Abstract
Centaurus A (Cen A) is one of the most famous galaxies hosting an active galactic nucleus (AGN), where the interaction between AGN activities and surrounding interstellar and intergalactic media has been investigated. Recent studies reported detections of Hα emission from clouds in the galactic halo toward the northeast and southwest of the nucleus of Cen A, suggesting that AGN jets may have triggered star formation there. We performed near-infrared line mapping of Cen A with the IRSF 1.4 m telescope, using the narrow-band filter tuned for Paβ, from which we find that Paβ emission is not detected significantly from either of the northeast or southwest regions. The upper limit of the Paβ/Hα ratio in the northeast region is compatible with that expected for a typical H ii region, in line with the scenario that AGNs have triggered star formation there. On the other hand, the upper limit of Paβ/Hα in the southwest region is significantly lower than that expected for a typical H ii region. A possible explanation of the low Paβ/Hα ratio in the southwest region is the scattering of Hα and Paβ photons from the center of Cen A by dust grains in the halo clouds. From the upper limit of Paβ/Hα in the southwest region, we obtain constraints on the dust size distribution, which is found to be compatible with those seen in the interstellar medium of our Galaxy.
1 Introduction
Active galactic nuclei (AGNs) are known to play an important role in the evolution of host galaxies through injecting kinetic energy into the surrounding interstellar and intergalactic media (ISM and IGM) via outflows or jets (Heckman & Best 2014, and references therein); however, it is not yet understood how this AGN feedback mechanism affects the host galaxy. It has often been considered that an AGN blows out cold gas from the host galaxy or prevents gas from cooling to suppress star formation (i.e. negative feedback; see, e.g., Fabian 2012; Harrison 2017). Alternatively, it is also suggested that AGN activity can compress gas and trigger star formation (i.e. positive feedback; see, e.g., Silk 2013; Zinn et al. 2013). Hence, the effect of AGN feedback is not likely to be straightforward (Zubovas & Bourne 2017; Shin et al. 2019).
Observational evidence for AGN-triggered star formation has been reported for nearby galaxies harboring AGNs, such as Minkowski’s Object and 3C 285. Minkowski’s Object is a star-forming peculiar object, located at the end of the radio jet from the nearby galaxy NGC 541, and it has long been suggested that its star formation is triggered by the radio jet (Brodie et al. 1985; van Breugel et al. 1985; Salomé et al. 2015). At a distance of nearly |$70\rm \, kpc$| from the galaxy center of 3C 285, van Breugel and Dey (1993) discovered a small Hα-emitting object near the eastern radio jet (3C 285/09.6), suggesting jet-induced star formation through compression of dense surrounding material.
Centaurus A (Cen A or NGC 5128), at a distance of 3.8 Mpc (Harris et al. 2010), is the radio galaxy nearest to us, hosting an AGN which emits powerful radio and X-ray jets on scales ranging from 1.35 kpc to 250 kpc (Israel 1998). Thus, Cen A is an important target for understanding the interaction between AGN jets and the surrounding ISM and IGM. At a distance of approximately 8 kpc from the nucleus, optically bright filaments and possible young stars therein are observed in regions close to the northeastern radio jet, which could be the result of jet–ISM interaction (Blanco et al. 1975; Osmer 1978; Graham & Price 1981; Morganti et al. 1991; Rejkuba et al. 2002; Crockett et al. 2012). Additionally, recent studies report detection of hydrogen recombination line emissions in the galactic halo around the jets, indicating jet-induced star formation. In the northeast of Cen A, Santoro et al. (2016) found an Hα-emitting region at about 15 kpc from the nucleus, suggesting that the AGN may indeed have triggered star formation there. On the opposite side of that region with respect to the nucleus, Keel et al. (2019) also found an Hα-emitting region in a halo cloud at about 12 kpc from the nucleus along the axis of the southwestern jet lobe, again suggesting positive AGN feedback taking place in this region.
