Outer Ionized Gas in Galaxy Group: Exchance Through Tidal Interaction or Accretion from Common Reservoirs?

Abstract

To clarify the problem of outer cold gas accretion onto disk galaxies, we performed the panoramic spectroscopy of six compact galaxy groups to search for intergalactic gas flows. The groups selected are partly known to possess HI data obtained in the 21 cm line, and most of them contain a member galaxy revealing decoupled kinematics of gas and stars and thus having recently experienced a gas accretion event. Fabry-Perot scanning interferometry performed at the Russian 6 m telescope has provided us with the group maps at H$\alpha$ emission-line intensity and with ionized-gas velocity maps. We detected several intergalactic ionized-gas flows and some tidal outer ionized-gas structures; but none of them can be a source of gas accretion onto neighboring galaxies with decoupled gas–star kinematics. Only in a single case, that of NGC 7465, we can relate the inner inclined gaseous disk with the outer gas inflow; but the origin of this gas stream remains unknown—it does not originate from the neighboring NGC 7463 or NGC 7464.

1. Introduction

The main driver of galaxy evolution was considered differently during the epoch of developing extragalactic astronomy research studies. Some 25–30 years ago, multiple mergers were thought to be decisive in shaping galaxy structure and star formation history. But now the point of view has finally established that the evolution of a disk galaxy is mainly driven by persistent accretion of outer cold material [1,2]. Initially this idea came from the understanding that neither the chemical evolution of, in particular, our galaxy, the Milky Way [3,4], nor the existence of the main sequence of star-forming galaxies, implying a stellar mass acquisition time of the order of the current Hubble time and gas depletion times much shorter [5,6,7], can be explained in any way besides the hypothesis that all the evolution is governed by a steady inflow of outer metal-poor cold gas. Later, a great deal of both empirical [8] and theoretical [9] evidence has been accumulated, supporting the idea that cold gas accretion drives the growth of galactic disks during all their evolution.

However, the dominant source of gas accretion is not yet clear. We can mention the following possibilities: (i) interaction with neighboring galaxies, including tidal gas exchange, satellite acquisition (multiple minor mergers), and even major mergers under specific orbital and spin configurations [10]; (ii) fueling by cosmological gas–dark matter filaments which are a part of the large-scale structure of the Universe—the so-called ‘cold accretion’ [9,11]; and (iii) ‘hot accretion’, provided by extended gas reservoirs cooling within galaxy dark matter halos [12,13,14]. Perhaps one of these may serve as the main scenario over the whole Universe evolution, or all three of them are becoming, one by one, especially important under various conditions and at various epochs. All these possibilities are currently discussed by the community, and experts in galaxy evolution try to establish observational arguments in favor of any of them by analyzing the total statistics of star–gas misalignment acquired with integral-field (IFU) spectroscopy surveys e.g., [15,16,17,18]. The arguments in favor of one or the other gas accretion source are the inner gas kinematics in relation to the kinematics of neighboring galaxies, the metallicity of the accreted gas (the filaments of the large-scale structure must supply the gas with primordial chemical composition), and the presence or absence of old stars connected by their kinematics to the accreted gas. The connection of star–gas misalignment to dynamical disturbance, visible either inside the galaxy or in its neighborhood, can be an argument in favor of the role of galaxy mergers in the outer gas supply [19,20].

The large IFU spectroscopic surveys undertaken and presented to the community in the last decade have provided a great deal of observational data on the gas–star kinematic misalignment phenomenon, which is unambiguous evidence of recent gas accretion. Indeed, if the gas presently observed rotates around the axis inclined to the stellar disk spin, this means that the present gas does not relate to the initial gas content fueling stellar disk formation. Hence, this gas has been accreted recently. In the meantime, integral-field spectroscopic surveys such as ATLAS-3D [21] or SAMI [22] demonstrate that half of early-type disk galaxies in the field reveal gas–star kinematic misalignment. In star-forming galaxies, the frequency of gas–star misalignment is lower [23], perhaps because their own gas diminishes the accreted gas’ visibility. These empirical results provide evidence in favor of the common occurrence of events of gas accretion onto disk galaxies beyond rich clusters.

However, direct observations in the course of gas accretion events are still deficient. While gas inflow along cosmological filaments or cold gas cloud accretion from the halos are rare and ambiguous when observed, there is, in the meantime, a good and rather easy approach to searching for intergalactic gas flows in compact galaxy groups, including interacting pairs and triplets. These flows are seen in the 21 cm radio line emitted by neutral hydrogen. Moreover, we can sometimes see these flows in the optical emission lines which are excited by shocks due to tidal gravitational effects [24]; by the UV radiation of young stars, whose formation may be provoked in situ by the same shocks [25]; and also by the ionization cone of active galactic nuclei triggered through galaxy interactions [26,27]. As for neutral-hydrogen flows, the most famous and well-studied case of such intergalactic cold gas flows is known in the group of M 81 [28,29]; but in fact, there exist at least a dozen nearby galaxy groups with a spectacular mapping of intergalactic gas flows. Currently, a deep survey of intergalactic neutral hydrogen is underway with a large Chinese radiotelescope, FAST, whose first beautiful results have been already published [30,31,32]. As for the emission-line gas flows in the interacting galaxy systems, there is a nice collection of Arp galaxies [33] studied in star formation tracers, H$\alpha$ emission lines, or UV radiation by Beverly Smith and coauthors [34,35,36].

