two tessellations were generated, one for [O III] λ5007 (with S/N = 30) and another one for [Fe VII] λ6087 (with S/N = 20). The S/N of the individual spaxels was calculated through a custom-made script fed into GIST. This script computes the signal integrated over a spectral window of 20 Å centred in the line of interest and subtracts the continuum signal measured in an equally wide window situated 30 Å to the red of the centre of the line. The noise was computed by quadratically summing the noise estimates in both the on and off windows. In both cases, only spaxels with S/N > 2 were selected for binning.

After the binning, GIST estimated the stellar contribution to each of the bins by calling pPXF as in Sect. 4.1. The characterisation of the stellar contribution is not critical because it is often subdominant compared to the line emission.

We instructed GIST to model the line emission with a single Gaussian component per line and per bin using pyGandALF (Bittner et al. 2019), which is the python implementation of GandALF⁶ (Sarzi et al. 2006; Falcón-Barroso et al. 2006). When studying [O III] λ5007, we only fitted the surroundings of the [O III] doublet, between λ = 4909 Å and λ = 5107 Å. For [Fe VII] λ6087 the window of the fit was between λ = 5987 Å and λ = 6187 Å. The shape of the previously fitted stellar contribution was modified with a second order Legendre multiplicative polynomial. When fitting the [O III] doublet, we tied the kinematics of the line and fixed the flux of [O III] λ4959 to be 0.336 times that of [O III] λ5007, following Storey & Zeippen (2000).

Figures 2, 3, 4, and 5 show the surface brightness (Σ), velocity (V), and velocity dispersion (σ) [O III] λ5007 and [Fe VII] λ6087 maps for the sample galaxies. The grey overlays correspond to the stellar continuum measured between λ = 5250 Å and λ = 5450 Å. The black overlays correspond to radio emission. For Mrk 1044 the radio data corresponds to 8.49 GHz emission from the NRAO/VLA Archive Survey (NVAS; Crossley et al. 2008). The beam size is of $0 \overset{″}{.} 36 \times 0 \overset{″}{.} 22$ $0{{\overset{\prime\prime}{.}}}36\times0{{\overset{\prime\prime}{.}}}22$ , and the central source is unresolved. For NGC 3783 we offer an 8.46 GHz image from the NVAS with a beam size of $0 \overset{″}{.} 68 \times 0 \overset{″}{.} 19$ $0{{\overset{\prime\prime}{.}}}68\times0{{\overset{\prime\prime}{.}}}19$ , which again yields an unresolved nucleus. The small extension to the south might be real, as a similar feature also appears in a 8.49 GHz NVAS image of the same field. For NGC 4593 the radio data corresponds to the 3 GHz continuum from the VLA Sky Survey (VLASS; Lacy et al. 2020). The large beam size of $3 \overset{″}{.} 4 \times 2 \overset{″}{.} 2$ $3{{\overset{\prime\prime}{.}}}4\times2{{\overset{\prime\prime}{.}}}2$ makes the central radio source unresolved. The 8.4 GHz data of NGC 7130 (Thean et al. 2000; Zhao et al. 2016) have a beam size of $0 \overset{″}{.} 60 \times 0 \overset{″}{.} 19$ $0{{\overset{\prime\prime}{.}}}60\times0{{\overset{\prime\prime}{.}}}19$ , and the inner parts of the radio jet are resolved. None of the images available in the NVAS shows kiloparsec-scale radio emission. Hence all the detected radio emission is found within the FoV of the MUSE datacubes.

Fig. 2.

Maps of the surface brightness, Σ (left column), velocity, V (middle column), and velocity dispersion, σ (right column) measured for [O III] λ5007 (upper row) and [Fe VII] λ6087 (lower row), respectively. The grey contours indicate the stellar continuum estimated in the range between λ = 5250 Å and λ = 5450 Å in the rest frame of the galaxy. The black contours correspond to the unresolved 8.49 GHz radio emission from NVAS with a beam size of $0 \overset{″}{.} 36 \times 0 \overset{″}{.} 22$ $0{{\overset{\prime\prime}{.}}}36\times0{{\overset{\prime\prime}{.}}}22$ and a position angle of $PA = - 19 \overset{°}{.} 93$ $\mathrm{PA}=-19{{\overset{\circ}{.}}}93$ . The separation between contours corresponds to factors of two in surface brightness. The purple circle indicates the FWHM of the MUSE data, $0 \overset{″}{.} 08$ $0{{\overset{\prime\prime}{.}}}08$ . The cross indicates the centre of the galaxy. The axes labels are in arcseconds. North is up and east is to the left.

Fig. 3.

Same as Fig. 2, but for NGC 3783. The black contours correspond to the 8.46 GHz radio emission from NVAS with a beam size $0 \overset{″}{.} 68 \times 0 \overset{″}{.} 19$ $0{{\overset{\prime\prime}{.}}}68\times0{{\overset{\prime\prime}{.}}}19$ and a beam position angle $PA = - 10 \overset{°}{.} 00$ $\mathrm{PA}=-10{{\overset{\circ}{.}}}00$ . The FWHM of the MUSE data is $0 \overset{″}{.} 12$ $0{{\overset{\prime\prime}{.}}}12$ . The insets in the lower-left corner of each panel show an enlarged version of the innermost $0 \overset{″}{.} 95 \times 0 \overset{″}{.} 95$ $0{{\overset{\prime\prime}{.}}}95\times0{{\overset{\prime\prime}{.}}}95$ (190 pc × 190 pc).

Fig. 4.

Same as Fig. 2, but for NGC 4593. The black contours correspond to the 3 GHz radio emission from VLASS. The latter emission has a beam size of $3 \overset{″}{.} 4 \times 2 \overset{″}{.} 2$ $3{{\overset{\prime\prime}{.}}}4\times2{{\overset{\prime\prime}{.}}}2$ with a position angle $PA = 27 \overset{°}{.} 10$ $\mathrm{PA}=27{{\overset{\circ}{.}}}10$ . The FWHM of the MUSE data is $0 \overset{″}{.} 12$ $0{{\overset{\prime\prime}{.}}}12$ .

Fig. 5.

Same as Fig. 2, but for NGC 7130. The radio contours (in black) correspond to 8.4 GHz data from Thean et al. (2000) reprocessed by Zhao et al. (2016). These radio data have a beam size of $0 \overset{″}{.} 60 \times 0 \overset{″}{.} 19$ $0{{\overset{\prime\prime}{.}}}60\times0{{\overset{\prime\prime}{.}}}19$ and an orientation PA ≈ 10°. The FWHM of the MUSE cube is $0 \overset{″}{.} 18$ $0{{\overset{\prime\prime}{.}}}18$ .

As proven by the study of the MUSE data of Mrk 1044 by Winkel et al. (2022), galaxies hosting a very bright central point source have the light of the unresolved nucleus scattered over a large fraction of the FoV. Following them, we estimated the shape of the PSF using QDeblend^3D (Husemann et al. 2013). This code extracts the broad line information from the data cubes and, under the assumption that the region emitting these lines is unresolved, reconstructs the PSF. Given their proximity in wavelength, we assumed that the PSF obtained from Hβ matches that in [O III] λ5007 and that the Hα PSF corresponds to that in [Fe VII] λ6087. We did not apply QDeblend^3D to the Type 2 AGN in NGC 7130 because it lacks broad lines. The effects of a central source are not very relevant in NGC 7130 because the nucleus is obscured.

In Fig. 6 we compare the circularly averaged surface brightness profile model of the PSF with that of emission depicted in the leftmost panels of Figs. 2, 3, and 4. We find that the distribution of the emission matches the PSF for Mrk 1044 in both [O III] λ5007 and [Fe VII] λ6087, which indicates that the emitters are not resolved. On the other hand, the emission in NGC 3783 and NGC 4593 is more extended than the PSF, so we are resolving both the NLR and the coronal-line emission region. This conclusion is strengthened by the fact that uniform kinematic maps, indicating a lack of resolved kinematic gradients, are seen in Mrk 1044 only.

Fig. 6.

Circularly averaged surface brightness profiles of the PSF models obtained with QDeblend^3D (in black) for the three Type 1 AGNs in our sample. The grey line indicates the circularly averaged emission in either [O III] λ5007 (left-hand column) or [Fe VII] λ6087 (right-hand column).

4.3. [Fe VII] λ6087 channel maps and spectroastrometry

Channel maps show surface brightness maps of specific wavelength channels, and are a powerful tool to unveil the projected distribution of gas moving at a given radial velocity. Historically, they have been used mostly in radio-astronomy, but the advent of optical and near-infrared integral-field spectroscopy has made it possible to produce them at other wavelengths.