On the other hand, there is the possibility of the scattering of Hα photons from a host galaxy by dust grains in halo clouds. At about |$4\rm \, kpc$| from the galaxy disk of M 82, strongly polarized Hα emission is observed, and its polarization pattern suggests that the Hα emission originates from the dust scattering of Hα photons coming from the central near-infrared (IR) nucleus (Scarrott et al. 1991; Yoshida et al. 2019). Therefore, in order to evaluate the effect of the AGN feedback accurately, it is important to investigate whether or not the hydrogen recombination line truly originates from star formation therein.
In the previous studies, only a small portion of Cen A was studied. In order to investigate the interaction between AGN feedback and the intergalactic clouds in Cen A, however, it is necessary to observe wider areas of Cen A along the axis of the jet with the hydrogen recombination lines tracing star-forming regions. Thus, in the present study we perform a wider-area mapping of Cen A along the axis of the jet using near-IR narrow-band filters. We obtain the physical parameters of the hydrogen recombination lines, combining our data with the previous studies to verify the possibility of dust scattering in the halo clouds of Cen A.
2 Observations and data analysis
Near-IR imaging toward Cen A was carried out in 2019 March, May, and June with the SIRIUS (Simultaneous InfraRed Imager of Unbiased Survey; Nagashima et al. 1999, Nagayama et al. 2003) camera on the InfraRed Survey Facility (IRSF) 1.4 m telescope at the South African Astronomical Observatory. We observed Cen A using the narrow-band filter tuned for Paβ 1.282 μm with the dithering of 10 frames and exposure time of 60 seconds per frame. The effective bandwidth of the filter for Paβ is 0.029 μm. In addition, to subtract the Paβ continuum we also observed Cen A using a notch filter with two peaks of transmittance at about 1.268 and 1.297 μm. The camera has a field of view of |${7{_{.}^{\prime}}7}\times {7{_{.}^{\prime}}7}$| with a pixel scale of |${0{_{.}^{\prime\prime}}45}$|. We observed five fields in total, and summarize the detail of the observation of each field in table 1. With the continuum filter we observed for twice the total exposure times listed in table 1.
Details of the IRSF observations for the present study.
| RA | Dec | Total exposure time | |||
|---|---|---|---|---|---|
| Region | (J2000.0) | (J2000.0) | (min) | Airmass | Date |
| Northeast | |${13^{\rm h}26^{\rm m}20{_{.}^{\rm s}}00}$| | −42°|${50^{\prime }00{^{\prime \prime }_{.}}0}$| | 30 | 1.02–1.24 | 2019 May 17 |
| Northeast2 | |${13^{\rm h}25^{\rm m}54{_{.}^{\rm s}}70}$| | −42°|${55^{\prime }33{^{\prime \prime }_{.}}8}$| | 30 | 1.02–1.39 | 2019 Mar 16 |
| Center | |${13^{\rm h}25^{\rm m}27{_{.}^{\rm s}}60}$| | −43°|${01^{\prime }08{^{\prime \prime }_{.}}8}$| | 10 | 1.12–1.25 | 2019 Mar 12 |