We have also recently presented a small sample of three interacting galaxy systems where the gas exchange is traced by optical emission lines [37]. Although the intergalactic gas in galaxy groups and pairs is mostly ionized by young massive stars producing H$\alpha$ emission as the strongest emission line in the optical spectra [38,39], the gas excitation mechanism is not important for our study. With the scanning Fabry–Perot interferometer, we measure the line-of-sight (LOS) velocities of gas filaments, and by comparing them to the LOS velocities of the galaxies, we attempt to establish the origin of these gas flows. Along with increasing numbers of observational data on the velocity distributions within these gaseous structures, some interesting questions have arisen. We cannot select a preferable direction of gas flows: from small satellites to its large host galaxy or from a central gas-rich galaxy to the satellites? Bright examples of both cases are found. For example, we have presented NGC 3921 destroying nearby satellites and UGC 1020 connected with a gaseous bridge to its small irregular satellite, still demonstrating a large amount of gas [37]. In the pair NGC 5719/NGC 5713, a large gas-rich galaxy gives its gas to the nearby early-type companion, with both having comparable masses [40]. Sometimes, it seems that several galaxies embedded in a common, large HI envelope receive their gas from this extended gaseous reservoir IC 718/IC 719 [41].

In the present work, we continue to study outer gaseous structures in interacting galaxy systems that may be treated as manifestations of gas exchange between galaxies of different masses and morphological types. Our choice of targets is biased toward galaxies whose inner parts demonstrate the gas–star misaligned rotation. We apply a 3D (panoramic) spectroscopy method to map emission-line intensities and line-of-sight velocities, which allowed us to inspect warm ionized-gas distributions and guess gas flow directions over large fields of view inhabited by interacting galaxies. The preference of optical panoramic spectroscopy over 21 cm neutral-hydrogen mapping is due to its higher spatial resolution. By leveraging a broad collection of kinematic, photometric, and spectral information, we hope to identify the sources of outer gas accretion in the galaxies selected. Since we are interested in the origin of these gas flows, for some galaxies, we also obtain long-slit spectra to measure their gas oxygen abundance.

2. The Sample

For our study, we selected six interacting galaxy systems at different stages of interaction, with the galaxy members having different stellar masses, but disk-like morphologies. The sample is small but diverse, presenting a range of possible interaction phenomena. Of course, we plan to extend our investigation further to a larger number of galaxy groups. The full-color images of our targets with their closest neighbors are shown in Figure 1, and their global characteristics retrieved from extragalactic databases are given in Table 1.

Table 1. The galaxies studied and their neighbors.

Galaxy	Type (NED a)	${\mathit{M}}_{\mathit{K}}$ (LEDA b)	${\mathit{M}}_{\mathit{B}}$ (LEDA b)	${\mathit{V}}_{\mathit{r}}$, $\mathbf{km}{\mathbf{s}}^{-1}$ (NED)	$\mathit{v}lg{\mathit{M}}_{*}$, ${\mathbf{M}}_{\odot }$	Separation, kpc (NED)
HCG 16, D = 55 Mpc c
NGC 835	Sab	−24.7	−20.77	3847	10.81 d	0
NGC 833	SABa	−24.2	−20.40	3854	10.76 d	13.3
NGC 838	S0/a	−23.9	−20.30	3849	10.41 d	49.2
Holmberg 802, D = 26.5 Mpc d
NGC 7463	SABb	−21.8	−19.43	2345	9.50	0
NGC 7464	E	−19.5	−18.22	1864	8.85	5
NGC 7465	S0	−22.8	−19.39	1979	9.77	19
KPG 565, D = 114.3 Mpc e
PGC 68555	Sbc	−23.4	−20.96	8369	9.944 e	0
PGC 68554	E	−23.3	−19.88	8047	10.33 g	29
UGC 9796, D = 76 Mpc a
UGC 9796	SBa b	−22.5	−19.11	5390	10.09 h	0
PGC 54478	Scd b	−21.55	−19.36	5471	9.51 h	31.5
NGC 6361, D = 60 Mpc d
NGC 6361	Sb	−24.8	−21.00	3797	10.91	0
PGC 60040	S0	−21.25	−18.45	3885	9.67	26.5
PGC 60446, D = 118 Mpc a
PGC 60446	E	−23.0	−19.63	8730	10.2 f	0
LEDA 2596531	S?	−20.6:: a	−18.53	8737	9.4	27

^a NASA/IPAC Extragalactic Database, http://ned.ipac.caltech.edu. ^b Lyon-Meudon Extragalactic Database, http://leda.univ-lyon1.fr. ^c The Extragalactic Distance Database, http://edd.ifa.hawaii.edu. ^d Leroy et al. [42]. ^e Yu et al. [43]. ^f Fraser-McKelvie et al. [44]. ^g Catinella et al. [45]. ^h Moffett et al. [46].