We constructed [Fe VII] λ6087 channel maps by merging three 1.25 Å spectral resolution elements together, so that each channel includes the integration of the light over a Δλ = 3.75 Å-wide window, which corresponds to Δv ≈ 180 km s⁻¹ for a z ≈ 0 object. This is slightly larger than the Δλ ≈ 2.6 Å spectral resolution expected at that wavelength (Sect. 3). The continuum was estimated from a 20 Å-wide window located 30 Å redwards of the line centre in the rest frame of the galaxy. The position of the window was selected so it would not contain any significant emission from redshifted components (see for reference the nuclear [Fe VII] λ6087 spectra in the middle panels of Figs. 11, 12, 13, and 14). We produced the channel maps for a window centred at the recession velocity of the galaxy, five blue windows, and five red windows, so the range of velocities between v ≈ −900 km s⁻¹ and v ≈ 900 km s⁻¹ was probed. This range was selected after checking that no significant emission is found at larger velocities. The exception is NGC 3783, where we are presenting the range between v ≈ −1300 km s⁻¹ and v ≈ 500 km s⁻¹ so as not to exclude some of the blueshifted gas. The channel maps are shown in Figs. 7, 8, 9, and 10.

Fig. 7.

[Fe VII] λ6087 channel maps of the circumnuclear region of Mrk 1044. The channels correspond to the sum of three spectral elements (covering 3.75 Å), whose central velocity in the rest frame of the galaxy is shown at the top of the panels. Photocentres were determined for each channel and their position is plotted in the lower-right panel if the signal within a diameter equal to twice the FWHM exceeds S/N = 6. Channels under this threshold are marked as having ‘No signal’. The photocentres in the lower-right panel are colour-coded according to the velocity of their channel map. The cross indicates the centre of the galaxy. The axes labels are in arcseconds. The grey contours indicating the stellar emission and the purple circles indicating the FWHM of the PSF are as in Fig. 2.

Fig. 8.

As Fig. 7, but for NGC 3783. The grey contours and the purple circles indicating the FWHM of the PSF are as in Fig. 3.

Fig. 9.

As Fig. 7, but for NGC 4593. The grey contours and the purple circles indicating the FWHM of the PSF are as in Fig. 4.

Fig. 10.

As Fig. 7, but for NGC 7130. The grey contours and the purple circles indicating the FWHM of the PSF are as in Fig. 5.

Fig. 11.

Fits to selected emission lines of the nuclear region of Mrk 1044 (aperture radius of $0 \overset{″}{.} 08$ $0{{\overset{\prime\prime}{.}}}08$ or 27 pc). The top panel covers the region with the complex comprising Hβ and the [O III] doublet, and the lower panel focuses on [Fe VII] λ6087. The original spectrum is shown in black, whereas the fit is shown in crimson. The different components of the emission lines are represented as Gaussians with their zero value situated at the lower limit of the displayed intensity range. The emission lines are colour-coded as dark blue for Hβ, turquoise for Fe II, Fe II], and [Fe II], violet for Cr II, purple for Ti II, pink for N I, green for [O III], light blue for Si II, and orange for [Fe VII] λ6087. The intensity of N I and Ti II is so small that they barely protrude above the horizontal axis.

Fig. 12.

Same as Fig. 11, but for NGC 3783 and an aperture radius of $0 \overset{″}{.} 06$ $0{{\overset{\prime\prime}{.}}}06$ (12 pc). This AGN has no Fe II, Fe II], [Fe II], Cr II, Ti II, N I, and Si II lines.

Fig. 13.

Same as Fig. 12, but for NGC 4593 and an aperture radius of $0 \overset{″}{.} 12$ $0{{\overset{\prime\prime}{.}}}12$ (20 pc).

Fig. 14.

Same as Fig. 12, but for NGC 7130 and an aperture radius of $0 \overset{″}{.} 18$ $0{{\overset{\prime\prime}{.}}}18$ (58 pc).

We used a spectroastrometric approach to assess the presence of [Fe VII] λ6087-emitting outflows. Spectroastrometry was first introduced by Beckers (1982) and a review on the technique can be found in Whelan & Garcia (2008). Briefly, it consists in finding the photocentre of the emission in different channels. Since this can be performed with a high angular precision, it becomes possible to detect kinematic structures at scales smaller than the FWHM of the data. We measured the photocentres using the routine centroid_sources from the python package photutils (Bradley et al. 2024). Once the photocentre was acquired, for each channel we performed aperture photometry of the source within a diameter equal to two times the FWHM of the datacube (see Table 1). We only considered the source to be reliably detected in a given channel if it exceeded S/N = 6 within a diameter equal to twice the FWHM.

The photocentres corresponding to each channel are shown in the lower-right panels of Figs. 7, 8, 9, and 10. We colour-coded them according to the velocity of the corresponding channel, where violet hues correspond to blueshifted velocities and red ones to redshifted gas. The position of the photocentres is also listed in Tables 3, 4, 5, and 6. For Mrk 1044, NGC 3783, and NGC 4593 the reference coordinates, with respect to which the shifts in coordinates Δα and Δδ were defined, were assumed from the photocentre as determined using IRAF’s imexamine in a white-light image. For NGC 7130, the central regions of the galaxy host several bright knots corresponding to star formation regions, so we instead manually selected the centre to be found at the centre of the biconic structure unveiled by Knapen et al. (2019) and Comerón et al. (2021).

Table 3.

Photocentre of the [Fe VII] λ6087 surface brightness distribution in different channels for Mrk 1044.

Table 4.

Photocentre of the [Fe VII] λ6087 surface brightness distribution in different channels for NGC 3783.

Table 5.

Photocentre of the [Fe VII] λ6087 surface brightness distribution in different channels for NGC 4593.

Table 6.

Photocentre of the [Fe VII] λ6087 surface brightness distribution in different channels for NGC 7130.

4.4. Spectra of selected emission lines of the nuclei

In order to characterise the nuclear [Fe VII] λ6087 emission we generated spectra of the central sources of the galaxies. We did so by integrating the light in an aperture centred in the AGN, and with a diameter equal to twice the FWHM. We first fitted the stellar underlying emission, as described in Sect. 4.1.

We fitted the emission lines in two wavelength ranges in the rest frame of the target: i) between 4710 Å and λ = 5110 Å to fit the complex of lines including Hβ and the [O III] doublet, and ii) between λ = 6007 Å and λ = 6167 Å to characterise the [Fe VII] λ6087 line. The reasons for fitting [O III] is to compare the coronal-line emission with that of lower-ionisation lines. Given the proximity of the [O III] doublet to Hβ, it is convenient to fit the Balmer line, especially if it has broad wings.

The emission lines were fitted with pyGandALF with as many components as required by an eyeball analysis. Since the nuclear profiles are rather complicated, obtaining a fit that converged to a reasonable solution required experimenting with the number of assumed components and the initial guesses of their kinematics. The intensities of the lines of the [O III] doublet were tied as done in Sect. 4.2. For NGC 3783, we used an order four polynomial to fit of the bluest spectral range because otherwise we would have required an [O III] component with σ > 2000 km s⁻¹ to properly describe the spectrum.

Given its NLSy1 nature, the nucleus of Mrk 1044 exhibits a dense forest of Fe II lines, requiring a specialised approach. When fitting the Hβ plus [O III] complex, we first attempted using the Fe II line template from Park et al. (2022). To do so, we converted the vacuum wavelengths of the template to air wavelengths using the vactoair2 function from the PyAstronomy package⁷ (Czesla et al. 2019), which in turn uses formulae for the refractive index of the air from Ciddor (1996). The Park et al. (2022) template includes permitted, semiforbidden, and forbidden Fe II lines, as well as lines of a few other metals (N I, Si II, Ti II, and Cr II) typically detected in NLSy1 AGNs.

Since the result of fit the Fe II forest in Mrk 1044 with the template was not satisfactory, we adopted another strategy based on how Park et al. (2022) obtained the amplitude of their components. We fixed the kinematics of the narrow components from Park et al. (2022) to the value obtained from fitting Fe IIλ4924 using the spectral range λ = 4890 Å to λ = 4950 Å and describing the blend of the weaker close-in-wavelength lines with a fourth-order multiplicative Legendre polynomial. Even so, there are large degeneracies for the kinematics of the broad lines, with different initial velocity guesses resulting in rather different results. We opted for a fit where the slope at the bluest side of the fitted range is properly reproduced by one of these broad lines. Another indication of the degeneracies involved in the fit is that an initial-parameter dependent subset of the weakest lines from Park et al. (2022) are given a zero amplitude by pyGandALF. Moreover, given the large amount of free parameters, the uncertainty in the flux of the weakest lines is likely to be very large.