| Southwest2 | |${13^{\rm h}25^{\rm m}01{_{.}^{\rm s}}70}$| | −43°|${04^{\prime }21{^{\prime \prime }_{.}}1}$| | 30 | 1.24–1.98 | 2019 Mar 17 |
| Southwest | |${13^{\rm h}24^{\rm m}35{_{.}^{\rm s}}00}$| | −43°|${10^{\prime }00{^{\prime \prime }_{.}}0}$| | 120 | 1.11–1.91 | 2019 Mar 15, Jun 15, 18, 20, 24 |
| Sky | |${13^{\rm h}26^{\rm m}54{_{.}^{\rm s}}00}$| | −43°|${17^{\prime }18{^{\prime \prime }_{.}}0}$| | 20 | 1.02–1.98 | All days above |
| RA | Dec | Total exposure time | |||
|---|---|---|---|---|---|
| Region | (J2000.0) | (J2000.0) | (min) | Airmass | Date |
| Northeast | |${13^{\rm h}26^{\rm m}20{_{.}^{\rm s}}00}$| | −42°|${50^{\prime }00{^{\prime \prime }_{.}}0}$| | 30 | 1.02–1.24 | 2019 May 17 |
| Northeast2 | |${13^{\rm h}25^{\rm m}54{_{.}^{\rm s}}70}$| | −42°|${55^{\prime }33{^{\prime \prime }_{.}}8}$| | 30 | 1.02–1.39 | 2019 Mar 16 |
| Center | |${13^{\rm h}25^{\rm m}27{_{.}^{\rm s}}60}$| | −43°|${01^{\prime }08{^{\prime \prime }_{.}}8}$| | 10 | 1.12–1.25 | 2019 Mar 12 |
| Southwest2 | |${13^{\rm h}25^{\rm m}01{_{.}^{\rm s}}70}$| | −43°|${04^{\prime }21{^{\prime \prime }_{.}}1}$| | 30 | 1.24–1.98 | 2019 Mar 17 |
| Southwest | |${13^{\rm h}24^{\rm m}35{_{.}^{\rm s}}00}$| | −43°|${10^{\prime }00{^{\prime \prime }_{.}}0}$| | 120 | 1.11–1.91 | 2019 Mar 15, Jun 15, 18, 20, 24 |
| Sky | |${13^{\rm h}26^{\rm m}54{_{.}^{\rm s}}00}$| | −43°|${17^{\prime }18{^{\prime \prime }_{.}}0}$| | 20 | 1.02–1.98 | All days above |
Details of the IRSF observations for the present study.
| RA | Dec | Total exposure time | |||
|---|---|---|---|---|---|
| Region | (J2000.0) | (J2000.0) | (min) | Airmass | Date |
| Northeast | |${13^{\rm h}26^{\rm m}20{_{.}^{\rm s}}00}$| | −42°|${50^{\prime }00{^{\prime \prime }_{.}}0}$| | 30 | 1.02–1.24 | 2019 May 17 |
| Northeast2 | |${13^{\rm h}25^{\rm m}54{_{.}^{\rm s}}70}$| | −42°|${55^{\prime }33{^{\prime \prime }_{.}}8}$| | 30 | 1.02–1.39 | 2019 Mar 16 |
| Center | |${13^{\rm h}25^{\rm m}27{_{.}^{\rm s}}60}$| | −43°|${01^{\prime }08{^{\prime \prime }_{.}}8}$| | 10 | 1.12–1.25 | 2019 Mar 12 |
| Southwest2 | |${13^{\rm h}25^{\rm m}01{_{.}^{\rm s}}70}$| | −43°|${04^{\prime }21{^{\prime \prime }_{.}}1}$| | 30 | 1.24–1.98 | 2019 Mar 17 |
| Southwest | |${13^{\rm h}24^{\rm m}35{_{.}^{\rm s}}00}$| | −43°|${10^{\prime }00{^{\prime \prime }_{.}}0}$| | 120 | 1.11–1.91 | 2019 Mar 15, Jun 15, 18, 20, 24 |
| Sky | |${13^{\rm h}26^{\rm m}54{_{.}^{\rm s}}00}$| | −43°|${17^{\prime }18{^{\prime \prime }_{.}}0}$| | 20 | 1.02–1.98 | All days above |
| RA | Dec | Total exposure time | |||
|---|---|---|---|---|---|
| Region | (J2000.0) | (J2000.0) | (min) | Airmass | Date |
| Northeast | |${13^{\rm h}26^{\rm m}20{_{.}^{\rm s}}00}$| | −42°|${50^{\prime }00{^{\prime \prime }_{.}}0}$| | 30 | 1.02–1.24 | 2019 May 17 |
| Northeast2 | |${13^{\rm h}25^{\rm m}54{_{.}^{\rm s}}70}$| | −42°|${55^{\prime }33{^{\prime \prime }_{.}}8}$| | 30 | 1.02–1.39 | 2019 Mar 16 |