Table 1. The galaxies studied and their neighbors.

Galaxy	Type (NED a)	${\mathit{M}}_{\mathit{K}}$ (LEDA b)	${\mathit{M}}_{\mathit{B}}$ (LEDA b)	${\mathit{V}}_{\mathit{r}}$, $\mathbf{km}{\mathbf{s}}^{-1}$ (NED)	$\mathit{v}lg{\mathit{M}}_{*}$, ${\mathbf{M}}_{\odot }$	Separation, kpc (NED)
HCG 16, D = 55 Mpc c
NGC 835	Sab	−24.7	−20.77	3847	10.81 d	0
NGC 833	SABa	−24.2	−20.40	3854	10.76 d	13.3
NGC 838	S0/a	−23.9	−20.30	3849	10.41 d	49.2
Holmberg 802, D = 26.5 Mpc d
NGC 7463	SABb	−21.8	−19.43	2345	9.50	0
NGC 7464	E	−19.5	−18.22	1864	8.85	5
NGC 7465	S0	−22.8	−19.39	1979	9.77	19
KPG 565, D = 114.3 Mpc e
PGC 68555	Sbc	−23.4	−20.96	8369	9.944 e	0
PGC 68554	E	−23.3	−19.88	8047	10.33 g	29
UGC 9796, D = 76 Mpc a
UGC 9796	SBa b	−22.5	−19.11	5390	10.09 h	0
PGC 54478	Scd b	−21.55	−19.36	5471	9.51 h	31.5
NGC 6361, D = 60 Mpc d
NGC 6361	Sb	−24.8	−21.00	3797	10.91	0
PGC 60040	S0	−21.25	−18.45	3885	9.67	26.5
PGC 60446, D = 118 Mpc a
PGC 60446	E	−23.0	−19.63	8730	10.2 f	0
LEDA 2596531	S?	−20.6:: a	−18.53	8737	9.4	27

^a NASA/IPAC Extragalactic Database, http://ned.ipac.caltech.edu. ^b Lyon-Meudon Extragalactic Database, http://leda.univ-lyon1.fr. ^c The Extragalactic Distance Database, http://edd.ifa.hawaii.edu. ^d Leroy et al. [42]. ^e Yu et al. [43]. ^f Fraser-McKelvie et al. [44]. ^g Catinella et al. [45]. ^h Moffett et al. [46].

Figure 1. The images in combined colors for the galaxies studied in this work, which were derived from the resource Legacy Survey [47].

Figure 1. The images in combined colors for the galaxies studied in this work, which were derived from the resource Legacy Survey [47].

In our study we consider galaxies for which there exist data of integral-field spectroscopy revealing some signs of kinematic decoupling between the inner gaseous disks and stellar disks. The decoupled gas–star kinematics give evidence for recent outer gas accretion and stimulate searches for outer gas accretion sources. Beom et al. [48] mention PGC 60040, PGC 60446, and PGC 68554 as galaxies with gas–star counterrotation based on IFU spectroscopy within the MaNGA survey [49]. NGC 7465 was found to have central gas–star kinematic decoupling in the ATLAS-3D survey [50]. UGC 9796 is known as a classical polar-ring galaxy [51], and its central gas–star kinematic decoupling is also seen in the MaNGA survey data. NGC 6361 and NGC 835 have been observed in the framework of the CALIFA IFU survey [52].

Some of these compact groups have also been observed in the 21 cm line. There exist detailed interferometric HI mapping data for HCG 16 [53,54], Holmberg 802 [55,56], and UGC 9796+PGC 54478 [57]. The integrated HI-flux for NGC 6361 was measured at the Green Bank Telescope and was retrieved by us from ‘The Extragalactic Distance Database’, http://edd.ifa.hawaii.edu. The compact pair KPG 565 (PGC 68554 and PGC 68555) was observed with the Arecibo radiotelescope within the GASS survey, and only the upper limit to its HI mass was obtained: $logM\left(HI\right)<8.68$ [45].

3. Observations

We performed panoramic spectroscopy of our interacting galaxy systems by using the scanning Fabry–Perot interferometer (FPI) operating as a part of the SCORPIO-2 multimode focal reducer [58] at the prime focus of the 6 m telescope BTA of the Special Astrophysical Observatory of the Russian Academy of Sciences (SAO RAS).

Scanning FPIs are intended for the 3D–spectroscopic investigation of the ionized-gas kinematics over the full field of view, whereas their spectral range usually contains only a single emission line cut by a narrow filter with an FWHM of about 15–30 Å. The targets were observed at the SAO RAS 6m telescope with the SCORPIO-2 in FPI mode by using the interferometer of the 186th order providing the H$\alpha$ spectral resolution of $FHWM=1.7$ Å (∼78 $\mathrm{km}{\mathrm{s}}^{-1}$) in the full spectral range (interfringe) of $\Delta \lambda =35$ Å. During the scanning process, we consequently obtained a few dozen interferograms probing different gaps between the FPI plates to trace homogeneously the interfringe, typically 40 frames in the vicinity of the redshifted H$\alpha$ emission line. The field of view was 6 arcmin. The log of observations listing the exposure times and the mean seeing values during exposure is presented in

Table 2. Log of the scanning FPI observations.