Mrk 1044 also has a rare line at λ ≈ 6150 Å that has seldomly been reported in the literature. The only explicit reference is in the list of lines used for the template of I Zw 1 from Véron-Cetty et al. (2004). If the identification is correct, the line is Fe IIλ = 6149 Å. The line also appears in the Fe II template from Pandey et al. (2024). The confidence in the identification of the line is strengthened by the fact that the fitted velocity (v ≈ −70 km s⁻¹) is not too far off those of the narrow Fe II lines in the bluest fitted range (v ≈ −50 km s⁻¹). The line could also be identified with Fe IIλ6155, but this would result in a velocity of v ≈ −360 km s⁻¹ that is incompatible with that of the other Fe II lines.

The results of the fits of the nuclear spectra are presented in Figs. 11, 12, 13, and 14. The fitted values (velocities, velocity dispersions, and fluxes of the lines), are shown in Tables 7, 8, 9, and 10. The errors were estimated from a set of hundred Montecarlo realisations, where the input spectrum was randomly modified according to the error estimates included in the MUSE datasets. These estimates do not include systematics and can be considered as a lower boundary to the real uncertainties. The accuracy at pinpointing the centre of a narrow emission line with a velocity dispersion of a few hundred of kilometres per second is probably of a spectral pixel (Δλ = 1.25 Å or ∼50 km s⁻¹) or better.

Table 7.

Properties of the emission lines in the nucleus of Mrk 1044.

Table 8.

Properties of the emission lines in the nucleus of NGC 3783.

Table 9.

Properties of the emission lines in the nucleus of NGC 4593.

Table 10.

Properties of the emission lines in the nucleus of NGC 7130.

5. Results

5.1. Mrk 1044

In Mrk 1044, both the [Fe VII] λ6087 and [O III] λ5007 emission are dominated by the central AGN at all radii (see Sect. 4.2). Only through a complex PSF-deblending procedure it becomes possible to characterise the underlying ionised disc (Winkel et al. 2022). Given the size of the PSF and the distance to the galaxy, the coronal-emitting area must have a radius of r ≈ 15 pc at most. Thus, the featureless kinematic maps shown in Fig. 2 reflect the properties of the unresolved central source, and indicate that the [O III] λ5007 emission is approaching with velocities of v ≈ −100 km s⁻¹. The [Fe VII] λ6087 emission is blueshifted as well, but moves faster, with v ≈ −350 km s⁻¹.

Channel maps of Mrk 1044 (Fig. 7) do not show an obvious shift of the position of the photocentre with velocity. We find tentative evidence (Table 3) that the centroids for the blueshifted channels are located to the south of the nucleus, whereas the centroid of the only redshifted channel is located to the north. The difference is subtle, since the north-south maximum separation is less than $0 \overset{″}{.} 03$ $0{{\overset{\prime\prime}{.}}}03$ (slightly more than a spaxel), or less than 10 pc. The lack of a clear separation in the blueshifted and redshifted photocentres might indicate that the AGN is seen nearly face-on (which would go in line with the Type 1 classification of the AGN). That, and the lack of resolved [Fe VII] λ6087 emission is consistent with the findings by Winkel et al. (2023), who propose that the AGN became active less than 10⁴ yr ago, meaning the outflowing gas has not yet had sufficient time to travel significant distances from the SMBH.

The interpretation of the nuclear spectra (Fig. 11 and Table 7) is complicated by the presence of the Fe II forest, especially in the bluest of the explored spectral ranges. We only detect one outflowing component for [O III] λ5007, with v ≈ −150 km s⁻¹, and which probably corresponds to the v = −144 ± 5 km s⁻¹ component in Winkel et al. (2023). We do not detect their broad line with v = −560 ± 20 km s⁻¹, but that purported component is very small in amplitude and could only be fitted by kinematically tying it to Hβ. The kinematics of the [O III] λ5007 outflow are not compatible with any of those of the Hβ components. The [Fe VII] λ6087 outflow has a velocity of v ≈ −370 km s⁻¹, which is not compatible with the kinematics of either [O III] λ5007 or Hβ. Mrk 1044 is the only galaxy in the sample for which we fit the nuclear [Fe VII] λ6087 spectrum with a single component.

Mrk 1044 is the only galaxy in the sample where nuclear kinematics of [Fe VII] λ6087 are completely incompatible with those of lines associated with the NLR, such as [O III] λ5007. We considered whether this might be due to a confusion and that we are actually observing a transition attributable to the Fe II forest. For that to be true, the line should share the velocity with other narrow lines in the Fe II spectrum, which would imply a rest frame wavelength of λ ≈ 6081 Å. No Fe II with this wavelength is listed in the NIST Atomic Spectra Database⁸ (Kramida & Ralchenko 1999). The template by Véron-Cetty et al. (2004) does not consider a line at that wavelength for other ions, such as Ti II, either. We thus conclude that a confusion is unlikely.

5.2. NGC 3783

The kinematic maps of NGC 3783 show great complexity (Fig. 3) and are hard to interpret. The [Fe VII] λ6087 emission is detected out to a radius of ∼200 pc, and shows a dipolar kinematical structure with a blueshifted and a redshifted side and matches the kinematics properties of that characterised in the [Si VI] λ1.96 μm coronal line maps by Müller-Sánchez et al. (2011). This structure is in appearance similar to that of NGC 7130 (Sect. 5.4). However, whereas in NGC 7130 the two sides of the dipole seem to be separated by the nucleus (which is obscured because the AGN is of Type 2), in NGC 3783 the blueshifted component has its centre at the position of the Seyfert 1.5 nucleus and the redshifted knot is found about $0 \overset{″}{.} 25$ $0{{\overset{\prime\prime}{.}}}25$ to the north and corresponds to a distinct peak in emission. The dipole is surrounded by [Fe VII] λ6087 emission cocoon that we detect out to about 1″. This envelope is affected by the scattered light of the central blueshifted source and the redshifted northern knot, which might contribute to the slight north-south gradient in velocity. The maps in Müller-Sánchez et al. (2011) do not display the extended coronal-emission cocoon that we detect here. This might be because of their choice of excluding regions with a flux density below 5% that of the peak.

The [Si II] λ1.96 μm kinematics were modelled by Müller-Sánchez et al. (2011) to be the superposition of a biconical outflow and a rotating disc. According to their model, the outflow is the dominant component. The fact that some of the coronal-line-emitting gas is rotating is attributed to either the outflow cone intersecting the disc or the disc being ionised by hard radiation filtering through a clumpy and leaky torus.

The dipole observed in [Fe VII] λ6087 is also seen in [O III] λ5007. It is surrounded by a large extension of ionised gas that we detect all the way out to more than 2″ (≈400 pc). Given that this extended emission has velocity that is close to zero, we interpret that it is tracing the ionised disc. The spiral-shaped region protruding out to 4″ to the north-east of the nucleus, with velocities v ≈ 40 km s⁻¹, corresponds to dust spirals hosted in the inner parts of the bar that can be seen in HST F547M images (Bentz 2009).

Within the innermost 2″ of NGC 3783 the σ of the [O III] λ5007 line is σ > 100 km s⁻¹ almost everywhere. Farther away from the centre, the velocity dispersion descends down to σ < 100 km s⁻¹. The relatively high velocity dispersion in the central region might be due to the superposition of a close-to-zero velocity disc and slightly blueshifted/redshifted gas expelled by the AGN. Indeed, emission lines with a separation of a few tens of km s⁻¹ cannot be resolved at the spectral resolution of MUSE, and would appear as an enhancement in σ. Another possibility is that the central engine is injecting energy into the disc medium, which would be indicative of AGN feedback effective at a scale of hundreds of parsecs.

The channel maps (Fig. 8) show that the centroid of the surface brightness is found at a similar position for all channels, except for the most blueshifted one. This is confirmed by examining Table 4, were we can see that, if we except the v = −1311 km s⁻¹ channel, all the centroids are at the same position within $0 \overset{″}{.} 005$ $0{{\overset{\prime\prime}{.}}}005$ in declination and $0 \overset{″}{.} 016$ $0{{\overset{\prime\prime}{.}}}016$ in right ascension, and are located slightly north of the centroid in white light (average $Δ δ = 0 \overset{″}{.} 009$ $\Delta\delta=0{{\overset{\prime\prime}{.}}}009$ or 2 pc, when excluding the outlier channel). There is a systematic trend for the most blueshifted channels to be to the east of the nucleus, whereas the redshifted ones tend to be to the west. The maximum east-west distance between centroids (excluding the outlier) is $0 \overset{″}{.} 018$ $0{{\overset{\prime\prime}{.}}}018$ , or less than 4 pc. These distances are well below the FWHM ( $0 \overset{″}{.} 06$ $0{{\overset{\prime\prime}{.}}}06$ ) of the data.