| Center | |${13^{\rm h}25^{\rm m}27{_{.}^{\rm s}}60}$| | −43°|${01^{\prime }08{^{\prime \prime }_{.}}8}$| | 10 | 1.12–1.25 | 2019 Mar 12 |
| Southwest2 | |${13^{\rm h}25^{\rm m}01{_{.}^{\rm s}}70}$| | −43°|${04^{\prime }21{^{\prime \prime }_{.}}1}$| | 30 | 1.24–1.98 | 2019 Mar 17 |
| Southwest | |${13^{\rm h}24^{\rm m}35{_{.}^{\rm s}}00}$| | −43°|${10^{\prime }00{^{\prime \prime }_{.}}0}$| | 120 | 1.11–1.91 | 2019 Mar 15, Jun 15, 18, 20, 24 |
| Sky | |${13^{\rm h}26^{\rm m}54{_{.}^{\rm s}}00}$| | −43°|${17^{\prime }18{^{\prime \prime }_{.}}0}$| | 20 | 1.02–1.98 | All days above |
The image data were processed with the standard data reduction by using the pyIRSF1 pipeline software, which includes dark subtraction, flat-fielding, sky subtraction, and dithered-image combining. We performed astrometric and photometric calibrations with the 2MASS Point Source Catalog (PSC; Skrutskie et al. 2006), assuming that the magnitudes of point sources are the same between the J-band and the narrow-band images of Paβ and the continuum. For photometry, we selected stars with J-band fluxes higher than 12.5 mag that were not saturated, and with errors smaller than 0.05 mag from the 2MASS PSC. Then, determining the sky regions in each field and subtracting the sky emission from the images, we adjusted the sky level of each observed field to be zero. Finally, we subtracted the Paβ-continuum image from the narrow-band image to obtain the Paβ intensity map of Cen A.
3 Results
Figure 1 shows the Paβ intensity map of Cen A obtained with the narrow-band filters, from which we confirm the presence of Paβ emission extending along the galaxy structure, but not clearly associated with the jet lobes. Figure 2 shows close-up images of the northeast and southwest regions where Hα emission was detected in the previous studies. In the northeast region, Santoro et al. (2016) spectroscopically estimated the Hα fluxes of Regions A and B to be 2.3 × 10−15 and |$3.8\times 10^{-15}\rm \, erg\, s^{-1}\, cm^{-2}$|, respectively. Similarly, in the southwest region, Keel et al. (2019) measured the Hα flux densities from the four Hα-bright local blobs Main, Main N, Diffuse, and North with integral-field spectroscopy to be 9.3 × 10−15, 1.3 × 10−15, 1.1 × 10−15, and |$1.0\times 10^{-15}\rm \, erg\, s^{-1}\, cm^{-2}$|, respectively. We performed aperture photometry with a radius of |${1{_{.}^{\prime\prime}}5}$| for Regions A and B, and for the four Hα-bright local blobs, using the present Paβ map. As a result, we find that the Paβ lines are not detected significantly from any of these regions (table 2); the 3 σ upper limits of the Paβ fluxes measured within the northeast and southwest regions are 7.2 × 10−16 and |$6.2\times 10^{-16}\rm \, erg\, s^{-1}\, cm^{-2}$|, respectively, where we used the sum of the photometry results for both regions.