Galaxy	Date	Emission Line	Exposure, s	Seeing, ″
PGC 60040/NGC 6361	21/22 September 2023	H$\alpha$	$40\times 180$	1.4
PGC 60446/LEDA 2596531	22/23 September 2023	H$\alpha$	$40\times 200$	2.0
PGC 68554/PGC 68555	19/20 September 2023	H$\alpha$	$40\times 240$	2.1
NGC 833/NGC 835/NGC 838	22/23 September 2023	H$\alpha$	$40\times 180$	2.5
NGC 7463/NGC 7464/NGC 7465	18/19 September 2023	H$\alpha$	$40\times 150$	2.3
UGC 9796/PGC 54478	22/23 September 2023	H$\alpha$	$40\times 180$	2.8

. Data were reduced using the software package described in detail by Moiseev and Egorov [59] and Moiseev [60]. After the primary reduction—airglow line subtraction, photometric and seeing variation corrections, and wavelength calibration—the frames were combined into data cubes, where a 40-channel spectrum was attributed to every spaxel in the field of view. The data cubes were rotated to the ‘standard’ orientation (N at the top and E to the left). The precise astrometry grid was created using the Astrometry.net project web interface http://nova.astrometry.net/. The maps in the nearby narrow continuum band and in the emission line (monochromatic images in the emission line H$\alpha$), the line-of-sight velocity fields, and the velocity dispersion maps were derived by fitting the line profiles by a one-component Voigt law, as described in Moiseev and Egorov [59].

To refine our analysis, for NGC 6361 and for the PGC 60446+LEDA 2596531 pair, we also obtained long-slit spectra deep enough to probe the emission lines in the outer parts of the galaxies (

Table 3. Log of the long-slit observations.

Galaxy	Spectrograph	b	$\mathbf{\Sigma }{\mathit{T}}_{\mathit{exp}}$, min	PA (Slit)	${\mathit{FWHM}}_{*}^{″}$
PGC 60446	SCORPIO-1/BTA	28 August 2024	120	37	1.4
NGC 6361	SCORPIO-1/BTA	29 August 2024	60	57	1.3
	TDS/2.5m	11 July 2023	80	43	2

). The reducer SCORPIO-1 [61] of the 6 m telescope of the SAO RAS was used, in long-slit mode, operating with the VPHG550G grism and the slit of 0.5″, providing the full optical spectral range and a spectral resolution of about 6 Å. NGC 6361 was also observed at the 2.5 m telescope of the Caucasian Mountain Observatory of SAI MSU with the two-armed long-slit spectrograph TDS [62], using the one arcsec slit aligned with the major axis. The spectral resolution was 2.6 Å in the red arm (5625–7420 Å) and 3.8 Å in the blue arm (3320–5795 Å).

The broad-band photometric data for the galaxies under consideration were obtained from the public resource Legacy Survey DECaLS [47].

4. Results

4.1. HCG 16

The inner kinematics of NGC 835 were analyzed using the IFU spectral data of the CALIFA survey [52]. The field of view of these data is one arcminute. We applied our software DETKA [63] to the 2D LOS velocity maps of the stars and of the ionized gas (the latter was traced by the emission lines H$\alpha$ and [NII]$\lambda$6583). The comparison of the kinematic major axes (the directions of the maximum LOS velocity gradient) with the orientation of the stellar disk line of nodes determined by isophote analysis under the assumption of an intrinsically round disk is presented in Figure 2. In general, the rotations of the stars and of the gas are consistent with each other, and both look circular over the major part of the galactic disk. However, in the radius range of $8^{″}$–${18}^{″}$, some indication of the noncircular motions caused by the possible presence of the bar strongly misaligned with the line of nodes can be noticed.

The results of our FPI observations of the HCG 16 group—H$\alpha$ emission-line flux distribution, the maps of the ionized-gas LOS velocities, and the velocity dispersion—are presented in Figure 3, together with the color map $g-r$, to trace possible sites of star formation. Although the most western member of the group, NGC 833, is not a low-massive galaxy at all—its stellar mass is quite comparable to that of NGC 835—its gaseous disk is destroyed completely. All tidal tails in this group evidently belong to NGC 833. But what is important to us is that the northern tidal gaseous flow cannot be accreted by NGC 835, though it is projected just onto the northern part of the disk of this large galaxy. The rotation of NGC 835 is quite regular and undisturbed, and the line-of-sight velocity difference between the northern part of the NGC 835 disk and the projected gaseous flow is about 700 $\mathrm{km}{\mathrm{s}}^{-1}$. This velocity difference between the northern gaseous flow and the northern part of the NGC 835 disk detected by us in the ionized-gas velocity distribution is quite similar to that found for neutral hydrogen in the 21 cm line by Jones et al. [54]. If, in the course of its orbital motion, NGC 833 receded away from us and decelerated (now, the systemic velocities of NGC 833 and NGC 835 are almost the same), then it would have to leave its gaseous debris well beyond the present position, some hundred kpc behind the plane of NGC 835. The blue rings on the velocity dispersion map of NGC 835 mark star-forming rings; as a whole, its gas velocity dispersion map demonstrates the cold dynamical status of the gaseous disk of this galaxy.