Another interesting feature of the channel maps is the blob of emission to the north of the nucleus seen in the redshifted channels. The contrast of Fig. 8 makes it the most obvious for the channels v = 335 km s⁻¹ and v = 518 km s⁻¹, but a careful examination of the images from which the figure was constructed shows that this bright region is detected for v = −30 km s⁻¹ and v = 152 km s⁻¹ too. This redshifted blob corresponds to the redshifted region in the [Fe VII] λ6087 and [O III] λ5007 velocity maps presented in Fig. 3. Thanks to the channel maps, it becomes obvious that this redshifted emission is distinct from that of the nucleus. The centroid of the blob is found $0 \overset{″}{.} 25$ $0{{\overset{\prime\prime}{.}}}25$ or 50 pc to the north of the nucleus. The fact that it is seen in both [Fe VII] λ6087 and [O III] λ5007 discards the possibility of a high-redshift interloper seen through the disc of NGC 3783.

The presence of the redshifted blob to the north of the nucleus seems to indicate a north-south alignment of the nuclear activity. Another hint of this orientation is the faint synchrotron extension observed to the south of the unresolved nuclear emission (Fig. 3). On the other hand, the subtle differences in the position of the centroids of the [Fe VII] λ6087 channel maps (Fig. 8) favour an east-west alignment. Given the projected distance to the nucleus, the northern redshifted blob was likely ejected from the nuclear zone longer ago than the nuclear gas. Some outflowing material has been found to be subjected to an apparent radial deceleration (Gabel et al. 2003; Scott et al. 2014). If this behaviour corresponds to a bulk deceleration of the material, it is possible that the same medium affecting the radial motions of the outflow is able to also produce a deflection, which could explain the change of orientation with radius. The same phenomenon was found by GRAVITY Collaboration (2021), who characterised a shift in the orientation of the outflow from PA ∼ −60° in the inner few parsecs to PA ∼ 10° at larger radii (see their Fig. 18). The consistency of the results indicates that spectroastrometry is a very powerful technique that in some cases can compete with the expensive interferometry from VLTI/GRAVITY.

The nuclear spectrum of NGC 3783 (Fig. 12 and Table 8) exhibits [Fe VII] λ6087 emission that can be fitted with two components. The most intense and narrowest one exhibits a velocity v ≈ −60 km s⁻¹, and the second component has v ≈ −320 km s⁻¹. The kinematics of the narrowest component are similar to those of the most blueshifted of the [O III] λ5007 components, perhaps hinting that they are emitted by the same region.

5.3. NGC 4593

The surface brightness maps in Fig. 4 show that the coronal emission is detected in a resolved circular region with a radius of roughly $0 \overset{″}{.} 5$ $0{{\overset{\prime\prime}{.}}}5$ (r ≈ 90 pc). The [O III] λ5007-emitting region also has circular symmetry, and is about 2″ (340 pc) in radius.

The [Fe VII] λ6087 velocity map in Fig. 4 displays negative velocities, which indicates that the emitters are outflowing and are probably located in the approaching side of a bicone. The velocity map is asymmetric, with the east side displaying larger velocities in module than the western one. The [O III] λ5007 line also shows an east-west velocity gradient, but the mean fitted velocity is Δv ≈ 150 km ⁻¹ smaller in module than that of [Fe VII] λ6087. Differences between the [Fe VII] λ6087 and the [O III] λ5007 emission are also seen in the velocity dispersion σ. Indeed, we find that [Fe VII] λ6087 has σ ≈ 250 km s⁻¹, whereas [O III] λ5007 has σ ≈ 150 km s⁻¹ or lower, depending on the region.

The maps in Fig. 4 are derived from single-Gaussian-component fits, and hence these differences might be related to the superposition along the line of sight of different components. For example, for [O III] λ5007, there is a rotating disc superposed to an outflow component (Mulumba et al. 2024) whose peak velocity (at about $0 \overset{″}{.} 5$ $0{{\overset{\prime\prime}{.}}}5$ to the east of the centre of the galaxy) is compatible with the values of v ≈ −250 km s⁻¹ from the [Fe VII] λ6087 map. The presence of the pervasive [O III] λ5007 outflow explains why velocities are found to be negative everywhere in the field, even when there is an underlying disc. The east-west velocity gradient in [O III] λ5007 is due to the combination of the imprint of the disc butterfly velocity pattern and the approaching side of the outflow bicone, which points towards the east or the south-east (Mulumba et al. 2024).

The channel maps in Fig. 9 do not show an obvious shift of the position of the peak surface brightness with velocity. Only through accurate determinations of the photocentre, it becomes possible to detect slight variations in the position (Table 5). We find evidence that the blueshifted channels are located to the south-east of the nucleus, whereas the only redshifted channel with good signal has the centroid located to the north-west. The alignment is consistent with orientation of the outflow found in Mulumba et al. (2024). The distance between the blueshifted and the redshifted photocentres is about $0 \overset{″}{.} 03 - 0 \overset{″}{.} 04$ $0{{\overset{\prime\prime}{.}}}03{-}0{{\overset{\prime\prime}{.}}}04$ (5 − 7 pc), which is a factor of at least three smaller than the FWHM.

The small projected distance between the photocentres of the redshifted and the blueshifted sides of the outflow and the lack of very sharp features in the [Fe VII] λ6087 kinematic map (such a clear dipole showing the approaching and receding side of an outflow; Fig. 4) hint at an almost end-on view of the bicone. Since the galaxy is nearly face-on, this implies that the bicone is almost perpendicular to the plane of the disc.

The nuclear [Fe VII] λ6087 spectrum of NGC 4593 (Fig. 13) has three components. The main peak corresponds to an approaching outflow with a velocity of v ≈ −160 km s⁻¹ (Table 9). This component has a [O III] λ5007 counterpart with a similar velocity of v ≈ −200 km s⁻¹. The [Fe VII] λ6087 nuclear spectrum displays an even more blueshifted component (v ≈ −620 km s⁻¹), which is fainter than the main one. It is possible that counterparts of the latter component in lines such as [O III] λ5007 are buried under the extended wings of the broadest components of the Balmer lines, which in the case of Hβ have a width in excess of σ = 1500 km s⁻¹. Finally, we tentatively discern a redshifted [Fe VII] λ6087 component whose velocity is compatible with that of the only redshifted channel with a detectable signal in Fig. 9. We do not detect any zero-velocity component (akin to the [O III] λ5007 component with v ≈ 20 km s⁻¹).

5.4. NGC 7130

The distribution of the [Fe VII] λ6087 emission in NGC 7130 is roughly circular and can be detected out to a radius of about $0 \overset{″}{.} 3$ $0{{\overset{\prime\prime}{.}}}3$ (100 pc; Fig. 5). In [O III] λ5007 the emission shows a blueshifted extension to the north-west, which coincides with the direction of the outflow (Comerón et al. 2021).

The [Fe VII] λ6087 velocity map shows a dipole, which is aligned with the north-south direction. The northern side is blueshifted and the southern side is redshifted. The morphology of the velocity map resembles that of the inner part of the [O III] λ5007 biconic outflow, although the velocity amplitude of the dipole is larger in [Fe VII] λ6087. This is because the single-Gaussian fit produces a weighted average between the actual velocity of the outflowing gas and that of the disc (for a detailed multi-component kinematic decomposition see Comerón et al. 2021). The [Fe VII] λ6087 dipole is oriented in the same direction as the 8.4 GHz radio emission. Therefore, the evidence indicates that the [Fe VII] λ6087 dipole corresponds to the innermost regions of the outflow.

The channel maps (Fig. 10) show that, other than for the zero-velocity channel, the centroid of the surface brightness distribution is not located at the centre of the galaxy. Table 6 indicates that the distance between the photocentres of the redshifted and the blueshifted lobes is of approximately $0 \overset{″}{.} 08$ $0{{\overset{\prime\prime}{.}}}08$ (≈23 pc). The clarity of the separation of the two lobes confirms the model presented in Fig. 14 from Comerón et al. (2021), where the axis of the outflow is shown to be at a small angle with respect to the plane of the disc. Because the disc is nearly face-on, this implies that the axis of the bicone is very inclined with respect to the line of sight.

The nuclear [Fe VII] λ6087 spectrum of NGC 7130 (Fig. 14) displays three distinct components. One is at rest with respect to the frame of reference of the galaxy. The two other components have either blueshifted or redshifted motions. The clarity and the symmetry of these kinematically displaced components makes them likely to trace the innermost parts of the bicone that are also seen through spectroastrometry. The three [Fe VII] λ6087 components have clear counterparts in [O III] λ5007 with similar kinematics (Table 10).