(a) Paβ + continuum, (b) continuum, and (c) continuum-subtracted Paβ maps of Centaurus A. The maps have been smoothed with a Gaussian kernel of |${2{_{.}^{\prime\prime}}7}$| in sigma, and the color scales are given in units of |$\rm erg\, s^{-1}\, cm^{-2}\, arcsec^{-2}$|. In panel (c), the contours show the Herschel 500 μm map from Auld et al. (2012), where the contour levels correspond to 0.10, 0.27, 0.72, 1.9, 5.2, 14, and |$37\rm \, mJy\, beam^{-1}$|. The red squares, whose sizes are 30″ × 45″ and 60″ × 90″, indicate the northeast and the southwest regions where the Hα emission was detected by Santoro et al. (2016) and Keel et al. (2019), respectively. (Color online)
Comparison of the Hα and Paβ line intensity maps of the northeast and the southwest regions of Centaurus A, which correspond to the red squares in figure 1, overlaid with the same contours of the Herschel 500 μm map as in figure 1. The Paβ maps have been smoothed with a Gaussian kernel of |${1{_{.}^{\prime\prime}}13}$| in sigma. The Hα maps of the northeast and southwest regions are both taken from Keel et al. (2019).
Photometry results for the Paβ fluxes in the southwest and northeast regions of Cen A.
| Northeast component | Region A | Region B | ||
|---|---|---|---|---|
| Observed Paβ flux |$\rm (erg\, s^{-1}\, cm^{-2})$| | (1.0 ± 1.7) × 10−16 | (0.7 ± 1.7) × 10−16 | ||
| Southwest component | Main | Main N | Diffuse | North |
| Observed Paβ flux |$\rm (erg\, s^{-1}\, cm^{-2})$| | (−0.6 ± 1.1) × 10−16 | (0.9 ± 1.0) × 10−16 | (8.0 ± 9.5) × 10−17 | (0.9 ± 1.1) × 10−16 |
| Northeast component | Region A | Region B | ||
|---|---|---|---|---|
| Observed Paβ flux |$\rm (erg\, s^{-1}\, cm^{-2})$| | (1.0 ± 1.7) × 10−16 | (0.7 ± 1.7) × 10−16 | ||
| Southwest component | Main | Main N | Diffuse | North |
| Observed Paβ flux |$\rm (erg\, s^{-1}\, cm^{-2})$| | (−0.6 ± 1.1) × 10−16 | (0.9 ± 1.0) × 10−16 | (8.0 ± 9.5) × 10−17 | (0.9 ± 1.1) × 10−16 |
Photometry results for the Paβ fluxes in the southwest and northeast regions of Cen A.
| Northeast component | Region A | Region B | ||
|---|---|---|---|---|
| Observed Paβ flux |$\rm (erg\, s^{-1}\, cm^{-2})$| | (1.0 ± 1.7) × 10−16 | (0.7 ± 1.7) × 10−16 | ||
| Southwest component | Main | Main N | Diffuse | North |
| Observed Paβ flux |$\rm (erg\, s^{-1}\, cm^{-2})$| | (−0.6 ± 1.1) × 10−16 | (0.9 ± 1.0) × 10−16 | (8.0 ± 9.5) × 10−17 | (0.9 ± 1.1) × 10−16 |
| Northeast component | Region A | Region B | ||
|---|---|---|---|---|
| Observed Paβ flux |$\rm (erg\, s^{-1}\, cm^{-2})$| | (1.0 ± 1.7) × 10−16 | (0.7 ± 1.7) × 10−16 | ||
| Southwest component | Main | Main N | Diffuse | North |
| Observed Paβ flux |$\rm (erg\, s^{-1}\, cm^{-2})$| | (−0.6 ± 1.1) × 10−16 | (0.9 ± 1.0) × 10−16 | (8.0 ± 9.5) × 10−17 | (0.9 ± 1.1) × 10−16 |
Assuming Case B recombination with the electron density |$n_{\rm e}=10^2\rm \, cm^{-3}$| and electron temperature |$T_{\rm e}=10^4\rm \, K$| for a typical H ii region, we estimate that the Paβ flux from the Hα flux of the previous study in the northeast region would be |$3.4\times 10^{-16}\rm \, erg\, s^{-1}\, cm^{-2}$|. Thus, the 3 σ upper limit of the observed flux (|$<\! 7.2\times 10^{-16}\rm \, erg\, s^{-1}\, cm^{-2}$|) is compatible with the Paβ flux expected from Case B recombination, indicating that jet-induced star formation may indeed have taken place. On the other hand, in the southwest region we similarly estimate the Paβ flux from the Hα fluxes to be |$7.2 \times 10^{-16}\rm \, erg\, s^{-1}\, cm^{-2}$|, assuming Case B recombination, which is not compatible with the observed 3 σ upper limit in the southwest region (|$<\! 6.2\times 10^{-16}\rm \, erg\, s^{-1}\, cm^{-2}$|). Thus, the Hα fluxes from this region may not be attributed to the emission of ionized gas in a star-forming region.