4.2. Holmberg 802

The inner kinematics of NGC 7465 have been studied in the ATLAS-3D survey [50,64], and they are very complex. The galaxy possesses a kinematically decoupled stellar core with the radius of some $5^{″}$, and within this core, the ionized gas counterrotates, while the molecular gas demonstrates the polar orientation of its kinematic major axis [65]—maybe bipolar outflows are present. But outside $R\approx 7^{″}$, the stellar component does not show any peculiarities (Figure 4): it rotates circularly on the plane of the galactic disk. The ionized gas reveals complex kinematics with possible non-circular motions.

The results of our FPI observations of the Holmberg 802 triplet—H$\alpha$ emission-line flux distribution, its maps of line-of-sight velocities, and velocity dispersion—are presented in Figure 5, together with the color map $g-r$, to trace possible sites of star formation. We see a long intergalactic gaseous flow; the main arm of it embraces NGC 7465 in the south and goes to NGC 7464—a dwarf elliptical, the smallest galaxy in the triplet. Evidently, both NGC 7464 and NGC 7465 accrete the gas just from this flow—because the elliptical NGC 7464 demonstrates intense star formation and because the gas–star kinematics are decoupled in NGC 7465 [65,66]. Moreover, the kinematic major axis (the line of nodes?) of the large-scale gaseous disk in NGC 7465 is directed toward the contact point of the outer gaseous flow with the galactic disk ($P{A}_{gas}\approx {120}^{\circ }$; Figure 5), which allows us to suppose that the inner gas in NGC 7465 may inherit the orbital momentum of the outer gas flow. Interestingly, if we apply the DETKA tilted-ring analysis to the most external part of the ionized gas related to NGC 7465, namely, to the southern subarm and to the NE clump, we obtain the parameters of the gas plane orientation within the framework of the circular rotation model: $P{A}_{gas}\approx {166}^{\circ }$, and $i={35}^{\circ }$. The former value is consistent with the orientation and rotation of the large-scale stellar disk of NGC 7465 [64]; but the kinematic inclinations are certainly different: $i_{*}={88}^{\circ }$[67] versus our $i_{H\alpha }={35}^{\circ }$. So, the outer gas is certainly off the plane of the stellar disk of NGC 7465. Since the total mass of neutral HI related to this flow is large, about 5 billion solar masses [56], the common point of view is that this intergalactic gas is provided by a “large spiral” [54]. However, the nearest large spiral, NGC 7463, cannot be a donor, because its systemic velocity differs from the line-of-sight velocity of the flow by some 400 $\mathrm{km}{\mathrm{s}}^{-1}$. So in this group, we have found an evident accretion of the intergalactic gas by NGC 7464 and NGC 7465, but the source of this gas flow remains unclear.

4.3. KPG 565

In Figure 6 we compare the orientation of the kinematic major axes of the stellar and ionized-gas components with the photometric major axis in PGC 68554. We applied our software DETKA (the tilted-ring analysis algorithm) to the MaNGA integral-field spectral data. The ionized gas is confined to the very center of the galaxy. It strictly counterrotates the stellar component, and both rotate probably on the main plane of the galaxy.

This group is known as a galaxy compact pair, PGC 68554+PGC 68555 [68]. However, our observations of the field in the narrow H$\alpha$ emission line redshifted to $V_{sys}\approx 8000$ $\mathrm{km}{\mathrm{s}}^{-1}$ revealed a lot (a half-dozen) of smaller galaxies over the area of $150\times 150$ kpc (Figure 7). Hence, it is, in fact, a compact galaxy group [69] with two central galaxies, PGC 68554+PGC 68555, of the E and $Sb$ types, respectively. Among the smaller emission-rich galaxies, two of them are cataloged in the NED: SDSS$J221905.16+130837.6$ in the south—it is of the Irr type, with $M_B=-17.96$ (HyperLEDA)—and SDSS$J221908.85+131040.5$ in the east—with $M_B=-17.18$. Both have spectra in the SDSS data. The northern edge-on galaxy is not cataloged. The group is evidently devoid of intergalactic neutral hydrogen [45], and we did not find any large-scale flows of ionized gas. The large spiral PGC 68555 has a possible tidal tail directed orthogonally to its highly inclined disk, propagating to the southeast. However, we do not see any gaseous bridge between PGC 68555 and PGC 68554; moreover, the difference in the line-of-sight velocities between PGC 68555 and PGC 68554 reaches almost 400 $\mathrm{km}{\mathrm{s}}^{-1}$. Evidently, PGC 68555 cannot be a donor of gas for PGC 68554; only PGC 68554 is detected as a galaxy with decoupled kinematics of the stars and ionized gas in the MaNGA survey [48]. The source of the outer gas accreted by PGC 68554 remains thus unclear.