6. Discussion and conclusion

We have used high-quality AO-assisted MUSE integral-field data to study the properties of the [Fe VII] λ6087 coronal line in the circumnuclear medium of AGN hosts. This was achieved by studying four nearby Seyfert galaxies, where we could resolve features with sizes down to a few tens of parsecs.

Our sample is composed of three AGN hosts with Eddington ratios λ_Edd ≳ 0.03, plus one object for which we do not have an accurate SMBH mass determination. These ratios are quite large, as the median ration for Seyfert 1 and Seyfert 2 AGNs is λ_Edd = 1.1 × 10⁻³ and λ_Edd = 5.9 × 10⁻⁶, respectively (Ho 2008).

We have found that the [Fe VII] λ6087 emission is resolved in three out of four galaxies in the sample (NGC 3783, NGC 4593, and NGC 7130), with sizes up to 200 pc for the radius of the coronal-line-emitting region. In the case of the fourth galaxy, Mrk 1044, the central point source is so intense that the PSF wings do not allow us to easily discern whether there is any resolved [Fe VII] λ6087 emission. To our knowledge, extended [Fe VII] λ6087 emission in local galaxies had only been found in IC 5063 (Fonseca-Faria et al. 2023), NGC 1068 (Rodríguez-Ardila et al. 2006), NGC 1386 (Rodríguez-Ardila et al. 2006), NGC 3783 (Rodríguez-Ardila et al. 2006), NGC 7130 (Knapen et al. 2019; Comerón et al. 2021), and the Circinus Galaxy (Rodríguez-Ardila et al. 2006; Rodríguez-Ardila & Fonseca-Faria 2020). When extended [Fe VII] λ6087 emission is found in our sample, it is smooth and symmetric with respect to the nucleus, except for a redshifted clump ≈50 pc to the north of the nucleus in NGC 3783. The finding of resolved [Fe VII] λ6087 emission implies that coronal emission is not necessarily confined to the innermost few parsecs and extends out to regions dominated by the NLR.

The kinematic maps of the circumnuclear gas (Figs. 2, 3, 4, and 5) indicate that the velocity dispersion of [Fe VII] λ6087 is typically ∼1.5 times larger than that of [O III] λ5007. A similar phenomenon was already reported by Rodríguez-Ardila et al. (2006) with spectra from 1″-wide slits for a sample of six nearby Seyfert galaxies that included NGC 3783. However, it is likely that the low velocity dispersions found in [O III] λ5007 are a consequence of a bias caused by an underlying cold disc. Indeed, for several of our galaxies there is a significant star-forming and low-velocity dispersion disc. This is evident from the spiral patterns seen in Fig. 3 for NGC 3783 in [O III] λ5007. In the case of NGC 7130, the multi-component fits of NLR lines in Fig. 5 in Comerón et al. (2021) describe outflow components with velocity dispersions of σ ≈ 400 km s⁻¹, that is, similar to those in [Fe VII] λ6087. This indicates that single-component maps may be insufficient to discern whether the coronal-line-emission kinematics are compatible with those of the NLR.

In order to check whether the coronal line kinematics match that of the lower-ionisation lines, we have also considered nuclear spectra (Sect. 4.4), where the star-forming disc must have a smaller effect than at larger radii. In two of the Seyfert 1 galaxies of our sample, namely NGC 3783 and NGC 4593, the most intense nuclear [Fe VII] λ6087 components (≳60% of the emission) have velocities compatible (within 50 km s⁻¹) with outflows seen in other lines such as [O III] λ5007. This finding, combined with the fact that the AGN is of Type 1 and enables a direct view of the central engine, indicates that in these two objects the dominant [Fe VII] λ6087 component is not associated with in-plane gas and forms part of an outflow seen in both low and high ionisation lines. The Type 2 AGN host NGC 7130 also has blueshifted and redshifted [Fe VII] λ6087 nuclear components with kinematics similar to those in [O III] λ5007, which indicates an association with the NLR, and possibly the inner parts of the biconic outflow.

Using spectroastrometry, we have found that for the Type 2 AGN host NGC 7130 the coronal lines show blueshifted and redshifted components that are clearly separated by the central engine (separation of the order of $0 \overset{″}{.} 08$ $0{{\overset{\prime\prime}{.}}}08$ or 23 pc). The alignment shown by the dipole is the same as that of the outflows traced by [O III] λ5007 and synchrotron radio emission. Hence, the data indicate that at least part of the coronal emission of NGC 7130 occurs within the innermost parts (a few tens of parsecs) of a biconic outflow. Much milder indications of separated blueshifted and redshifted components are found for NGC 3783 and NGC 4593 (separations between $0 \overset{″}{.} 02$ $0{{\overset{\prime\prime}{.}}}02$ and $0 \overset{″}{.} 04$ $0{{\overset{\prime\prime}{.}}}04$ ). The fact that these galaxies are significantly closer than NGC 7130 (d ≈ 40 Mpc versus d ≈ 70 Mpc), indicates that the physical projected separation between the blueshifted and the redshifted lobes is at least four times larger in NGC 7130. This finding fits nicely with the unified model for AGNs (Antonucci 1993). Indeed, in the case of the Type 2 AGN NGC 7130, the axis of the bicone has a large angle with respect to the line of sight (see the toy model in Fig. 14 in Comerón et al. 2021), so it lies close to the plane of the sky, and the longitudinal structure is easier to resolve. On the other hand, NGC 3783 and NGC 4593 are Type 1 AGNs, where the axis of the bicone is foreshortened. If the [Fe VII] λ6087 emission in the Type 1 AGN host Mrk 1044 were indeed unresolved, the argument would be reinforced.

The study by Müller-Sánchez et al. (2011) also indicated that, for at least four out of the seven Seyfert galaxies in their sample, the coronal emission from the outflow is dominant over emission coming from other sources. Our sample and theirs have in common NGC 3783, a galaxy where they find traces of a coronal-emitting disc, whose emission is subdominant compared to that of the outflow. They hypothesised that the disc is ionised by either the cone intersecting the plane of the galaxy or by hard radiation leaking through the AGN torus. Discerning possible [Fe VII] λ6087-emitting discs in the other two galaxies with resolved emission (NGC 4593 and NGC 7130) would require dedicated modelling that is outside the scope of this paper. However, both for NGC 4593 and NGC 7130 the galaxies are nearly face-on and the amplitude of their projected rotation curves is of only a few tens of kilometres per second (see disc-outflow decompositions in Mulumba et al. 2024; Comerón et al. 2021, respectively), which is much smaller than the maximum projected mean [Fe VII] λ6087 velocity (≳100 km s⁻¹ as seen in Figs. 4 and 5), so a dominant outflow emission seems likely.

The analysis of the small sample presented here has shown that in at least some Seyfert galaxies, most of the coronal [Fe VII] λ6087 emission is rather compact (tens of parsecs in size). We find evidence that some of the [Fe VII] λ6087 emission is co-spatial and shares the kinematics with lower-ionisation lines ([O III]), and probably comes from the inner few tens of parsecs of the outflows. Our high-angular-resolution study is in line with previous works where at least part of the coronal emission is found to come from the NLR gas (e.g. Mazzalay et al. 2010; GRAVITY Collaboration 2021). This might hint at a common origin for the ionisation mechanism of the high- and low-excitation lines in the inner parts of outflows.

Further clues of a possible link between [Fe VII] λ6087 emission and the inner parts of the outflow could come from finding systematic alignments with radio jets, as seen at a much larger scale for the powerful radio-galaxies with extended kiloparsec-scale [Fe VII] λ6087 emission in Dabhade et al. (in prep.). Unfortunately, all of our galaxies have a very compact radio emission and only for NGC 7130, and perhaps NGC 3783, we have radio data with enough angular resolution to confirm the alignment. Additionally Fonseca-Faria et al. (2023) also found an alignment between the [Fe VII] λ6087 emission and the radio jet for the Seyfert 2 galaxy IC 5063. Better-sensitivity and higher-resolution observations are required to characterise the details of the radio jet (see e.g. Orienti & Prieto 2010). If the link between extended [Fe VII] λ6087 emission and radio jet were found to be ubiquitous, it would be evidence that the coronal emission is caused by shocks driven by the interaction between the jet and the gas in the AGN bicone.