4 Discussion
We obtain the observed upper limits of the Paβ/Hα ratios of 1.2 × 10−1 and 4.9 × 10−2 in the northeast and southwest regions, respectively, which are compared with those predicted by models. In particular, we obtain the strongest constraint on the upper limit of the Paβ/Hα ratio of 3.5 × 10−2 for Main in the southwest region, which is the brightest Hα blob, and discuss its implications below.
First, we calculate the Paβ/Hα ratios expected for Case A and B recombinations for a typical H ii region, as shown in table 3, which are not consistent with the observed upper limit of the Paβ/Hα ratio. Even if we assume Case B recombination for wide ranges of ne and Te of 102–|$10^6\rm \, cm^{-3}$| and 5000–20000 K, respectively, the expected Paβ/Hα ratios exceed 5.3 × 10−2 (Osterbrock & Ferland 2006). Thus, we find that neither Case A nor Case B satisfies the observed upper limit of the Paβ/Hα ratio in the southwest region. We then considered the effect of dust extinction; however, it would further increase the Paβ/Hα ratio observed, because Paβ emission is less attenuated by dust compared to Hα emission.
Comparison of the observed and model-predicted Paβ/Hα ratios.
| Dust scattering‡ | |||||
|---|---|---|---|---|---|
| Observed 3 σ upper limit | Case A* | Case B† | Silicate | Graphite | |
| Paβ/Hα | <3.5 × 10−2 (Main southwest) | 7.6 × 10−2 | 5.7 × 10−2 | (3.0 ± 0.3) × 10−2 | (5.2 ± 0.4) × 10−2 |
| Dust scattering‡ | |||||
|---|---|---|---|---|---|
| Observed 3 σ upper limit | Case A* | Case B† | Silicate | Graphite | |
| Paβ/Hα | <3.5 × 10−2 (Main southwest) | 7.6 × 10−2 | 5.7 × 10−2 | (3.0 ± 0.3) × 10−2 | (5.2 ± 0.4) × 10−2 |
Low density limit, |$T_{\rm e}=10^4\rm \, K$| (Osterbrock & Ferland 2006).
|$n_{\rm e} = 10^2\rm \, cm^{-3}$|, |$T_{\rm e} = 10^4\rm \, K$| (Osterbrock & Ferland 2006).
Scattering efficiency from Laor and Draine (1993). A single dust size of 0.35 μm is assumed. The scattered Paβ/Hα photons originate from the Case B recombination lines from the host galaxy.
Comparison of the observed and model-predicted Paβ/Hα ratios.
| Dust scattering‡ | |||||
|---|---|---|---|---|---|
| Observed 3 σ upper limit | Case A* | Case B† | Silicate | Graphite | |
| Paβ/Hα | <3.5 × 10−2 (Main southwest) | 7.6 × 10−2 | 5.7 × 10−2 | (3.0 ± 0.3) × 10−2 | (5.2 ± 0.4) × 10−2 |
| Dust scattering‡ | |||||
|---|---|---|---|---|---|
| Observed 3 σ upper limit | Case A* | Case B† | Silicate | Graphite | |
| Paβ/Hα | <3.5 × 10−2 (Main southwest) | 7.6 × 10−2 | 5.7 × 10−2 | (3.0 ± 0.3) × 10−2 | (5.2 ± 0.4) × 10−2 |
Low density limit, |$T_{\rm e}=10^4\rm \, K$| (Osterbrock & Ferland 2006).
|$n_{\rm e} = 10^2\rm \, cm^{-3}$|, |$T_{\rm e} = 10^4\rm \, K$| (Osterbrock & Ferland 2006).