4.4. UGC 9796

The polar-ring galaxy UGC 9796 consists of two radially separated components, both with quite regular rotation (Figure 8, left). In the inner, S0-like part, the stars rotate circularly on the galactic disk plane. However, at $R>{15}^{″}$, the isophote major axis turns by some ${70}^{\circ }$, reflecting the dominance of another extended stellar disk, and the outer ionized gas traced in the H$\alpha$ emission line by our FPI measurements rotates on this new galactic plane. Interestingly, the rotation of the inner ionized gas, which we analyzed with the [NII]$\lambda$6583 emission line in the MaNGA spectra, corresponds to the orientation of the outer galactic plane, so that it is quasipolar with respect to the inner galactic disk (Figure 8, left).

PGC 54478 is a late-type spiral, with some morphological peculiarities in the central part. However, its ionized gas rotates circularly on the galactic plane (Figure 8, right).

Although the polar-ring galaxy UGC 9796 has a gas-rich neighbor, LEDA 54478, only 30 kpc to the east, its polar-ring origin is unlikely to relate to this neighbor: the planes of the ring and of the LEDA 54478 disk are roughly orthogonal to each other, and we did not find any gaseous flows between UGC 9796 and LEDA 54478. Our conclusion is confirmed by the data obtained by Cox et al. [57], who have mapped the whole group of UGC 9796 in the 21 cm line: though the total mass of neutral hydrogen in the group is rather high, up to $23\times {10}^9$ solar masses, all this gas is distributed among six member galaxies, with no signal having been detected in the field between the galaxies. In such a configuration, the most probable origin of the large polar ring around UGC 9796 may be the merger of a gas-rich satellite from an orbit with a high orbital momentum (Figure 9).

4.5. NGC 6361

A giant spiral galaxy, NGC 6361 is the central galaxy of a group; its nearest neighbor, the S0-galaxy PGC 60040, has been indicated as a galaxy with counterrotating gas within the MaNGA survey [48]. In Figure 10 we show the kinematic major axes of the stars and of the ionized gas in PGC 60040 in comparison with its photometric major axis (the probable galactic disk line of nodes). The stars and the gas counterrotate with respect to each other on the common plane, but both are off the main galactic plane. It seemed natural to suppose that only NGC 6361 provided the gas for its small early-type companion.

We obtained the Fabry–Perot panoramic spectral data for both galaxies (Figure 11). We did not find any ionized-gas flow connecting directly NGC 6361 and PGC 60040, but the large galaxy demonstrates clearly extended tidal arms, probably off the galactic plane highly inclined to our line of sight.

To probe the excitation mechanism of the gas disturbed by the interaction in NGC 6361, we performed long-slit spectroscopy and plotted the strong emission-line flux ratios onto a diagnostic Baldwin–Phillips–Terlevich (BPT) diagram, which allowed us to distinguish emission-line sites excited by young stars from those excited by shock waves [70]. The dividing lines are taken from Kewley et al. [71], where the theoretical limits for the gas ionized by star formation are calculated, and from Kauffmann et al. [72], where the empirical locus for the galactic nuclei excited by young stars is derived from the data of the SDSS survey. When the emission-line knots under consideration fall onto the sequence by Kauffmann et al. [72] and to the left from Kewley’s sequence [71], we can treat them as being excited by young stars. Then, we can use strong-line indicators to obtain the oxygen abundance of the gas; these indicators are calibrated through the statistics of well-studied HII regions.

As we can see in Figure 12, almost all the emission-line knots in NGC 6361 probed by our long-slit spectroscopy are indeed HII regions. Only two regions, ${44}^{″}$ to the south and a large area ${14}^{″}$ to the north from the center, reveal some signs of pollution by shock-excited gas. For all of the regions certainly excited by young stars, except these two, we calculated the individual oxygen abundance by using the indicators $N2\equiv log(\left[\mathrm{NII}\right]\lambda 6583/H\alpha )$ and $O3N2\equiv log(\left[\mathrm{OIII}\right]\lambda 5007/H\beta )-log(\left[\mathrm{NII}\right]\lambda 6583/H\alpha )$ calibrated by Pettini and Pagel [73]. The results are presented in

Table 4. Oxygen abundance of HII regions in NGC 6361 derived from our long-slit data.

The HII Region	$12+log\left(\mathbf{O}/\mathbf{H}\right)\left(\mathit{N}2\right)$	$12+log\left(\mathbf{O}/\mathbf{H}\right)\left(\mathit{O}3\mathit{N}2\right)$
${84}^{″} S$	8.62	8.67
${60}^{″} S$	8.59	8.55
${38}^{″} S$	8.615	8.59
${22}^{″} S$	8.56	8.84
${13}^{″} S$	8.70	8.68
${26}^{″} N$	8.66	8.78
${31}^{″} N$	8.615	8.645
${34}^{″} N$	8.69	8.705
${57}^{″} N$	8.60	8.51

; they cover the radial range from $R={14}^{″}$ (3.6 kpc, the inner star-forming ring) to $R={84}^{″}$ (20.5 kpc, the outer tidal arm).