Our study underlines the importance of combining AO modules with integral-field instrumentation. Thanks to spectroastrometry, it now becomes possible to study the morphology of the circumnuclear gas at scales below $0 \overset{″}{.} 2$ $0{{\overset{\prime\prime}{.}}}2$ (tens of parsecs or less at the distance of the galaxies in the sample), a feat that was until recently reserved to space- or interferometry-based studies. In particular, we have been able to obtain an orientation of the outflow of NGC 3783 at a few parsec scale that is compatible with that obtained with VLTI/GRAVITY.

Data availability

All the datacubes used in this paper can be downloaded from the ESO archive as Phase 3 products. They can be accessed through the following hyperlinks: Mrk 1044, NGC 3783, NGC 4593, and NGC 7130.

¹

http://leda.univ-lyon1.fr/

²

http://www.astro.gsu.edu/AGNmass/

³

https://pypi.org/project/ppxf/

⁴

https://abittner.gitlab.io/thegistpipeline/index.html

⁵

https://www-astro.physics.ox.ac.uk/~cappellari/software/#binning

⁶

https://www.star.herts.ac.uk/~sarzi/PaperV_nutshell/PaperV_nutshell.html

⁷

https://github.com/sczesla/PyAstronomy

⁸

https://physics.nist.gov/asd

Acknowledgments

We thank J. A. Acosta-Pulido for helping to identify the Fe IIλ6149 line. SC acknowledges funding from the State Research Agency (AEI-MICIU) of the Spanish Ministry of Science, Innovation, and Univerities under the grant ‘The relic galaxy NG 1277 as a key to understanding massive galaxies at cosmic noon’ with reference PID2023-149139NB-I00. This research has made use of the NASA/IPAC Extragalactic Database, which is funded by the National Aeronautics and Space Administration and operated by the California Institute of Technology. We acknowledge the usage of the HyperLeda database (http://leda.univ-lyon1.fr). This research made use of Photutils, an Astropy package for detection and photometry of astronomical sources (Bradley et al. 2024).