Scattering efficiency from Laor and Draine (1993). A single dust size of 0.35 μm is assumed. The scattered Paβ/Hα photons originate from the Case B recombination lines from the host galaxy.
It is possible that the jet-driven shock excites the optical emission line. For example, Sutherland, Bicknell, and Dopita (1993) suggested that the inner emission-line filaments in Cen A at approximately 8 kpc from the nucleus could be produced by interaction between the jet and the surrounding ISM. Actually, we estimate the Paβ/Hα ratio assuming shock excitation with ne and Te of 102–104 and |$500\rm \, K$|, respectively, to find that the resultant Paβ/Hα ratio is larger than 2.1 × 10−2 (Osterbrock & Ferland 2006), which is compatible with the observed 3 σ upper limit of the Paβ/Hα ratio. In this shock excitation model, however, a strong radio jet needs to be present in the vicinity of the emission-line filaments. In the northeast region of Cen A, Santoro et al. (2015) claim that the jet–cloud interaction is unlikely to be important because the jet is too diffuse in this region. Thus, jet-driven shock excitation may not be taking place in the southwest region as well, where the galactic cloud is |$12\rm \, kpc$| away from the nucleus and the jet is as diffuse as in the northeast region.
Yet another possibility to explain the observed upper limit of the Paβ/Hα ratio in the southwest region is dust scattering of the Hα and Paβ photons emitted from the center of the host galaxy (table 3). In this case, the scattering efficiency of Paβ photons is expected to be significantly lower than that of Hα photons. Indeed, in the southwest region, Auld et al. (2012) identified a dust cloud in the 500 μm emission, which excludes the possibility that the 500 μm emission is due to synchrotron radiation. A scattering scenario would also be useful to explain the low dust temperature and the lack of UV emission observed by Auld et al. (2012) for the dust cloud in the southwest region. On the other hand, Keel et al. (2019) found that the optical spectrum in the southwest region is consistent with photoionization in typical H ii regions. Since scattering preserves local line ratios, this would suggest scattering not so much of AGN radiation as that from the inner starburst disk.
Dust-scattered Hα emission from the host galaxy is actually observed in the galactic halo of M 82 (Yoshida et al. 2019). Assuming that this is also the case in Cen A, we measured the Paβ flux of the host galaxy with an aperture radius of |${2{_{.}^{\prime}}5}$|. The resultant Paβ flux is |$(8.0\pm 0.7)\times 10^{-13}\rm \, erg\, s^{-1}\, cm^{-2}$| received by us and |$(8.0\pm 0.7)\times 10^{-8}\rm \, erg\, s^{-1}\, cm^{-2}$| by dust in the southwestern cloud with a distance of 12 kpc from the galaxy to the cloud. Using these results, we calculated the Hα flux of the host galaxy for Case B recombination, and then we estimated the Paβ and Hα fluxes scattered by dust in the southwestern cloud.
First, in order to get a rough picture, we tentatively used silicate and graphite dust of a constant size of 0.35 μm in radius, for which the scattering coefficients at the wavelengths of the Paβ and Hα lines, QPaβ and QHα, are taken from Laor and Draine (1993). As a result, we find that the estimated Paβ/Hα ratio, both scattered by the silicate and graphite dust of 0.35 μm in size in the southwestern cloud, would be (3.0 ± 0.3) × 10−2 and (5.2 ± 0.4) × 10−2, respectively (table 3). Hence, silicate dust scattering is compatible with the 3 σ upper limit of the observed Paβ/Hα ratio in the southwest region, whereas graphite dust scattering is not.