One can see an undisturbed radial metallicity gradient within the disk of NGC 6361; only the southern tidal arm, probed half-way to PGC 60040, gives a hint of some metallicity overabundance as far from the center as at 20.5 kpc. The metallicity gradient in the disk of NGC 6361, between the radii of ${13}^{″}$ (3.8 kpc) and ${60}^{″}$ (17.5 kpc), appeared to be $-0.012\pm 0.006$ dex per kpc, which seems to be quite normal for a giant spiral galaxy of Sb; for example, in the Milky Way, the young disk oxygen abundance gradient is $-0.027$–$-0.07$ dex per kpc [75,76,77], and in the M 31 disk, it is –0.023 [78]. So the interaction of NGC 6361 with PGC 60040, though it produced the outer tidal arms, has not mixed the gas along the radius in the disk of the galaxy.

4.6. PGC 60446

In PGC 60446, the central ionized gas strictly counterrotates with respect to the stars (Figure 13), and both are probably confined to the inner disk, with a radius of some ${12}^{″}$, inclined with respect to the outer body of the galaxy. If PGC 60446 is indeed an elliptical as HyperLEDA states, the discordance between the inner disk major-axis orientation and the main symmetry axis of the outer galaxy may be a hint of the intrinsic triaxial shape of the galactic stellar body.

In this tight galaxy pair, PGC 60446 is an early-type galaxy, probably an elliptical one, demonstrating the decoupled kinematics of the ionized gas and its stellar components; and its northern neighbor, LEDA 2596531, is a blue late-type galaxy, with very open spiral arms, one of which is directed toward PGC 60446. Our natural expectation was that LEDA 2596531 represented a gas donor to PGC 60446. However, the H$\alpha$ emission-line intensity map obtained with our Fabry–Perot scanning interferometer showed that in the late-type LEDA 2596531, only the nucleus emits ionized-gas radiation—the arms are invisible in the emission lines (Figure 14). The blue spiral arms of LEDA 2596531 lack current star formation!

Beyond the circumnuclear gaseous disk in PGC 60446, which is counterrotating with respect to the stellar component, the galaxy demonstrates the presence of a separate ionized-gas clump $7^{″}$ to the south of the nucleus—this shift is almost perpendicular to the line of nodes of the circumnuclear gaseous disk. Despite its position at the kinematic minor axis, the line-of-sight velocity of the clump is blueshifted with respect to the systemic velocity of PGC 60446. Evidently, since the clump does not participate in the global galaxy rotation, it was recently accreted from an off-plane direction. Using the lucky fact that the clump is caught by our long-slit spectroscopy, we calculated the strong emission-line flux ratios for it and put them into a BPT diagram (Figure 15).

As one can deduce from inspecting Figure 15, the gaseous clump $7^{″}$ to the south from the nucleus of PGC 60446 is certainly excited by young stars—perhaps, it is the only star-forming region in this elliptical galaxy. The estimate of the gas metallicity made through the calibrations of strong-line indicators by Pettini and Pagel [73] gives $12+log\left(O/H\right)\left(N2\right)=8.56$ and $12+log\left(O/H\right)\left(O3N2\right)=8.50$—typical oxygen abundance values in the star-forming regions of early-type galaxies fueled by accreted gas [79]. The nucleus of LEDA 2596531 appears to be in the AGN area of the BPT diagram; despite the narrow $FWHM$ of its emission lines, the strong [OIII] lines imply the Sy2 classification. Since PGC 60446 is known to also possess a Seyfert nucleus (its nuclear emission lines are broad), we deal here with a pair of Seyfert galaxies whose active nuclei are perhaps triggered by outer gas accretion.

5. Discussion and Conclusions

When we see a galaxy belonging to a compact group with decoupled kinematics of gas and stars, the first idea is that the decoupled gas has been accreted from a neighboring galaxy. When observations in the 21 cm HI line reveal intergalactic cold neutral hydrogen between galaxies in compact groups, the first idea is that this gas has been removed from some members of the group by interaction. However, this process must leave individual galaxies with gas deficiency, while a significant amount of gas must pass from galaxies into the intergalactic medium. Using this idea, there have been attempts to connect the gas deficiency of individual galaxies in compact groups to the dynamical status of a group [54]; however, these attempts have not been very successful. In this particular field, Jones et al. [54] considered 38 Hickson compact groups mapped in the 21 cm line by VLA, and one of these group was just HCG 16. Among the five members of HCG 16, the gas deficiency was stated by Jones et al. [54] only for NGC 833 and NGC 835—so they were the primary candidates for the role of donors for the intergalactic HI of HCG 16, representing a linear large-scale HI flow passing through the whole group, from NGC 833 to NGC 839 [53].