References

Antonucci, R. 1993, ARA&A, 31, 473 [Google Scholar]
Arévalo, P., Uttley, P., Lira, P., et al. 2009, MNRAS, 397, 2004 [CrossRef] [Google Scholar]
Armus, L., Mazzarella, J. M., Evans, A. S., et al. 2009, PASP, 121, 559 [Google Scholar]
Atwood, B., Baldwin, J. A., & Carswell, R. F. 1982, ApJ, 257, 559 [NASA ADS] [CrossRef] [Google Scholar]
Bacon, R., Accardo, M., Adjali, L., et al. 2010, in Ground-based and Airborne Instrumentation for Astronomy III, eds. I. S. McLean, S. K. Ramsay, & H. Takami, SPIE Conf. Ser., 7735, 773508 [Google Scholar]
Bacon, R., Conseil, S., Mary, D., et al. 2017, A&A, 608, A1 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Barr, P., Willis, A. J., & Wilson, R. 1983, MNRAS, 202, 453 [NASA ADS] [Google Scholar]
Barr, P., Clavel, J., Giommi, P., Mushotzky, R. F., & Madejski, G. 1987, Variability of Galactic and Extragalactic X-ray Sources, 43 [Google Scholar]
Beckers, J. M. 1982, Opt. Acta, 29, 361 [NASA ADS] [CrossRef] [Google Scholar]
Behar, E., Rasmussen, A. P., Blustin, A. J., et al. 2003, ApJ, 598, 232 [Google Scholar]
Bentz, M. 2009, The Black Hole Mass– Bulge Luminosity Relationship for the Nearest Reverberation-Mapped AGNs, HST Proposal ID 11661. Cycle 17 [Google Scholar]
Bentz, M. C., & Katz, S. 2015, PASP, 127, 67 [Google Scholar]
Bentz, M. C., Street, R., Onken, C. A., & Valluri, M. 2021, ApJ, 906, 50 [Google Scholar]
Bittner, A., Falcón-Barroso, J., Nedelchev, B., et al. 2019, A&A, 628, A117 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Bradley, L., Sipőcz, B., Robitaille, T., et al. 2024, https://doi.org/10.5281/zenodo.596036 [Google Scholar]
Burnell, S. J. B., & Culhane, J. L. 1979, MNRAS, 188, 1P [NASA ADS] [CrossRef] [Google Scholar]
Busko, I. C., & Steiner, J. E. 1988, MNRAS, 232, 525 [NASA ADS] [CrossRef] [Google Scholar]
Buta, R. J., Sheth, K., Athanassoula, E., et al. 2015, ApJS, 217, 32 [Google Scholar]
Cappellari, M. 2017, MNRAS, 466, 798 [Google Scholar]
Cappellari, M. 2023, MNRAS, 526, 3273 [NASA ADS] [CrossRef] [Google Scholar]
Cappellari, M., & Copin, Y. 2003, MNRAS, 342, 345 [Google Scholar]
Cappellari, M., & Emsellem, E. 2004, PASP, 116, 138 [Google Scholar]
Chapman, G. N. F., Geller, M. J., & Huchra, J. P. 1985, ApJ, 297, 151 [NASA ADS] [CrossRef] [Google Scholar]
Chelouche, D., & Netzer, H. 2005, ApJ, 625, 95 [NASA ADS] [CrossRef] [Google Scholar]
Ciddor, P. E. 1996, Appl. Opt., 35, 1566 [Google Scholar]
Clavel, J. 1983, MNRAS, 204, 189 [Google Scholar]
Clavel, J., Joly, M., Collin-Souffrin, S., Bergeron, J., & Penston, M. V. 1983, MNRAS, 202, 85 [Google Scholar]
Comerón, S., Knapen, J. H., Beckman, J. E., & Shlosman, I. 2008, A&A, 478, 403 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Comerón, S., Knapen, J. H., Beckman, J. E., et al. 2010, MNRAS, 402, 2462 [Google Scholar]
Comerón, S., Knapen, J. H., Ramos Almeida, C., & Watkins, A. E. 2021, A&A, 645, A130 [EDP Sciences] [Google Scholar]
Contini, M., Radovich, M., Rafanelli, P., & Richter, G. M. 2002, ApJ, 572, 124 [NASA ADS] [CrossRef] [Google Scholar]
Cordier, D., Pietrinferni, A., Cassisi, S., & Salaris, M. 2007, AJ, 133, 468 [NASA ADS] [CrossRef] [Google Scholar]
Crenshaw, D. M., Kraemer, S. B., Boggess, A., et al. 1999, ApJ, 516, 750 [NASA ADS] [CrossRef] [Google Scholar]
Crossley, J. H., Sjouwerman, L. O., Fomalont, E. B., & Radziwill, N. M. 2008, in Observatory Operations: Strategies, Processes, and Systems II, eds. R. J. Brissenden, & D. R. Silva, SPIE Conf. Ser., 7016, 70160O [Google Scholar]
Czesla, S., Schröter, S., Schneider, C. P., et al. 2019, Astrophysics Source Code Library [record ascl:1906.010] [Google Scholar]
Davies, R. L., Rich, J. A., Kewley, L. J., & Dopita, M. A. 2014, MNRAS, 439, 3835 [Google Scholar]
De Marco, B., Adhikari, T. P., Ponti, G., et al. 2020, A&A, 634, A65 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
De Robertis, M. M., & Osterbrock, D. E. 1984, ApJ, 286, 171 [NASA ADS] [CrossRef] [Google Scholar]
De Rosa, A., Piro, L., Fiore, F., et al. 2002, A&A, 387, 838 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
de Vaucouleurs, G., de Vaucouleurs, A., Corwin, Herold G. J., et al. 1991, Third Reference Catalogue of Bright Galaxies (New York, NY, USA: Springer) [Google Scholar]
Denney, K. D., Bentz, M. C., Peterson, B. M., et al. 2006, ApJ, 653, 152 [NASA ADS] [CrossRef] [Google Scholar]
Dey, A., Schlegel, D. J., Lang, D., et al. 2019, AJ, 157, 168 [Google Scholar]
Dietrich, M., Kollatschny, W., Alloin, D., et al. 1994, A&A, 284, 33 [NASA ADS] [Google Scholar]
Dopita, M. A., Pereira, M., Kewley, L. J., & Capaccioli, M. 2002, ApJS, 143, 47 [NASA ADS] [CrossRef] [Google Scholar]
Du, P., Hu, C., Lu, K.-X., et al. 2015, ApJ, 806, 22 [NASA ADS] [CrossRef] [Google Scholar]
Eddington, A. S. 1926, The Internal Constitution of the Stars (Cambridge: Cambridge University Press) [Google Scholar]
Erkens, U., Appenzeller, I., & Wagner, S. 1997, A&A, 323, 707 [NASA ADS] [Google Scholar]
Esquej, P., Alonso-Herrero, A., González-Martín, O., et al. 2014, ApJ, 780, 86 [Google Scholar]
Evans, I. N. 1989, ApJ, 338, 128 [NASA ADS] [CrossRef] [Google Scholar]
Falcón-Barroso, J., Bacon, R., Bureau, M., et al. 2006, MNRAS, 369, 529 [Google Scholar]
Fonseca-Faria, M. A., Rodríguez-Ardila, A., Contini, M., Dahmer-Hahn, L. G., & Morganti, R. 2023, MNRAS, 524, 143 [CrossRef] [Google Scholar]
Gabel, J. R., Crenshaw, D. M., Kraemer, S. B., et al. 2003, ApJ, 595, 120 [NASA ADS] [CrossRef] [Google Scholar]
Gabel, J. R., Kraemer, S. B., Crenshaw, D. M., et al. 2005, ApJ, 631, 741 [NASA ADS] [CrossRef] [Google Scholar]
Gelbord, J. M., Mullaney, J. R., & Ward, M. J. 2009, MNRAS, 397, 172 [NASA ADS] [CrossRef] [Google Scholar]
George, I. M., Turner, T. J., & Netzer, H. 1995, ApJ, 438, L67 [NASA ADS] [CrossRef] [Google Scholar]
Ghosh, K. K., & Soundarajaperumal, S. 1993, A&A, 273, 397 [NASA ADS] [Google Scholar]
Giannuzzo, E. M., & Stirpe, G. M. 1996, A&A, 314, 419 [NASA ADS] [Google Scholar]
Glass, I. S. 1992, MNRAS, 256, 23P [NASA ADS] [CrossRef] [Google Scholar]
Glidden, A., Rose, M., Elvis, M., & McDowell, J. 2016, ApJ, 824, 34 [NASA ADS] [CrossRef] [Google Scholar]
González Delgado, R. M., Pérez, E., Tadhunter, C., Vilchez, J. M., & José Miguel Rodríguez-Espinosa. 1997, ApJS, 108, 155 [NASA ADS] [CrossRef] [Google Scholar]
González Delgado, R. M., Heckman, T., Leitherer, C., et al. 1998, ApJ, 505, 174 [CrossRef] [Google Scholar]
Goodrich, R. W. 1989, ApJ, 342, 224 [Google Scholar]
GRAVITY Collaboration (Amorim, A., et al.) 2021, A&A, 648, A117 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Green, A. R., McHardy, I. M., & Lehto, H. J. 1993, MNRAS, 265, 664 [NASA ADS] [Google Scholar]
Hinshaw, G., Weiland, J. L., Hill, R. S., et al. 2009, ApJS, 180, 225 [Google Scholar]
Ho, L. C. 2008, ARA&A, 46, 475 [Google Scholar]
Ho, L. C., & Kim, M. 2014, ApJ, 789, 17 [NASA ADS] [CrossRef] [Google Scholar]
Husemann, B., Wisotzki, L., Sánchez, S. F., & Jahnke, K. 2013, A&A, 549, A43 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Husemann, B., Singha, M., Scharwächter, J., et al. 2022, A&A, 659, A124 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Kaastra, J. S., Mehdipour, M., Behar, E., et al. 2018, A&A, 619, A112 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Kaspi, S., Brandt, W. N., Netzer, H., et al. 2001, ApJ, 554, 216 [NASA ADS] [CrossRef] [Google Scholar]
Kilborn, V. A., Forbes, D. A., Koribalski, B. S., Brough, S., & Kern, K. 2006, MNRAS, 371, 739 [NASA ADS] [CrossRef] [Google Scholar]
Knapen, J. H., Comerón, S., & Seidel, M. K. 2019, A&A, 621, L5 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Kollatschny, W., & Fricke, K. J. 1985, A&A, 143, 393 [NASA ADS] [Google Scholar]
Kormendy, J., Cornell, M. E., Block, D. L., Knapen, J. H., & Allard, E. L. 2006, ApJ, 642, 765 [NASA ADS] [CrossRef] [Google Scholar]
Kraemer, S. B., Crenshaw, D. M., Filippenko, A. V., & Peterson, B. M. 1998, ApJ, 499, 719 [NASA ADS] [CrossRef] [Google Scholar]
Kraemer, S. B., Crenshaw, D. M., & Gabel, J. R. 2001, ApJ, 557, 30 [NASA ADS] [CrossRef] [Google Scholar]
Kramida, A., & Ralchenko, Y. 1999, NIST Atomic Spectra Database, NIST Standard Reference Database 78 (National Institute of Standards and Technology) [Google Scholar]
Krongold, Y., Longinotti, A. L., Santos-Lleó, M., et al. 2021, ApJ, 917, 39 [NASA ADS] [CrossRef] [Google Scholar]
Kroupa, P. 2001, MNRAS, 322, 231 [NASA ADS] [CrossRef] [Google Scholar]
Lacy, M., Baum, S. A., Chandler, C. J., et al. 2020, PASP, 132, 035001 [Google Scholar]