Then we assumed a more realistic dust size distribution such as that in Mathis et al. (1977, hereafter MRN77), with QPaβ and QHα in Laor and Draine (1993), the result of which is shown in table 4. We adopted the ranges of silicate and graphite dust sizes of 0.025–0.25 μm and 0.005–1 μm, respectively (MRN77). Table 4 also shows the results for power-law indices different from that of MRN77. As compared to table 3, the Paβ/Hα ratios in table 4 are relatively small due to the presence of dust smaller than 0.35 μm in size, and therefore the observed result favors the presence of such small-sized dust. Consequently, the size distribution following the MRN77 power-law function can reproduce a Paβ/Hα ratio compatible with the observed upper limit. As shown in table 4, as the power-law index is steeper, the compatibility with our observational result is more secure. Actually, in the galactic halo, Hirashita and Lin (2020) showed a relatively high abundance in small grains with a radius of less than 0.03 μm with their numerical simulation, which may be caused by dust shattering during the galactic outflow processes. In order to give a stronger constraint on the dust size distribution in this region, we need further observations of Paβ with higher sensitivity.
Paβ/Hα ratios expected from dust scattering with different power-law indices of the dust size distribution.
| Power-law index | Observed 3 σ upper limit | −3.5 (MRN77) | −2.5 | −4.0 |
|---|---|---|---|---|
| Paβ/Hα | <3.5 × 10−2 (Main southwest) | 1.1 × 10−2 | 1.8 × 10−2 | 8.7 × 10−3 |
| Power-law index | Observed 3 σ upper limit | −3.5 (MRN77) | −2.5 | −4.0 |
|---|---|---|---|---|
| Paβ/Hα | <3.5 × 10−2 (Main southwest) | 1.1 × 10−2 | 1.8 × 10−2 | 8.7 × 10−3 |
Paβ/Hα ratios expected from dust scattering with different power-law indices of the dust size distribution.
| Power-law index | Observed 3 σ upper limit | −3.5 (MRN77) | −2.5 | −4.0 |
|---|---|---|---|---|
| Paβ/Hα | <3.5 × 10−2 (Main southwest) | 1.1 × 10−2 | 1.8 × 10−2 | 8.7 × 10−3 |
| Power-law index | Observed 3 σ upper limit | −3.5 (MRN77) | −2.5 | −4.0 |
|---|---|---|---|---|
| Paβ/Hα | <3.5 × 10−2 (Main southwest) | 1.1 × 10−2 | 1.8 × 10−2 | 8.7 × 10−3 |
5 Conclusion
We have conducted near-IR line mapping of Cen A with IRSF, using the narrow-band filter tuned for the Paβ line, to study the interaction between the AGN activity and the surrounding IGM in the northeast and the southwest regions of the halo of Cen A. In previous studies, Hα emission was detected in these regions (Santoro et al. 2016; Keel et al. 2019), suggesting that AGN-triggered star formation may have taken place therein. Our observations, however, do not detect Paβ emission, and in the southwest region the upper limit of the Paβ/Hα ratio is significantly lower than that expected for a typical H ii region. One possibility to explain the observed ratio is scattering of Hα and Paβ photons from the center of the host galaxy by dust in the galactic halo. If this is the case, the Hα fluxes detected in these regions are not relevant to AGN-induced star formation, but just reflect the presence of dusty clouds where the Hα photons originating from the galactic center of Cen A are scattered by the dust grains therein. According to the scattering scenario, the observed 3 σ upper limit of the Paβ/Hα ratio is consistent with the size distribution of the interstellar dust in our Galaxy. Finally, our result suggests the importance of observing Paβ emission in addition to Hα to identify the origin of Hα emission in the halo of a galaxy.
Acknowledgements
We are grateful to the referee for giving us useful comments. The IRSF project is a collaboration between Nagoya University and the SAAO supported by Grants-in-Aid for Scientific Research on Priority Areas (A) (nos. 10147207 and 10147214) and the Optical & Near-Infrared Astronomy Inter-University Cooperation Program, from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan and the National Research Foundation (NRF) of South Africa. K.M. is financially supported by MEXT/JSPS KAKENHI grant number 17K18019.
Footnotes
References