To verify if our galaxies in the groups are HI-deficient, we employ the known scaling relations, which represent correlations between integrated galaxy properties and the stellar mass. Figure 16 shows the scaling relations for large unbiased disk galaxy samples connecting HI mass and stellar mass and connecting star formation rate and stellar mass (‘main galaxy sequence’) established by the data of the recent, large radio surveys xGASS [80] and WHISP [81]. Not all of our galaxies have measured HI content; but almost all that are covered by the HI surveys have quite normal HI content for their stellar masses (Figure 16). Only PGC 68554, which is a moderate-luminosity elliptical galaxy having acquired its gas perhaps through the minor merging of a small satellite, is below the median trend (though within the total scatter of this scaling relation). We do not confirm the gas deficiency of NGC 835—and our conclusion is more natural than that by Jones et al. [54], because the intergalactic HI flow seen to the north of NGC 835 has quite different line-of-sight velocity from the northern part of the NGC 835 disk. As for the scaling relation involving the star formation rate, only two galaxies fall below the main sequence—they are NGC 833, which has suffered from the tidal disturbance of its large gaseous disk, and PGC 60040, which, similarly to PGC 68554, has probably acquired too little gas to start normal star formation (we know nothing about the individual HI mass of PGC 60040).

The main question of what is the source of the accreted gas remains vague at this stage of our study. The last successes in assembling the statistically relevant samples of galaxies with decoupled kinematics of gas and stars achieved in the IFU surveys SAMI and MaNGA have generated the hypothesis that the outer gas sources may be various and multiple [15,16,82]. As is known, the gas accreted from an arbitrary direction tends to occupy the main symmetry planes of the recipient galaxy, so the decoupled gas systems prefer to lie on the galactic plane, demonstrating counterrotation, or to form a polar ring/disk [23,83]. If we concentrate on counterrotating gaseous systems, they may be inner or outer, with or without current star formation [18]. For this variety of situations, two main scenarios have been proposed: cold planar accretion from a filament for the inner counterrotating disks and minor merging for the outer counterrotating disks [16,18]. If, initially, the recipient galaxy is gas-rich, the central gas compression during the accretion event may provoke a star formation burst. Outer star formation in galaxies with gas counterrotation is not implied by any scenario, but it can be observed in lenticular galaxies [84]. The scenario with gas accretion from filaments in galaxies with inner counterrotation is in contrast with the rather high metallicity of this counterrotating gas excited by young stars [17,18].

Figure 16. The scaling relation positions of the group galaxies considered in this work. The gray shadow regions mark the scatter of data from the original papers, the best-fit lines were also present in the works cited below. Left panels: The HI vs. $M_{*}$ relation [80] and the main sequence $sSFR$ vs. $M_{*}$ [85], both based on the xGASS survey. The red points indicate weighted average values, and the solid line is a main sequence with $\pm 0.3$ dex borders (the dotted lines). In the right column, the same scaling relations for dwarf galaxies are from the WHISP survey [81]. The black line indicates the best fit to the $M_{HI}$ values contained within the stellar disk, whereas the magenta line corresponds to the total $M_{HI}$, including the wide outskirts of the galaxies. The black squares show the positions of our galaxies.

Figure 16. The scaling relation positions of the group galaxies considered in this work. The gray shadow regions mark the scatter of data from the original papers, the best-fit lines were also present in the works cited below. Left panels: The HI vs. $M_{*}$ relation [80] and the main sequence $sSFR$ vs. $M_{*}$ [85], both based on the xGASS survey. The red points indicate weighted average values, and the solid line is a main sequence with $\pm 0.3$ dex borders (the dotted lines). In the right column, the same scaling relations for dwarf galaxies are from the WHISP survey [81]. The black line indicates the best fit to the $M_{HI}$ values contained within the stellar disk, whereas the magenta line corresponds to the total $M_{HI}$, including the wide outskirts of the galaxies. The black squares show the positions of our galaxies.

In our sample we have four galaxies with central counterrotation and one galaxy with a large polar ring. Having analyzed, by measuring the H$\alpha$ emission line, the intergalactic ionized-gas distributions and velocities of these six early-type galaxies in compact groups, we may conclude that none of the galaxies with decoupled gas–star kinematics (and thus with recent gas accretion) has received its gas from neighboring non-dwarf galaxies. Thus, we exclude at least one of the popular scenarios. The only case where we can state that we see an intergalactic gas flow feeding the decoupled gaseous disk is the case of NGC 7465; but we cannot identify the origin of this intergalactic gas flow. When large reservoirs of neutral gas are observed, as it takes place in HCG 16, they are too large to be filled by individual galaxies members of groups, because the galaxies members of the group HCG 16 do not look strongly gas-deficient. Even the large massive polar ring of UGC 9796 does not demonstrate any connection to the late-type galaxies of its group [57]. Hence, the most attractive hypothesis for the gas accretion sources if we consider our small sample of early-type galaxies in compact groups remains the minor merging of gas-rich satellites.