Levenson, N. A., Weaver, K. A., Heckman, T. M., Awaki, H., & Terashima, Y. 2005, ApJ, 618, 167 [NASA ADS] [CrossRef] [Google Scholar]
Lira, P., Arévalo, P., Uttley, P., McHardy, I., & Breedt, E. 2011, MNRAS, 415, 1290 [NASA ADS] [CrossRef] [Google Scholar]
Lu, Y., & Wang, T. 2000, ApJ, 537, L103 [NASA ADS] [CrossRef] [Google Scholar]
Makarov, D., Prugniel, P., Terekhova, N., Courtois, H., & Vauglin, I. 2014, A&A, 570, A13 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Mallick, L., Alston, W. N., Parker, M. L., et al. 2018, MNRAS, 479, 615 [NASA ADS] [Google Scholar]
Mao, J., Mehdipour, M., Kaastra, J. S., et al. 2019, A&A, 621, A99 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Maran, S. P., Crenshaw, D. M., Mushotzky, R. F., et al. 1996, ApJ, 465, 733 [NASA ADS] [CrossRef] [Google Scholar]
Markowitz, A. 2005, ApJ, 635, 180 [NASA ADS] [CrossRef] [Google Scholar]
Markowitz, A. G., & Reeves, J. N. 2009, ApJ, 705, 496 [NASA ADS] [CrossRef] [Google Scholar]
May, D., Rodríguez-Ardila, A., Prieto, M. A., et al. 2018, MNRAS, 481, L105 [NASA ADS] [CrossRef] [Google Scholar]
Mazzalay, X., Rodríguez-Ardila, A., & Komossa, S. 2010, MNRAS, 405, 1315 [NASA ADS] [Google Scholar]
Mazzalay, X., Rodríguez-Ardila, A., Komossa, S., & McGregor, P. J. 2013, MNRAS, 430, 2411 [NASA ADS] [CrossRef] [Google Scholar]
Menzies, J. W., & Feast, M. W. 1983, MNRAS, 203, 1P [Google Scholar]
Moorwood, A. F. M., Lutz, D., Oliva, E., et al. 1996, A&A, 315, L109 [NASA ADS] [Google Scholar]
Mulchaey, J. S., Regan, M. W., & Kundu, A. 1997, ApJS, 110, 299 [NASA ADS] [CrossRef] [Google Scholar]
Mullaney, J. R., Ward, M. J., Done, C., Ferland, G. J., & Schurch, N. 2009, MNRAS, 394, L16 [NASA ADS] [CrossRef] [Google Scholar]
Müller Sánchez, F., Davies, R. I., Eisenhauer, F., et al. 2006, A&A, 454, 481 [CrossRef] [EDP Sciences] [Google Scholar]
Müller-Sánchez, F., Prieto, M. A., Hicks, E. K. S., et al. 2011, ApJ, 739, 69 [Google Scholar]
Mulumba, D., Knapen, J. H., Comerón, S., et al. 2024, A&A, 690, A277 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Muñoz Marín, V. M., González Delgado, R. M., Schmitt, H. R., et al. 2007, AJ, 134, 648 [CrossRef] [Google Scholar]
Murayama, T., & Taniguchi, Y. 1998, ApJ, 503, L115 [Google Scholar]
Negus, J., Comerford, J. M., Müller Sánchez, F., et al. 2021, ApJ, 920, 62 [NASA ADS] [CrossRef] [Google Scholar]
Negus, J., Comerford, J. M., Sánchez, F. M., et al. 2023, ApJ, 945, 127 [NASA ADS] [CrossRef] [Google Scholar]
Netzer, H., Kaspi, S., Behar, E., et al. 2003, ApJ, 599, 933 [NASA ADS] [CrossRef] [Google Scholar]
Norris, R. P., Allen, D. A., Sramek, R. A., Kesteven, M. J., & Troup, E. R. 1990, ApJ, 359, 291 [NASA ADS] [CrossRef] [Google Scholar]
Ohta, K., Aoki, K., Kawaguchi, T., & Kiuchi, G. 2007, ApJS, 169, 1 [NASA ADS] [CrossRef] [Google Scholar]
Oke, J. B., & Sargent, W. L. W. 1968, ApJ, 151, 807 [NASA ADS] [CrossRef] [Google Scholar]
Oliva, E., Salvati, M., Moorwood, A. F. M., & Marconi, A. 1994, A&A, 288, 457 [NASA ADS] [Google Scholar]
Orienti, M., & Prieto, M. A. 2010, MNRAS, 401, 2599 [NASA ADS] [CrossRef] [Google Scholar]
Osmer, P. S., Smith, M. G., & Weedman, D. W. 1974, ApJ, 189, 187 [NASA ADS] [CrossRef] [Google Scholar]
Pal, M., & Naik, S. 2018, MNRAS, 474, 5351 [NASA ADS] [CrossRef] [Google Scholar]
Paltani, S., & Courvoisier, T. J. L. 1994, A&A, 291, 74 [NASA ADS] [Google Scholar]
Pandey, A., Martínez-Aldama, M. L., Czerny, B., Panda, S., & Zajaček, M. 2024, ArXiv e-prints [arXiv:2401.18052] [Google Scholar]
Park, D., Barth, A. J., Ho, L. C., & Laor, A. 2022, ApJS, 258, 38 [NASA ADS] [CrossRef] [Google Scholar]
Phillips, M. M., Charles, P. A., & Baldwin, J. A. 1983, ApJ, 266, 485 [Google Scholar]
Pier, E. A., & Voit, G. M. 1995, ApJ, 450, 628 [NASA ADS] [CrossRef] [Google Scholar]
Pietrinferni, A., Cassisi, S., Salaris, M., & Castelli, F. 2004, ApJ, 612, 168 [Google Scholar]
Pietrinferni, A., Cassisi, S., Salaris, M., & Castelli, F. 2006, ApJ, 642, 797 [Google Scholar]
Pietrinferni, A., Cassisi, S., Salaris, M., Percival, S., & Ferguson, J. W. 2009, ApJ, 697, 275 [Google Scholar]
Pietrinferni, A., Cassisi, S., Salaris, M., & Hidalgo, S. 2013, A&A, 558, A46 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Prieto, M. A., Marco, O., & Gallimore, J. 2005, MNRAS, 364, L28 [NASA ADS] [Google Scholar]
Prieto, M. A., Reunanen, J., Tristram, K. R. W., et al. 2010, MNRAS, 402, 724 [Google Scholar]
Prieto, A., Rodríguez-Ardila, A., Panda, S., & Marinello, M. 2022, MNRAS, 510, 1010 [Google Scholar]
Rafanelli, P., & Schulz, H. 1983, A&A, 117, 109 [NASA ADS] [Google Scholar]
Ramírez, J. M., Bautista, M., & Kallman, T. 2005, ApJ, 627, 166 [CrossRef] [Google Scholar]
Reis, R. C., Fabian, A. C., Reynolds, C. S., et al. 2012, ApJ, 745, 93 [NASA ADS] [CrossRef] [Google Scholar]
Reunanen, J., Kotilainen, J. K., & Prieto, M. A. 2003, MNRAS, 343, 192 [NASA ADS] [CrossRef] [Google Scholar]
Riffel, R. A., Storchi-Bergmann, T., Winge, C., et al. 2008, MNRAS, 385, 1129 [CrossRef] [Google Scholar]
Rodríguez-Ardila, A., & Fonseca-Faria, M. A. 2020, ApJ, 895, L9 [CrossRef] [Google Scholar]
Rodríguez-Ardila, A., Prieto, M. A., Viegas, S., & Gruenwald, R. 2006, ApJ, 653, 1098 [Google Scholar]
Rose, M., Tadhunter, C. N., Holt, J., Ramos Almeida, C., & Littlefair, S. P. 2011, MNRAS, 414, 3360 [NASA ADS] [CrossRef] [Google Scholar]
Rose, M., Elvis, M., & Tadhunter, C. N. 2015, MNRAS, 448, 2900 [NASA ADS] [CrossRef] [Google Scholar]
Ruschel-Dutra, D., Storchi-Bergmann, T., Schnorr-Müller, A., et al. 2021, MNRAS, 507, 74 [NASA ADS] [CrossRef] [Google Scholar]
Sarzi, M., Falcón-Barroso, J., Davies, R. L., et al. 2006, MNRAS, 366, 1151 [Google Scholar]
Scott, A. E., Brandt, W. N., Behar, E., et al. 2014, ApJ, 797, 105 [CrossRef] [Google Scholar]
Shields, J. C., & Filippenko, A. V. 1990, AJ, 100, 1034 [NASA ADS] [CrossRef] [Google Scholar]
Sriram, K., Agrawal, V. K., & Rao, A. R. 2009, ApJ, 700, 1042 [NASA ADS] [CrossRef] [Google Scholar]
Stirpe, G. M., de Bruyn, A. G., & van Groningen, E. 1988, A&A, 200, 9 [NASA ADS] [Google Scholar]
Storchi-Bergmann, T., McGregor, P. J., Riffel, R. A., et al. 2009, MNRAS, 394, 1148 [NASA ADS] [CrossRef] [Google Scholar]
Storey, P. J., & Zeippen, C. J. 2000, MNRAS, 312, 813 [NASA ADS] [CrossRef] [Google Scholar]
Stuik, R., Bacon, R., Conzelmann, R., et al. 2006, New Astron. Rev., 49, 618 [CrossRef] [Google Scholar]
Thean, A., Pedlar, A., Kukula, M. J., Baum, S. A., & O’Dea, C. P. 2000, MNRAS, 314, 573 [Google Scholar]
Thomson, J. J. 1906, Conduction of Electricity Through Gases (Cambridge, UK: Cambridge University Press) [Google Scholar]
Thuan, T. X. 1984, ApJ, 281, 126 [NASA ADS] [CrossRef] [Google Scholar]
Tody, D. 1986, in Instrumentation in Astronomy VI, ed. D. L. Crawford, SPIE Conf. Ser., 627, 733 [NASA ADS] [CrossRef] [Google Scholar]
Tombesi, F., de Marco, B., Iwasawa, K., et al. 2007, A&A, 467, 1057 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Tully, R. B. 2015, AJ, 149, 171 [NASA ADS] [CrossRef] [Google Scholar]
Ursini, F., Petrucci, P. O., Matt, G., et al. 2016, MNRAS, 463, 382 [NASA ADS] [CrossRef] [Google Scholar]
Vasudevan, R. V., & Fabian, A. C. 2009, MNRAS, 392, 1124 [CrossRef] [Google Scholar]
Vazdekis, A., Coelho, P., Cassisi, S., et al. 2015, MNRAS, 449, 1177 [Google Scholar]
Véron-Cetty, M. P., & Véron, P. 2010, A&A, 518, A10 [Google Scholar]
Véron-Cetty, M. P., Joly, M., & Véron, P. 2004, A&A, 417, 515 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Walter, R., & Courvoisier, T. J. L. 1992, A&A, 266, 57 [NASA ADS] [Google Scholar]
Ward, M., & Morris, S. 1984, MNRAS, 207, 867 [Google Scholar]
Whelan, E., & Garcia, P. 2008, in Jets from Young Stars II, eds. F. Bacciotti, L. Testi, & E. Whelan, 742, 123 [NASA ADS] [CrossRef] [Google Scholar]
Winge, C., Pastoriza, M. G., & Storchi-Bergmann, T. 1990, Rev. Mex. Astron. Astrofis., 21, 177 [Google Scholar]
Winge, C., Pastoriza, M. G., Storchi-Bergmann, T., & Lipari, S. 1992, ApJ, 393, 98 [NASA ADS] [CrossRef] [Google Scholar]
Winkel, N., Husemann, B., Davis, T. A., et al. 2022, A&A, 663, A104 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Winkel, N., Husemann, B., Singha, M., et al. 2023, A&A, 670, A3 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
Winkler, H. 1992, MNRAS, 257, 677 [NASA ADS] [Google Scholar]
Xu, Y., Pinto, C., Rogantini, D., et al. 2023, MNRAS, 523, 2158 [NASA ADS] [CrossRef] [Google Scholar]
Zhao, Y., Lu, N., Xu, C. K., et al. 2016, ApJ, 820, 118 [Google Scholar]

All Tables